SEEA EEA Tech Rec Consultation Draft 8.1 Dec2015 Final

SEEA Experimental Ecosystem Accounting:
Technical Recommendations
Consultation Draft – December 2015
Prepared as part of the joint UNEP / UNSD / CBD project on Advancing

Natural Capital Accounting funded by NORAD
Preface / Foreword
Key points
 Description of project and broad motivation (in addition to SDGs)
 Status of SEEA EEA TR in relation to SEEA EEA, to Global Strategy
 Process for developing SEEA EEA TR incl expert forum
 Links to other materials for testing of SEEA EEA
 Next steps at international level
 Connection to national work – implementation frameworks/plans, diagnostic tools,
SEEA Implementation Guide,
ii
Acknowledgements
To be drafted by UNSD
iii
Contents
Preface / Foreword ii
Acknowledgements iii
Contents iv
List of abbreviations and acronyms viii
1. Introduction 1
1.1 Definition and role of ecosystem accounting 1
1.2 Scope and purpose of SEEA EEA Technical Recommendations 2
1.2.1 Connection to the SEEA EEA 2
1.2.2 Connection to other materials 3
1.2.3 Audience for the Technical Recommendations 4
1.2.4 The scope of the Technical Recommendations 4
1.3 Links between the Technical Recommendations and other initiatives 5
1.3.1 International and national initiatives 5
1.3.2 Corporate initiatives 6
1.4 Structure of the Technical Recommendations 7
2. Main aspects of ecosystem accounting 8

2.1 Introduction 8
2.2 The SEEA EEA model and key accounting principles 8
2.2.1 The ecosystem accounting model 8
2.2.2 Assets and services 10
2.3 The steps in compiling ecosystem accounts 10
2.3.1 Introduction 10
2.3.2 Summary of compilation steps 12
2.3.3 Key considerations in compiling ecosystem accounts 14
2.3.4 What constitutes ecosystem accounting? 16
3. Spatial infrastructure and definition of units for ecosystem accounting 17

3.1 Introduction 17
3.2 The framework for delineating spatial areas for ecosystem accounting 18
3.2.1 Ecosystem units (EU) 18
3.2.2 Basic spatial units (BSU) 20
3.2.3 Geographical aggregations 22
3.3 Data sources, classifications and methods for delineating spatial units 24
3.4 Key issues and challenges in delineating spatial units for ecosystem accounting 25
3.5 Recommendations 26
iv
3.5.1 Recommended activities and approaches for testing 26
3.5.2 Issues requiring ongoing research 28
4. The ecosystem accounts 29

4.1 Introduction 29
4.2 Ecosystem extent accounts 32
4.3 Ecosystem condition accounts 34
4.4 Ecosystem services supply and use account 35
4.4.2 The ecosystem services supply table 38
4.4.3 The ecosystem services use table 39
4.5 Ecosystem monetary asset account 40
4.6 Integrating ecosystem accounting information with standard national accounts 44
5. Accounting for flows of ecosystem services 47

5.1 Introduction 47
5.2 The definition of ecosystem services 48
5.3 The classification of ecosystem services 52
5.4 The role and use of biophysical modelling 53
5.4.2 Overview of biophysical modelling approaches 54
5.5 Data sources, materials and methods for measuring ecosystem service flows 55
5.5.2 Data sources 55
5.5.3 Measuring the supply of ecosystem services 56
5.5.4 Recording the beneficiaries of ecosystem services 58
6. Accounting for ecosystem assets 61

6.1 Introduction 61
6.2 Dimensions in the measurement of ecosystem assets 62
6.3 Compiling measures of ecosystem condition 64
6.3.2 Different approaches to the measurement of ecosystem condition 64
6.3.3 Developing indicators of individual ecosystem characteristics 66
6.3.4 Aggregate measures of condition 68
6.3.5 Determining a reference condition 70
6.4 Developing the concept of ecosystem capacity 71
6.4.1 Defining ecosystem capacity 71
v
6.4.2 Linking ecosystem capacity and ecosystem degradation 73
6.5.1 Recommended steps for testing and experimentation on ecosystem assets 74
6.5.2 Recommendations in relation to ecosystem capacity 76
7. Thematic accounts 78
7.1 Introduction 78
7.2 Accounting for land 79
7.2.2 Relevant data and source materials 80
7.2.3 Key issues and challenges in measurement 81
7.2.4 Recommended activities and research issues 81
7.3 Accounting for water related stocks and flows 82
7.4 Accounting for carbon related stocks and flows 84
7.5 Accounting for biodiversity 85
7.5.2 Why account for biodiversity? 85
7.5.3 Assessing ecosystem and species diversity 86
7.5.4 Suitability of assessment approaches for biodiversity accounting 86
7.5.5 Implementing biodiversity accounting 88
7.5.6 Limitations and issues to resolve 89
7.5.7 Recommendations for testing, refining and validating 89
7.6 Other thematic accounts and data on drivers of change in ecosystem condition 90
8. Valuation in ecosystem accounting 92

8.1 Introduction 92
8.2 Valuation principles for ecosystem accounting 93
8.3 Relevant data and source materials 94
8.4 Key issues in valuation 97
vi
9. Integrating ecosystem accounting with standard economic data 101
9.1 Introduction 101
9.2 Steps required for full integration with the national accounts 102
9.3 The role of combined presentations 103
9.4 Integrated ecosystem-economic accounting structures 104
9.4.1 Extended supply and use tables 104
9.4.2 Integrated ecosystem accounts: full sequence of institutional sector accounts
and balance sheets 107
9.5 Key issues in integration 107
9.5.1 Measurement of net present value 107
9.5.2 Allocation of ecosystem degradation to economic units 108
9.5.3 Valuation of ecosystem degradation 109
9.5.4 Integrating balance sheet valuations 110
9.5.5 Application of integrated approaches for individual ecosystem assets 111
9.6 Alternative approaches to integration 111
Annexes 114
A1: Clarifications of SEEA EEA 114
A2: Summary of selected natural capital accounting initiatives 117
A3: Listing of project research papers 120
A4: UK Guidance for ecosystem accounts scoping studies 121
A5: Key features of a national accounting approach to ecosystem measurement 122
References 130
vii
List of abbreviations and acronyms
BSU basic spatial unit

CBD Convention on Biological Diversity
CICES Common International Classification of Ecosystem Services
CPC Central Production Classification
CSIRO Commonwealth Science and Industrial Research Organisation (Australia)
DEM digital elevation model
EC European Commission
EE-IOT environmentally-extended input-output tables
EEZ exclusive economic zone
ENCA QSP Ecosystem Natural Capital Accounts: Quick Start Package
ESVD Ecosystem Services Valuation Database
EU ecosystem unit
Eurostat Statistical Office of the European Union
FAO Food and Agriculture Organization of the United Nations
FEGS final ecosystem goods and services
FRA FAO Forest Resource Assessment
GDP gross domestic product
GHG greenhouse gas
GIS geographic information system
GRI Global Reporting Initiative
IIRC International Integrated Reporting Council
IPBES Intergovernmental Platform on Biodiversity and Ecosystem Services
IPCC Intergovernmental Panel on Climate Change
IRWS International Recommendations on Water Statistics
ISO International Organization for Standardization
IUCN International Union for the Conservation of Nature
2
km square kilometre
LCCS Land Cover Classification Scheme
LPI Living Planet Index
LUCAS Land Use and Cover Area Survey
LULUCF land use, land use change and forestry
2
m square metre
3
m cubic metre
MA Millennium Ecosystem Assessment
MAES Mapping and Assessment of Ecosystems and their Services
viii
MEGS Measuring Ecosystem Goods and Services
NBSAP National Biodiversity Strategic Action Plan
NCP National Capital Protocol
NEP net ecosystem productivity
NESCS National Ecosystem Services Classification System
NNI Norwegian Nature Index
NPP net primary productivity
NPV net present value
NSDI national spatial data infrastructure
OECD Organisation for Economic Cooperation and Development
PIM perpetual inventory model
PSUT physical supply and use table
SANBI South African National Biodiversity Institute
SCBD Secretariat for the Convention on Biological Diversity
SDG Sustainable Development Goals
SEEA System of Environmental-Economic Accounting
SEEA EEA System of Environmental-Economic Accounting Experimental Ecosystem
Accounting
SIDS small island developing states
SNA System of National Accounts
SUT supply and use tables
SWAT Soil and Water Assessment Tool
TEEB The Economics of Ecosystems and Biodiversity
UK United Kingdom
UK NEA UK National Ecosystem Assessment
UN United Nations
UNECE United Nations Economic Commission for Europe
UNEP United Nations Environment Programme
UNSC United Nations Statistics Commission
UNU-IHDP University of the United Nations / International Human Dimensions on
Poverty Programme
US United States
US EPA United States Environment Protection Agency
USLE Universal Soil Loss Equation
WAVES Wealth Accounting and the Valuation of Ecosystem Services
WCMC UNEP World Conservation Monitoring Centre
WWF World Wildlife Fund
ix
1. Introduction
1.1 Definition and role of ecosystem accounting

1.1 Ecosystem accounting is a coherent and integrated approach to the
measurement of ecosystems and the flows of services from them into economic and
other human activity. Ecosystem accounting complements, and builds on, the
accounting for environmental assets as described in the System of Environmental-
Economic Accounting (SEEA) Central Framework. In the SEEA Central Framework,
environmental assets are accounted for as individual resources such as timber
resources, soil resources and water resources. In ecosystem accounting, the
accounting approach recognises that these individual resources function in
combination within a broader system.
1.2 A prime motivation for ecosystem accounting is that separate analysis of
ecosystems and the economy does not adequately reflect the fundamental relationship
between humans and the environment. In this context, the SEEA Experimental
Ecosystem Accounting (SEEA EEA) provides a platform for the integration of
relevant information on ecosystem extent, ecosystem condition, ecosystem services
and ecosystem capacity, with information on economic and other human activity and
the associated beneficiaries (households, businesses and governments).
1.3 The integration of ecosystem and economic information is intended to
mainstream information on ecosystems within public and private decision making
and hence there must be a strong relevance of the information set to current issues of
concern. The general nature of the ecosystem accounting described here is for
application at a national level. That is, linking information on multiple ecosystem
types and multiple ecosystem services with macro level economic and planning
decisions. At the same time, it will prove highly relevant to apply ecosystem
accounting approaches at the regional or lower levels. This may demonstrate more
easily the potential for ecosystem accounting to aid in the development of responses
to specific policy themes or issues. A reduced scale focus may be of particular
interest in the development of pilot studies on ecosystem accounting.
1.4 The accounting approach outlined in SEEA EEA extends and complements a
range of other ecosystem and biodiversity measurement initiatives in a number of
important ways.
 First, the SEEA EEA framework includes accounting for the changes in
ecosystem condition (including changes in biodiversity) and the flows of
ecosystem services. Often, measurement of these two aspects of ecosystems is
undertaken in separate fields of research.
 Second, the SEEA EEA framework encompasses measurement in both
biophysical terms (e.g. hectares, tonnes) and in monetary terms where flows of
ecosystem services are ascribed monetary valuations through various market and
non-market valuation techniques. The valuation of ecosystem services also
supports the valuation of ecosystem assets.
 Third, the SEEA EEA framework is designed to facilitate comparison and
integration with the economic data prepared following the System of National
Accounts (SNA). This leads to the adoption of certain measurement boundaries
and valuation concepts that are not systematically applied in ecosystem
measurement. Their use facilitates the mainstreaming of ecosystem information
with standard measures of income, production and wealth that is required for
analysis of, for example, sustainability and green economy issues.
1
 Fourth, the aim of the SEEA EEA framework is to provide a broad, cross-cutting
perspective on ecosystems at a country or large, sub-national level. In principle,
while many of the concepts can be applied at a detailed level, the intent is to
provide a broad picture to enable integration with the broad picture of the
economy from the national accounts. Since many ecosystem measurements are
conducted at a detailed, local level, there is an important methodological
challenge to utilize these data to provide a national view. It is relevant then to
balance the scope of the accounting exercise (sub-national, national, global) with
the appropriate scale of analysis and information.
1.5 Building an integrated set of information concerning ecosystem extent and
condition, ecosystem services and economic activity can provide a platform for
discussion and integration between the various perspectives, disciplines and related
initiatives that are involved in this area of work.
1.6 By its nature, ecosystem accounting is an inter-disciplinary undertaking with
each discipline (statistics, economics, national accounts, ecology, geography, et al.)
bringing its own perspective and language. However, in order to obtain the benefits
from an integrated approach, institutional co-ordination is required to support the
compilation and use of accounting solutions.
1.7 The SEEA EEA has emerged from work initiated by the international
community of official statisticians and their development of the SEEA Central
Framework. While there has long been recognition of ecosystems in the context of
environmental-economic accounting, and recognition of the need to account for the
degradation of ecosystems, the approach described in the SEEA EEA has only
emerged in recent years. Its design is attributable to the relatively recent development
of concepts of ecosystem services. With these concepts, it has been possible to
develop accounting for ecosystems using, as a starting point, the accounting
approaches developed for recording economic activity and individual environmental
stocks and flows (e.g. water, energy, timber resources, greenhouse gas (GHG)
emissions).
1.8 Given its recent development, ecosystem accounting is considered to be an
emerging area of work. While it shows considerable potential as an integrating
framework, further discussion and testing of concepts and methods is required. These
SEEA EEA Technical Recommendations are intended to support ongoing discussion
and testing.
1.2 Scope and purpose of the SEEA EEA Technical Recommendations

1.2.1 Connection to the SEEA EEA
1.9 The SEEA Experimental Ecosystem Accounting: Technical
Recommendations (Technical Recommendations) provides a range of content to
support testing and research on ecosystem accounting. It complements the SEEA
EEA endorsed by the United Nations Statistical Commission (UNSC) in 2013. The
SEEA EEA describes a framework for ecosystem accounting and provided an initial
platform for discussion and collaboration on ecosystem measurement issues.
1.10 Since the SEEA EEA’s drafting in 2012, there has been further discussion
and testing of concepts and engagement with a broader range of interested experts.
The core conceptual framework remains solid but some additional issues,
interpretations and approaches have arisen. Thus, advances in thinking on specific
topics, for example on ecosystem capacity, have been introduced to ensure that the
content is as up-to-date as possible in this rapidly developing field.
2
1.11 These Technical Recommendations should not be considered to reflect the
final word on ecosystem accounting since further testing and discussion in this
emerging field is required. Thus, they provide additional background, context and
clarification to the concepts outlined in SEEA EEA with the intent of increasing
understanding of the ecosystem accounting approach. A summary of the main
differences and clarifications between the SEEA EEA and these Technical
Recommendations is provided in Annex 1.
1.2.2 Connection to other materials

1.12 The Technical Recommendations also incorporate findings reflected in a
range of other materials on ecosystem accounting that have developed over the past
few years. Examples include Ecosystem Natural Capital Accounts: A Quick Start
Package (ENCA QSP) (Weber, 2014); Guidance Manual on Valuation and
Accounting of Ecosystem Services for Small Island Developing States (UNEP,
2015); Designing Pilots for Ecosystem Accounting (World Bank, 2014); and
Mapping and Assessment of Ecosystems and their Services (MAES) (2nd report)
(Maes, et al., 2014). These materials have been developed by different agencies and
in different contexts but have an important role to play in the testing of SEEA EEA
by providing technical options and communicating the potential of a national
accounting approach to ecosystem measurement. A short overview of these different
documents is provided in Annex 2 and other initiatives are described in section 1.3.
1.13 As described in SEEA EEA, there are often strong connections between
accounting for ecosystem condition and ecosystem services, and accounting for
individual ecosystem components such as water and land. Consequently, work on
ecosystem accounting should take advantage of the range of materials that have been
developed relating to the measurement of water resources (including SEEA Water),
forests and timber, fisheries, and land. While these materials have not generally been
developed with ecosystem accounting in mind, they will support the development of
relevant estimates and accounts, especially in terms of describing methods and data
sources. Also, these documents describe potential applications of accounting that can
provide a useful focus for compilers.
1.14 As part of the broader ecosystem accounting project (in which developing
these Technical Recommendations is one output), there have been a range of
materials and outputs that have been developed that support the testing and research
on ecosystem accounting. National testing plans have been described for seven
countries and a range of entry level training materials have been developed. Also,
research papers on important measurement topics have been prepared by ecosystem
accounting experts. A full listing of these project research papers is included in
Annex 3.
1.15 Throughout the Technical Recommendations, references to these documents
and other relevant material are included as appropriate. Consequently, the Technical
Recommendations should reflect somewhat of a reference guide in addition to being
an up-to-date description of the state of ecosystem accounting.
3
1.2.3 The audience for the Technical Recommendations
1.16 The primary audience of the Technical Recommendations are those people
working on the compilation and testing of ecosystem accounting and those providing
data to those exercises, perhaps as part of separately established ecosystem and
biodiversity monitoring and assessment programs. Since ecosystem accounting is a
multi-disciplinary exercise, and requires the integration of data from multiple sources,
testing will require the development of arrangements involving a range of agencies
including national statistical offices, environmental agencies and scientific institutes.
The Technical Recommendations are intended to be applicable to all of these groups
although it is accepted that different people will have different levels of
understanding about different parts of the ecosystem accounting model.
1.17 The content should also assist those who will use the information that
emerges from sets of ecosystem accounts in terms of understanding the broad
ecosystem accounting model, the relevant definitions and terms, and the types of
approaches to measurement. However, potential applications of ecosystem accounts
and possible tools for analysis using ecosystem accounting are not the focus of this
document.
1.2.4 The scope of the Technical Recommendations

1.18 All aspects of ecosystem accounting as described in SEEA EEA are within
scope of the Technical Recommendations. However, more emphasis has been placed
on measurement in biophysical terms than on issues concerning monetary valuation
and integration into the standard national accounts. This balance reflects that work
over the past few years in the context of SEEA EEA has tended to focus on
biophysical measurement of land and ecosystem condition. The balance also reflects
a pragmatic view that the valuation of ecosystem services and ecosystem assets
requires a strong understanding of the relevant stocks and flows in biophysical terms.
Consequently, considering the accounting issues in biophysical terms can be
considered a necessary first step.
1.19 It is recognized that there is a substantial field of expertise and experience on
the valuation of ecosystem services. It is less clear that there have been significant
advances in linking this knowledge to the challenge of valuation for ecosystem
accounting purposes – a challenge raised substantively in the SEEA EEA Chapter 5.
While some developments will be reported on in this document, this area requires
further work both in testing valuation approaches in an accounting context and in
discussion among relevant experts (mainly in accounting and economics) to broaden
the understanding of the valuation of ecosystem services for accounting purposes.
1.20 Since the field of ecosystem accounting is quite new and is likely to advance
quickly given the range of testing underway, the Technical Recommendations cannot
be considered a final document but rather a summary at a point in time. However, it is
intended that a process will be undertaken to update the SEEA EEA in the coming 3-
5 years. This would take advantage of all relevant conceptual and practical
development, and aim to put in place the first international statistical standard for
ecosystem accounting. The active participation of the research and academic
communities involved in ecosystem related measurement and analysis would be
welcomed in this process. Also, on an ongoing basis it is proposed that relevant
guidance be updated and shared.
4
1.3 Links between the Technical Recommendations and other ecosystem measurement
related initiatives
1.3.1 International and national initiatives
1.21 As noted in Section 1.2, the content of the Technical Recommendations is
based on the conceptual ecosystem accounting model described in SEEA EEA. In
turn, the conceptual model complements the accounting for environmental assets in
the SEEA Central Framework and the accounting structures themselves are
applications of the principles and structures described in the SNA. Thus, the
Technical Recommendations are based on national accounting conventions and
accounting approaches to the organization of information.
1.22 At the same time, the ongoing testing and development of ecosystem
accounting as reflected in the Technical Recommendations continues to demonstrate
that this area of accounting is not a straightforward application of national accounting
principles. The primary reason for this is that ecosystems are not standard assets in
the way generally conceived in traditional economic accounting. Instead, ecosystems
are characterized by having multiple owners, generating multiple services and having
the potential to regenerate themselves in the future.
1.23 The second reason is that the information set required for the compilation of
a full set of ecosystem accounts is very diverse and not well coordinated at national
level. On the whole, economic statistics are quite well coordinated by a small number
of leading institutions (e.g. national statistics offices, central banks, taxation offices).
The lack of organisation of the underlying information needed for ecosystem
accounting has meant that ecosystem accounting is just one among a number of
environmental information initiatives. These include work on the Framework for the
Development of Environment Statistics (FDES), work on National Biodiversity
Strategic Action Plans (NBSAPs) and work in the context of measuring sustainable
development (e.g. indicators for the UN Sustainable Development Goals (SDGs)).
For the Technical Recommendations and those compiling ecosystem accounts, it
suggests that initial focus should be placed on gathering and adapting existing data
before moving to the collection of additional data as data gaps and analytical
priorities are established.
1.24 Finally, since ecosystem accounting is a relatively new field, it is natural that
different approaches and perspectives are developing. There is thus a range of
documents describing approaches that are essentially ecosystem accounting, even if
some of the content is not fully aligned with the conceptual model described in the
SEEA EEA. Since these documents provide useful background methods and data
sources for SEEA based ecosystem accounting, Annex 2 provides a short summary of
these documents.
1.25 In addition to these documents, there is an increasing body of work that is
testing the conceptual model for ecosystem accounting as described in the SEEA
EEA. Projects are taking place at national level and sub-national level, and being
undertaken as part of international initiatives (for example, the UNEP study The
Economics of Ecosystems and Biodiversity (TEEB) (TEEB, 2010), by national and
provincial governments, by non-government organisations (for example, work led by
Conservation International in Peru) and by academia (for example, the work at
Wageningen University, including Sumarga and Hein, 2013; Remme et al., 2014,
Schroter et al., 2014).
1.26 Also, there are an increasing number of examples of projects and initiatives
focused on the measurement of particular aspects of ecosystems. Various studies are
underway on accounting for biodiversity, land, water and carbon. Sometimes this
work is completed in awareness of the SEEA EEA framework, sometimes not. It is
5
likely that these various studies are relevant in a SEEA EEA context, particularly
with regard to connecting the measurement outcomes of ecosystem accounting work
with policy discussion that is often framed around such themes as biodiversity,
carbon, land, soil and water.
1.27 In combination, the initiatives noted in Annex 2 provide an information base
to support ecosystem accounting. Inevitably, when drawing information from
multiple sources, the results will need to be tailored and adjusted to suit the
requirements of accounting and to ensure integration.
1.3.2 Corporate initiatives

1.28 In parallel with the work on developing environmental-economic accounting
as a complement to the SNA, there is an equally long history of work on the
integration of environmental information into corporate accounting. By and large,
these two streams of accounting have not interacted in a significant way. While there
are differences between national and corporate accounting, there appear more
similarities than differences and a joining of efforts in this space would be a positive
step forward.
1.29 To this point however, the integrated ecosystem accounting approach
described in SEEA EEA has not been applied in corporate accounting. Efforts at
environmental or natural capital accounting have either focused on (i) integrating the
costs of residual flows (emissions, pollutants, etc.) into current accounting structures
or (ii) focused on reporting environmental and natural capital issues as a complement
to the standard suite of accounts. This work may be placed within the broader
contexts of corporate sustainability reporting and corporate social responsibility.
1.30 This second approach has developed considerable momentum via the Global
Reporting Initiative (GRI) and the International Integrated Reporting Council (IIRC)
but neither of these approaches yet incorporates an integrated approach to accounting
similar to ecosystem accounting.
1.31 Work on integration into standard accounting structures is being developed
within the Natural Capital Protocol (NCP) project led by the Natural Capital
Coalition, and the work on the Natural Capital Declaration being co-ordinated by the
UNEP-Finance Initiative. However, at this stage, whether ecosystem accounting type
measurement approaches will be incorporated is unclear. Also, research funded by
the Natural Capital Committee of the United Kingdom (UK) has proposed a corporate
natural capital accounting model whereby the value of ecosystems is incorporated on
a company’s balance sheet using the net present value of ecosystem services – thus
following the broad logic of SEEA EEA.
1.32 In relation to the testing and development of ecosystem accounting at a
national level, there are a number of reasons for establishing a relationship between
these two branches of accounting. First, in many cases understanding the
environment-economic relationship requires assessment of public goods.
Consequently, the development of corporate accounting requires information beyond
the boundaries of their own operations. Second, there may be a good opportunity for
the public sector to improve their collection of data on the environment through
appropriate coordination with the business community. Third, the business
community relies on public data, such as the national accounts, to understand its
wider operating environment both nationally and globally. Widely developed
ecosystem accounts should be able to offer similar advantages in terms of
standardised approaches to assessing operational risks and opportunities. Fourth, joint
6
development and exchange should help to more quickly advance the research agenda
especially via a common understanding of terms and concepts.
1.4 Structure of Technical Recommendations

1.33 The Technical Recommendations are structured in the following way.
 Chapter 2 covers the general principles of ecosystem accounting with a summary
of the ecosystem accounting model described in SEEA EEA and the paths that
might be chosen to compile ecosystem accounts.
 Chapter 3 summarises the various accounting units and their classification.
 Chapter 4 describes the main types of ecosystem accounts.
 Chapter 5 introduces accounting for flows of ecosystem services with a
description of some of the key boundary and classification related issues and the
relationships to other concepts such as benefits and wellbeing.
 Chapter 6 considers the issue of accounting for ecosystem assets in a holistic way
which, in particular, involves dealing with the aggregation of information and the
measurement of condition, capacity and degradation.
 Chapter 7 provides an introduction to accounting for various thematic accounts
related to ecosystems namely land, carbon, water and biodiversity.
 Chapter 8 summarises monetary valuation from an ecosystem accounting
perspective.
 Chapter 9 updates the discussion in the SEEA EEA Chapter 6 on the integration
of ecosystem and economic information via the accounting framework.
7
2. Main aspects of ecosystem accounting
Key points in this chapter

The core ecosystem accounting model from the SEEA EEA provides a robust framework for
placing, in context, information on ecosystem assets, ecosystem services, the benefits
generated from ecosystem services and well-being.
It is important to understand the analytical and policy questions of most relevance such that
compilation can be focused in the appropriate areas.
There are five broad compilation steps within ecosystem accounting. Each step provides
useful information for analytical and policy purposes.
There is no single path that must be followed in the compilation of ecosystem accounts.
However, as a general observation, the initial focus is on measurement in physical terms and
then on valuation in monetary terms.
Physical measures of ecosystem extent and condition, and measures of the supply of
ecosystem services may be compiled in parallel since there will be a close relationship
between the selection of indicators to measure ecosystem condition and the use of the
ecosystem as reflected in the basket of ecosystem services.
The accounting structures presented in the Technical Recommendations can be adapted to
support varying levels of detail – e.g. by providing information at different scales from
municipal to national.
To gain the most benefit from the ecosystem accounting approach, it is important to
continuously assess coherence between different ecosystem accounts. A number of iterations
through the accounting system are likely to be appropriate.
Further, the true value of ecosystem accounting will emerge when accounts are compiled on
an ongoing basis such that a time series of coherent information can be analysed and
relationships and trends established.
2.1 Introduction
2.1 This chapter provides an overview of ecosystem accounting; relevant details
are provided in the following chapters, and in associated training materials and
research papers. The chapter complements the text in SEEA EEA Chapter 2 by
providing additional descriptions of key elements of the ecosystem accounting model.
In doing so, the section also provides some additional material to reflect the ongoing
developments in ecosystem accounting.
2.2 The SEEA EEA model and key accounting principles

2.2.1 The ecosystem accounting model
2.2 The SEEA EEA ecosystem accounting model has five main components that
are reflected in Figure 2.1. Starting at the bottom of Figure 2.1, the model is based
around accounting for an ecosystem asset (1) that is represented by a spatial area.
The delineation of the area that defines an ecosystem asset is required for accounting
purposes and should be considered a statistical representation of ecosystems, which
by their nature are not discrete systems.
8
2.3 Conceptually, ecosystem assets are represented by ecosystem units (EU). EU
are contiguous areas of different types distinguished according to different
characteristics including vegetation, climate, soil type, hydrology and use. To support
the delineation of EU, the measurement of ecosystem services and the integration of
multiple data sets, basic spatial units (BSU) should be delineated, often through the
formation of a standard or reference grid. Generally, accounting will be undertaken at
more aggregated level, for example at the level of a state or country, or in terms of
broad ecological areas such as bioregions or river basins. These larger areas
comprising multiple EU are known as geographical aggregations.
2.4 Each ecosystem asset has a range of relevant ecosystem characteristics and
processes (2) that together describe the functioning of the ecosystem. The accounting
model proposes that the stock and changes in stock of ecosystem assets is measured
by considering the ecosystem asset’s extent and condition which can be done using
indicators of the relevant ecosystem asset’s area, characteristics and processes.
Figure 2.1 Ecosystem accounting model
Source: SEEA EEA Figure 2.2, UN et al 2014b
2.5 Each ecosystem asset generates a set or basket of ecosystem services (3)
which, in turn, contribute to the production of benefits (4). Benefits may be goods or
services (products) currently included in the economic production boundary of the
SNA, referred to as SNA benefits; or they may be benefits received by individuals
that are not produced by economic units (e.g. clean air). These are referred to as non-
SNA benefits. Both SNA and non-SNA benefits contribute to individual and
societal well-being (5).
2.6 The chain of relationships from ecosystem assets to well-being is at the core
of the SEEA EEA. While there remain some important issues of definition in terms of
the boundaries between different components, and there are measurement challenges;
9
the core model reflecting the relationships between ecosystem assets, ecosystem
services, benefits and individual and societal well-being remains strong.
2.2.2 Assets and services

2.7 At the centre of the ecosystem accounting model is the distinction between
ecosystem assets and ecosystem services. The former are the stocks within the
accounting system and the latter are flows. The distinction is an application of the
separation in standard accounting between capital and income.
2.8 By accounting for both of these components and presenting them in a single
integrated model, two key advantages accrue:
 First, a significant volume of data can be placed in context and integrated
in both bio-physical and monetary terms
 Second, issues of sustainability can be considered since the capacity of an
ecosystem asset to deliver services is considered separately from the
flows of ecosystem services themselves.
2.9 There are a number of approaches to ecosystem measurement that focus on
either the assessment of ecosystem assets or on the flows of ecosystem services.
Those that focus on ecosystem assets tend to work in bio-physical terms. Those that
focus on ecosystem services tend to also incorporate monetary valuations. While this
information is undoubtedly of value and relevance, the issue of why ecosystem assets
are important for economic and human activity is commonly not considered
quantitatively.
2.10 On the other hand, approaches that focus on ecosystem services, particularly
those targeting monetary valuation of ecosystem services, can tend to infer or assume
a connection to the underlying ecosystem assets which generate the services. This is
consistent with standard accounting and economics where the value of an asset is
considered to be equal to its discounted future income stream. However, using this
assumption in ecosystem accounting ignores significant issues of the complex, non-
linear relationship between ecosystem assets and the services they generate.
2.11 The significance of the SEEA ecosystem accounting model, and the general
relevance of accounting approaches, thus lies in: (i) requiring measurement of both
assets and services; and (ii) recognizing the connection between these two key
components. Indeed, the development of policy relevant information sets requires an
overall understanding of an ecosystem’s condition, interactions and connections such
that suitable recommendations can be formed.
2.3 The steps in compiling ecosystem accounts

2.3.1 Introduction
2.12 In principle, ecosystem accounts will provide information that is relevant in a
range of policy and analytical contexts. However, in the initial development and
testing phase, it will likely be necessary to understand a more limited number of
specific purposes or questions for which ecosystem accounts might be compiled. The
type of policy question will help determine the scale of the accounts - either national
or sub-national (e.g. water catchment, province, habitat type, etc.) - and the type of
data needed. Over time, and building on the initial testing, the development of a more
complete set of national level ecosystem accounts can be envisaged through
progressive development, extension and integration. Further, the development of an
10
initial set of ecosystem accounts is likely to spark awareness of additional potential
applications.
2.13 Determining the appropriate scope and scale for a set of ecosystem accounts
must be a matter of discussion at the country or provincial level. It is anticipated that
the content of the Technical Recommendations will support that discussion, although
it is recognised that other factors will need to be taken into account. Following the
general principles of SEEA implementation (see SEEA Implementation Guide (UN,
2013)), the discussions should involve all relevant stakeholders, data users, account
compilers and source data holders. Note that the information in the Technical
Recommendations is appropriate for discussions on both the commencement of pilot
studies and the establishment of nation-wide programs of work.
2.14 Generally, the compilation of ecosystem accounts will start using physical
measures of stocks of ecosystem assets and flows of ecosystem services and
associated benefits. From there, compilation would proceed to valuation and
integration with standard economic accounts.
2.15 The conceptual model for ecosystem accounting shown in Figure 2.1
provides a general description of the relationships between the different stocks and
flows. However, it does not provide a sense of how a compilation of ecosystem
accounts might proceed. This section provides an overview of the steps involved in
compiling ecosystem accounts. Later chapters in these Technical Recommendations
provide more detail on the various types of accounts, related measurement issues and
recommendations.
2.16 The broad steps in ecosystem accounting are shown in Figure 2.2. The first
set of steps involves accounting in physical terms and the second set of steps is in
monetary terms. While it is useful to see this sequencing, the reality of accounting is
that there will be multiple iterations through the accounts and further, that the precise
starting point may vary. By way of example, Annex #4 shows the steps applied in the
UK for the development of ecosystem accounts for broad habitat types. Other projects
will likely follow a different sequencing. Whatever sequence is adopted, the reality of
accounting exercises is that iteration will be required in order to develop a coherent
integration of data across the various accounts.
11
Figure 2.2 Broad steps in ecosystem accounting
a. Steps in physical terms
b. Steps in monetary terms
2.3.2 Summary of compilation steps

2.17 Within the general considerations noted above, the following paragraphs
describe the main steps that must be incorporated within any pathway. The sequence
of steps used is reflected in the structure of the Technical Recommendations.
2.18 Step 1: For ecosystem accounting, the first important step is to delineate the
ecosystem assets that are to be the focus for the accounts. In principle, these areas
should cover the entirety of a country’s terrestrial area (including inland waters) and,
as appropriate, relevant coastal and marine areas – perhaps extending to a country’s
exclusive economic zone (EEZ). Chapter 3 discusses the issues of delineating and
classifying ecosystem assets for ecosystem accounting purposes.
2.19 Information on the total area of different types of ecosystem assets, often
measured in hectares, is presented in an ecosystem extent account. This account
presents an opening and closing area by type of ecosystem, together with information
on the additions and reductions in area. A key aim with this account is to measure the
change over time in the composition of ecosystem types within a country. The
ecosystem extent account is described in Chapters 4 and 6 with relevant information
also in the discussion of land accounts in Chapter 7.
2.20 Step 2: Using the breakdown of ecosystem types determined for the
ecosystem extent account, the next step is to compile the ecosystem condition
account. This account records information on the various characteristics that reflect
the condition or state of an ecosystem, and over time, will provide information on
trends in ecosystem degradation or enhancement. The set of relevant characteristics
will depend both on the type of ecosystem (i.e. indicators for forests will likely be
different compared to indicators for coastal ecosystems) and on the use of the
12
ecosystem since the way in which an ecosystem is used will usually have a direct
effect on the way in which its condition may change.
2.21 Chapters 4 and 6 discuss the compilation of ecosystem condition accounts in
more detail. Chapter 7 discusses the compilation of information on land, carbon,
water and biodiversity using accounting approaches since these data may be relevant
in monitoring the condition of many ecosystems.
2.22 Step 3: The next step involves the measurement of ecosystem services in
physical terms1. This is completed by considering each ecosystem asset in turn and
determining the relevant ecosystem services and appropriate indicators. This task
should be conducted using a classification of ecosystem services, such as the
Common International Classification of Ecosystem Services (CICES) (Haines-Young
and Potschin, 2013). A classification can provide a checklist to ensure appropriate
coverage.
2.23 This step involves the estimation of both the supply of ecosystem services
from each ecosystem asset and the use of those services by various beneficiaries.
Together, the information on supply and use are used to compile an ecosystem
services supply and use table. To support integration with the national economic
accounts the beneficiaries in ecosystem accounting are grouped in the same way as
for the economic accounts – i.e. by industry group and by institutional sector. The
possible approaches to measurement are discussed in Chapter 5.
2.24 Step 4: There are many examples of the valuation of ecosystem services and
it is a necessary step for certain types of integration with the standard national
accounts, such as adjusted gross domestic product (GDP) and extended measures of
net wealth.
2.25 There are two main targets for valuation in ecosystem accounting. The first
target is the valuation of ecosystem services by applying relevant prices to the
physical flows of ecosystem services measured in Step 3. This permits the
compilation of ecosystem service supply and use tables in monetary terms. In
some cases, particularly for provisioning services, it may be relevant to use broad
scale national level measurements of output as a starting point for estimating
ecosystem service values, whereas for other services, particularly regulating services,
ecosystem specific measurements will be required.
2.26 The second target is the valuation of ecosystem assets. This is done by
estimating the net present value (NPV) of the future flow of all ecosystem services
from each ecosystem asset. There are, of course, many challenges in this step
(discussed further in chapters 8 and 9). A particularly important one is estimating the
future flow of ecosystem services and hence the extent to which current ecosystem
services supply can be maintained. This requires an assessment of ecosystem capacity
which reflects the connection between ecosystem condition and ecosystem services.2
2.27 The value of ecosystem degradation will be related to the change in the net
present value of ecosystem assets. Opening and closing values for ecosystem assets
and changes in those values over an accounting period are presented in an ecosystem
monetary asset account.
1
In this context, “physical” means “non-monetary” and measurement in physical terms encompasses
ecosystem services that reflect flows of materials and energy, flows of services related to the regulation of
an ecosystem, and flows related to cultural services. (SEEA EEA para 3.2)
2
The measurement of ecosystem capacity may be facilitated by the description of baseline scenarios, i.e.,
describing expected changes in ecosystem condition given relevant business-as-usual assumptions.
13
2.28 Step 5: The final step involves the use of information on ecosystem services,
ecosystem assets and ecosystem degradation from the accounts described above, to
integrate environmental and economic data and augment the current, standard
national accounts. This may be done in a number ways including:
i. The compilation of combined presentations where data on ecosystem
condition and services in physical terms are presented alongside standard
economic data, such as value added, employment, and costs of
environmental restoration.
ii. The full extension of the ecosystem services supply and use table in
monetary terms to also include all products. This approach can be used to
show the integration of ecological and economic supply chains.
iii. The compilation of an extended sequence of accounts where standard
economic measures such as GDP, national income, and national saving
are adjusted for the cost of ecosystem degradation. Adjusted measures
may also be derived by institutional sector and industry.
iv. The estimation of a national balance sheet in which the value of
ecosystem assets is incorporated with the value of other assets and
liabilities to derive extended measures of national wealth.
2.29 There are challenges in all of these areas that are discussed at more length in
Chapter 9 on integrated ecosystem-economic accounts.
2.30 Alongside the compilation of ecosystem accounts, it will be relevant to
compile thematic accounts such as accounts for land, carbon, water resources and
species diversity. These accounts organise data on themes that are commonly of
interest from a policy perspective. At the same time, they will each have linkages to
some aspects of ecosystem measurement and data from thematic accounts may
directly support the compilation of ecosystem accounts.
2.31 Eventually the complete cross-reading of ecosystem and thematic accounts is
what should ensure consistency of the whole accounting system. The key aspects of
national accounting approaches to measurement are summarised in Annex 5. Annex 5
has been included to provide an overview of the key elements of the national
accounting approach that underpins the compilation of ecosystem accounts described
here. This material is highly relevant to those who have not practised national
accounting and should provide a context in which what constitutes ecosystem
accounting can be understood.
2.3.3 Key considerations in compiling ecosystem accounts

2.32 Six key considerations emerge in understanding the set of ecosystem
accounts as presented in the Technical Recommendations. First, it is a set of accounts
with each account containing specific pieces of information applicable to one part of
the ecosystem accounting model outlined in Chapter 2. There is not one single
“ecosystem account” and it would be inconsistent with accounting principles to force
information on all stocks and flows into a single account while attempting to also
retain notions of internal consistency and coverage.
2.33 Second, as far as possible, the accounts are designed to link together such
that information can be readily compared between accounts. Thus, while there is
more than one account, and each can stand alone in accounting terms, there are
14
relationships between the accounts that can be highlighted by structuring the
information appropriately.
2.34 Indeed, in some cases, the same information may be contained in two
accounts, as it is relevant in both contexts. For example, the accumulation of biomass
for accessible timber will be relevant in the measurement of the supply of ecosystem
services and also in understanding the change in the ecosystem asset, i.e. the forest.
This does not represent double counting but rather the appropriate placement of
information to reflect different accounting identities.
2.35 Third, a very specific design feature of the ecosystem accounts is that
ultimately the information should be able to be integrated with the standard national
accounts that record economic activity. This design feature does not impact on all
accounts but is particularly relevant for accounts concerning ecosystem services.
2.36 Fourth, the accounting structures presented should not be considered
unchangeable with regard to the level of detail they contain. For example, the
accounts concerning ecosystem assets tend to be structured to show high-level EU
types (e.g. forests) within a given geographical aggregation. In practice, it may be
most relevant to provide finer detail only for some specific EU types (e.g. by type of
forest). The accounting principle of working from the outside-in (see Annex 5)
implies that rearrangement of information inside a boundary is perfectly appropriate
and the level of detail should be determined based on analytical and policy
requirements and with regard to data availability.
2.37 Fifth, the accounts described in this chapter present information for one
accounting period, usually one year. The length of the accounting period determines
the points chosen to measure the opening and closing stocks. Flows are measured in
terms of observed changes between the opening and closing of the accounting period.
2.38 Most commonly however, the interest in accounting information stems from
the presentation of time series of information, i.e. for multiple accounting periods.
Presuming that time series of accounts are compiled, including, for example, accounts
for two accounting periods 5 years apart, users of accounting information are likely to
require a re-organisation of the information such that time is one of the dimensions
recorded. In practice, this is an issue of data management and dissemination rather
than of concept. Compilers should feel free to restructure the accounts described here
in such a way to best suit the presentation and analysis of data, in reference to the
associated policy questions.
2.39 Sixth, the structure of accounts will generally represent a level of detail
suitable for presentation and analysis of outputs from accounting. It represents the
level of detail at which accounting relationships (e.g. supply and use, balancing end
of period stocks and changes in stocks) are applied. However, it will generally be
necessary for underlying information to be compiled at different, usually lower,
levels of aggregation before entry into the accounts.
2.40 In the case of ecosystem accounting, it is likely to be ideal to compile data at
an appropriately detailed level, e.g. by BSU, and then aggregate to the relevant EU
level for accounting purposes. This does not require that accounts are developed at
the BSU level but rather that the input data and the output data are managed
separately. Making this distinction is essential if changes to the source of the input
data are to be managed effectively, without affecting the integrity of the time series of
data contained in the accounts themselves. Indeed, changes to input data should be
considered a normal and common situation.
2.41 With all this in mind the ecosystem accounts described in Chapter 4 should
be taken as a guide to the types of information that can be organized following an
accounting logic. Countries are encouraged to compile accounts using structures that
15
are most appropriate to understanding the relationship between ecosystems and the
economy in their country. However, to support ongoing dialogue and international
comparison, it is essential that these structures, classifications, concepts and resulting
indicators are coherent with the core presented in the Technical Recommendations. If
variations are used, these should be described and presented with the accounts.
2.42 The design of the accounts will benefit from further testing and discussion in
terms of both the relevant compilation approaches and the most appropriate levels for
analysis and communication of results.
2.3.4 What constitutes ecosystem accounting?

2.43 Given that there are a number of accounts within scope of the ecosystem
accounting system, a reasonable question is which accounts constitute ecosystem
accounting? Further, do all of the accounts need to be compiled? The response to
these questions has two main aspects. First, ecosystem accounting is as much an
approach to measurement as it is a set of accounts. Ecosystem accounting embodies
important underlying aspects of national accounting by establishing broad and
comprehensive boundaries and standardised relationships between different stocks
and flows. In this context, ecosystem accounting is an approach to measurement that
goes well beyond the measurement of individual ecosystems or the valuation of
individual ecosystem services. It provides a framework for discussing the relationship
between the economy and the environment.
2.44 Second, in the context of this comprehensive approach, it must be accepted
that all of the accounts described above cannot be completed at once. As compilation
takes place and understanding increases, the set of ecosystem accounts will become
more advanced and complete. However, at each point along the way, the accounts
that have been compiled will be relevant for particular policy purposes and analysis.
Thus, it is not necessary to complete the full set of accounts for the information to
become relevant.
2.45 Based on current experience, a reasonable first level of attainment in terms of
ecosystem accounting would be the compilation of accounts in physical terms for
ecosystem extent, ecosystem condition and ecosystem services supply and use. These
three accounts form the basis for the remaining accounts and, in their own right,
cover the key elements of the ecosystem accounting model in Figure 2.1.
2.46 In compiling the ecosystem condition account and the ecosystem services
supply and use table in physical terms, a challenge is the derivation of meaningful
aggregates that support a broad assessment across ecosystems. For ecosystem
services, aggregation is possible using monetary valuation techniques. It is in this
context that valuation and, ultimately integration with the standard economic
accounts, may have most relevance. It is often assumed that valuation in monetary
terms is an accounting requirement. However, the SEEA perspective is that all of the
accounts described embody national accounting principles and structures, and hence
work towards the meaningful mainstreaming of environmental information into
economic and other decision making.
16
3. Spatial infrastructure and definition of units for ecosystem
accounting

Ecosystem accounting requires the delineation of areas within a country into mutually
exclusive units that represent ecosystem assets.
Ecosystem units (EU) are the spatial areas that form the conceptual base for accounting and
the integration of relevant statistics.
Basic spatial units (BSU) support the delineation of EU and the integration of multiple
datasets. For ecosystem accounting, BSUs are assumed to be internally homogenous in terms
of their biophysical properties. BSUs may be delineated through the formation of a spatial
grid covering the extent of a country.
Accounts will generally be produced for relatively large administrative areas, such as a
provinces, states or countries; or in relation to a large ecological areas such as bioregions or
river basins. These larger areas that are used for accounting and reporting are referred to as
geographical aggregations.
The delineation of EU will require the use of multiple data sources. As a starting point for
delineation, land cover data, grouped into relatively broad land cover types may be
appropriate.
Testing of methods for the delineation of EU and the formation of BSUs is required. A focus
should be placed on determining the appropriate scale of information required for integration
of data and for accounting and policy use.
3.1 Introduction
3.1. The SEEA EEA applies the definition of ecosystems from the Convention on
Biological Diversity (CBD) – “ecosystems are a dynamic complex of plant, animal
and micro-organism communities and their non-living environment interacting as a
functional unit” (CBD, 2003, Article 2, Use of Terms). Ecosystems are not
straightforward to delineate spatially for measurement purposes. In ecological terms,
they may be defined at a range of spatial scales, and hence it is often difficult to
identify clear boundaries and they may overlap in spatial terms. For statistical and
accounting purposes, it is necessary to clearly differentiate ecosystem assets as
discrete units. Hence, the boundaries described here should be considered a statistical
abstraction from an ecological reality.
3.2. Ideally, in ecosystem accounting, there is consideration of all ecosystems
within a country. To support this scope, it is necessary to delineate all areas within a
country such that there are no gaps or overlaps – i.e. an approach that is mutually
exclusive and collectively exhaustive. A key objective in delineating areas is to be
able to classify a country into different types of ecosystems. Different types of
ecosystems will supply different ecosystem services. Hence, understanding the
multiple and changing structures of different types of ecosystems is an important
aspect of ecosystem accounting.
17
3.3. Different types of spatial areas need to be delineated to co-ordinate and
present the information required for ecosystem accounting. For this purpose, a
framework covering three types of spatial areas has been developed. The purpose of
the framework is two fold. First, it allows the organization of information into
separate entities that can then be compared and aggregated. This is akin to the role of
a units model in economic statistics where economic units (businesses, households
and governments) are distinguished by their types of economic activity and legal
structure. These economic units can then be grouped into relevant types, such as
industry classes and institutional sectors. Second, the spatial framework and the
associated classifications provide a basis for structuring and presenting data on
ecosystem extent, condition, and ecosystem services.
3.4. This chapter outlines the approach taken in SEEA EEA to delineate spatial
areas for the purpose of ecosystem accounting, building on the discussion in SEEA
EEA Section 2.3. The conceptual and practical aspects of delineating spatial areas are
described in section 3.2. In section 3.3 a summary of data sources, classifications and
methods is provided. In section 3.4 key measurement issues are described and section
3.5 provides recommendations for testing and areas for ongoing research.
3.2 The framework for delineating spatial areas for ecosystem accounting
3.5. The conceptual model for ecosystem accounting involves the integration of
three types of areas. First, areas representing ecosystem assets which lie at the heart
of the conceptual ecosystem accounting model. Second, small areas that can be used
to integrate the detailed geo-spatial data that is required for ecosystem accounting.
Third, larger areas that represent aggregations of smaller areas and will correspond to
the level of spatial detail likely to be reflected in ecosystem accounts. Each of these
three types of areas is described in this section.
3.2.1 Ecosystem Units (EU)

3.6. Conceptually, for accounting purposes, each area representing a different
type of ecosystem is considered to represent an ecosystem asset. Each of these
individual areas is considered an EU. Ecosystem assets are considered to be bounded
spatially with each asset comprising all of the relevant biotic and abiotic components
within those bounds that are required for it to function and to supply ecosystem
services. An EU can be considered to reflect the spatial boundary for accounting and
statistical purposes that supports measurement of the CBD’s definition of an
ecosystem quoted above.
3.7. The EU is the conceptual unit that underpins the ecosystem accounting
model. It is the EU that represents the ecosystem that supplies a mix of ecosystem
services and which can be considered to have a condition in terms of a combination
of structure, function and process. In principle, an EU will be differentiated from
neighbouring EUs by the extent to which the relationships between biotic and abiotic
components within an EU are stronger than the relationships with components
outside of the EU.
3.8. EUs are contiguous areas representing individual ecosystems. In practice, it
may be relevant to analyse accounting variables, such as ecosystem condition and
ecosystem service supply, at a more aggregated level reflecting information for EUs
of the same type. For example, ecosystem account users may be interested in
information on the ecosystem services supplied by all EUs of type “deciduous
forest”, rather than in services from individual patches of deciduous forests.
18
3.9. Ideally, EU will be delineated based on various characteristics including
vegetation structure and type, physiognomy, climate, hydrology, soil characteristics
and topography. These characteristics may be used alone or in combination. The
choice will be dependent on the country and the data available.
3.10. Additionally, it may be relevant to use information on land management as
part of the delineation of EUs. This may be particularly helpful in understanding the
flows of ecosystem services that are most likely from a particular area. For example,
it may be useful to distinguish between protected forests that are not logged and
other, ecologically similar forests in which logging in permitted. It is also noted that
maps that delineate land within a country according to different land management
regimes (for example protected areas and water catchments) may be readily available
and can be used to support the establishment of spatial areas for ecosystem
accounting.
3.11. In terms of delineation using ecological characteristics, the breadth of data
covering the characteristics noted above may not be available to specify EU
sufficiently well across all areas within a country. Therefore, where various
ecological data are not available, a land cover based delineation of EUs can be used
as a starting point. This raises the practical question of which land cover types should
be considered and at what level of detail.
3.12. For land cover based EUs, it is recommended that the most coarse level of
aggregation should be based on the interim land cover classification of the SEEA
Central Framework which has 15 types/classes as shown in Table 3.1.3 Note that a
class for sea and marine areas has been incorporated to ensure appropriate coverage
for all of a country’s area that may be included within ecosystem accounting. (For a
more detailed description see the SEEA Central Framework, Annex 1, Section C.)
3.13. Each of these 15 types may be used to represent a type of EU. Generally,
across a country, there will be a number of different areas of the same EU type. For
example, there will be different areas of forest in different parts of a country. Each
separate forest is considered a separate EU but is classified to the same EU type.
3.14. The use of high level land cover information as a starting point for ecosystem
accounting is likely to be amenable to measurement in most countries at the national
level. Data and analysis at this aggregated level of detail will also be useful to convey
broad trends in the changing structure of ecosystem types.
3.15. However, the use of a limited number of EU types will also limit the
sophistication of the questions that can be answered using the accounting
information. Ideally, and as appropriate within a country, more detailed
classifications should be integrated with the broad land cover classes such that a
nested or hierarchical set of EU types is developed. This could be done progressively
as the understanding of ecosystem accounting develops and as the availability of
information increases.
3.16. The availability of more detailed EU type information would also support a
more straightforward link to ecosystem services supply. For example, ecological
detail concerning the type of perennial crop, or the hydrological properties of a forest,
3
The SEEA EEA proposed a set of 16 classes for land cover/ecosystem functional units. These
classes were developed as an application of the interim Land Cover classes presented in the
SEEA Central Framework by combining land cover information with information on land use.
Since there may be various ways in which land use and land cover information may be
combined, it is now considered that for the task of attributing land cover characteristics to the
BSU level and hence for delineating broad EUs, the starting point should be the land cover
classification of the SEEA Central Framework.
19
would facilitate identifying and analysing the ecosystem services supplied by those
EUs.
3.17. As the number of different EU types increases, the sizes of EUs will
generally decrease. For example, individual areas of forest will be larger than areas of
different types of forest, for example, classified according to species or altitude.
While additional detail may be useful for better understanding the changing
composition of types of ecosystems, this outcome may be problematic for deriving
other measures of ecosystems, for example ecosystem services, since information
may not be available in relation to detailed types of EU. Further, as the size of EU
becomes smaller, often the accuracy of the information attributed to it will also
reduce (depending on the data source). In practice, a balance must be found between
the number of EU types that are identified and the availability of information.
3.18. To support the improvement of accounts over time and the integration of
data, it is recommended that the classification used for EU types is nested. That is,
finer level classes providing more detail should sit within broader classes, perhaps of
land cover type, eco-region or biome. Further, use of a common set of land cover
types at the highest level in all ecosystem accounting work would support cross
country comparison and support exchange of methods.
Table 3.1 Land cover classes

Description of classes
Artificial areas (including urban and associated areas)
Herbaceous crops
Woody crops
Multiple or layered crops
Grassland
Tree-covered areas (forests)
Mangroves
Shrub-covered areas
Shrubs, and/or herbaceous vegetation, aquatic or regularly flooded
Sparsely natural vegetated areas
Terrestrial barren land
Permanent snow and glaciers
Inland water bodies
Coastal water bodies and intertidal areas
Sea and marine areas
Source: SEEA Central Framework Table 5.12 (UN et al., 2014a)
3.2.2 Basic spatial units (BSU)

3.19. While EU represent the underpinning conceptual unit for accounting and
statistical purposes, the delineation of EU in practice is unlikely to be completed
directly from available information. Further, much of the information that needs to be
attributed to EU or to types of EU, will not be available at the EU level directly.
Consequently, another type of spatial area is needed to support compilation. For
ecosystem accounting, this area is termed a basic spatial unit (BSU).
3.20. A BSU is a small spatial area that is a geometrical construct. The purpose of
delineating BSUs is to provide a fine level frame to which a range of different
information can be attributed. Once different data have been integrated to the same
spatial scale then many aggregation and integration possibilities emerge.
3.21. The most common approach to delineating BSUs is to form a grid of
appropriate coverage and cell-size (ideally <100m) that is overlayed on a large area
20
or country. This forms a reference grid. In this context, a BSU corresponds to a grid-
cell in geo-information disciplines and a grain in landscape ecology.
3.22. A reference grid, and its constituent BSU, is used to ensure spatial data
quality, for example in terms of ensuring consistent coverage and structure by
stopping shifting boundaries, removing gaps, etc.
3.23. It is also necessary to establish and agree on a reference coordinate system or
datum. A datum is a set of reference points on the earth's surface against which
position measurements are made, and (often) an associated model of the shape of the
earth to define a geographic coordinate system. A reference coordinate system based
on a common spatial projection is important for accounting purposes to ensure all
spatial datasets can be combined for analysis. Ecosystem accounting relies on being
able to specify the spatial location of EUs and requires the integration of different
spatial data sets. In these contexts, a reference grid with an agreed datum is an
essential tool for spatial analysis and reporting. It is necessary that all spatial data
layers, including grid and vector data are converted to the same reference coordinate
system for analysis.
3.24. Where a reference grid is established, a key question is what size the grid
squares should be for ecosystem accounting purposes. A general principle for the
reference grid is to use a grid size that is as small as possible within computational
capabilities. It is recommended that grid squares of 100m be considered the
maximum size. Grid squares down to 10m and smaller are now possible in some
countries, but whether delineation at that level of detail is required or appropriate for
ecosystem accounting should be informed by the use of the accounts for decision
making. In most cases, grid square sizes ranging between 25m and 100m will be
suitable.
3.25. Once a reference grid delineating BSUs has been established, attention must
turn to the attribution of information to the BSUs. Note that when attributing
information to a single BSU, it is assumed that, within the boundary of a BSU, its
structure and processes are homogenous. In principle, a range of information can be
attributed to each BSU, including EU type, land cover, soil type, elevation and other
biophysical information.
3.26. Spatial data are available in two formats – grids and vectors. With grid data,
it is likely that different types, or spatial layers, of information that will be used for
accounting will be produced using different grid sizes. For example, satellite data are
often mapped using 30m grids, or perhaps 250m grids when mapping larger areas or
for national analysis.
3.27. It is also possible to use vector data for the ecosystem extent account, and as
analytical basis for the data analysis of the other accounts. This has been tested in the
Netherlands, where five different maps depicting land cover, vegetation and land use
have been combined to create a vector map of the ecosystem units.
3.28. The smallest EUs measure 10 to 20 square meters, but the size of the EU
goes up to several hectares in some cases. It is assumed that the individual units are
homogenous in terms of ecosystem type, and ecosystem service flows are calculated
for each EU. The use of a vector format can be useful for the analysis of linear
elements in the landscape, in particular narrow elements which may not be covered
accurately using a raster map, such as roadsides, hedgerows or streams.
3.29. Generally, individual spatial data sets have been stored in vector format due
to their smaller file size. However, combining vector data sets is computationally
intensive and requires assumptions be made about the process adopted for adjusting
overlapping areas between data layers. More recently, there has been a move to the
use of raster data since computational capabilities are no longer limiting. By adopting
21
a reference grid, and converting all vector data to raster format, the time required to
analyse many data layers can be significantly reduced.
3.30. Once information is attributed to the BSUs of the reference grid, it is possible
to aggregate BSUs. One particular objective is to delineate EUs by combining
contiguous BSUs that have the same or similar characteristics (e.g. the same
vegetation type). In practice, the formation of a comprehensive set of BSU, often in
the form of a reference grid, is the path that will be taken to delineating EU rather
than a direct form of delineation. Other types of aggregations are considered in the
next section.
3.2.3 Geographical aggregations

3.31. Conceptually, it is possible to develop a set of ecosystem accounts for an
individual EU, such as an individual forest, wetland or farming area. This would be
akin to developing accounts for individual businesses. However, aside from the
measurement challenges this might entail if applied to all EUs within a country, the
general ambition of ecosystem accounting in the SEEA is to provide more general
guidance as to the changes in ecosystem related stocks and flows.
3.32. To provide this larger picture of ecosystems, it is necessary to consider
aggregations of EUs that both provide information (i) at a scale relevant for policy
monitoring and analysis, and (ii) at a scale where the accuracy of the information is
considered acceptable.
3.33. At the most aggregated level, this involves accounting at the national level,
i.e. covering all EUs within a country but it may also be relevant to create
aggregations (i) of EUs within specific sub-national administrative areas; (ii) of EUs
within hydrologically defined areas within a country (such as river basins); and (iii)
for other areas of policy interest (such as protected areas). Commonly, these
geographical aggregations will reflect contiguous areas, such as administrative areas
or river basins, but this is not a requirement for accounting purposes.
3.34. Within each of these geographical aggregations there will be multiple types
of EU, e.g. combinations of forests, rivers, agricultural areas, etc. The resulting
account structures, as described in Chapter 4, will be such that measures of ecosystem
extent, ecosystem condition and ecosystem services, will present information for
aggregations reflecting types of EU. For example, for a given sub-national
administrative area, an ecosystem extent account would show the changing total area
of each type of EU. It would not show the changing area of each individual EU.
3.35. There is no single classification structure for geographical aggregations.
However, it is important to ensure that a single account covers all EUs within a
country or large administrative area (e.g. state or province) boundary in a mutually
exclusive way. Thus, there may be a national-level ecosystem account that is
structured to reflect river basins; and another national level ecosystem account that is
structured to reflect sub-national administrative areas. In all cases, the sum of the
areas of all component geographical aggregations must equal the total area being
accounted for. It is also important that the reference grid provide coverage of the
geographical aggregations.
3.36. In some cases, ecologically, a single EU may cross administrative and
national boundaries. For example, large forests and lakes that cross the boundary of
neighbouring countries. In these cases, the EU should be delineated such that they
remain within the geo-political boundaries. In other cases, a single EU may be
classified to different geographical aggregations, for example to a river basin and to
22
an administrative region. Incorporating a single EU into these different geographical
aggregations will be relevant depending on the question being addressed.
3.37. The relationships between EU, BSU and geographical aggregations are
shown in Figure 3.1.
Figure 3.1 Relationships between spatial areas for ecosystem accounting
Source: (adapted from SEEA EEA Figure 2.4 (UN et al., 2014b)
3.38. In practice, a balance must be struck between providing a highly detailed

typology of EUs and providing information that can be readily translated to support
trend analysis at an aggregate level and which is of suitable accuracy. Approaches to
delineating spatial units are described in section 3.3.
3.39. The framework for delineating spatial units described above focused on
ecological characteristics. It is also possible to delineate spatial areas using
characteristics such as land ownership and land management. Since all data layers
will be snapped to a reference coordinate system, it is possible to combine these non-
ecological layers in different ways for accounting purposes. For instance, it is
possible to combine information on land tenure (ownership) with EU and river basin
information to understand the tenure of EUs within each river basin. The potential to
utilise land management information in the delineation of EU has also be noted
earlier in the chapter.
3.40. In many countries, cadastres are well established, administratively defined
units that are delineated on the basis of land ownership. Information from cadastre-
based datasets can be attributed to BSUs and, consequently, linked to information on
EUs, economic activity, land use and other socio-economic information.
3.41. The use of cadastre-based information is likely to be meaningful in terms of
understanding the link to policy initiatives, particularly in those countries where land
is under private ownership. However, it is not recommended that land ownership data
be directly applied to delineate EU since this information may raise sensitivities in
many countries.
23
3.3 Data sources, classifications and methods for delineating spatial units
3.42. The delineation of spatial units will involve the use of a range of spatial
information relating to:
 Land cover and land use
 The topography of the country (coastline, digital elevation model (DEM),
slopes, river basins and drainage areas)
 Vegetation, habitats and species composition
 Soil resources
 Meteorological data
 Bathymetry (for coastal areas)
 Administrative boundaries
 Population, built-up areas and settlements
 Transport and communication (roads, railways, power lines, pipelines)
3.43. Using these types of information, it is possible to construct maps for a given
country outlining different EUs as objects in the landscape. Object based
classification tools for spatial data can facilitate the mapping process. In practice, it
will be necessary to put in place a common spatial infrastructure and reference grid
that ensures the data can be integrated in a common spatial framework.
3.44. All spatial layers must have the same area coverage (in terms of information)
as that of the reference grid. It some instances, layers may have only been partially
populated with data. In these cases, the unpopulated areas of each spatial layer need
to be classified as either “no data” or “unclassified’, or the data need to be modelled
in some way, to ensure consistent coverage and reporting.
3.45. With these data sources and tools in place, there is a range of choices
available for delineating the spatial units needed for ecosystem accounting,
depending on scale (i.e. the level of spatial detail) and thematic detail (the number of
classes in the classification). The following considerations are relevant, beyond those
described in section 3.2.
3.46. First, there is no standardised method for delineating EUs. The approaches
used will depend, in part, on the extent to which information is available that can be
attributed to the BSU level and hence be grouped to form EUs, or whether individual
EUs can be directly mapped. Wherever possible, it is recommended that ecological
principles are followed since EUs are considered the units that function to supply
ecosystem services. Habitat/biotope and vegetation classification methods are
expected to offer the most suitable inputs for EU delineation.
3.47. The most obvious choices for geographical aggregations are nationally
defined statistical areas since they permit integration with national level economic,
social and demographic data. Statistical areas will also commonly correspond best to
the level of coverage of government decision-making. Depending on the decision
making context, other boundaries will also be relevant including river basins,
ecozones, landscapes and view-scapes, water provisioning areas, flood/storm
protection areas and protected areas (e.g. national parks).
3.48. In all instances, geographical aggregations must provide exhaustive and
mutually exclusive coverage of a country or sub-national area to ensure a clear
reference area for ecosystem accounts. For example, reporting for river basins should
still be undertaken with complete coverage of the country so that the total area of all
river basins is equal to the total area of the country.
24
3.49. There may be times when decision making and policy needs requires a focus
on select types of EU, for example wetlands. In compiling accounts for these
purposes, it is appropriate for the account to show not only forest areas but also the
“residual” non-wetland areas to ensure that the sum is still equal to the total area of
the country.
3.4 Key issues and challenges in delineating spatial units for ecosystem accounting
3.50. Approaches to delineating spatial areas for ecosystem accounting are still
developing and there are a number of issues that should be kept in mind.
3.51. First, there are a number of methods that can be used to delineate spatial units
through the integration of several data sources. The choice of method should take into
consideration the policy and decision making requirements.
3.52. Second, the framework including BSU, EU and geographical aggregations
has been developed in the context of terrestrial areas. Some work has commenced on
the application of the model to marine areas (e.g. South Africa (Driver, 2012),
Mauritius (Weber, 2014b), UK (UK NEA, 2011)) and to river systems (e.g. South
Africa (Driver, 2012) and Australia (Victorian DSE, 2013)). Further work is needed
to appropriately incorporate the atmosphere and airsheds; to deal with linear features
such as coastlines and hedgerows; and to account for the zones between different
ecosystem types – known as ecotones – which may have specific ecological
properties and supply specific ecosystem services. Particularly for linear features,
further testing is needed to assess whether the solution may lie in providing data at
higher resolution or whether an alternative approach is needed.
3.53. Third, for EUs, the choice of classification and the associated level of detail
is particularly important for the preparation of accounts. As explained further in
Chapter 4, the accounts to be compiled in the first stage of ecosystem accounting –
the ecosystem extent account, ecosystem condition account and the ecosystem
services supply and use table – are all structured based on data by type of EU. Since
each EU represents an ecosystem asset, measures of condition should be able to be
developed at the EU level which, in turn, should require an understanding of the
relevant characteristics in the supply of ecosystem services at that level. How
effectively these considerations may be brought into the delineation process requires
testing and research.
3.54. A related point is the extent to which the EUs are consistent with ecological
factors. If the EU are to represent ecosystem assets for accounting purposes, it is
reasonable to suppose that they would also reflect spatial areas that ecologists would
consider to be appropriate functional units.
3.55. Fourth, ideally the delineation of spatial units should take into account the
need to scale information both upwards and downwards, particularly in relation to the
attribution of information to the BSU level. Delineating units in a manner that
requires a heavier burden of assumptions to support scaling would likely reduce the
general quality of the accounts.
3.56. Fifth, it is likely to be the case that the delineation of spatial units will
involve the use of remote sensing data including satellite data. This is an important
step forward but is not without its challenges, particularly in the context of compiling
a consistent time series for accounting purposes. Further, there is the related
challenge of uncertainty in spatial interpretation and the need for validation and
ground truthing. These challenges are not unique to ecosystem accounting but
developing methods for adapting remote sensing data for ecosystem accounting
purposes is an important area for testing.
25
3.5 Recommendations
3.5.1 Recommended activities and approaches for testing
3.57. Based on the approach described in this chapter a number of points emerge as
being steps that countries can focus on in testing and experimentation in ecosystem
accounting. A clear theme in these recommendations is that work to establish the
spatial units required for ecosystem accounting should be undertaken within a
broader context of work to establish a national spatial data infrastructure that would
support integration of environmental and socio-economic data.
3.58. The starting point is the need to develop a reference grid/coordinate system
and a common spatial projection (i.e. grid size). This is required to ensure all data
layers no matter what their source can be snapped to a common basis and be
combined to produce statistics and accounts. Further, all data layers must be fully
populated for the country or area of interest. As noted above if they are not then
unpopulated areas of the layer need to be classified as “no-data” or unclassified.
3.59. Work on the delineation of spatial units should be undertaken at the national
level as part of the development of a national spatial data infrastructure (NSDI). Even
where the initial intent is to develop ecosystem accounts for a sub-national region,
there will be significant gains in articulating a national approach to the delineation of
spatial units. In particular, it will ensure that any initial work can be integrated into
subsequent larger scale measurement and it is likely that there are some production
efficiencies in undertaking the delineation task for all areas within a country at the
same time.
3.60. The development of an NSDI requires hardware with sufficient processing,
storage and back-up capacity, geographic information system (GIS) software4, and
integration of the following recommended data layers:
 Official country boundaries, official administrative and statistical boundaries for
collection of official statistics (in a shapefile of polygons)
 A common geographic projection, compatible with an international projection
(e.g. WGS 84)
 A reference grid with a common projection, applied as a BSU structure, and used
to organise and manage spatial data consistently and allocate all spatial input
datasets5 (it may also be used to prepare data cubes for storing and extracting
data)
 Elevation and topography data, based on a Digital Elevation Model (DEM). If no
detailed country-level data are available the global ASTER dataset6 can be used.
The DEM data are important to distinguish the elevation and slope of the BSU.
 Land cover data
 Additional data layers as available including
o Land management/use
o Vegetation type
o Soil and geology data
4
Freeware, such as QGIS, or commercial - ArcGIS
5
It is recommendable to identify the grid size of highest resolution that will be applicable. This needs to be
consistently nested into lower resolution sizes.
6
http://asterweb.jpl.nasa.gov/gdem.asp
26
o Hydrological data related to rivers, lakes, streams, coastal and marine
areas
o Data on urban infrastructure, including cities, villages, industrial zones,
and transport (rail, road), that is needed for assessing ecosystem
condition and understanding ecosystem use (e.g. relevant for mapping
fragmentation and other impacts)
3.61. Most likely the delineation of spatial units will rely, in the first instance, on
the use of land cover information. This will provide a strong starting point for the
compilation of ecosystem accounts. Over time, it is recommended that land cover
information be combined with information on other characteristics such as hydrology,
soil, vegetation types and road networks, to support delineation of EUs.
3.62. As part of the development of an NSDI and, ultimately a national register or
listing of ecosystem units, it will be important to understand spatial areas that have
already been delineated by government agencies for administrative purposes, for
example land management, river basin and catchment management and nature
conservation purposes. In some cases these existing spatial boundaries may provide a
suitable starting point for ecosystem accounting.
3.63. Further, it is relevant to develop an understanding of the hierarchy of
ecosystem/landscape/eco-region/biome units that are relevant for the country. This
can be done in reference to existing products (e.g. World Wildlife Fund (WWF) eco-
regions) and rules to convert (aggregate/disaggregate) between them (particularly
between EU and geographical aggregations). Overall, testing and experimentation
should reveal what are the most relevant data sources for the delineation of units for
ecosystem accounting ensuring that the outcome provides a spatially exhaustive
coverage of a country.
3.64. One area for testing is assessing the appropriate scale of analysis for different
analytical purposes. In particular it will be important to test whether data should be
maintained at the scale at which they are collected or transformed to a common
spatial scale for subsequent compilations. Also, the effect of the size of the BSU in
terms of measures of ecosystem condition and ecosystem services is an important
area for testing.
3.65. Other key aspects in testing and experimentation are:
 Assessing the accuracy and uncertainty in the use of remote sensing data for
ecosystem accounting purposes. This will normally require use of ground-
truthing (point) data.
 Establishing appropriate quality control criteria and assessment protocols
 The use of soil science to delineate spatial units, for example the potential
role of soil classifications and links to the SEEA Central Framework.
 Criteria for, and testing, other intermediate spatial units (such as landscapes,
viewscapes and river units)
 Linking levels of spatial units with specific information that may be
attributed to a unit for certain types of analysis.
 Understanding how spatial units are treated in various ecosystem services
models.
3.66. It is important to recognise that the ecosystem accounting approach will be
most useful when developed over a period of time. Indeed, given the potential
complexity of the accounts, a step-by-step approach and learning by doing are
27
required. In addition, data will be more useful when both spatial and intertemporal
(trend) analyses are possible.
3.5.2 Issues requiring ongoing research

3.67. In addition to testing the delineation of spatial units at country level, there are
some issues that require more substantive research and investigation. These include:
 The delineation of units for freshwater, coastal and marine ecosystems
 The treatment of airsheds, hydrological networks and other connective
phenomena within the spatial units framework
 The treatment of subterranean ecosystems (such as caves and ground water
systems)
3.68. The creation of units that are closely linked to ecological principles to aid
ecologically focused analysis. While some ongoing research is required, overall, the
work on the delineation of spatial units for ecosystem accounting is largely a matter
of testing and application of the principles as outlined in SEEA EEA and expanded
on in this chapter.
28
4. The ecosystem accounts

There are three main types of ecosystem accounts: accounts for ecosystem assets, accounts
for ecosystem services and integrated accounts which presented ecosystem accounting
information with standard economic and national accounts data.
Thematic accounts on themes such as land, biodiversity, carbon and water provide important
context and supporting information for ecosystem accounts.
The structure of ecosystem accounts follows the general structures of supply and use tables
and asset accounts as described in the SEEA Central Framework and the SNA.
There are important connections between the different ecosystem accounts which permit the
description of a single integrated picture of ecosystems.
In compilation, there is a progression in compilation from extent accounts, to condition and
ecosystem services accounts in physical terms, to ecosystem services supply and use tables in
monetary terms, to full integrated accounts.
The appropriate structure and detail for the ecosystem condition requires further consideration
following testing.
It is possible to compile the accounts at different levels of detail. It may be appropriate, as a
starting point, to compile information at a broader level within the proposed accounts, and
then, over time, add finer levels of detail as relevant for policy and analysis.
4.1 Introduction
4.1 The presentation of the main ecosystem accounts helps to frame much of the
discussion concerning data sources and compilation methods that follows in the
remainder of Technical Recommendations. This chapter describes the set of nine
ecosystem accounts as shown in Table 4.1. The different accounts have been grouped
as to whether they are accounting for ecosystem assets, for ecosystem services or
relate to the integration of ecosystem accounting data with the standard economic
accounts. While each account stands alone, there are also important connections
between the accounts. These connections reflect the accounting relationships between
stocks and flows.
4.2 The designs and structures of the accounts listed in Table 4.1 reflect, or are
adaptations of, the accounting structures of the SNA and the SEEA Central
Framework. Further testing of appropriate structures is still required, building on the
descriptions provided here.
29
Table 4.1: The ecosystem accounts
Ecosystem extent account
Accounts for
Ecosystem condition account
ecosystem assets
Ecosystem monetary asset account
Accounts for Ecosystem services supply and use table – physical terms
ecosystem services
Ecosystem services supply and use table – monetary terms
Integrated accounts* Combined presentations

Extended supply and use table
Sequence of accounts for institutional sectors
National and sector balance sheets
* These accounts reflect the integration of ecosystem accounting based information with information
from the standard set of national accounts
4.3 From a SEEA perspective, one ambition is to integrate information on

ecosystems with the standard national accounts. However, it is clear that achieving
this ambition requires iterations through a series of steps. On the whole, these steps
reflect following a path from the ecosystem extent account and ecosystem condition
account, on to the measurement of the supply and use of ecosystem services in
physical terms, then the valuation of ecosystem services in monetary terms, followed
by the valuation of baskets of expected future flows of ecosystem services to value
ecosystem assets, and finally, the integration of these data into the standard economic
accounts. This sequence is shown in Figure 4.1.
4.4 From a purely theoretical accounting perspective, this transition through the
various accounts is quite logical. However, in practice it is better to consider each
account as reflecting one module within an integrated system. Consequently, it is
possible to start the compilation process at different points and work iteratively
among the accounts such that a consistent and single picture emerges.
4.5 It should not be expected that all accounts will be compiled in the initial
stages and developing a comprehensive coverage of all of the accounts within the
ecosystem accounting framework will take some time.
30
Figure 4.1 Connections between ecosystem accounts
NB: The dotted line around boxes 2a and 2b indicates that the development of these two accounts may
often be completed in parallel, and iteration between them is appropriate in developing a single best
picture.
4.6 Special mention must be made of thematic accounts. These accounts cover
themes such as land cover, carbon, water and species diversity7 and are linked to
ecosystem accounting in two ways. First, they are accounts that record information on
themes that are of particular policy interest and hence can play an important role in
monitoring developments in, for example, carbon emissions, water scarcity and
biodiversity loss. The connection to ecosystem accounting in this context comes from
recognising that, by understanding the dynamic of the links between ecosystem assets
and ecosystem services, policy options that are aimed at these themes may be better
informed and targeted.
4.7 Second, the information in the thematic accounts may be a source of
information for the compilation of ecosystem accounts, especially for the estimation
of ecosystem extent and condition and the measurement of ecosystem services.
4.8 Each of the thematic accounts has a structure that is described in the SEEA
Central Framework or has a similar asset account based structure (as is the case for
carbon and species diversity accounts that are described in the SEEA EEA). Other
potential thematic accounts include accounts for soil resources, nutrients, air, timber
resources and fish resources. Chapter 7 provides a more complete introduction to four
thematic accounts - land, water, carbon and species diversity.
7
Following the CBD definition, biodiversity can be considered in terms of genetic, species and ecosystem
diversity. The assessment of ecosystem diversity is supported by the compilation of ecosystem extent
accounts. Genetic diversity has not been a focus of ecosystem accounting to this point.
31
4.2 Ecosystem extent accounts
4.9 The starting point for ecosystem accounting is most likely organizing
information on the extent or area of different ecosystem types within a country. This
is important for four reasons. First, the task of defining the ecosystems of interest for
accounting purposes is by no means straightforward and a balance between scale of
analysis, available data and policy questions will need to be found. It is very
appropriate to start this discussion by examining the most conceptually
straightforward issue of the definition of ecosystem assets and the delineation of their
extent.
4.10 Second, the organisation of information required to establish an ecosystem
extent account is likely to be a good entry point for establishing the national level
spatial infrastructure, i.e. NSDI, required for ecosystem accounting. As described in
more detail in Chapter 3, the delineation of spatial units will require the co-ordination
of a range of information. Ecosystem extent accounts are a first application in this
process.
4.11 Third, the structure of the ecosystem extent account, as shown below, gives a
clear indication of the nature of accounting for assets in a SEEA context. The
requirement to produce a time series of data to allow meaningful comparison between
the opening and closing of an accounting period is clear. This may be challenging
because distinguishing changes in ecosystem type over time is difficult, especially
where the changes are relatively local. It is also likely to be more complex as the
number of ecosystem types increases. The challenge may be met by using higher
aggregations of ecosystem types – for example using only 4 or 5 highest level classes,
but, in turn, this may limit the type of analysis that can be undertaken.
4.12 Fourth and finally, while the ecosystem extent account provides a clear base
for the development of the other ecosystem accounts, it also provides important
information in its own right. For example, ecosystem extent accounts should provide
an assessment of ecosystem diversity at a national level. Commonly, higher level
extent accounts will be based primarily on land cover information. Ideally, it will be
the case that all countries will be able to report changes in ecosystem extent at a
common level of detail on a regular basis.
4.13 The structure of a basic ecosystem extent account is shown in Table 4.2. The
structure of the rows reflects the basic logic of asset accounts as described in the
SEEA Central Framework with an opening extent (in hectares or km2), closing
extent, additions and reductions. The columns reflect the chosen classification for
ecosystem types. The proposed structure here uses high level EU types based on the
interim land cover classification in the SEEA Central Framework. Additional sub-
classes may be added depending on the ecosystem types of most relevance within a
country.
32
Table 4.2 Ecosystem extent account (hectares)
4.14 The alignment of the high level EU types with the SEEA Central Framework
land cover classes recognizes that the use of land cover information is the most likely
starting point for ecosystem accounting in practice. To maintain consistency over
time as ecosystem accounting develops, it is recommended that any more detailed
classification of ecosystem unit types be “nested” within these higher level land cover
classes.
4.15 From an accounting perspective, it is important to recognize that the total
area of a country is unlikely to change over an accounting period and hence the total
area recorded in the right hand column should be the same for the opening and
closing stock. This number should remain irrespective of the number of different
types of EU that are introduced into the table.
4.16 Ecosystem types of national interest may not coincide with one or more of
the 15 high level classes. For example, no singular definition of wetlands can be
applied at an aggregate country level, especially from the perspective of land cover. It
is likely that, in most countries, the majority of wetlands are within EU type 9 but
areas of wetlands may also be present in other land cover types, e.g. grasslands.
Where relevant for analysis, a separate aggregation of finer level ecosystem types
may be compiled, but for international comparison, data should also be presented
according to the 15 high level classes.
4.17 Information in an ecosystem extent account would be usefully presented in
maps using different colours for different types of EU. In the ecosystem extent
account presented in Table 4.2 there is no requirement that the areas of each type of
ecosystem be contiguous. That is, the total area of, for example, grassland, will occur
in various “patches” across a country and the data in Table 4.2 represents an
33
aggregation of all of the different patches. Mapping the information can more readily
highlight issues of fragmentation of ecosystem types and possible connections
between ecosystem types that are not apparent when the information is presented in a
traditional table format.
4.3 Ecosystem condition accounts

4.18 A central feature of ecosystem accounting is organizing biophysical
information on the condition of different ecosystem assets across a country. The
ecosystem condition account in Table 4.3 is compiled in physical terms using a
variety of indicators for selected characteristics.
4.19 Generally, it will be relevant to compile these accounts by type of EU within
a relevant geographical aggregation (as shown in Table 4.3). This is so because each
type of EU (e.g. tree-covered areas, grasslands, mangroves, etc.) will have distinct
characteristics that should be taken into account in assessing condition. This approach
also recognizes that much information on ecosystem condition is available by type of
ecosystem rather than in reference to landscape scales or administrative boundaries.
Consequently, harnessing available scientific information and expertise may be more
readily achieved through a focus on EU types.
4.20 There is a range of measurement issues and challenges in the compilation of
ecosystem condition accounts. Indeed, it is reasonable to conclude that there is still
much to learn about the structure and compilation of these accounts. Particular issues
concern the selection of characteristics for different ecosystem types, the relevant
indicators for different characteristics, the potential to aggregate across different
characteristics to derive an overall measure of condition for a single EU, the
aggregation of condition measures for multiple EUs of the same type, and the
approach to recording changes in ecosystem condition over time. These issues are
discussed in more detail in Chapter 6. Chapter 6 also provides recommendations for
testing and research related to ecosystem condition accounts.
4.21 The structure of the ecosystem condition account in Table 4.3 is focused on
recording information at a single point in time, i.e. it presents information on the
condition of different types of ecosystem assets at one time and does not show
changes in condition over time. Each characteristic shown in Table 4.3 uses one
column. In practice, it may be that each characteristic is best measured using more
than one indicator in which case, multiple sub-headings for each characteristic would
be needed. In some cases, multiple indicators may be combined to form an overall
index of condition for a single characteristic. By way of example, for the ecosystem
type of inland water bodies and the characteristic of water, indicators of river flow
and pollutant levels may be relevant.
34
Table 4.3 Ecosystem condition account – end of accounting period
4.22 Underpinning these accounts will be information from a variety of sources. In

some cases, source data may itself be organized following accounting approaches.
Examples include information concerning land cover, water resources, nutrients,
carbon and biodiversity. Accounts about these themes are discussed in Chapter 7.
4.4 Ecosystem services supply and use table

4.4.1 Introduction
4.23 The supply of ecosystem services by ecosystem assets and the use of these
services by economic units, including households, is one of the most important
aspects of ecosystem accounting. These are the flows that reflect the link between
ecosystems and economic and human activity. The supply and use table records the
actual flows of ecosystem services supplied by ecosystem assets and used by
economic units during an accounting period. The data relate to a given geographical
aggregation and should be structured by type of ecosystem service. The table may be
compiled in both physical and monetary terms.
4.24 To interpret the supply and use table, it is important to distinguish between
economic units and ecosystem assets in relation to the supply and use of ecosystem
services. Following the ecosystem accounting model described in Chapter 2, only
ecosystem assets can supply ecosystem services that are then received by economic
units.
4.25 This distinction may seem quite abstract. For example, in the case of
agriculture or forestry related ecosystem services, it may seem that it is the farmer or
forester as the economic unit that supplies the ecosystem service. However, for
accounting purposes, in these instances, the farmer and forester should be seen as
supplying benefits (crops, timber, etc.) that are produced by combining inputs from
other economic units (fuel, fertiliser, etc.) and inputs from ecosystem assets (i.e. the
ecosystem services). Consistent with this explanation, ecosystem assets cannot
produce or use benefits (or products in the case of the supply and use account shown
below).
4.26 The structure of the supply and use account is shown in Table 4.4. The
various quadrants, labelled A – H reflect the following information.
35
 A: No data are recorded in this quadrant as in concept economic units
cannot supply ecosystem services.
 B: In this quadrant the supply of ecosystem services by type of EU is
recorded.
 C: This quadrant is the equivalent of the standard physical supply and use
table showing the supply of products by different economic units. This
reflects the production of benefits to which the ecosystem services
contribute. The scope of products is all goods and services produced in
an economy.
 D: No data are recorded here as, in concept, EUs cannot supply products.
 E: Here the use of ecosystem services by types of economic units is
recorded. This includes both the use of ecosystem services as input to
further production and the use of ecosystem services as final
consumption.
 F: At this stage, it is not anticipated that data would be recorded here as it
represents the use of ecosystem services by other EUs – i.e. intermediate
ecosystem services. If these flows were to be recorded then the supply of
ecosystem services in quadrant B would need to have an equivalently
larger scope.
 G: This quadrant is the equivalent of the standard physical supply and use
table showing the use of products by different economic units.
 H: No data are recorded here as, in concept, EUs cannot use products.
4.27 As described, the structure of the ecosystem services supply and use table
incorporates flows of products. This supports the joint presentation of data on both
the ecosystem services used by economic units, and the products (SNA benefits) to
which those ecosystem services contribute. In terms of the quadrants of Table 4.4, the
output of products such as livestock would be recorded in the first column of
quadrant C (under agriculture) and the use of ecosystem services (e.g. grass
consumed directly by livestock) would be recorded in the same column in quadrant E.
(The original supply of those ecosystem services would be recorded in quadrant B.)
As desired, quadrant G could record inputs of products such as fertilizer or veterinary
costs. In effect, each column pertaining to type of economic unit can be used to
record key elements of a production function that includes ecosystem services as
inputs.
4.28 The basic structure of the account comes from the design of physical supply
and use tables (PSUT) in the SEEA Central Framework. There are two principle
alterations. First, rather than showing just one column representing the environment,
there are multiple columns each representing a type of EU. In effect, each EU is
considered a unit supplying ecosystem services.
4.29 Second, in the SEEA Central Framework PSUT there are three types of flows
– natural inputs, products and residuals. In the ecosystem services supply and use
account there are just two – ecosystem services and products. In some cases, the
flows relating to natural inputs are replaced by flows relating to provisioning services
– e.g. the harvest of biomass from natural forests. However, particularly for
regulating and cultural services, there are additional flows that are not recorded in a
standard PSUT table, for example air filtration services and opportunities for
recreation.
36
Table 4.4 Ecosystem services supply and use table
4.30 Residual flows are not ecosystem services but reflect physical flows from
economic units into the environment. In cases where the residual flows comprise
waste or pollution, ecosystems may play a regulating and filtering role in reducing the
impact of residual flows on ecosystem condition. In these cases, a flow of regulating
services from ecosystems to economic units should be recorded in quadrant B. These
flows of regulating services will be related to the residual flows but are a different
concept.
4.31 The accounting for residual flows is described in Chapter 3 of the SEEA
Central Framework. This accounting treatment could be modified and appended to
the supply and use table shown above. Thus, relevant residual flows might be
recorded as being supplied by different economic units (immediately below block C)
and then shown as being used/received by different ecosystems (immediately below
block H). The addition of this information may assist in conveying a broader picture.
However, care should be taken since, as noted above, information concerning
residuals is related but different from ecosystem services and hence there is no
37
defined accounting relationship to be conveyed by including both residual flows and
ecosystem services in a single table.
4.32 Another difference from the PSUT is that, generally, a PSUT will follow the
flows of a single type of resource or material – e.g. energy or water. In the ecosystem
services supply and use account there are multiple services and products. Thus, in
terms of balancing the table, the focus should be on ensuring that the supply and use
for each row is balanced, i.e. for each ecosystem service. Indeed, in physical terms
this is the only balance that can be determined.
4.33 Although the supply of ecosystem services is shown as only emerging from
ecosystems (quadrant B), there may be interest in attributing the supply of services to
particular economic units. For example, ecosystem services that are inputs to the
production of farming outputs might be attributed to agricultural businesses. This
alternative presentation could be recorded in quadrant A. Note however, that while
this recording is likely to be reasonably straightforward for provisioning services, it
will be more difficult to attribute the supply of regulating services to specific
economic units.
4.34 The potential extent of this supply and use table is all products within the
production boundary. One way of seeing this table is thus as an extended supply and
use table that contains flows of all products between industries, as well as flows of
ecosystem services. Viewing the account in this way is consistent with the idea,
inherent in the ecosystem accounting model, of expanding the SNA production
boundary to include the supply and use of ecosystem services. The table facilitates
recording this additional supply and use. In Table 4.4, this potential expansion to
incorporate a full supply and use table would imply expanding (considerably) the
quadrants of the tables labelled C and G. A longer description of this proposed
extended supply and use table, including its distinction from environmentally
extended input-output tables (EE-IOT)8, is presented in Chapter 9.
4.35 The current focus for the compilation of the ecosystem services supply and
use table is on flows of final ecosystem services – i.e. those ecosystem services where
there is a direct interaction with economic units or people. Conceptually, the design
of the ecosystem services supply and use account can also be applied to record the
flows of ecosystem services between different ecosystem units – so-called
intermediate services. An example is water regulation services from an upstream
forest provided to a reservoir from which drinking water is subsequently abstracted.
In principle, these flows could be recorded as being supplied by ecosystem units in
quadrant B (i.e. total supply recorded in quadrant B would increase) and recorded as
used by other ecosystem units in quadrant F.
4.4.2. The ecosystem services supply table

4.36 A likely challenge in compiling the supply table will be attributing the supply
of ecosystem services to a specific EU. This is unlikely to be an issue for
provisioning services, but it may be of concern for regulating services and some
cultural services in cases where the service is provided through a combination of
ecosystem types.
4.37 Given this, it is recommended that, as a first step in accounting for ecosystem
services, compilers create a table showing which ecosystem services are likely to be
supplied from different EU types for their country or target geographical area. In
undertaking this task, it is relevant to use a classification of ecosystem services such
as CICES as a type of checklist. It is to be expected that for some services,
8
For an introduction to EE-IOT see SEEA Applications and Extensions (UN, 2014)
38
particularly regulating services such as carbon sequestration, the same service will be
supplied by more than one EU type. Also for some ecosystem services the service
will be supplied as a result of the combined production of neighbouring EU types.
For example, cultural services supplied in a mixed landscape setting. In these cases,
some allocation between EU types will be required.
4.38 It will be relevant to use this initial table as a discussion document to obtain
input from various experts. It is important that the development of such a table is
informed by people experienced in considering the link between ecosystems and
economic and human activity. This should ensure that commonly overlooked services
are not ignored.
4.39 This table would also serve as a basis for scoping and prioritising the
required work, and comparing compilation exercises across countries (for example
comparing lists of ecosystem services attributed to forests). Completing such a table
is also a good expression of the accounting approach of working from the outside-in9,
in contrast to the “bottom-up” measurement of selected ecosystem services for
specific ecosystem types.
4.40 The proposed ecosystem services supply table (Table 4.4) has columns
reflecting the various ecosystem types and rows reflecting the range of different
ecosystem services, in this case classified following CICES. In this table there is no
direct recording of the beneficiaries or users of ecosystem services, this takes place in
the ecosystem services use table. It will be relevant to also compile information on
the combination of EU type, ecosystem services and beneficiaries at the same time.
4.41 The choice of indicators for measuring the flows of different ecosystem
services is discussed in Chapter 5 and relevant data sources and examples are
provided in that chapter. Recommendations for countries for testing and research are
also discussed in Chapter 5.
4.42 The ecosystem services supply table shown in Table 4.4 can be compiled in
physical terms and in monetary terms, when valuation is possible. When compiled in
physical terms each ecosystem service will be measured using a different unit of
measurement. One consequence is that there can be no aggregation of ecosystem
services over different ecosystem service types because the relative importance of
individual ecosystem services cannot be immediately determined. Aggregation within
a single row to obtain a total flow from all types of ecosystem units is possible. In
practice, for some services, compilation may involve using aggregate, country or
geographical area level information for a single ecosystem service and then allocating
that information to the EU type level.
4.43 The ecosystem services supply table can be compiled in monetary terms,
usually by applying appropriate prices to the physical flows of each ecosystem
service. Direct measurement of values may be possible for some provisioning
services. The ecosystem services supply table shown in Table 4.4 can then be
extended with additional rows to record the total flows of ecosystem services. The
estimation of prices for ecosystem services is discussed in Chapter 8.
4.4.3 The ecosystem services use table

4.44 An important difference between the supply and the use tables is that the
focus of the use table is on the link between ecosystem services and different types of
beneficiaries, while the supply table focuses on the supply from types of EU.
Beneficiaries include economic units classified by industry, government sector and
9
For an explanation of the “outside-in” approach of accounting see Annex 3.
39
household sector units, following the common conventions of organising the national
accounts.
4.45 The focus on beneficiaries arises because, while the supply of ecosystem
services can be directly linked to a spatial area (e.g. to an EU), there is no
requirement that the location of the beneficiary is the same as the location of the area
from which the ecosystem service is supplied. This is especially the case for
regulating services but also for some cultural services.
4.46 While a precise link between beneficiaries and the spatial areas from which
ecosystem services are supplied may be difficult to define, it is likely to be useful to
consider, for different ecosystem services, whether the beneficiaries are, in general
terms, local, national or global. For example, in the case of most provisioning
services, the direct beneficiaries will be located within the supplying spatial area (e.g.
farmers, foresters, fishermen, water supply companies). This will also be true of
many cultural services where there is a recreational or touristic component. However,
for many regulating services the beneficiaries will be located in neighbouring
ecosystems (for example air filtration) or will be global beneficiaries (for example
with respect to carbon sequestration).
4.47 Given the lack of a definitive spatial link, the structure of the ecosystem
services use table must be guided by possible uses and analysis of data. The choice
made here, is to structure the ecosystem services use table showing the allocation of
the total supply of each ecosystem service to the various economic units. This
allocation provides the first sense of a link to the national accounts datasets.
4.48 As for the ecosystem services supply table, the use table may be compiled in
both physical and monetary terms. In physical terms, entries will be limited to
measures of indicators for each ecosystem service. Note that, since supply must equal
use, the unit of measurement applied for each ecosystem service must be the same in
both the supply and use table in order for a balance to be obtained.
4.49 In monetary terms, entries for the total use of ecosystem services can be
derived both for individual ecosystem service types and for total use by each
beneficiary. The estimation of prices for ecosystem services is discussed in Chapter 8.
4.50 The presentation of the tables outlined here may suggest that the supply of
ecosystem services would necessarily be compiled before measuring the use of
ecosystem services. In practice, the reverse may be the case or at least compilation of
the supply and use estimates should take place in an iterative fashion. For example,
measures of provisioning services are likely to be estimated based on measures of the
extraction of materials or energy from the environment by economic units, i.e. a use
perspective. It is then this perspective then drives the estimation of supply. Since for
all final ecosystem services there must be some link to economic units and other
human activity, there is a strong case for compiling both the supply and use of
ecosystem services in tandem.
4.5 Ecosystem monetary asset account

4.51 The SEEA Central Framework uses the asset account structure to record
information on stocks and changes in stocks (additions and reductions) of individual
environmental assets such as mineral and energy resources, timber resources, water
resources, etc. This standardised approach to recording information about specific
asset types is a particularly useful way of structuring relevant information about
changes in the asset base.
40
4.52 When focusing on individual environmental assets, it is possible to develop
asset accounts in both physical and monetary terms since the units of measurement in
physical terms can be consistently recorded in a single account. For example, all
timber resources can be measured in both cubic metres and in a common currency
unit.
4.53 For ecosystem assets, their measurement in physical terms (as undertaken in
the ecosystem condition account) is a more complex process, requiring the integration
of data on a range of characteristics each with different units of measure.
Consequently, at this stage, the ecosystem condition account is envisaged as
reflecting estimates only in terms of the opening and closing condition for an
accounting period. Defining entries for additions and reductions in condition requires
further consideration.
4.54 Accounting for ecosystem assets in monetary terms would appear more
tractable, since a single unit of currency is used. However, the complexities in
accounting for the changes in assets remains.
4.55 In the ecosystem monetary asset account, the opening and closing stocks of
ecosystem assets are estimated using the net present value of the future stream of
each ecosystem service – covering provisioning, regulating and cultural services. It is
assumed that the individual services are mutually exclusive and can be aggregated.
4.56 Assuming that the net present value for each type of service is separable, it is
possible to consider the total value and changes in value for each ecosystem service
flow separately. For provisioning services, such as timber extraction, accounting for
the additions and reductions is conceptually straightforward. In this case, the value of
provisioning services will reflect a resource rent. Therefore accounting for the
changing value of provisioning services is equivalent to the advice provided in the
SEEA Central Framework and the SNA on accounting for individual environmental
assets. Applying this approach is possible since the underlying physical flows (e.g. of
timber resources) can be accounted for using a single metric.
4.57 For regulating and cultural services, the link to underlying physical flows
may be less clear. For regulating services, the issue is that the supply of the service
does not only depend upon the extent and condition of the ecosystem, but also upon
other driving factors that will not be stable over time. For example, air filtration is a
function of the extent and type of vegetation and its leaf area index, but it is also
influenced by air pollution levels that can be spatially and temporally heterogeneous.
Thus, the higher the concentration of atmospheric particulate matter, the higher the
amount of particulate matter that is captured by the vegetation.
4.58 In addition, most regulating services depend upon a variety of ecosystem
components and processes, with each of them providing only limited information on
the capacity of the ecosystem to supply the service over time. Recording additions
and reductions in ecosystem assets for regulating services is thus likely to involve
different types of indicators for different services. These indicators may reflect
ecosystem characteristics, such as leaf area index per BSU in the case of air filtration
or the density of pollinators in the case of pollination, or they may reflect outputs of
the ecosystem, for instance, amount of water made available for irrigation throughout
the year in the case of the water regulating service provided by upland forests.
Establishing accounting entries for changes in ecosystem assets related to regulating
services is therefore not straightforward.
4.59 For cultural services, it may also be challenging to find appropriate physical
indicators for quantifying the services in terms of the underlying ecosystem assets.
Cultural services can involve a passive (enjoying without visiting through
information obtained from various media) or active (visiting) interaction with an
41
ecosystem. Thus describing the link between the condition of the ecosystem in
physical terms and its capacity to supply individual cultural services is difficult to
define in general terms. For example, in the case of recreation, a natural park may
physically allow access to large number of people, but at some point the recreational
experience gained per individual will decline because of overcrowding. It is however,
hard to quantify this decreasing marginal return.
4.60 Overall, policy-relevant physical indicators about ecosystem assets can be
defined for most provisioning services, but for many regulating and cultural services
such physical indicators are more difficult to define. The challenge to define physical
indicators also impacts on the ability to develop estimates of changes in assets in
monetary terms.
4.61 It is therefore envisaged that, in the first phases of ecosystem accounting,
ecosystem monetary asset accounts should focus on estimating the opening and
closing value of the stock of ecosystem assets. This estimation can be followed by
applying the standard national accounting technique of net present value, where the
opening and closing stock value of an ecosystem asset can be estimated by using
assumptions about the future flows of ecosystem services and discounting these flows
to provide a current, point in time, estimate of their value.
4.62 The relevant accounting structure is shown in Table 4.5. The entries in the
rows have been simplified to very basic asset account entries. If more detail is
required to account for changes in assets, particularly those related to provisioning
services, then additional entries similar to those from the SEEA Central Framework
could be incorporated. In the columns, different presentations are possible given that
the data are in monetary terms. That is, a single asset account may relate to an
individual ecosystem asset (e.g. a specific grassland), to a type of ecosystem (e.g. all
tree-covered areas), or to an administrative region or country.
Table 4.5 Ecosystem monetary asset account (currency units)
4.63 Entries in the asset account for ecosystems go beyond the measurement
requirements of the ecosystem services supply account in monetary terms by
incorporating the use of net present value techniques and assumptions about the
42
future. That is, the focus is on the measurement of the value of ecosystem assets as
distinct from ecosystem services. In measurement terms, this represents an increase in
uncertainty given the general challenges of net present value based estimation.
4.64 Nonetheless, such estimates are useful for making decisions about alternative
uses of ecosystem assets since they provide a consistent basis for comparison. In
addition, estimates in monetary terms can be integrated with valuations for other
types of assets to provide more complete assessments of net wealth.
4.65 Using the data recorded within an asset account, it is possible to derive an
estimate of ecosystem degradation in monetary terms. In general terms, ecosystem
degradation will reflect the decline in the value of an ecosystem asset over an
accounting period (i.e. between opening and closing positions), where the decline is
considered to be due to the use of ecosystems by economic units. This should be seen
as quite distinct from the change in the net present value of the asset.
4.66 In national accounting terms, the concept of ecosystem degradation has a
specific role. It represents the capital cost that should be attributed to a user of an
ecosystem asset in generating an income stream. Thus, degradation should not
include changes in the value of the asset that arise for other reasons. In particular,
reductions in asset value due to unforeseen events, that are not part of the use of the
asset in production (e.g. due to natural disasters), are not considered part of
degradation for accounting purposes. Further, it is possible that the value of an asset
changes solely due to changes in prices. These are considered revaluations for
accounting purposes and are separately recorded.
4.67 These distinctions are reflected in Table 4.5 where the series of entries
between opening and closing stock are characterised for different types of assets.
Note that for ecosystems, depletion will be a subset of degradation, in that depletion
refers only to the loss of future income associated with provisioning services from an
ecosystem – e.g. timber resources from a forest. Degradation will encompass losses
from provisioning services and other ecosystem services. An important aspect in the
table is that there is a consistency of treatment in the accounting framework between
consumption of fixed capital (depreciation of produced assets), depletion and
degradation.
Table 4.6 Accounting entries for depletion and degradation

Accounting entry
Type of asset Opening Transactions Other changes in Revaluations Closing
stock volume stock
Produced assets Gross fixed Primarily physical Changes in
capital appearance and value between
(incl. cultivated formation disappearance of opening and
biological resources (investment) assets closing stock
e.g. plantations, solely due to
livestock) Depreciation - discoveries changes in
Natural resources Depletion - catastrophic losses prices of assets
Ecosystem assets Degradation - reappraisals of

stock
4.68 However, within this accounting construct, further consideration of exactly

how ecosystem degradation should be defined is required, building on the discussion
of this issue in SEEA EEA Chapter 4. Relevant issues, discussed further in section
6.4 and 9.5, include:
43
i. the treatment of complete changes (conversions) in the use of an ecosystem – for
example from a forest area to an agricultural area;
ii. the treatment of situations where economic activity, including household
consumption, has an indirect and potentially delayed impact on ecosystem
conditions – for example, the impacts arising from human-induced climate
change;
iii. the treatment of declines in condition of ecosystems that are not direct suppliers
of final ecosystem services – for example, remote forests.
4.69 Another consideration is the role of capacity in the measurement of
degradation. Recent discussion (see Hein et al., 2015) suggests that the change in
capacity may represent an appropriate measure of ecosystem degradation but further
consideration of this alternative and the associated accounting implications is
required. (See section 6.4 for further discussion)
4.6 Integrating ecosystem accounting information with standard national accounts

4.70 The discussion of the ecosystem services supply and use account highlighted
the potential for information on ecosystem services to be integrated with information
presented in standard supply and use or input-output tables. However, aside from this
potential, the accounts described in the previous sections do not involve integration of
information on ecosystem assets or services with the standard national accounts.
Since one of the motivations for the development of ecosystem accounting is
integration with the standard national accounts, this section introduces how this step
might be achieved. A more complete discussion is presented in Chapter 9 and the text
in SEEA EEA Chapter 6 is also of particular relevance.
4.71 Building on the SNA framework, there are two main types of accounts that
can be adapted to integrate ecosystem accounting information. The first is integrated
institutional sector accounts and the second is balance sheets. Institutional sector
accounts, commonly referred to in national accounting as the sequence of accounts,
record information on the generation and distribution of income, saving and
investment by institutional sectors (e.g. household saving), and transactions in
financial assets and liabilities.
4.72 The intent from an ecosystem accounting perspective is to integrate estimates
of ecosystem services in monetary terms and estimates of ecosystem degradation (as
a cost of capital equivalent to consumption of fixed capital) into the standard
sequence of accounts. The type of presentation that emerges from this integration is
shown in Table 4.6, taken directly from SEEA EEA Annex A6.
4.73 SEEA EEA Table A6.1 presents simplified versions of two models (A and
B). In the example, presented for a farm, a single ecosystem asset provides a mix of
ecosystem services (total of 110) of which 80 are used by the farmer and 30 are the
final consumption of households. 10 All SNA production of the farmer (200) is
recorded as final consumption of households. For simplicity, no other production,
intermediate consumption or final consumption is recorded. It is to be noted that in
the generation of ecosystem services, there is no recording of “inputs” from the
ecosystem. Such recording is not required for the purposes of developing a sequence
of accounts focused on economic units.
10
The allocation is based on the assumed composition of the ecosystem services. Thus, the
value of 80 for ecosystem services may be considered inputs to agricultural production and the
value of 30 may be considered regulating services, such as air filtration, used by households.
44
4.74 In both models, the rise in GDP occurs in relation only to the final
consumption of ecosystem services that relate to non-SNA benefits. Many ecosystem
services will be indirectly included in measures of final consumption when they are
used by enterprises in the production of standard SNA outputs (e.g. food, clothing,
recreation).
4.75 Measures of GDP may be adjusted for both consumption of fixed capital and
ecosystem degradation, thus providing degradation adjusted net domestic product.
4.76 In drafting the SEEA EEA, there was no clear choice about the structure of
the sequence of accounts with a decision needed on whether (i) ecosystems should be
treated as producing units in their own right; or (ii) treated as assets owned and
managed by existing economic units. At this stage, no decision on this issue has been
made. However, discussion on other issues in ecosystem accounting suggests treating
ecosystem assets as producing units is the preferable option, on balance. This would
imply a sequence of accounts akin to Model A in Table 4.7. The largest as yet
unresolved issue in compiling a sequence of accounts is the attribution of ecosystem
degradation to economic units. This is a topic that will remain on the research agenda.
The second type of accounts are balance sheets, in which the opening and closing
values of ecosystem assets in monetary terms, as recorded in the ecosystem monetary
asset account, are integrated with the values of asset and liabilities recorded in the
standard balance sheet of the SNA. Such an integration would lead to the derivation
of extended measures of national and sector net wealth.
4.77 The integration of ecosystem asset values would seem a relatively
straightforward step. However, for a variety of reasons, it is likely to be quite
complex. There are two main challenges that are described at more length in SEEA
EEA Chapter 6. First, in a full SNA and SEEA Central Framework balance sheet,
there will already be values recorded for natural resources, such as timber and fish.
Since the value of these resources is embedded in the value of ecosystem assets it will
be necessary to appropriately ensure the removal of double counting of these
resources. This will also apply to various cultivated biological resources such as
orchards and vineyards.
4.78 Second, in many countries, the value of land will be recorded on the SNA
balance sheet estimated in terms of its market price. Since there is a generally well-
established market in land, balance sheet values may be obtained more directly than
by using net present value techniques as applied in resource accounting. It is likely to
be the case that the market values of land, particularly agricultural land, will capture
the value of some ecosystem services, to some extent. However, they are unlikely to
capture a full basket of ecosystem services, particularly those that have clear public
good characteristics and longer term benefits. Also, the land value may well reflect
aspects that are not ecosystem services in nature – for example, the location and the
value of alternative uses.
4.79 From a national accounting perspective, the development of a sequence of
accounts and balance sheets represents an important objective that helps to motivate
the development of other parts of the ecosystem accounting framework. At the same
time, it is clear that: (i) work is needed to progress the development of the ecosystem
accounts which must underpin the integrated accounts described here; and (ii) that
further research and testing is needed to meet the challenges posed by integration.
Consequently, it is recommended that countries focus their efforts on developing
ecosystem extent and condition accounts and ecosystem services supply and use
accounts. There is tremendous value in these accounts in their own right.
45
Table 4.7 Simplified sequence of accounts for ecosystem accounting
Model A Model B
Farmer Household Ecosystem Total Farmer Household Total
Production and generation of
income accounts
Output – Products 200 200 200 200
Output – Ecosystem services 110 110 30 30
Total Output 200 110 310 230 230
Int. consumption – Products 0 0 0 0 0

Int. consumption –Ecosystem services 80 0 80 0 0
Gross value added 120 110 230 230 230
Less Consumption of fixed capital 10 10 10 10

(SNA)
Less Ecosystem degradation (non- 15 15 15 15
SNA)
Degradation adjusted Net Value 110 95 205 205 205
Added
Less Compensation of employees – 50 50 50 50

SNA
Degradation adj. Net Operating 60 95 155 155 155
Surplus
Allocation and use of income

accounts
Degradation adj. Net Operating 60 95 155 155 155
Surplus
Compensation of employees 50 50 50 50
Ecosystem transfers 80 30 -110 0 -30 30 0
Disposable income 140 80 -15 205 125 80 205
Less Final consumption – Products 200 200 200 200

Final consumption – Eco. serv. 30 30 30 30
Degradation adjusted net saving 140 -150 -15 -25 125 -150 -25
Source: SEEA EEA Table A6.1 (UN et al., 2014b)
46
5. Accounting for flows of ecosystem services

In ecosystem accounting, ecosystem services are considered to reflect the contribution that
ecosystems make to the benefits received by economic units and people from the
environment. It is therefore important to distinguish clearly between ecosystem services and
benefits.
Generally, most focus for national level accounting is on final ecosystem services. All final
ecosystem services have a direct link between ecosystems and economic units.
Intermediate ecosystem services are important for understanding relationships and
dependencies between ecosystems and can be incorporated into the ecosystem accounting
model.
The use of a classification of ecosystem services, such as CICES, is an important aspect in
compiling estimates of ecosystem services flows.
In almost all cases it will be necessary to undertake modelling to estimate flows of ecosystem
services. The modelling may be either spatial or temporal involving a dynamic systems
approach.
In measuring the supply of ecosystem services it will be useful to also consider the use of
ecosystem services by different beneficiaries including households, business and
governments.
In some cases, biodiversity may be considered a cultural ecosystem services but generally,
biodiversity is best considered as a characteristic of ecosystem assets that can be degraded or
improved over time, and which underpins the supply of ecosystem services.
5.1 Introduction
5.1. Ecosystem services are the glue that enables the connection to be made
between ecosystem assets on the one hand and economic production and
consumption on the other. Their measurement is thus central to the ambition to
integrate environmental information fully into the existing national accounts.
5.2. However, recognition of the potential role of the ecosystem services concept
in an accounting context has followed the development and testing of the concept in
other disciplines. As a result, the reality is that there are multiple definitions,
alternative measurement boundaries and classifications, and a wide array of
measurement methods available in relation to ecosystem services. The SEEA EEA
attempted to chart a course through the various discussions on ecosystem services. It
made a range of choices about the definition and measurement of ecosystem services
for accounting purposes.
5.3. This chapter summarises the main points from the SEEA EEA concerning
ecosystem services, discusses possible refinements to the SEEA EEA discussion, and
describes the main measurement issues and remaining challenges.
47
5.2 The definition of ecosystem services
5.4. Because of the ambition to integrate measures of ecosystem services with the
standard national accounts, the measurement scope and definition of ecosystem
services in the SEEA EEA must be defined in the context of the SNA production
boundary. This boundary sets the scope for the measurement of GDP and related
measures of production, income and consumption. It is noted that this means that
ecosystem services related to illegal activity or subsistence production are included in
scope since both of these types of activity are included in the production boundary of
the SNA.
5.5. An important part of the rationale for measuring ecosystem services is that
much economic production (for example in agriculture, forestry and fisheries) utilizes
inputs directly taken from ecosystems but these inputs are not recorded in the
standard accounting framework. In these situations, the logic of the SEEA EEA, is
that ecosystem services should be differentiated from the goods and services that are
produced. Thus, the ecosystem services represent the contribution of the ecosystem to
the production of those goods and services. In effect, this sets up an extended input-
output or supply chain that includes ecosystem assets as suppliers, whose contribution
was previously not explicitly recognised. Defining the contribution of ecosystems is
particularly important in valuation but also relevant in ensuring that physical
measures of ecosystem service flows are consistently described and appropriate
indicators chosen.
5.6. A second important part of the rationale for measuring ecosystem services is
the understanding that there are many benefits that economic units, and society more
generally, receive from functioning ecosystems, and that a full and proper accounting
would incorporate this production of services by ecosystems, and the consumption of
them in economic and human activity.
5.7. With these two rationales in mind, the SEEA measurement of ecosystem
services recognizes all of the additional production by ecosystems. If accounting had
been starting from a zero base of information on ecosystem services, then it would be
possible that measurement would be simply limited to this scope. However, as noted,
the measurement of ecosystem services has a longer and wider history and
consequently the following factors need to be taken into account.
5.8. Distinguishing ecosystem services and benefits: The SEEA EEA accounting
model makes a clear distinction between ecosystem services and the benefits to which
they contribute. From an accounting perspective, the distinction is meaningful since
 it allows description of the relationship between final ecosystem service flows
and existing flows of products currently recorded in the SNA
 it recognizes the role of human inputs in the production process and that the
contribution of final ecosystem services to benefits may change over time (e.g.
due to changes in the methods of production)
 it helps in identifying the appropriate target of valuation since the final
ecosystem services that contribute to marketed products (e.g. crops, timber, fish,
tourism services) will only represent a portion of the overall value of the
corresponding benefits.
5.9. For these reasons, the principle of distinguishing between final ecosystem
services and benefits is appropriate for accounting. It is also consistent with the
approach taken in TEEB (2010), Boyd and Banzhaf (2012), Haines-Young and
Potschin (2013) and the UK National Ecosystem Assessment (NEA) (2011), although
48
the precise definitions and terms applied for final ecosystem services and benefits
vary in the different cases.
5.10. In practice, particularly at large scales, the explanation and application of this
distinction can be challenging. The issues arise differently for provisioning services,
regulating services and cultural services. For provisioning services, the challenge is
fully describing the various ecosystem services involved in supplying cultivated
biological resources 11 . These outputs, including crops, plantation timber, and
aquaculture, are considered benefits produced as a combination of final ecosystem
services and human inputs. Further, since the balance of inputs between final
ecosystem services and human inputs will vary by production process (e.g. between
irrigated and rainfed agriculture), it means that using a measure of output from
production as a measure of the ecosystem service will be misleading.
5.11. For regulating services, there are generally no direct human inputs consumed
in the production of benefits, although there may be costs incurred in maintaining an
ecosystem such that the supply of regulating services can be supported.
Consequently, the quantity of final ecosystem services may be equal to the quantity
of the benefit, as in the case of, for example, carbon sequestration services supplied
by a forest. However, in other cases, there is a distinction between final services and
benefit. For example, in the case of air filtration services, the benefit is reduced risk
(to the local population) of respiratory and cardio-vascular diseases through cleaner
air. The ecosystem service in this example is the capture of air-borne pollutants.
5.12. The challenge is to appropriately describe the ecosystem service such that the
focus of measurement is appropriate. The focus in describing the ecosystem service
should be a description of ecosystem processes that reflect the contribution of the
ecosystem to the production of the benefit – i.e. what is the ecosystem doing?
5.13. For cultural services, the contribution of ecosystems is relatively passive in
that it is commonly the ecosystem providing opportunities for people to engage in
activities, learning experiences and the like. Costs may be incurred to facilitate people
benefiting from these services, such as the construction of cycling or hiking paths,
visitor facilities, etc. Often, cultural services are conceptualised in terms of the
benefits that people receive from the engagement and hence the challenge for
ecosystem accounting is to distinguish the contribution that represents the ecosystem
service among the various benefits.
5.14. Distinguishing final and intermediate ecosystem services: The distinction

between final and intermediate services reflects the principles of national accounting
where aggregate production is measured by netting out flows along the supply chain.
This ensures that double counting of outputs that become inputs to subsequent
production is removed. In the context of ecosystem accounting, this means that cases
where ecosystems provide services to a neighbouring ecosystem (e.g. via pollination,
water filtration or soil retention) these should be considered intermediate ecosystem
services, i.e. inputs to the supply of other ecosystem services.
5.15. While straightforward in theory, the complexity of ecosystems means that it
can be difficult to make this distinction in practice. Further, at an aggregate level a
focus on only final ecosystem services is appropriate but this may not be the case
when considering the contribution of individual ecosystems. This is especially so if
an ecosystem’s primary function is to support neighbouring ecosystems. A typical
11
Cultivated biological resources is a term from the SNA that supports the distinction between biological
resources (e.g. timber, fish, animals, etc.) the growth of which is considered the output of a process of
production – i.e. cultivated; and those whose growth is the result of natural processes – i.e. natural
biological resources.
49
example would be the services provided by upstream forests in regulating
downstream water abstraction.
5.16. In the SEEA EEA, the flows between ecosystem assets, if recorded, were
described as inter-ecosystem flows and in turn these flows were equated with
intermediate services. (UN et al., 2014b, 2.29) However, recording only the tangible,
physical flows between ecosystems (e.g. of water) does not serve to highlight the true
nature of the dependencies among ecosystems. Indeed, there are many ecosystem
services both final and intermediate for which there are no movements in physical
terms. For example, the regulation of water flows by a forest does not require that
there is a movement of materials between ecosystems (other than the water itself). Of
course, the actual measurement of the ecosystem service may rely on actual and
modelled physical flows.
5.17. For intermediate services, the accounting issue is not that flows of services
between ecosystem assets cannot be recorded in the system. Rather, the issue is that
there are potentially many different intermediate services that all require large
amounts of data to record them in the accounts.
5.18. There is a general sense that it is not advisable to attempt to measure all
flows and dependencies between ecosystems and, indeed, current ecological
knowledge would seem to suggest this was not practical in any event. Consequently,
it is an open question as to which intermediate ecosystem services should be
considered within scope of ecosystem accounting. Possible criteria to determine a
measurement boundary for accounting include those services considered critical for
ecosystem functioning and ecosystem services supply, and services affected by
ecosystem management.
5.19. The following observations are relevant to advancing this discussion and
supporting testing.
 One of the most important and common inter-ecosystem flows is water
and hence it is likely that some of the most important intermediate
ecosystem services are related to flows of water.
 A second area of likely importance is the provision of habitat services,
where the contribution of these services is embodied in the mature animal
that is an input to final ecosystem services, commonly in a separate
ecosystem. For example, the habitat services provided by mangroves as
nurseries for fish that are subsequently caught in open seas.
 One means by which the scope of intermediate services may be
“contained” is to ensure recording only of those intermediate services
from another ecosystem asset that are considered a direct input to a final
ecosystem service, i.e. they can be described in the context of an
ecological production function.
 It would be appropriate, for accounting purposes, to ignore the services
flowing within the bounds of an ecosystem asset, since these services will
be embodied within the final ecosystem services generated by the asset.
 Based on these last two observations, the recording of intermediate
services will be directly affected by the scale of analysis. With smaller
ecosystem assets (in terms of area) there will be an increased likelihood
of intermediate services being recorded.
 The recording of intermediate services would seem most useful for the
purposes of supplying management information. In aggregate, at national
level, it is likely that most intermediate services will offset each other,
since ultimately their value is embodied in final ecosystem services.
50
However, recognizing the relative importance of different ecosystems
within a country is likely to be very relevant for management purposes.
 Increasing the measurement scope to include only certain intermediate
services causes no specific issues in terms of accounting structure. The
changes needed would be to recognize only additional service types and
also to recognize flows between ecosystem assets, in addition to those
flows of final ecosystem services from ecosystems to economic and
human activity. In concept, these may be treated as imports and exports
of ecosystem services.
5.20. The treatment of other environmental goods and services: As noted in the
SEEA EEA Table 2.3, not all flows from the bio-physical environment to the
economy and society can be considered ecosystem services. There are a range of so-
called “abiotic” services reflecting flows we receive in the form of mineral and
energy resources: flows of energy such as solar, wind, wave and geo-thermal energy;
solar energy for photosynthesis; oxygen for combustion; air for respiration: and more
generally, the space for people to live.
5.21. Since the focus of the SEEA EEA is on accounting for ecosystems, these
various flows are not incorporated in the ecosystem accounting model. Many of these
flows are accounted for in specific accounts described in the SEEA Central
Framework (e.g. mineral and energy accounts, energy supply and use tables and land
use accounts). At the same time, the spatially explicit approach outlined in the SEEA
EEA may mean that it is relevant to consider incorporating measures of abiotic
services to consider the full range of benefits from a defined area. The extension of
the accounting tables to consider this aspect has not been developed at this stage.
5.22. The link between biodiversity and ecosystem services: On the whole, the
perspective taken for ecosystem accounting in the SEEA EEA is that biodiversity is a
feature most directly relevant in measurement of the condition of ecosystem assets.
This is distinct from an alternative conception that biodiversity is an ecosystem
service supplied by ecosystem assets. Thus, measures of biodiversity, whether of
ecosystem diversity or species diversity (the inclusion of genetic diversity measures
has not yet been examined), are considered to relate primarily to the stocks
component in the accounting model. This approach is consistent with a view that
biodiversity can be degraded or enhanced over time, an attribute that only applies to
stocks and not to flows (i.e. ecosystem services). The exact nature of the relationship
between biodiversity and ecosystem condition is a matter of some uncertainty.
Indeed, given that biodiversity is a many layered concept it would be possible to
consider that biodiversity might be the asset that is the focus of accounting rather than
ecosystem assets. This issue is discussed further in section 7.5.
5.23. At the same time, it is recognised that there are some aspects of biodiversity,
especially species diversity, that can supply final ecosystem services. This includes,
for example, the value of recreational services from wildlife related activities, where
people gain benefit from experiencing the diversity of nature. In addition, people may
appreciate, and therefore value, elements of biodiversity as such, for example when
they take an interest in the conservation of endemic and/or iconic species. In this
latter case, specific elements of biodiversity (e.g. related to the conservation of
species) could be considered representing a ‘final use’ of biodiversity. Given these
potential links to both ecosystem assets and ecosystem services, it is relevant to
recognise that measures related to biodiversity may be appropriate indicators in a
51
variety of accounts, including ecosystem condition accounts and ecosystem services
supply and use tables.
5.24. The treatment of ecosystem disservices: Ecosystem disservices arise in cases

where the interaction between ecosystems and humans is considered to be “bad”.
Usually this refers to the effects of things such as pests and diseases that emerge from
ecosystems and negatively affect economic production and human life. The SEEA
EEA recognises the frequent discussion on the measurement of ecosystem disservices
but does not propose a treatment in accounting terms.
5.25. Unfortunately, accounting principles do not work well when trying to make a
distinction between products that may be considered as either “goods” and “bads”.
Indeed, accounting, as distinct from economics, does not focus on the welfare effects
of use, and focuses instead on the activity associated with the generation of products
and the associated patterns of consumption. As a consequence, all flows between
producers and consumers have positive values in the accounts, irrespective of their
possible welfare effects. The positive values also reflect that, in accounting, values
must be able to be separated into price and quantity elements. It is difficult to
envisage either prices or quantities being negative and thus recording negative
estimates of value is not possible.
5.26. A related matter is the treatment in ecosystem accounting of negative
externalities, such as carbon emissions, where economic and human activity leads to
declines in the condition of ecosystems. Any associated environmental flows,
pollutants, emissions, etc. are not considered ecosystem disservices and their negative
impacts on welfare are not captured directly in the accounting system.
5.27. For both disservices and negative externalities, work is ongoing to outline the
appropriate treatment in the context of the ecosystem accounting model. It is also
noted that the SEEA Central Framework provides accounting approaches for
recording flows of emissions and other residuals to support measurement in these
areas.
5.3 The classification of ecosystem services

5.28. The classification of ecosystem services is an important aspect of
measurement since classifications can provide important guidance to ensure that an
appropriate breadth and depth of measurement is undertaken or, at least, that
individual measures are understood within a broader context.
5.29. The classification included in the SEEA EEA is the CICES, version 3
(Haines-Young and Potschin, 2013). It was considered an interim version.
Subsequent releases have been made with the latest being CICES version 4.3 and
consultation on the CICES is ongoing.
5.30. CICES has been adopted for work on the European Union’s MAES project
(Maes, et al., 2014) but alternative approaches to the classification of ecosystem
services have also been developed. Over time, it will be necessary to consider the
different merits and roles that might be played by the different classifications.
Perhaps the most important alternative approach is the work by the United States
Environmental Protection Agency (US EPA) on final ecosystem goods and services
(FEGS) (Landers and Nahlik, 2013) and the associated National Ecosystem Service
Classification System (NESCS) (Landers and Nahlik, 2012). This work places
52
attention on the links between ecosystem types and the classification of beneficiaries
from the final services supplied by those ecosystem types.
5.31. One of the most important roles of a classification of ecosystem services is
that it can be used to frame a discussion on the measurement and relative significance
of ecosystem services. In effect, a classification can operate as a checklist and be
applied in initial discussions by considering each EU type and noting those ecosystem
services that are considered most likely to be generated from that EU. The resultant
“baskets” of services for each EU type can aid in discussion of the role of accounting,
the structuring of information, the assessment of resources required for compilation
and generally communicating the message about the breadth of the relationship
between ecosystems and economic and human activity.
5.32. One finding from work on ecosystem services is that the choice of words
used to describe an ecosystem service can have a significant impact on how it is
visualized and understood by those involved. In particular, for regulating services the
choice of words to distinguish the benefit that people receive (e.g. reduced risk of
landslide) from the corresponding ecosystem service (e.g. soil retention) can be
material in the selection of measurement approaches and in valuation. Further
discussion across the full suite of ecosystem services, and the related benefits, is
required to ensure that the measures and the concepts are appropriately aligned.
5.33. There is common misunderstanding of the role of classifications with regard
to the distinction between final and intermediate ecosystem services. Put simply, it is
not the case that ecosystem services must be neatly classified between those that
contribute directly to economic and social beneficiaries and those that support the
ongoing functioning of ecosystems. For example, when water is abstracted from a
lake it would be considered final if the beneficiary was a household but intermediate
if consumed by a wild deer.
5.34. A similar situation arises in economic statistics. The classification of
products (e.g. following the international standard CPC) includes, appropriately, a
large number of products that may be considered either intermediate or final
depending on the beneficiary. For example, the purchase of bread is considered final
if purchased by a household but intermediate if purchased by a restaurant. However,
the CPC appropriately only contains one product, bread, rather than two (or more)
products.
5.35. Given this, for accounting purposes, the CICES and other classifications of
ecosystem services, must be used in conjunction with an understanding of the
beneficiaries that are within scope of the measurement concept. Without clearly
defining the beneficiaries, there is likely to be an overestimation of the quantity of
ecosystem services by adding together the intermediate ecosystem services reflecting
the operation of an ecosystem, and the “final” ecosystem services that are
contributions to economic and social beneficiaries.
5.36. These considerations on the role of classifications are important in
developing agreed accounting structures both in the case of ecosystem services alone
and in the context of integrating measures of ecosystem services within standard
accounting structures such as input-output and supply and use tables.
5.4 The role and use of biophysical modelling

5.4.1 Introduction
53
5.37. Biophysical modelling, in the context of this guidance document, is defined
as the modelling of biological and/or physical processes in order to understand the
biophysical elements to be recorded in an ecosystem account. These elements are part
of either ecosystem asset measurement (including ecosystem condition and the
ecosystem’s capacity to generate services) or ecosystem services measurement. In
this chapter, the focus is on ecosystem services.
5.38. The intention here is to provide some general guidance on the types of
biophysical modelling approaches that can be used to analyse ecosystem service
flows, as distinct from models that can be used to understand ecosystem processes
(e.g. nutrient cycling, energy flows). In the scientific literature, a wide range of
different modelling approaches has been described in the fields of ecology,
geography, and hydrology. Many of them are potentially relevant to ecosystem
accounting depending upon the environmental characteristics, the uses of the
ecosystem, the scale of the analysis, and the available data. It is impossible to
describe all these different modelling approaches in one document. This chapter
provides an overview of the different approaches, and their main uses for the
biophysical modelling of ecosystem services.
5.39. An important aspect of applying biophysical models in ecosystem accounting
is recognising the nature of the connections between ecosystem service flows and the
condition of the relevant ecosystem asset. This connection is reflected in the concept
of ecosystem capacity. Although the definition of ecosystem capacity remains a
matter of ongoing discussion (see section 6.4), it is accepted broadly, that modelling
ecosystem service flows must take into consideration the current and expected
condition of the ecosystem and its various functions and processes.
5.4.2 Overview of biophysical modelling approaches

5.40. The two most relevant forms of modelling are spatial and temporal modelling
techniques. Spatial modelling is required to produce maps of ecosystem services for a
complete EU or geographical aggregation, including at national level. Thus, where
data is lacking in relation to some spatial areas, spatial modelling can fill the gaps.
Spatial modelling is most commonly undertaken using GIS packages such as Arc GIS
and Quantum GIS. There are also several ecosystem services specific modelling tools
such as ARIES, MIMES and InVEST.
5.41. Within the general GIS packages, spatial modelling tools include the use of
look-up tables, and the application of statistically based approaches such as Maxent
(Philips, et al., 2006). There is also a range of geostatistical interpolation techniques,
such as “kriging”, that use statistical algorithms to predict the value of un-sampled
pixels on the basis of values for nearby pixels in combination with other
characteristics. The basic interpolation methods use simple interpolation algorithms,
for instance nearest-neighbour interpolation, but there are more sophisticated
geostatistical tools that also consider sets of correlated variables. For instance, timber
productivity may be related to productivity in nearby pixels, but in a more
comprehensive approach, it may also be related to factors such as soil fertility or
water availability for which spatial maps are available. Critical in applying
geostatistics is that a sufficiently large sample size is available, and that samples are
representative of the overall spatial variability found. Examples of the applications of
these approaches and related technical references can be found in, for example,
Sumarga and Hein (2013), Remme, et al. (2014) and Schröter, et al. (2014).
5.42. In ecosystem accounting, temporal modelling is required to estimate the
capacity of an ecosystem to generate ecosystem services. The modelling approach
most consistent with coming to an understanding of flows of ecosystem services is a
54
dynamic systems approach, which can also be applied in combination with spatial
models. A dynamic systems approach is based upon modelling a set of state (level)
and flow (rate) variables in order to capture the state of the ecosystem over time,
including relevant inputs, throughputs and outputs. Dynamic systems models use a
set of equations linking ecosystem state, management and flows of services. For
instance, a model may include the amount of standing biomass (state), the harvest of
wood (flow), and the price of wood (time dependent variable).
5.43. The systems approach can contain non-linear dynamic processes, feedback
mechanisms and control strategies, and can therefore deal with complex ecosystem
dynamics. However, it is often a challenge to understand these complex dynamics,
and their spatial variability, and data shortages may be a concern in the context of
ecosystem accounting that requires large scale analysis of ecosystem dynamics and
forecasted flows of ecosystem services.
5.44. In some cases, spatial and temporal modelling approaches need to be
combined. For instance, process based models are generally required to model
regulating services such as erosion control, or ground and surface water flows.
Erosion, and erosion control, is often modelled with the USLE (Universal Soil Loss
Equation), although its reliability outside of the USA (where it was developed) has
proven to be variable. Other examples of process based models are the hydrological
models such as SWAT (Soil and Water Assessment Tool) and SedNet from the
Australian Commonwealth Scientific and Industrial Research Organisation (CSIRO).
These models are both temporally and spatially explicit, and use a dynamic systems
modelling approach integrated in a GIS environment.
5.5 Data sources, materials and methods for measuring ecosystem service flows
5.5.1 Introduction
5.45. SEEA EEA Annex A3 provides some stylised figures to help articulate the
measurement required to estimate flows of ecosystem services. The figures included
in that annex only relate to selected services but the basic logic of the models can be
applied more generally. Of particular importance, is recognising the distinction
between the ecosystem service and the associated benefit.
5.46. It will generally be helpful for measurement purposes to distinguish clearly
between provisioning, regulating and cultural services. For this task the use of a
classification of ecosystem services, such as CICES, can serve as a useful checklist.
Further, it is likely to be useful to consider the measurement of ecosystem services in
relation to broad ecosystem types such as forests, wetlands, and agricultural areas.
(Note that the use of broad ecosystem types may also be useful in accounting for
change over time and reducing uncertainty in spatial measurement.)
5.47. A useful structuring of indicators is presented in Chapter 5 of the European
Union’s MAES project report (Maes et al., 2014). In this chapter, indicators for
different ecosystem services are mapped out within four broad ecosystem types –
forest, cropland and grassland, freshwater and marine. A review of this material
highlights the likely broad range of data sources that will need to be considered in
generating a full coverage of ecosystem services.
5.5.2 Data sources

5.48. Data sources will be different in each country. It is suggested that important
national data holders be engaged in the process of compiling the accounts. They will
55
be able to advise on data availability and quality. The following list suggests some
government departments (using generic titles) and the data they may hold:
 National Statistical Offices: Agricultural production (crops and livestock);
health statistics (incidence of environmentally-related diseases); Population
data
 Departments of Natural Resources: Timber stock and harvest; biomass harvest
for energy; water supply and consumption; natural disaster statistics (floods,
landslides, storms); land cover (to estimate carbon stock and sequestration);
remote sensing (to estimate primary production);
 Water management and related agencies: Water stocks and flows, abstraction
rates, hydrological modeling
 Departments of Agriculture: Erosion potential, biomass harvest;
 Departments of Forestry: Forest stock and harvest; carbon sequestration;
 Departments of Environment and Parks: Iconic species habitats, visitors to
natural areas
5.49. Local academic and government researchers may have conducted studies for
specific regions of the country or for specific services. As well, international
organizations (e.g. UNEP, SCBD, World Bank) may have conducted studies on
specific locations or services. These should also be reviewed and considered for
integration into the ecosystem services supply and use table.
5.50. Databases storing research on ecosystem valuation will also include
information on the physical aspects of the ecosystems they value. For example, if the
fish harvest in one lake is estimated to be 500 tonnes per year, the research is likely
looking to estimate the economic value of those fish. For the purposes of the
ecosystem services supply and use table, it is useful to know that 500 tonnes per year
was harvested from a specific lake.
5.51. Two broad based ecosystem valuation databases that should be investigated
for country-specific (or region-specific) data are the EVRI (Environmental Valuation
Reference Inventory, at www.evri.ca) and the ESVD (Ecosystem Service Valuation
Database at http://www.fsd.nl/esp/80763/5/0/50) that emerged from the TEEB 2010
study. Other service-specific or region-specific databases or projects should also be
investigated.
5.52. Depending on the resources, including time, available, it may be feasible to
collect new data. New data collection activities could include:
 Ecological field studies to determine location-specific production functions
for ecosystem services. Such studies could, for example, collect data on water
purification services of wetlands.
 Socio-economic surveys could be conducted on a national scale to better
understand how people and businesses use ecosystem services (e.g., water
withdrawals, visits to recreational sites)
 Case studies could be conducted on target populations (e.g., households near
forest areas) to better understand their use of ecosystem services (e.g.,
biomass for fuel, food gathering, sources of water).
5.53. While there is an increasing amount of information and examples of
measurement of ecosystem service flows, a challenge is likely to lie in adapting and
scaling the available information for ecosystem accounting purposes. The issue of
scaling is considered below (section 5.6). From a practical perspective, it is sufficient
56
to note here that, when accounting for multiple ecosystem services, the aim must be
to measure the supply of ecosystem services at a broad landscape scale (ideally up to
national level) and also over a series of accounting periods. As appropriate,
adjustments to ensure that measures of different ecosystem services relate to common
spatial areas and the same time periods should be made.
5.5.3 Measuring the supply of ecosystem services

5.54. The measurement of provisioning services can generally be linked to
measures commonly available in statistical systems. Data on the production of crops,
livestock, other agricultural products, forestry products and fisheries products are all
of direct relevance in the estimation of provisioning services. It is relevant to recall
however, that the crops and agricultural products themselves are considered benefits
in the SEEA EEA model, and the ecosystem services are the contributions to the
production of those benefits.
5.55. In some cases, data may be available at a fine level of spatial detail, for
example from an agricultural census. In other cases it may be necessary, if sub-
national accounts are to be compiled, to allocate national or regional level estimates
to the spatial areas being used for ecosystem accounting using spatial modelling
techniques.
5.56. For some cultural services, particularly those relating to tourism and
recreation, the use of available administrative and survey based information is also
appropriate. The measurement of non-use cultural services is more problematic and is
considered further in section 5.6.
5.57. For regulating services, some specific suggestions for measurement using
bio-physical models are suggested in Table 5.1). These suggestions are intended as
guide only.
Table 5.1: Ecosystem services metrics and mapping methods for selected ecosystem
services
Ecosystem Potential metric Description
Service
Carbon storage Ton of carbon (or Carbon storage includes storage in vegetation (above ground, root, dead wood,
carbon-dioxide) per and litter carbon) and soil carbon. Soil carbon may be low compared to
hectare or square vegetation carbon, as in some types of low fertility tropical forest soils, or it
kilometre. may be by far the largest component of total carbon storage, as in peatland
soils in deep peat (World Bank, 2014). Above ground carbon can be measured
with radar remote sensing, but the measurement of below-ground carbon with
remote sensing techniques is generally not possible. Instead, for this part of the
carbon stock, soil sampling and interpolation of data points is required.
Carbon maps are increasingly available for different parts of the world and
the capacity to map above ground carbon stock globally is increasing with the
launch of the Sentinel radar satellites.
Carbon Ton of carbon (or Carbon sequestration can be related to net ecosystem productivity (NEP), i.e.
sequestration carbon-dioxide) the difference between net primary productivity (NPP) and soil respiration.
sequestered per year, NPP can be derived from the Normalized Difference Vegetation Index
per hectare or per square (NDVI) that can be measured using remote sensing images. However, care
kilometre. needs to be taken that the relationship between NDVI and NPP is well
established for the ecosystems involved, and that accuracy levels are
calculated based on sample points. It is often difficult to find credible values
for the spatially very variable soil respiration rate, which depends on bacterial
and fungi activity which are in turn guided by the local availability of organic
matter (e.g. fallen leaves), temperature, moisture, etc.
It will be relevant to understand estimation approaches used in the compilation
of IPCC based greenhouse gas inventory estimates for the LULUCF.
Maintaining mm water Rainfall patterns depend on vegetation patterns at large scales. For instance, it
rainfall patterns evapotranspiration per has been estimated that maintaining rainfall patterns in the Amazon at current
hectare per year, mm levels requires maintaining at least some 30% of the forest cover in the basin.
57
rainfall generated per Reductions in rainfall in the Western Sahel and the Murray Darling Basin in
hectare per year. Australia have also been correlated to past losses of forest cover. This is a
significant ecosystem service, however the value of individual pixels is
difficult to establish since it requires understanding large scale, complex
climatological patterns, large scale analyses of potential damage costs, and
interpolations of values generated at large scales to individual pixels with
detailed climate-biosphere models.
Water - water storage capacity Water regulation includes several different aspects, including (i) flood control;
regulation in the ecosystem in m3 (ii) maintaining dry season flows; and (iii) water quality control – e.g. by
per hectare (or in mm); trapping sediments and reducing siltation rates). Temporal, i.e. inter-annual
- difference between and intra-annual, variation is particularly important for this service. Modelling
rainfall and evapo- this service is often data-intensive and also analytically complex.
transpiration in
m3/ha/year;
Flood Surface water modelling Flood protection depends on linear elements in the landscape that act as a
protection can be deployed to buffer against high water levels (e.g. a mangrove, dune or riparian system).
analyze reductions in Modelling this service requires modelling flood patterns and the influence of
flood risk, expressed the vegetation. It is also necessary to define the benchmark against which the
either as reduction in reduction in risk can be assessed.
probability of It may not always be necessary to model flood protection in physical terms in
occurrence, reduction in order to understand the monetary value of the service - in particular in those
average duration of the areas where it is certain that natural systems, if lost, would be replaced by
flood, or reduction in artificial ones (e.g. a dyke), as would be the case in most of the Netherlands,
water level depending for instance. In this case, valuation may be done on the basis of a replacement
on context cost approach that does not require understanding the physical service in full.
Erosion and - difference between There is relatively much experience with modelling this service. Erosion
sedimentation sediment run-off and models can be integrated in a catchment hydrological models (such as SWAT
control sediment deposition in or CSIRO SedNet, both freeware) to predict sediment rates. In SWAT, a
ton/ha/year in the watershed is divided into Hydrological Response Units (HRUs), representing
current ecosystem state homogeneous land use, management, and soil characteristics. Erosion rates
compared to a situation need to be estimated for each HRU, for instance on the basis of the MUSLE or
with no plant cover RUSLE erosion models or alternatively SWAT landscape can be used which
includes grid based land cover units.
Water Amount of excess Various hydrological models, including SWAT include modules that allow
purification nitrogen and or estimating the nutrient loads in rivers as a function of streamflow, discharge,
phosphorous removed in temperature, etc. Nitrogen is broken down by bacterial activity, phosphorous
the ecosystem is typically removed in ecosystems by binding to the soil particles. Modelling
these processes requires large datasets, preferably with daily time-steps, of
nutrient concentrations in various sampling stations along the river course.
5.5.4 Recording the beneficiaries of ecosystem services

5.58. Within the ecosystem accounting model all benefits must have a
corresponding beneficiary. Given that ecosystem services are “contributions to
benefits” this implies that all ecosystem services also have a corresponding
beneficiary. Using broad national accounting categories, these beneficiaries can be
grouped as being corporations, governments and households, including beneficiaries
that are both resident in a country and those in the rest of the world.
5.59. Other groupings of economic units that might be considered include
 Industry groupings whereby individual establishments or businesses are
grouped into those that undertake similar activities such as agriculture or
manufacturing.
 Allocation of use of ecosystem services by household income levels
 Distinguishing between rural and urban use of ecosystem services
 Distinguishing between those services that are used locally, nationally or
globally.
58
5.60. To the extent that these alternative groupings of users can be identified
during the data collection stages, there is the potential to develop information sets
relevant to a broader range of policy questions.
5.61. When measuring the supply of ecosystem services and mapping out the
supply across ecosystem types (e.g. forests), it is likely to be useful to consider the
link to beneficiaries. This approach has been extensively applied in the development
of FEGS and NESCS by the US EPA (Landers and Nahlik, 2012 and 2013).
5.62. To support integration with the national accounts and its tables such as input-
output tables, it is recommended that the matching of ecosystem services to
beneficiaries use the classification of beneficiaries used by the national accounts,
either by institutional sector or by industry/economic activity.
5.6 Recommendations
5.63. Since the release of the Millennium Ecosystem Assessment (MA) in 2005
(MA, 2005) there has been an explosion in the number of studies on ecosystem
services. The studies have focused on many aspects of definition and measurement
and have involved researchers from a range of disciplines. Subsequent work in the
context of the TEEB Study (2010), the MAES initiative (Maes, et al., 2014) and the
Inter-governmental Platform on Biodiversity and Ecosystem Services (IPBES) have
reinforced the potential of the ecosystem services approach in understanding the
relationship between humans and the environment. Ecosystem accounting has built
upon this work and research.
5.64. From an official statistics perspective, the number of alternative estimates
and the variety of methods and definitions presents a conundrum. The challenge from
a national-level, statistical and accounting perspective is how to draw the knowledge
together to form a comprehensive and coherent data set on trends in ecosystem
service flows. The descriptions in the SEEA EEA should be seen as a first attempt to
support this type of data coordination.
5.65. For countries seeking to undertake pilot studies in ecosystem accounting, the
most appropriate initial advice is that there is a large body of work on which
estimates of ecosystem services flows can be based. At the same time, it is unlikely
that estimates of ecosystem services for specific ecosystems in each country will have
been developed in a relatively standardised way. Consequently, it is the role of the
ecosystem accountant to bring together the available expertise and research.
Advancing the measurement of ecosystem services is a matter for testing rather than
research.
5.66. It is recommended that the ecosystem accounting model be used to build an
understanding of the gaps in information, either because certain ecosystem services
have not been measured or because ecosystem services from certain ecosystem types
have not been measured. The accounting model can play an important role in
identifying data gaps.
5.67. This may be done by determining a list of priority ecosystem services based
on existing national land and water management, and nature conservation practices.
For this task:
 The US EPA FEGS-CS and associated NESCS can be applied as analytical tools
as they contain a broad set of ‘origin points’ of services, linked to types of
ecosystems and beneficiaries
59
 CICES can be used as check-list, and as a framework to facilitate international
comparability.
5.68. Once a set of priority services has been determined it will be relevant to
quantify and map the ecosystem services in terms of both ecosystem services supply
(from ecosystem units) and use (by beneficiaries, including businesses, households
and governments).
5.69. Data relating to flows of provisioning services may be available from
national agriculture, forestry, fishery and water agencies. Ideally, these data would be
sourced from accounting frameworks such as the SNA or SEEA Central Framework.
Data for some regulating services may be obtained from thematic accounts, such as
for carbon sequestration.
5.70. If no data on observed ecosystem flows is available, new data
collection/generation (including inventories, remote sensing, spatial modelling and
other sources) may be required. Data collection should be developed in a way that
provides consistent estimates across the different service types (e.g. of similar detail,
quality, error and uncertainties), as well as correctly embedded in the ecological and
land use processes. Alternative modelling tools include:
 For soil and water-related process and service modelling tools consider USLE
and associated models, and SWAT, SedNet, HEC and others
 For carbon-related process and service modelling tools consider CASA (Potter, et
al, 2012)
 For biodiversity and other process and service modelling tools consider
MAXENT
5.71. Depending on the nature of the data gaps, the use of benefit transfer functions
can be considered and cautiously tested. Generally, it will be important to develop an
understanding of uncertainties when defining, classifying, quantifying and mapping
ecosystem services and to prepare validation/quality control data and protocols.
5.72. Notwithstanding the priority for testing to be the focus of current activity,
there are some areas of research that would support this testing work. Primary among
these is resolution of issues concerning the definition and classification of ecosystem
services. This work has advanced well and the relevant boundary issues are quite well
delineated. However, further consultation leading to decisions or treatments is needed
to put in place a classification of ecosystem services that is, at least, appropriate for
ecosystem accounting purposes.
5.73. The second key area of research is articulation of the role of intermediate and
supporting services in ecosystem accounting. A related task is specifying what are
best called ecological production functions or value chains – i.e. the sequence of
ecosystem processes and characteristics, possibly across ecosystem types, that leads
to the supply of a final ecosystem service. Although it is not anticipated that a
complete catalogue of all such production functions would be established in the short
to medium term, research in this direction would be of direct benefit to applying
ecosystem services and ecosystem accounting measures to policy questions.
60
Chapter 6: Accounting for ecosystem assets

Three main dimensions are considered in the measurement of ecosystem assets: ecosystem
extent, ecosystem condition and expected ecosystem service flows.
Ecosystem extent can be measured in straightforward fashion in terms of the area of different
ecosystem units.
Ecosystem condition is more challenging to measure requiring the selection of relevant
characteristics and then selection of appropriate indicators and metrics of each characteristic.
For some characteristics in certain ecosystem types, condition metrics are well established.
A key challenge for ecosystem accounting is developing a full coverage of measures in a
manner that support aggregation and comparison.
Reference condition approaches are one technique for developing measures that can be
monitored over time and can be compared across ecosystem types and across countries.
Determining reference conditions for multiple ecosystem types and more than one country is
not straightforward.
Testing is required, especially regarding ecosystem condition measurement.
Ecosystem capacity is emerging as an important concept in ecosystem asset measurement.
Ecosystem capacity provides a link between ecosystem services and ecosystem assets and
supports the assessment of sustainability in ecosystem use and trade-offs between different
baskets of ecosystem services.
6.1 Introduction
6.1. Accounting for ecosystem assets is a fundamental component of ecosystem
accounting. Without accounting for ecosystem assets, ambitions to understand and
monitor the changes in the natural capital base and hence consider issues of
sustainability are not achievable. Further, understanding the connections between the
characteristics of ecosystem assets and the services that are supplied, can form the
basis for better planning and management of natural capital.
6.2. This chapter builds on the initial discussion of accounting for ecosystem
assets in Chapter 4 of the SEEA EEA. When that chapter was drafted, there were
many concepts and ideas about how ecosystem assets might be described and
measured, and, in many respects, the text of the SEEA EEA represents a first attempt
at synthesising approaches to environmental and ecosystem assessment within a
national accounting framework.
6.3. This chapter assumes that ecosystem assets are represented by spatial areas
called EU, as outlined in Chapter 3. EU represent ecosystem assets while
geographical aggregations are combinations of EU for the purpose of reporting and
analysis. Note that a geographical aggregation could be an aggregation of specific EU
types (e.g. deciduous forests), or a geographical aggregation could be all EUs
aggregated within a specific ecological boundary (e.g. a river basin) or an
administrative boundary (e.g. a country or local government area).
61
6.4. It may be intuitively appealing to consider that certain geographical
aggregations are assets in their own right, for example all forests within a country.
However, from a stricter accounting and measurement perspective it will be more
consistent and useful to consider only EU as ecosystem assets. This approach is more
strongly aligned with the definition of ecosystem assets in the SEEA EEA as being
“spatial areas containing a combination of biotic and abiotic components and other
characteristics that function together” (SEEA EEA 4.1).12
6.5. The treatment of EU as being ecosystem assets does not imply that individual
EU themselves are the appropriate level of reporting for ecosystem accounting
purposes. In national economic accounting, reporting takes place at the industry (e.g.
manufacturing) or institutional sector level (e.g. households) rather than at the level
of individual economic units. Similarly, it is likely that accounts will be structured to
present different types of EU within a single geographical aggregation, e.g. a country.
A key issue for further research is the level of detail at which accounts can provide
accurate information, as a function of landscape properties and the quality of input
data.
6.2 Dimensions in the measurement of ecosystem assets

6.6. SEEA EEA Chapter 4 outlines a number of dimensions that are relevant to
the measurement of ecosystem assets. The three primary dimensions are ecosystem
extent, ecosystem condition, and expected ecosystem services flows. A dimension or
concept that has become increasingly of interest from an accounting perspective is
that of ecosystem capacity. SEEA EEA notes that “the capacity of an ecosystem asset
to generate a basket of ecosystem services can be understood as a function of the
condition and the extent of that ecosystem” (SEEA EEA, 4.1). While SEEA EEA
does not provide a measurement definition for ecosystem capacity, there is
recognition that it provides a linking point in the measurement of ecosystem assets
and ecosystem services.
6.7. The measurement of each of the dimensions of ecosystem assets must take
into account the appropriate scale of analysis. This is likely to be considerably
affected by the number of different types of EU that are used. Where accounting is
undertaken using the 15 classes of land cover to represent the types of EU, this will
provide a broad and relatively coarse sense of changes in ecosystem assets. For a
range of analytical purposes, it is likely to be necessary to incorporate additional EU
types and hence provide a richer understanding of the structure and changes in the
structure of ecosystem assets within a country.
6.8. This section introduces the different dimensions of ecosystem assets noted
above, with an extended discussion on the measurement of ecosystem condition in
section 6.3 and a discussion on ecosystem capacity in section 6.4.
6.9. The most straightforward measurement dimension is ecosystem extent. The
preparation of ecosystem extent accounts, introduced in Chapter 4, is the appropriate
starting point for ecosystem accounting since they will reflect fundamental choices on
the delineation of spatial areas and also provide important information on the
changing composition of ecosystem asset types at an aggregate level. For example,
these accounts will support the derivation of indicators of ecosystem diversity.
12
Some studies consider that biodiversity, habitats, individual species or individual ecosystem services are
“assets”. Since the objective in ecosystem accounting is to consider the ecosystem as an integrated,
functioning unit, the focus is on EU as assets rather than specific characteristics.
62
6.10. It is this second feature that is perhaps the most significant in accounting
terms. Because accounts about ecosystem extent are compiled in a common unit of
measurement, often hectares, this permits aggregation and comparison at larger
scales. Thus comparisons can be made between the relative proportions of different
ecosystem asset types and the changes in these proportions over time. It is not as
straightforward to undertake this type of comparison for the condition of ecosystem
assets.
6.11. The second dimension is ecosystem condition. “Ecosystem condition reflects
the overall quality of an ecosystem asset in terms of its characteristics” (SEEA EEA
2.35). The measurement of ecosystem condition, discussed at more length in section
6.3, requires the selection of specific characteristics and then measurement of
relevant indicators pertaining to those characteristics.
6.12. Once indicators are measured, the task from an accounting perspective is to
develop methods that support comparison and aggregation. Being able to understand
the relative significance of different ecosystem assets is core to the accounting
approach. The general approach to this task outlined in the SEEA EEA is the
comparison of indicators to benchmark or reference condition. Guidance on this step
is provided in section 6.3.
6.13. The third dimension concerns expected ecosystem services flows. The
concept of expected ecosystem services flows is an application of standard capital
accounting to the area of ecosystems. It stands somewhat distinct from the experience
to date in measuring ecosystems either in terms of their extent and condition, or in
terms of the actual flows of ecosystem services in a given period.
6.14. Since the measurement of expected flows is forward looking and relates to a
basket of ecosystem services, it relies on an understanding of the link between the
future condition of ecosystem assets and a basket of services, and also on measuring
an entire basket of services for different ecosystem types.
6.15. The concept relates to the actual flows of ecosystem services that are
considered most likely to occur in future accounting periods. It does not reflect the
sustainable yield of ecosystem services, unless a sustainable yield is expected. Rather,
the actual (or observed) flow refers to the quantity of ecosystem services (measured
in terms of tonnes, m3, number of visitors, etc.) that flow from an ecosystem asset to
a beneficiary during an accounting period. The actual flow is different from estimates
of flows of ecosystem services that are based on consideration of alternative scenarios
(e.g. if prices for resources were higher or population distributions were different) or
under different assumptions (e.g. if the ecosystem asset is used sustainably).
6.16. Given this definition, the concept of expected ecosystem service flows is
applied by estimating what the flows of actual ecosystem services are likely to be in
future accounting periods. In terms of the asset as a whole, some mixture or basket of
ecosystem services must be assumed for estimation to take place.
6.17. Ultimately, from an ecosystem accounting perspective, the main ambition is
for measures of ecosystem extent, condition, capacity and expected service flows to
be able to be reconciled to provide a consistent picture of each ecosystem asset both
in its own right and in comparison with other ecosystem assets.
6.18. One perspective on ecosystem asset measurement not mentioned above
concerns measurement in monetary terms through the valuation of ecosystem
services. In concept, measurement in monetary terms permits aggregation and
comparison among ecosystem assets, as well as supporting the integration of
information on ecosystem assets with data on other assets currently included in the
national accounts balance sheets, e.g. produced assets. The measurement of
ecosystem extent and ecosystem condition in monetary terms is not possible (at least
63
not directly) and hence the focus for the valuation of ecosystem assets is on the
valuation of expected ecosystem service flows.
6.19. In the SEEA EEA, the valuation of ecosystem assets is undertaken by
deriving the net present value of all future expected ecosystem service flows from an
asset. There is a range of conceptual and practical challenges in valuation including
(a) developing future scenarios of condition; (b) modelling the capacity to supply
services under those conditions; (c) determining the expected basket of ecosystem
services given that capacity; and (d) selecting appropriate discount rates.
6.20. Progress toward the full valuation of ecosystem assets is a medium to longer
term objective. A more complete discussion of the relevant valuation issues is
presented in Chapters 8 and 9.
6.3 Compiling measures of ecosystem condition

6.3.1 Introduction
6.21. The intent of an ecosystem condition account is to bring together a range of
information about the condition of different ecosystem assets. In general, most
environmental accounting, and indeed most measurement activities, tend to focus on
specific characteristics in individual or multiple ecosystem assets. Thus, there may be
monitoring on, for example, carbon, water, timber, soil, species diversity, landscape
characteristics (such as fragmentation or relative attractiveness), ecosystem resilience
and integrity.
6.22. In the SEEA EEA, the general ambition in accounting for ecosystem
condition is to bring together the relevant pieces of information to provide an overall
assessment of the various ecosystem characteristics. The general tone of the SEEA
EEA was that this was a two-step process whereby individual indicators for different
characteristics were defined and then weighted together. Discussion since the release
of SEEA EEA has highlighted the following points that provide a broader context for
the measurement of ecosystem condition.
6.23. First, it is relevant to recognise different motivations for the measurement of
ecosystem condition which each requires the use of a different measurement
approach. There are two main motivations:
a. That ecosystem condition indicators can be used to indicate the general
condition, state or health of an ecosystem and the relevant trends in
condition. These indicators may reflect policy priorities (e.g.
preservation of native habitat); ecosystem functioning (e.g. deposition
levels of acidifying compounds versus critical loads for such
compounds); or the capacity of ecosystems to generate one or more
services (e.g. attractiveness of the landscape for tourism).
b. That spatial information concerning ecosystem condition is needed for
the measurement and modelling of ecosystem services supply. For
example, to analyse erosion risks and the ecosystem service of erosion
control, it is necessary to have information on, among other things,
slope, slope length and soil type. By themselves, indicators of these
characteristics are not necessarily policy relevant, and so they may not
change in the time frames relevant for ecosystem accounting.
6.24. For the compilation of the ecosystem condition account, the focus in the
Technical Recommendations is on condition indicators that are relevant for
conveying information on the state and trends in ecosystems. These may include
64
condition indicators that reflect the capacity of an ecosystem to supply ecosystem
services, thus enabling a connection to be made between ecosystem condition and
ecosystem service flows.
6.25. Second, although it is clear that assessment of the condition of an ecosystem
asset must involve consideration of a set of ecosystem characteristics, there is no
definitive set of characteristics that can be determined a priori, even for a given
ecosystem type. In part, this reflects that the assessment of condition should take into
consideration, or implicitly place emphasis on, the likely uses of an ecosystem asset.
Since there is a range of possible uses of an ecosystem, there will also be a number of
combinations of ecosystem condition indicators. Determining the appropriate set of
characteristics is a particularly important task for testing in ecosystem accounting.
6.26. In general, the types of indicators being considered in the measurement of
ecosystem condition focus on measuring ecosystem process and function. However,
there are two other groups of indicators that may be relevant in the measurement of
ecosystem condition, or at least will be relevant in ecosystem measurement more
generally.
a. The first group concerns the recording of relatively fixed characteristics
of ecosystem assets such as measures of soil type, slope, altitude, climate
and rainfall. Such measurements will provide important context for
measures of condition and also may be important inputs in the modelling
of ecosystem services.
b. The second group are more difficult to label but pertain to measures of
impacts or pressures on the environmental state, for example, measures
of pollution, emissions or waste. They were labelled “enabling factors” in
SEEA EEA in the sense that, without the presence of these flows, there
would be no associated ecosystem service being delivered. Accounting
for these flows is described in the SEEA Central Framework although
more spatial detail will be required for ecosystem accounting purposes.
While primarily needed for measuring regulating services, they will also
be relevant in the assessment of ecosystem condition. For example, in
understanding the drivers for change in condition over time. Finally, this
group of indicators will also be of particular interest from a policy
monitoring perspective.
6.27. Third, in some cases, it maybe useful to compile composite indicators where
a range of indicators is combined to reflect different elements of ecosystem condition.
Having determined a suitable set of indicators, there is no natural, apriori weighting
of the indicators that might be used to estimate the overall condition of an ecosystem
asset. Rather, a measure of the overall condition requires a view to be taken on the
relative importance of the different ecological processes involved, or the different
purposes for use of an ecosystem asset. It would be possible to give each indicator
equal weight, or perhaps to determine weights via surveys of ecosystem users, but
this does not overcome the underlying issue.
6.28. It will be useful to compare condition indicators to benchmark or reference
conditions. This approach is discussed at more length later in this chapter. Condition
indicators may also be attributed to in specific classes on the basis of an assessment
against standard criteria. This is sometimes done with classes reflecting different
levels of soil fertility or land suitability for a specific purpose.
65
6.3.2 Different approaches to the measurement of ecosystem condition
6.29. There are three broad approaches that can be tested. The first is a “top-down”
approach where indicators for a small set of generally applicable ecosystem
characteristics across a country, are combined to form an overall condition measure.
The combination uses assumptions on the relative importance of each characteristic
and correlations among them.
6.30. This is the approach adopted for the ENCA QSP (Weber, 2014) where
indicators of carbon, water, biodiversity and ecosystem potential are measured for all
ecosystem types in a country and then combined to form a single index, the ECU,
using carbon as the primary weighting variable.
6.31. The second approach is a “bottom-up” approach in which different
characteristics are determined for different ecosystem types and perhaps, also for
different uses of ecosystem types. This is the approach that has been used by SANBI
in South Africa (Driver, 2012), MEGS in Canada (Statistics Canada, 2013), in the
Norwegian Nature Index (Alasken, et al., 2012) and the Wentworth Group in
Australia (Cosier and McDonald, 2010)13. In theory, it may be possible to combine
the various indicators of the different characteristics to provide aggregate measures of
condition but this step is not generally taken. Perhaps the closest to undertaking this
step is the work of the Wentworth Group through their development of the “econd” as
a reference condition based indicator and in the aggregate indexes of the Norwegian
Nature Index.
6.32. The third approach is a variation on the bottom-up approach and involves
selecting the condition indicators through a direct link to the basket of ecosystem
services for a given ecosystem asset. This is the approach used in the UK NEA
(2011) and relates to the concept of measuring ecosystem capacity (see section 6.4)
where the ability for an ecosystem asset to continue to produce a given ecosystem
service is a function of the ecosystem condition. An extension of this approach is that
the relative importance of different condition measures could be determined on the
basis of relative value shares of ecosystem services (i.e. those services with a higher
share of total supply of ecosystem services in monetary terms received higher
weight).
6.33. All three approaches are plausible ways forward for the measurement of
ecosystem condition. Testing is required to understand whether there is a significant
difference in the results from the use of different approaches, and which approaches
might be most appropriate for ecosystem accounting purposes. This issue is discussed
further in section 6.5.
6.34. To support the measurement of condition, it is likely to be useful to prepare
data relating to drivers of change in ecosystem condition. Information on population
growth, climate change related variables, economic growth and similar indicators will
help to provide a broad context for measuring and interpreting ecosystem condition.
6.3.3 Developing indicators of individual ecosystem characteristics

6.35. The SEEA EEA points to a number of different characteristics and indicators
(see for example Table 2.3).
6.36. The development of indicators to assess condition for particular purposes for
particular ecosystem types, is a relatively well developed area of research. For a
given characteristic, often the research enables the relative importance of the different
13
Note that the ENCA QSP approach also supports the use of additional indicators beyond an initial
standard set.
66
factors to be weighted to provide an appropriate composite index. The issue is not
whether indicators of specific characteristics of ecosystem condition can be
measured, but rather which characteristics are relevant and how the indicators might
be combined.
6.37. Ideally, information on each selected characteristic would be measured or
downscaled to the BSU level. In many cases this may be possible and indeed, for
some ecosystem characteristics, such as those pertaining to soil retention and water
flows, there may be significant spatial variability that must be considered.
6.38. However, there will be some situations in which this may make little
conceptual sense or imply assumptions in downscaling that are not appropriate. For
example, in measuring ecosystem condition for the purpose of providing habitats,
fragmentation and connectivity are key factors. These factors are measureable only at
a multiple ecosystem asset or landscape level. Attribution of information on
fragmentation and connectivity to the BSU level is possible, but the information
remains meaningful only at the EU and/or landscape level.
6.39. One type of indicator not mentioned in SEEA EEA but which is worthy of
further consideration are holistic indicators of ecosystem health, resilience and
integrity. To the extent that scientific research has established an overall indicator
that relates well to the concept of ecosystem condition, then it may be that such
indicators can be applied directly for ecosystem accounting purposes for particular
ecosystem types.
6.40. It is not expected that the development of various individual indicators for
each ecosystem type would require the measurement of a vast number of
characteristics for every ecosystem. From an ecosystem accounting perspective, the
intention remains to provide a broad indication of the level and change in condition
rather than to fully map the functioning of every ecosystem. In this regard, a key
element of accounting is monitoring change over time and hence a focus on those
characteristics that reflect changes in ecosystem condition is an important
consideration. Based on assessments of the various projects referred to above, it
seems that for most ecosystem types a set of 4-6 indicators can provide a good set of
information to enable assessment of the overall condition of an ecosystem asset.
6.41. In terms of data sources, these will vary depending on the indicator selected.
In the areas of carbon, water and species diversity, a range of potential data sources is
introduced in Chapter 7. The ENCA QSP also proposes many data sources in these
areas. In many cases, satellite based data are likely to be useful information especially
in providing the breadth of data across different ecosystem assets that is required for
ecosystem accounting purposes.
6.42. Four considerations that might be used in selecting indicators are (i) the
sensitivity of ecosystem services supply to the indicator; (ii) the degree to which the
indicator reflects the overall health of the ecosystem or key processes within it; (iii)
data availability; and (iv) the possibility to generate new data cost effectively.
6.43. Compilers are encouraged to consider the work described in the project
research papers, the outcomes from testing in different projects, and most
importantly, to engage with national experts on ecosystems and biodiversity
measurement noting that there may be different experts for different ecosystem types.
In this regard, the ecosystem condition account is likely to be the primary account
through which engagement with the ecological community can be fostered.
6.3.4 Aggregate measures of condition
67
6.44. Given that indicators of individual characteristics are available (as just
described), the next question for ecosystem accounting concerns the aggregation of
indicators to obtain overall measures of ecosystem condition for a single ecosystem
asset and for multiple ecosystems. As noted in the introduction to this section, the
development of overall measures of the condition of ecosystem assets remains a
challenge in measurement terms.
6.45. In considering aggregation, the main objective in ecosystem accounting is to
provide decision makers some macro-level information that supports understanding
the relative state and importance of different ecosystems. Usually, there will be a
limited set of resources available to influence ecosystem condition and capacity and
hence choices must be made among a range of investment options.
6.46. Ideally, macro-level information would give a sense of the overall condition
of each ecosystem relative to each other and also relative to relevant thresholds. For
example, there may be more concern about an ecosystem whose condition was
considered relatively close to collapse.
6.47. In moving from individual indicators of specific characteristics to
information on relative overall condition, a continuum of information can be
described. Moving along the continuum reflects the use of additional information and
assumptions. In general terms, as the range of ecosystem types increases, it is likely
to be more difficult to make comparisons.
6.48. The continuum is as follows:
i. At the most basic level, there will be information on individual
characteristics that can be measured directly. For example, the pH level
of soil or the cubic metres of biomass in a forest can both be measured in
absolute terms. Looking at these measures over time can provide
information on the ecosystem’s condition.
ii. For some characteristics, it may be necessary to compare the chosen
metric with a known baseline, standard or threshold, to be able to infer
something about ecosystem condition. For example, measures of water or
air quality will rely on both direct measures of pollutants and an
understanding of how the estimate compares to a relevant standard.
iii. At the third level, still for a single characteristic, a composite indicator
could be formed whereby a number of indicators (related to the same
characteristic) are weighted together (as discussed in the previous
section). For this composite indicator to be meaningful, it is necessary
that the different indicators are measured or interpreted in relation to a
common baseline or standard. Note that for many characteristics, in
different ecosystem types, measurements at these first three levels are
well developed in the literature.
iv. At the fourth level, the aim is to consider, within a specific ecosystem,
some combination of indicators, each relating to a different characteristic.
Two issues arise, first the selection of characteristics, and second the
means of comparison. The first issue has been discussed in the previous
section. The solution to the second issue, as proposed in the SEEA EEA,
is the use of reference conditions, whereby indicators for each
characteristic are compared to the same reference condition for that
ecosystem.
v. At the final level, assuming that measures of ecosystem condition exist
for each ecosystem (i.e. at level four), it is necessary to find a means of
comparing different ecosystem types. Again, the use of reference
68
conditions is a possible way forward where, in this case, all ecosystems
are compared to a single reference condition. An extension at this level
would be comparisons between countries.
6.49. Given the desire in accounting for macro-level information, it is particularly
in the last two levels that the challenges in measurement and interpretation arise. A
focus is thus on the extent to which a single reference conditions can be used to both
compare within ecosystems, and to compare across ecosystems.
6.50. Assuming that an appropriate reference condition can be determined, the next
step is to normalise each indicator. This is commonly done using the reference
condition reflecting a “score” of 100 and the actual condition being between 0 and
100. A related approach used by SANBI (Driver, 2012), among others, is to grade
ecosystems on a scale of A – E (or similar), with A representing a characteristic
associated with a reference or near reference condition ecosystem and E representing
a characteristic within a heavily degraded ecosystem.
6.51. Establishing reference conditions and normalising scores is another task that
should be conducted in close consultation with national experts in ecosystems and
biodiversity. Indeed, it may well be the case that there are existing bodies of work in
government agencies, research bodies and universities that can be used or built upon
to support this type of assessment. The use of reference conditions is well known in
the ecological literature and it should be considered as an adaptation for ecosystem
accounting purposes rather than reflecting the use of an entirely new measurement
approach.
6.52. While the use of reference conditions is well known and practiced, the
precise choice of a reference or benchmark condition for accounting purposes
requires further testing. A short summary of the issues is provided in the following
sub-section.
6.53. As noted earlier in this section, the second step of weighting together
indicators of the different characteristics is more complex and less developed. The
ambition is not new from a statistical perspective (consider for example the Human
Development Index (UNDP, 2014)) but, as for socio-economic indicators, the
weighting of different ecosystem condition indicators is a matter of debate.
6.54. By far the easiest solution is to give each indicator equal weight in an overall
measure. However, this may not be appropriate from an ecological perspective with
different characteristics possibly playing a relatively more important role. Also, equal
weighting may not reflect the relative importance of different characteristics in the
supply of ecosystem services, or take into account various thresholds and non-
linearities which may apply in aggregating indicators that relate to different aspects of
condition.
6.55. At this stage, no clear pathways forward have emerged but there are a
number of areas for testing and research described below in section 6.5.
6.3.5 Determining a reference condition
6.56. As noted above, determining an appropriate reference condition is not
straightforward and can be the matter of considerable debate. The discussion here
provides a short summary of the key points from an ecosystem accounting
perspective.
6.57. A common starting point for determining a reference condition is application
of the idea of natural or pristine condition where the reference condition reflects the
condition of the ecosystem asset if it had been unaffected by human activity. In many
cases the application of this reference condition is done by selecting a point in time at
a pre-industrial stage. In Australia, for example, the year 1750 is commonly used.
69
6.58. A positive feature of this approach is that it places all ecosystem assets on a
common footing and “distance from natural” can be interpreted relatively
equivalently irrespective of the type of ecosystem. That is, it is possible to compare
ecosystems that are either extremely diverse, such as rainforests, or much less diverse
such as deserts.
6.59. Unfortunately, what constitutes a natural ecosystem can lead to significant
debate particularly in those countries where human influence on the landscape has
been evident for thousands of years. For example, almost all of Europe may be
considered to have been forested at one point in time, but the use of this as a
reference condition for the current mix of ecosystem types may not be appropriate.
6.60. Another concern that arises is that the reference condition can be mistakenly
interpreted as a target or optimal condition. A clear distinction should be made
between reference and target conditions. A reference condition should be used solely
as a means of estimating relative condition and comparing across ecosystem
characteristics and ecosystem types. Target conditions, on the other hand, should be
developed through participatory processes, taking into account economic, social and
environmental considerations. For example, in urban areas where the actual condition
would be likely very low or zero relative to a reference condition of the previous
natural state of that area, it would be inappropriate to suggest that the target condition
should be the natural state. This would imply a social choice of relocating an urban
area. On the contrary, it would be expected that information on the actual and
reference condition would be useful input to a discussion of target conditions.
6.61. For accounting purposes, it may be sufficient to simply select the condition at
the beginning of the accounting period as a reference condition and measure the
actual condition relative to that point in time. A variation on this approach is to select
the condition at the point in time at which the accounts commence.
6.62. The difficulty with this approach is that ecosystems that may have been
heavily degraded in the past will be compared from the same starting point as those
that have not been degraded at all – i.e. both would be given a reference condition of
100.
6.63. On balance then, some degree of discretion in the selection of a reference
condition is required. In making a decision, an important consideration is the question
of scale of analysis. In general, it will be more challenging to determine a reference
condition as the scale of analysis gets larger since there are more factors affecting the
analysis to take into account. Thus, if the intent is to only measure the condition of a
specific characteristic (e.g. soil condition) of a specific ecosystem (e.g. open
grasslands) then the choice of reference condition may be made taking only that
characteristic and ecosystem type into account.
6.64. However, where there is a desire to compare multiple characteristics and
multiple ecosystems, then a relevant reference condition will not be readily apparent.
For some countries, it may be that a pre-industrial time point gives an appropriate
reference condition, since there is a point in the not too distant past where a relative
common understanding of change from a reference condition can be understood. As
noted though, such a choice will likely not be appropriate where the current landscape
mix of ecosystem types has, to varying degrees, been evident for centuries. Assuming
a national level reference condition can be determined, a remaining challenge is to
find a reference condition that allows comparison across countries. Given the
diversity of landscape development patterns this choice is not at all obvious.
6.65. Pending further testing of different reference conditions, it is recommended
that in the development of ecosystem condition accounts for a given country, that a
point in time be selected, preferably at least 30 years previous, to allow the
70
development of the relevant metrics of current condition and the application of the
reference condition approach. For the measurement of change over time within a
country, this is a pragmatic starting point and ensures that discussions focus on
perhaps the more challenging measurement issue of actually selecting the indicators
and maintaining ongoing time series. Using a relatively distant reference point, rather
than the beginning of the accounting period, will better support the assessment of
distance from thresholds for ecosystem assets.
6.66. Where comparison across countries is required, then it will be necessary for
the measurement community to test options that are both meaningful for comparison
purposes and also feasible for implementation. One point for testing would be the
potential for a country to use one reference condition domestically, but to use a
different reference condition for international comparison.
6.4 Developing the concept of ecosystem capacity

6.4.1 Defining ecosystem capacity
6.67. Earlier in this chapter, it was noted that the SEEA EEA describes three main
ecosystem asset concepts: ecosystem extent, ecosystem condition and expected
ecosystem service flows. Ecosystem capacity was the main ecosystem asset concept
not dealt with in SEEA EEA. This concept is implicit in making the connection
between ecosystem assets and ecosystem services but the nature of this connection
was not articulated in SEEA EEA. This was for two reasons:
 First, there was recognition that the link between ecosystem assets and
ecosystem services is hard to describe, particularly in terms of the link
between changes in overall ecosystem condition and the supply of individual
ecosystem services. Notions of threshold effects, resilience, ecosystem
dynamics and other non-linear factors are important to consider.
 Second, since the concept of capacity was considered to relate to the overall
ecosystem asset, a requirement in measuring capacity was defining an
expected basket of ecosystem services. Discussion on how such a basket
should be defined was not conclusive.
6.68. Since the release of SEEA EEA in 2013, it has become increasingly apparent
that the concept of ecosystem capacity is a central one for explaining the ecosystem
accounting model and applying the model in practice. This is especially the case in
relation to developing information sets that can support the discussion of
sustainability. An initial definition of ecosystem capacity is – the ability of an
ecosystem to generate an ecosystem service under current ecosystem conditions and
uses at the maximum yield or use level that does not negatively affect the future
supply of the same or other ecosystem services (Hein et al., 2015).
6.69. In the context of developing this definition, the following points have
emerged. They are listed here to frame an ongoing discussion about ecosystem
capacity in the context of ecosystem accounting and to encourage further dialogue on
this topic. In particular, it discusses issues in applying the concept of capacity to
different types of ecosystem services, i.e. provisioning, regulating and cultural
services.
i. Ecosystem capacity is a function of ecosystem extent and condition and
maintaining condition is an essential feature in the measurement of capacity.
ii. The supply of one ecosystem service can reduce the ecosystem’s capacity to
supply other ecosystem services. For instance, timber harvesting may reduce
71
recreational opportunities in a forest. A distinction is therefore needed
between (a) an ecosystem asset’s capacity to supply an individual ecosystem
service, i.e. there is a measure of capacity corresponding to each ecosystem
service within the basket; and (b) an ecosystem asset’s capacity to supply a
basket of ecosystem services as a whole.
iii. Ideally, to take into account the systemic nature of ecosystem service supply,
each individual service capacity measure will be a function of the overall
ecosystem asset condition, thus bringing together the two conceptualizations
just outlined.
iv. For ecosystem accounting, ecosystem capacity should be considered in
relation to a current basket of ecosystem services and a specific ecosystem
asset. It can be related to the sustainable flow of ecosystem services under
current ecosystem conditions and ecosystem uses, and with respect to the
current basket of ecosystem services. Estimates of ecosystem capacity may
relate to flows of ecosystem services that are lower, higher or equal to actual,
observed flows. In most cases, defining sustainable ecosystem use levels
requires specific consideration of extractive ecosystem uses (e.g. logging,
fishing), and involves analysing the basket of ecosystem services under the
assumption that these extractive uses are brought to a sustainable level.
v. Each individual service capacity may be considered as a sustainable yield or
flow relevant to the specific ecosystem service and taking into account the
use levels of other ecosystem services supplied in by the ecosystem. The
capacity measure should therefore reflect the estimated stream of annual
service flows for the forthcoming accounting period, given the extent and
condition of the ecosystem asset at that time, and under the constraint that the
extent and condition remain unchanged over the accounting period.
vi. For ecosystem accounting, capacity is related to the actual basket of
ecosystem services supplied. Thus, capacity requires the presence of users of
ecosystem services. Capacity therefore differs from the ability of an
ecosystem asset to supply ecosystem services independently from the
potential use of those services by beneficiaries. This could be labelled
theoretical or potential ecosystem supply (e.g. Bagstad et al., 2014). It may
also differ from the basket of ecosystem services that would be obtained
under optimal ecosystem management, which could be labelled ‘the
capability of an ecosystem to supply services’. Both theoretical/potential
supply and capability are relevant concepts for ecosystem management but
would not necessarily underpin ecosystem accounting estimates.
vii. In cases where high levels of use of the ecosystem asset take place, e.g.
through high levels of extraction or pollution, it is expected that the condition
of the asset will fall and hence that actual flows of ecosystem services will be
higher than the sustainable flow. This set of circumstances would reflect
ecosystem degradation.
6.70. Considering capacity as being measurable in terms of individual ecosystem
services is an important step forward in an accounting context, since it permits a
direct link to discussions of sustainable yield and flow that are well established in
biological models and resource economics. However, there remain significant
challenges in understanding the links between measures of capacity for individual
services and overall ecosystem condition.
6.71. An interim step may be the measurement of the potential of an ecosystem to
supply ecosystem services, without requiring a link to be made to the likely use of
ecosystem services by beneficiaries. However, for some ecosystems, particularly
72
remote ecosystems, measures of such potential or “theoretical” supply may provide a
significant overstatement of the availability and value of ecosystem services for
economic and human activity.
6.72. Suggestions for taking this work forward in an ecosystem accounting context
are described in section 6.5.
6.4.2 Linking ecosystem capacity and ecosystem degradation

6.73. From an accounting perspective, an important and emerging aspect of
ecosystem capacity measurement concerns the link between ecosystem capacity and
ecosystem degradation. In the SEEA EEA, ecosystem degradation is defined in
relation to the decline in condition of an ecosystem asset as a result of economic and
other human activity (SEEA EEA 4.31). This aligns with the approach in the SEEA
Central Framework for the definition of depletion of natural resources and in the
SNA for consumption of fixed capital (depreciation) of produced assets. Table 4.6
presents the linkages between these accounting entries.
6.74. The emerging idea is that while ecosystem degradation is clearly related to
declining condition, it should be defined more specifically as reflecting a decline in
ecosystem capacity as a consequence of human activity (in line with the accounting
definition of degradation). Thus ecosystem degradation would be measured as the
decline in ecosystem capacity over an accounting period, or in other words, the
decline in the ecosystem asset’s ability to supply ecosystem services due to a decline
in ecosystem condition. This would mean that, at the beginning and end of the
accounting period, an assessment would be needed as to the future flow of ecosystem
services if, from that point forward, the supply of ecosystem services did not lead to a
decline in condition.
6.75. Standard concepts of depreciation and depletion are developed in relation to
assets with finite lives and hence, the measurement of the capacity for sustainable
supply of capital services does not arise explicitly. As a result, how a capacity based
definition of degradation can be incorporated in a manner that remains consistent
with the accounting principles of the SNA and SEEA Central Framework requires
further investigation.
6.76. Overall, the role of measures of ecosystem capacity in an accounting context
requires further discussion. As indicated in the SEEA Central Framework, discussion
of depletion, in an accounting setting it is necessary to measure the depletion and
degradation of renewable biological resources with reference to the potential for
infinite regeneration. This issue does not arise in accounting for produced assets or
non-renewable resources since, for accounting purposes, it is sufficient to assume that
over some period of time the asset will be completely used up.
6.77. In determining the appropriate asset life for renewable resources, the concept
of capacity may be particularly relevant since it supports consideration of the effects
of current activity on the long term usefulness of the asset. Further, in the context of
measuring degradation, the emerging conceptual discussion is that ecosystem
degradation occurs when actual ecosystem service flows (in particular, provisioning
services) exceed the ecosystem’s capacity to supply that service. Therefore, where
capacity can be quantified, it can be used in accounting as a measure to analyse
whether flows of ecosystem services can be sustained in the future (see Schröter, et.
al., 2014).
6.78. While ecosystem degradation may be most appropriately measured in terms
of changes in capacity, degradation will also be reflected in measures of changes in
ecosystem condition and, depending on how the ecosystem is used, in flows of
73
ecosystem services (since the expected flow of ecosystem services will ultimately
decrease over time as a result of ecosystem degradation).
6.79. There have been proposals to develop asset accounts for ecosystem capacity
At this point, an ecosystem capacity account has not been defined. Instead, the
emphasis is placed on the measurement of ecosystem capacity for individual services
such that there can be a more complete understanding of the extent to which current
patterns of use differ from patterns of use that would leave the condition of the
ecosystem asset unchanged.
6.80. From an ecosystem accounts compilation perspective, the need for further
discussion on ecosystem capacity in no way limits the potential to compile most other
ecosystem accounts. Indeed the compilation of these various accounts (extent,
condition, ecosystem services supply and use) will be important in providing the
measurement experience and detail for the refinement of measures of ecosystem
capacity that have been discussed.
6.5 Recommendations
6.5.1 Recommended steps for testing and experimentation on ecosystem assets
6.81. As described in Chapter 4, the compilation of an ecosystem extent account is
the likely starting point for all ecosystem accounting work. In the process of
compiling an extent account, the relevant spatial units should be clearly identified,
and it will be possible to develop maps and tables showing the changes in
composition of ecosystem extent in a continuous, spatially- and temporally-explicit
way at the country level. These information should be consider important outputs of
ecosystem accounting in their own right and can be linked to relevant socio-economic
data such as population change, production and income data, employment
information to provide a sense of the potential of accounting for environmental
assets.
6.82. Consistent with the advice provided in Chapter 4, at a minimum ecosystem
extent accounts should be developed at the level of the top level land cover types as
described in the SEEA Central Framework land cover classification. However,
wherever possible and as appropriate, finer level breakdowns of these land cover
types should be developed including integrating this with existing ecological
classifications. An important outcome from testing the measurement of ecosystem
extent is determining the level of detail needed to provide broad trends in changes in
ecosystems at the country level.
6.83. Specific advice concerning the measurement of ecosystem extent is to
measure the extent14 of the EU identified (and mapped) following recommendations
in Chapter 3. An important (and still challenging) consideration is the choice of an
appropriate spatial resolution to ensure balanced representation of dominant versus
smaller ecosystem units. In this context is it necessary to:
 Apply land cover/use change to measure the extent of ecosystems, with additions
and reductions, in spatially- and temporally-explicit way
 Report, where appropriate, the extent measures in a way that allows international
comparability, for example, by aggregating to the global biome/ecoregion defined
by WWF, and also reclassifying and aggregating the ecosystem types to match
the 15 land cover classes from the SEEA Central Framework
14
Change in volume, length or other physical measures may be more relevant for ecosystem like rivers and
lakes
74
 Combine area (raster or vector-polygon) measurements and linear (vector)
measurements for the smallest features, such as streams and hedgerows to
prevent double counting of areas. Key principles are to maintain the same total
national area and ensure that no gaps are left.
 Consider special area-corrections where coastal lines have a high fractal
dimension, and in mountain slopes terrains.
6.84. The measurement of ecosystem condition following the concepts in the

ecosystem accounting model is a complex task due the need to consider multiple
ecosystems and multiple characteristics. At the same time, there is sufficient research
in the general area of condition measurement to suggest that the testing of different
approaches in pilot ecosystem accounting projects is quite possible and should be
pursued.
6.85. An initial question in taking this work forward is whether a top down or
bottom up approach should be used. In the initial phases of testing it is recommended
that a top down approach be used following the steps just below. Over time, it may
then be appropriate to expand the set of indicators for different ecosystem types
within the same spatial units architecture.
 Measure the main ecosystem characteristics/stocks on land including soil,
biomass (composed of C, N, P and others), species/habitats abundance and
conservation, and water. These can be estimated applying the models
recommended in Chapter 5.
 Carbon, nutrients and biodiversity measures will be applicable to
virtually all ecosystems, while water- and soil-related ones for terrestrial
ecosystems.
 Cross-cutting estimates related to land, water and forest should be
consistent with the national accounts developed for the Central
Framework
 Choose an appropriate reference period for the condition measure, or
alternatively use the ‘opening stock’
 Apply measurement of condition with reference to the supply and use of
ecosystem services
 Record and report on the variability and sources of error in the data.
6.86. Beyond top down approaches, and where resources are available, it is likely
to be more ecologically well founded if bottom up approaches can be tested – i.e.
developing measurement specific characteristics for different ecosystem types.
6.87. An important observation is that a broader set of characteristics can be
considered than included in the ecosystem condition account described in the SEEA
EEA. Thus, in addition to the characteristics of vegetation, biodiversity, soil, water
and carbon, it is recommended that consideration also be given to developing
condition indicators for air and ecosystem integrity and health (including for example
indicators of fragmentation, naturalness, and ecosystem diversity).
6.88. While the table provides a good starting point for a testing program, it is not
intended that every indicator proposed in the table be tested in all countries. In the
planning phase there are two important steps that should be taken. First, for each
ecosystem type a connection should be made between the common uses of the
ecosystem and the most relevant characteristics – i.e. not all characteristics will be
relevant in the measurement of condition of all ecosystem types. It is noted that
answering these questions would be supported by understanding the relevant
75
ecosystem services that are likely to be supplied from a given ecosystem. Second, an
assessment must be made regarding data availability for the different indicators.
6.89. To support the comparison of different ecosystems types within a country it
is recommended that a single reference condition approach be used. Different
principles for determining a reference condition can be applied, including the
principle of naturalness. However, given the difficulty of applying this principle in a
number of countries and the practical issue of defining naturalness, it is
recommended that, as a starting point for ecosystem accounting, a single reference
point be selected, preferably at least 30 years previous. This will allow the
development of the relevant metrics of current condition and the application of the
reference condition approach.
6.90. For the measurement of change over time within a country this is a pragmatic
starting point and ensures that discussions focus on the more challenging
measurement issue of actually selecting the indicators and maintaining ongoing time
series. Where available, it is also likely to be relevant to understand target conditions
and hence understand the movement towards or away from the desired state.
6.91. On the whole, these recommendations are ones that can be tested in
applications at country and regional level. One area for research is the development
of overall indexes of condition for ecosystem assets either based on aggregation of
indicators for selected characteristics or using some alternative approach, for example
isolating a key characteristic in ecological terms. Further research is also required on
the choice of reference condition for ecosystem accounting purposes.
6.5.2 Recommendations in relation to ecosystem capacity

6.92. For ecosystem capacity, it is increasingly evident that this concept has an
important role to play in ecosystem accounting, both in terms of interpreting the links
between ecosystem services and ecosystem condition and in terms of the definition of
6.93. Given that a definition of ecosystem capacity is still emerging, the advice
here is that the measurement of ecosystem capacity should be considered to be a topic
of ongoing research but with a very high priority. The principle aims in the short term
should be (i) to reach a common understanding of the definition of ecosystem
capacity and its relationship to other related concepts; (ii) to articulate the role of
ecosystem capacity within the accounting system, primarily with respect to defining
6.94. To support this research into ecosystem capacity, it would be beneficial for
those countries and agencies undertaking testing of ecosystem accounting to consider
questions relating to the links between flows of ecosystem services and measures of
ecosystem condition. These links should, in any event, be a part of any testing since it
is generally accepted that the measurement of condition must integrate information
on the management and use of ecosystems and that modelling the flows of ecosystem
services, particularly regulating services, will involve the use of information on
ecosystem condition.
6.95. One way in which these relationships might be developed and tested is to
undertake analysis for different scenarios of future conditions. Such information and
methods are likely to be of direct interest for policy purposes in any event and may be
usefully defined to support an understanding of ecosystem capacity.
6.96. Finally, it is observed that understanding and measuring capacity will benefit
from a more complete articulation of the “cascade” model that links ecosystem
structure, composition and function to ecosystem processes and flows of ecosystem
76
services. Understanding these connections which are relevant in a number of parts of
the ecosystem accounting model should be seen as an important aspect of testing.
77
7. Thematic accounts

Thematic accounts are standalone accounts on topics of interest in their own right and also of
direct relevance in the measurement of ecosystems and in assessing policy responses.
Thematic accounts include accounts for land, carbon, water and species diversity.
Land accounts can focus on land use, land cover and land ownership. The development of
land accounts provides a platform for measurement and is commonly the basic entry point for
ecosystem accounting.
In water accounting measurement at the river basin level is important for ecosystem
assessment. Working at this level will require the use of hydrological models but this work
can also underpin the estimation of relevant ecosystem services such as water filtration and
soil retention.
Accounting for stocks of carbon can provide a strong base for co-ordinating information on
carbon and complements measurement within the Inter-governmental Panel on Climate
Change (IPCC) framework and the UN REDD+. The data can support measurement of
ecosystem condition and ecosystem services such as carbon sequestration and storage.
Accounting for biodiversity considers both ecosystem and species diversity. Biodiversity is
considered primarily a characteristic of ecosystem assets rather than an ecosystem service. In
accounting terms this permits recognising declines or improvements in biodiversity over time
and links to the capacity of ecosystems to supply ecosystem services.
In all cases - land, water, carbon and biodiversity - there is a broad range of information and
measurement methodologies available. The challenge for ecosystem accounting is the
assessment and integration of these data and methods within the SEEA EEA framework.
7.1 Introduction
7.1. The ecosystem accounts described in Chapter 4 provide a coherent coverage
of information pertaining to ecosystem assets and ecosystem services. At the same
time, from both an analytical and a measurement perspective, it can be challenging to
focus only on a systems perspective. More commonly, our view of ecosystems, and
our policy responses, are framed using themes that concern specific aspects of the
economic – environment relationship. Four main themes that are evident are land,
water, carbon and biodiversity. This chapter provides an introduction to accounting in
relation to these themes.
7.2. The incorporation of a thematic focus in the context of ecosystem accounting
provides two benefits. First, it enables a closer link to be drawn between the
compilation of ecosystem accounts and the likely areas of policy response – for
example in terms of land management, management of river basins, carbon emissions
policy and maintenance of protected areas. Second, often the data that are used to
understand trends in thematic areas can also be used to compile ecosystem accounts.
7.3. It is relevant to note that while measurement in each of the four main themes
is relatively well advanced, the work on the SEEA has highlighted the potential to use
accounting approaches to (i) improve the co-ordination of data and (ii) recognise
links between the themes.
78
7.4. In the case of two themes – land and water – the SEEA Central Framework
and the SEEA Water provide the conceptual grounding for accounting. For carbon, as
a single element, it is actually quite well suited as a subject for accounting. It has thus
been relatively straightforward to consider adapting the measurement of carbon into a
broad accounting structure. The relevant concepts are described in the SEEA EEA.
For biodiversity, the situation is developing. SEEA EEA section 4.5 introduces
relevant ideas for accounting for biodiversity but more testing is needed.
7.5. Accounts for land, water, carbon and biodiversity contain much relevant
information in their own right. Consequently, compilers of ecosystem accounts are
encouraged to seek opportunities to promote and use the information presented in
these thematic accounts to support discussion of environmental-economic issues. In
particular, information from the thematic accounts, when presented in the context of
ecosystem measures, can provide a more tangible hook for users when making links
between ecosystems and policy choices.
7.6. This chapter provides a summary of the relevant accounting issues for each
of these four areas.
7.2 Accounting for land

7.2.1 Introduction
7.7. Accounting for land, particularly land cover, will be a common starting point
for compilers of ecosystem accounts, given the focus on terrestrial ecosystems. A
distinction is made here between land accounting and ecosystem extent accounts.
Land accounting is considered to encompass compilation of a variety of accounts
utilising different classifications of land including land use/management, land cover,
and land ownership. In applying these classifications links to standard SNA
classifications of industry (ISIC) and institutional sector. Land accounting will
include standard asset account structures and also change matrices and tables that
cross classify land. These various aspects of land accounting are covered in the SEEA
Central Framework Chapter 5. Ecosystem extent accounts on the other hand are a
specific account recording the area and change in area of different EU types. Ideally,
EU types would be nested within a land cover classification, hence providing a link
between land accounts and ecosystem extent accounts.
7.8. As part of the accounts compilation process, the information from land cover
accounts can be used to help define the relevant spatial areas, to determine the extent
of different ecosystem types at a broad level, to support understanding the links
between ecosystem services supply and the beneficiaries of those ecosystem services
and finally, to facilitate the scaling of other data to finer and broader levels of detail.
7.9. From an analytical and policy perspective, information on land cover can, at
a national scale, provide important information on trends in deforestation,
desertification, urbanisation and similar forms of landscape change. As recognised in
ecosystem accounting, understanding these types of changes is not sufficient for
understanding the effects on ecosystem condition or flows of ecosystem services but
it is a relevant starting point.
7.10. The total area of a country may also be classified according to land
use or land ownership criteria. An interim land use classification is provided in the
SEEA Central Framework (Table 5.11 and Annex I). Land ownership may be
classified by institutional sector (corporations, government, households) or by
industry (agriculture, manufacturing, retail, etc.). In some cases, a reasonably clear
connection can be made between different classifications of land – for example there
79
will often be a clear link between tree-covered areas and forestry. However, it is not
possible for a simple integration of land cover and land use classes to be described.
7.11. Information on land use and land ownership will be important in
understanding the connection between ecosystem assets and the beneficiaries of
ecosystem services. For that reason, it is recommended that, where possible, accounts
for land use and land ownership be compiled following the advice in the SEEA
Central Framework. A useful output for ecosystem accounting may be a table which
cross-classifies land cover and land use at a given point in time. Such a table would
highlight the relative significance of different land cover types to specific uses.
7.12. Land accounts can also provide an important tool to link environmental and
socio-economic data, essentially providing a means by which policy can be placed in
a spatial context. A key link here is recognising that implementation of policy to
maintain and restore ecosystem condition is likely to require the involvement of land
holders. Hence, understanding the connection between land ownership, current use
and the relevant ecosystems can provide the means by which decisions on appropriate
policy interventions can be made.
7.13. Generally, the initial focus of land accounting is on terrestrial areas of a
country, including freshwater bodies. Within this scope land must be classified into
various classes (type of cover, type of use, or type of owning economic unit). Often
there will be relevant national classifications and datasets but alignment or
correspondence to international classifications is a positive step. Chapter 3 discusses
issues of classification in more detail.
7.14. The basic structure of a land account follows the structure of an asset account
as described in the SEEA Central Framework. That is, there will be an opening stock,
additions and reductions in stock and a closing stock. Ideally, changes in stock over
an accounting period would be separated into those that are naturally driven and those
due to human activities. Both the SEEA Central Framework and the SEEA EEA
describe the structure of a land cover and land use accounts.
7.15. In addition to an asset account, information on land cover and land use may
be organised in the form of a change matrices which show how, over an accounting
period, the composition of land has changed. An example of such a matrix for land
cover is provided in the SEEA Central Framework, Table 5.14 (UN et al., 2014a).
7.2.2 Relevant data and source materials

7.16. In terms of data requirements, a distinction has to be made between dynamic
and permanent features. Dynamic features include information on land use, land
cover and vegetation type. Permanent features include information on administrative
boundaries, ecological regions, and river basins.
7.17. The compilation of accounts will generally require bringing these various
data together using GIS systems to produce data for a country as a whole. The
ambition in accounting terms is to generate harmonised maps, in time series, such
that the stock and changes in stock can be consistently accounted for.
7.18. Materials to support land accounting include the SEEA Central Framework,
the SEEA EEA and the ENCA QSP. The ENCA QSP in particular has an extensive
discussion of land cover accounting and associated data sources and methods.
7.19. Additional support and guidance is available in looking at country examples
and case studies. Relevant examples include the work of the European Environment
Agency (Weber, 2011), the Australian Bureau of Statistics (2015), Statistics Canada
80
(2013), the Victorian Department of Sustainability and Environment (2015) and in
Mauritius (Weber, 2014).
7.2.3 Key issues and challenges in measurement

7.20. There is a range of measurement challenges in land accounting. An
immediate challenge is being able to integrate the various data to produce harmonised
geo-databases and, for accounting purposes, measures of change over time. This
requires careful consideration of scale and classification in combining different data
sets.
7.21. In general terms, higher levels of detail will be better but will also have
higher resource costs. Balancing the resources available with the degree of accuracy
required will be important. A relevant issue in this context is understanding
approaches to the validation of data, particularly since much data will be derived
from remote sensing, including satellite imagery. Ideally, some degree of sampled
ground truthing should be undertaken.
7.22. An integrated approach involving sampled reference points to measure land
use and land cover across Europe, the Land Use and Cover Area Survey (LUCAS)
has been developed in recent years by Eurostat. This approach may provide
additional ideas for measurement approaches at national level.
7.23. The approach to classifying land is particularly important for communicating
messages on the changing composition of land at national level. For land cover, there
is now an ISO standard 15 that underpins the Land Cover Classification Scheme
(LCCS version 3) as developed by the FAO (FAO and GLCN, 2009). This provides a
structure by which each type of land cover around the world can be consistent
classified. It thus provides a way of linking the various classifications that are in use
in different countries and regions.
7.24. While this provides a base level classification tool, more varied have been the
approaches to the formation of higher level classes that can be used to summarise
detailed classes in meaningful ways. There are a number of options, one of which is
the interim land cover classification presented in the SEEA Central Framework.
Definition of a broadly accepted set of high level (say 10-15) classes of land cover
(and the associated definitions of these classes) would be an significant step forward
in coordinating information and would underpin greater alignment in ecosystem
accounting discussions and applications.
7.25. With regard to the classification of land use, the SEEA Central Framework
describes an interim land use classification based on work on agriculture, forestry and
fisheries land use by the FAO, and UNECE work on the classification of land use for
all economic activities.
7.2.4 Recommended activities and research issues

7.26. It is recommended that countries develop land accounts as an integral part of
a suite of national environmental-economic accounts. In their own right, land
accounts provide important information on environmental trends. Also, their
compilation requires the organisation of spatial data which in turn provides the inputs
15
ISO 19144-2:2012 specifies a Land Cover Meta Language (LCML) expressed as a UML
metamodel that allows different land cover classification systems to be described based on the
physiognomic aspects.
http://www.iso.org/iso/iso_catalogue/catalogue_tc/catalogue_detail.htm?csnumber=44342
81
for the delineation of spatial units and ecosystem accounting. Finally, a focus on land
provides a platform for integrating environmental and socio-economic data.
7.27. A number of relevant areas for testing in relation to land accounting are
presented in chapters 3 and 6 in relation to the delineation of spatial units and the
compilation of ecosystem extent accounts. In terms of areas for research, the main
issues concern (i) finalising appropriate classifications for land cover and land use
beyond the interim classifications of the SEEA Central Framework; and (ii)
determining the best approaches to account for linear features, such as rivers, beaches
and hedgerows.
7.3 Accounting for water related stocks and flows

7.3.1 Introduction
7.28. Water is a fundamental resource. It is essential for all life and underpins the
production of food, fibre and energy in many countries. The management of water,
including taking into account cross-boundary flows (e.g. in relation to the Nile
River), and the joint ownership of surface water bodies (e.g. in relation to Lake
Victoria), is an important focus for many governments around the world.
7.29. Accounting for stocks and flows of water is a key feature of both the SEEA
Central Framework and the SEEA EEA. While both of these documents promote
accounting for water at the river basin level, this is the focus in ecosystem
accounting. This short section is intended only to provide direction to relevant
technical and compilation materials rather than reproduce or summarise the content
of those materials.
7.30. Accounting for water is relevant to ecosystem accounting in a number of
ways. First, water is a key feature of ecosystems and hence the measurement of the
stocks and changes in stocks of water resources is a relevant aspect in the
measurement of ecosystem condition.
7.31. Second, there are a number of ecosystem services which relate directly to
water. These include the provisioning service of water when it is abstracted for use
(irrigation, drinking, hydropower), the regulating role of water in filtering pollutants
and other residual flows, and the cultural services associated with water such as
fishing and other recreational activities. In addition, there are a number of ecosystem
services to which water is linked, for example, the regulation of water flows to
provide flood protection and the filtration of water by the soil in catchments.
7.32. Measurements in all of these areas are ultimately important within a complete
set of ecosystem accounts. The accounts of the SEEA Central Framework for water
focus on two areas – (a) the supply and use of water and (b) the asset account for
water. They provide the basis for accounting for water.

7.33. There are many relevant materials to support the compilation of water
accounts. Aside from the content in the SEEA Central Framework and the SEEA
EEA, there is also SEEA Water (UN, 2012b) and the associated International
Recommendations on Water Statistics (IRWS) (UN, 2012a). Chapter 6 of the ENCA
QSP has much relevant information.
7.34. There is a wide range of data sources, including global data sets that might be
considered for use in water accounting. To date, over 50 countries have trialled the
82
development of SEEA based water accounts. Consequently, there is an increasing
body of knowledge and experience in water accounting that can be drawn on.

7.35. There remain some specific challenges in accounting for water, especially in
an ecosystem accounting context. Linked to the issue of defining spatial units, there is
the need for clarity on the delineation of wetlands with the scale of analysis being a
particular area of concern. Many wetlands may be quite small but disproportionately
important within larger land cover types (for example, in grasslands). The ability to
recognise these areas and hence better understand the stocks and flows of water
resources is important.
7.36. On a related note, integrating information on groundwater within the
ecosystem accounting framework requires further consideration given that generally
the ecosystem accounts have considered surface water resources. The same comment
also applies in the case of information on stream flow and water yield. While these
are not standard SEEA accounting items, indicators on these aspects of the
hydrological system may be relevant in understanding the system and more
completely assessing ecosystem condition.
7.37. Given that flows of water are often key pathways between different
ecosystems, more work is needed to understand and account for flows of ecosystem
services between ecosystem assets that are related to water. For example,
understanding water flows is relevant in measuring the service of soil retention in the
upper reaches of river basins. SEEA EEA largely ignored flows between ecosystems
but further reflection suggests that incorporating certain intermediate ecosystem
services is required.
7.38. A general challenge in water accounting from a national accounts perspective
is that, often, national level data on water resources are not overly meaningful.
Instead, information at a river basin level is required. While it may be straightforward
to propose measurement at this level of detail, developing estimates at a river basin
level will be resource intensive. Further, in some situations, sub-annual data may be
needed to understand the dynamics of seasonal fluctuations in water supply and water
use. Such information may be overlooked if working with annual or long-term
averages.

7.39. The main conclusion in relation to accounting for water resources is that
there is a wide array of information and examples of water accounting in practice to
support countries that wish to start work in this area. Further, there are many datasets
that can provide a starting point for compilation. Testing the compilation of water
accounts should therefore be given a very high priority.
7.40. A few areas in which further research might be conducted include accounting
for dependencies between ecosystem units within river basins (including flows of
intermediate ecosystem services), advancing discussion on the valuation of water
resources and accounting for water quality at a broad scale.
7.4 Accounting for carbon related stocks and flows

7.4.1 Introduction
83
7.41. Carbon has a central place in ecosystem and other environmental processes
and hence accounting for carbon stocks and transfers between them must be seen as
an important aspect of environmental-economic accounting. This short section is
intended only to provide direction to relevant technical and compilation materials
rather than to reproduce or summarise the content of those materials.
7.42. Accounting for carbon in the SEEA commenced in the context of accounting
for carbon in forests and for greenhouse gas emissions. With the development of the
SEEA EEA, the scope of carbon accounting has been broadened and, as described in
the SEEA EEA, ideally it encompasses carbon stocks and flows and their changes for
all parts of the carbon cycle and all carbon pools. Thus it covers geocarbon,
biocarbon, atmospheric carbon, carbon in the oceans and carbon accumulated in the
economy. In practice, the focus of carbon stock accounting at this stage is on
biocarbon and geocarbon.
7.43. The measurement of stocks and transfers of carbon can support discussion of
many policy relevant issues. These issues include the analysis of greenhouse gas
emissions, sources of energy, deforestation and land use change, loss of productivity
and biomass, and sources and sinks of carbon emissions. Since carbon is also a
common focus of policy response, for example carbon taxes, its direct measurement
is of high relevance.
7.44. In ecosystem accounting, information on stocks and flows of carbon can be
used in two main areas. First, as part of the measurement of ecosystem condition.
One broad approach is to use changes in net ecosystem carbon balance (net primary
production less human appropriation) as an indicator of ecosystem condition. This
single indicator can capture changes in soil, vegetation and other biomass. Second,
information on carbon stocks and flows relates directly to the ecosystem services of
carbon sequestration and carbon storage.

7.45. The structure of a carbon stock account is presented in SEEA EEA Chapter 4.
The compilation of this account, with a focus on biocarbon and geocarbon, involves
the collection of (i) data on land vegetation/cover and the rates at which different
land/vegetation cover types sequester and store carbon in above and below ground
biomass; (ii) data on the carbon content of soils; and (iii) information on sub-soil
fossil fuel resources. A particularly relevant source is information compiled by
countries as part of reporting to the IPCC.
7.46. Advice on the compilation of carbon accounts is summarised in SEEA EEA.
A more detailed explanation is provided by Ajani & Comisari (2014) which describes
the development of a carbon account for Australia including discussion of the
relevance and application of the account.
7.47. A number of aspects of carbon accounting are also reflected in the SEEA
Central Framework. For example, air emission accounts will include flows of GHG
emissions; and mineral and energy resource asset accounts will record stocks and
changes in stocks of sub-soil fossil fuel resources. For SEEA EEA purposes, it is
relevant to compile such accounts at a finer level of spatial detail.
7.48. The ENCA QSP provides a detailed discussion on accounting for changes in
biocarbon at national scale including a discussion about global datasets and
measurement challenges. Of particular relevance, is the work undertaken on the
measurement of carbon through the FAO Forest Resource Assessment (FRA) (FAO,
2015) which is conducted every five years. The FRA asks for estimates of carbon
84
stock for forests including above and below ground carbon stocks. These data may
provide a useful starting point for compiling a time series of carbon accounts.

7.49. Compared to other areas of measurement, the measurement issues in relation
to carbon are relatively limited, although challenges exist in using point
measurements for estimation of large scale areas. In large part, this reflects the
substantial resources that have been applied to this measurement task within the IPCC
processes. Nonetheless, there remain important issues of data quality to consider. In
large part these relate to being able to measure carbon stocks accurately across the
wide variety of vegetation and soil types since different carbon content ratios will
apply in different situations. Related to this, the sourcing of information via either
remote sensing or using local sources requires balancing between coverage and
accuracy.

7.50. Given the high policy relevance of carbon and the comparably large
resources currently directed at measuring stocks and flows of carbon at national level,
it is recommended that countries support the development of carbon accounts. The
preparation of these accounts can provide information on broad trends in
environmental change and also provide insight into the requirements of bringing data
together from a variety of sources.
7.5 Accounting for biodiversity

7.5.1 Introduction
7.51. Biodiversity (the diversity of ecosystems, species and genes) plays an
essential role in supporting human well-being. Biodiversity maintains functioning
ecosystems that in turn deliver ecosystem services such as food, the regulation of our
climate and aesthetic enjoyment.
7.52. The SEEA EEA provides a framework to measure and link ecosystem service
flows supported by biodiversity and other characteristics (e.g. soil type, altitude) with
the economy and other human activities. It also allows comparison and integration of
data on ecosystem services with other economic and social data. Biodiversity
accounts can help build an understanding of the relationship between biodiversity and
economic activity by integrating data in a spatially explicit manner.
7.53. On the whole, the perspective taken for ecosystem accounting in the SEEA
EEA is that biodiversity is a feature that is directly relevant in measurement of the
condition of ecosystem assets. Measures of biodiversity, whether of ecosystem
diversity or species diversity (the inclusion of genetic diversity measures has not yet
been examined), are considered to relate primarily to the stocks component in the
accounting model. This approach is consistent with a view that biodiversity can be
degraded or enhanced over time, an attribute that applies only to stocks and not to
flows (i.e. ecosystem services). In this context, the connection between biodiversity
and ecosystem functioning may often be difficult to make. This is related to both to
the limitations of ecosystem dynamics models as well as data gaps for many
ecosystems world-wide.
7.54. In the framing of the SEEA EEA, species diversity may be considered as a
85
characteristic of an individual or connected EUs, and ecosystem diversity would
emerge from assessment of the diversity of EU types. People may appreciate and
therefore value specific elements of biodiversity, for example when they take an
interest in the conservation of endemic and/or iconic species. This is reflected, for
instance, in the creation of protected areas in many countries. However, these species
can only survive in the long-term when the overall condition of the ecosystems in
which they occur is maintained.
7.55. In order to reflect the multi-layered relation between biodiversity, ecosystem
functioning, ecosystem services and the human appreciation of ecosystems, a range
of biodiversity asset indicators should be considered. Species indicators may be
selected on the basis of the importance of species for specific ecosystem processes,
for being indicative of ecosystem quality or functioning, or because the species
represent specific aspects that people appreciate in biodiversity, such as the
occurrence or abundance of endemic and/or iconic species.
7.56. Integrated accounting for aspects of biodiversity is still developing and
experimentation on biodiversity accounting by countries is less advanced than for
water or carbon accounting. To advance work, a paper was commissioned as part of
the project - project research paper #10 (UNEP-WCMC, 2015). This section
summarizes the findings of that paper.
7.5.2 Why account for biodiversity?

7.57. To deliver sustainable development, national accounting systems need
information on the foundation for sustainable economic growth provided by
ecosystems and their services. Biodiversity accounting provides a methodology to
help understand the contribution of biodiversity (both ecosystem and species
diversity) to human well-being and the economy, by explicitly considering its role as
a determinant of the condition of ecosystem assets that are essential for the supply of
ecosystem services.
7.58. Biodiversity data is incorporated into the SEEA EEA framework via a
thematic, species diversity account and also in the ecosystem extent account that can
support analysis of ecosystem diversity. The species diversity account provides
information that can be used in the measurement of the condition of ecosystem assets.
The spatial nature of the accounts allows statistics on biodiversity to be examined
against economic and social statistics in a spatially consistent manner.
7.59. Notwithstanding the connection made here between biodiversity
measurement and the measurement of ecosystem assets, there are situations in which
measures of biodiversity can be indicators of flows of final ecosystem services. For
example, the value of recreational services from wildlife related activities, where
people gain benefit from experiencing the diversity of nature. In these circumstances,
it is relevant to recognise that measures related to biodiversity may be appropriate
indicators in a variety of accounts, including ecosystem condition accounts and
ecosystem services supply and use tables.
7.60. Biodiversity accounting also provides opportunities for the harmonization of
national level biodiversity data alongside other reporting mechanisms, such as the
CBD through the implementation of National Biodiversity Strategies and Action
Plans (NBSAPs) and reporting on the SDGs.
7.5.3 Assessing ecosystem and species diversity
86
7.61. Assessments of biodiversity generally consider ecosystem and species
diversity due to the cost and complexity of assessing genetic diversity. However, that
is not to say that genetic diversity is not important and could not be integrated into an
accounting framework in the future.
7.62. Ecosystem diversity may be assessed using information on ecosystem extent
as described in Chapter 6. Extent measures are based on data on land cover, land use,
habitat and other ecosystem data, commonly sourced from satellite remote sensing.
Within the SEEA EEA framework, these data inputs also provide spatial information
for delineating ecosystems (as assets) on the basis of common characteristics. Many
countries have their own ecosystem classification standards and methods for mapping
them, and work is progressing towards an internationally accepted ecosystem
classification.
7.63. Species diversity is the focus of this section, in part due to the availability of
data and also because species measures can be used to approximate the status of
biodiversity, in the form of biodiversity indicators. Development of a biodiversity
account using species data should move beyond simple counts of the number of
species (the species richness) and include the population size of each species (the
species abundance) as this provides more information on the status of species.
7.64. Developing measures of species diversity is resource intensive and has
methodological challenges. A complete inventory of a country’s species is not
possible and so species to be included in a biodiversity account will need to be
prioritized. When selecting species, the broader the representation of taxonomic
groups (e.g. plants, birds, mammals etc.), the better the account will estimate overall
biodiversity. In addition, some species (e.g. keystone species) are better indicators of
biodiversity and ecological condition than others.
7.5.4 Suitability of assessment approaches for biodiversity accounting

7.65. Biodiversity data needs to meet the following criteria to be suitable for
informing biodiversity accounting. It should:
 Be available at a spatial resolution suitable for accounting. This allows data to be
mapped to individual ecosystem units.
 Be temporally relevant. This informs net changes in the stock of biodiversity
between the opening and closing of accounting periods.
 Be comparable to a common reference condition. This allows the comparison of
biodiversity measures against a benchmark indicative of a balanced state and
aids aggregation of different types of biodiversity data.
 Be possible to aggregate the measures to provide a composite indicator of the
condition of biodiversity relevant to ecosystem functioning (e.g., via the
Simpson Index or aggregation using a common reference condition). The change
in this composite indicator between accounting periods provides an indication of
the net biodiversity balance.
 Be comparable over space and time. This allows direct comparison of
biodiversity stocks in different ecosystem types.
7.66. When measuring ecosystem diversity, remote sensing and associated
mapping of land cover, use, habitat or other characteristics can provide data that meet
most of these criteria for informing biodiversity accounting. To best support
integration, this work should be undertaken in the context of the spatial areas defined
for ecosystem accounting.
87
7.67. The International Union for the Conservation of Nature (IUCN) Red List of
Ecosystems (Keith, et al., 2013) will, in due course, meet these criteria by generating
measures of ecosystem condition based on risk of ecosystem collapse. The spatial
resolution will be high enough for national accounting (anticipated to be at least
equivalent to a 250m resolution. The first global assessment (scheduled for 2025) will
provide a baseline which may then be used as a reference condition. Assessments are
likely to be repeated on a 5-year basis. The application of the quantitative categories
and criteria will ensure consistency and comparability between countries and over
time.
7.68. In regard to organizing species diversity data, three main approaches are
noted here. First, the IUCN Red List of Threatened Species (IUCN, 2014) which
measures extinction risk. Application of the IUCN Red List categories and criteria
ensures consistency in assessment over space, over time, and between assessors.
While originally designed for global assessments, methods are available to allow
disaggregation of the Red List Index to national levels. Downscaling of the global
Red List to national levels can be complemented with national red lists, where these
exist. It is suggested that both the global Red List and national red lists are used to
ensure as broad and relevant a coverage as possible.
7.69. Second, the Norwegian Nature Index (NNI) (Certain and Skarpaas, 2010;
Alasken, et al., 2012) uses indicators from a variety of species groups and major
ecosystem types that measure deviation from a reference state. The NNI produces a
single ‘value’ that provides information on ecosystem condition. The methodology
involves a series of aggregations, first within spatial units, and then across spatial
units. The NNI incorporates expert judgment, monitoring-based estimates, and
model-based estimates, so the method can be used in both data rich and data poor
areas.
7.70. Third, the Living Planet Index (LPI) (WWF, 2014) aggregates species
population trend data from different sources and across multiple spatial scales. The
methodology involves a series of aggregations in order to avoid bias induced by
including only well-known taxonomic groups and well-studied locations. With
systematic monitoring of species abundance, the data lends itself for incorporation in
a biodiversity account. For those countries that lack this systematic data, the
methodology can still yield a single ‘value’ to support assessment of ecosystem
condition.
7.5.5 Implementing biodiversity accounting

7.71. A key starting point for biodiversity accounting is to identify biodiversity-
related policy priorities to help determine what information should be compiled,
covering plants, animals and to a lesser extent fungi. Guidance on the selection
process is provided below. This step will also establish the required resolution of data
(both spatial and temporal) necessary to address these priorities.
7.72. Establishing an inventory of all existing monitoring data will help identify
any ‘data-gaps’ for national biodiversity accounting. Countries should consider their
reporting obligations to regional processes and biodiversity-related
conventions/agreements, such as the CBD or the Ramsar Convention. Identifying
data gaps could inform a protocol for further data gathering (e.g., via monitoring or
modelling approaches).
7.73. An initial step in the biodiversity accounting process is to delineate
ecosystem assets spatially (e.g. by type of EU) on the basis of similarities in
ecological and ecosystem characteristics. This can be considered the foundation of a
biodiversity account in the broadest sense, as it provides information on ecosystem
88
diversity. Capturing further information on ecosystem diversity on the basis of
ecological variations within EU can improve these accounts further. Additional
biodiversity accounts can then be developed at the type of EU level. In order to
facilitate analysis and reporting, information on biodiversity can be aggregated across
EU to larger scales (e.g., across administrative boundaries or ecological features, such
as river basins). Testing is required on the characteristics and scales that are most
appropriate for the delineation of spatial areas for biodiversity accounting.
7.74. For the construction of ‘species accounts’ coverage should be based on
policy priorities. For example, a focus may be on economically important species
(e.g. game species) or species associated with ecosystem functioning (e.g. keystone
species).
7.75. More than one biodiversity account may be required in order to answer the
range of biodiversity relevant policy questions. For instance, information on
biodiversity relevant to ecosystem functioning may require a different accounting
structure than information on species extinction risk. In creating a species account,
analysts should consult with ecologists to ensure meaningful data is collected and
collated.
7.76. Building on these comments, a number of approaches to biodiversity
accounting may be followed that vary in complexity and resource requirements. They
are presented here as three tiers of accounts.
 ‘Tier 1’ accounts capture information on the ecosystem characteristics used
to define different types of EU (or important areas of biodiversity habitat)
and measure their extent. These measures can be weighted using input
indicators of species diversity.
 ‘Tier 2’ accounts capture information on species richness, extinction risk and
potentially other characteristics (e.g. species health) for ecosystem and
other spatial areas.
 ‘Tier 3’ accounts capture information on species abundance within ecosystem
units and other spatial areas.
7.77. While primary monitoring data is the ideal for assigning biodiversity
information to ecosystem units, this is unlikely to be available at the spatial resolution
required for ecosystem accounting. A number of approaches exist for upscaling or
downscaling data on biodiversity. These include habitat modelling, land use
modelling, species-area curves and expert judgment approaches. A portfolio of these
approaches will be required to inform biodiversity accounting. It is important
however, that any application of these approaches is supported by regular updating of
primary monitoring data.
7.5.6 Limitations and issues to resolve

7.78. In their present state, the majority of potential global datasets do not provide
the temporal or spatial resolution necessary to inform national biodiversity
accounting. Further, developing biodiversity accounts that are globally comparable is
likely to be challenging, particularly when relative measures of biodiversity are
employed. This is because a consistent reference condition is required.
7.79. While a single biodiversity indicator can provide an overall indication of
ecosystem condition, it is unlikely to be useful in informing the link to ecosystem
service supply. This is because there will be different aspects of biodiversity that will
be relevant to different ecosystem services. Consequently, a broad suite of
biodiversity indicators is likely to be required. For those species considered to
provide an ecosystem service in their own right, (e.g. for their existence or aesthetic
89
enjoyment) information in a species diversity account can inform ecosystem service
supply estimates directly.
7.80. Ultimately, the value of the contribution of biodiversity to ecosystem service
supply would be extremely useful to record in the ecosystem accounting framework.
There exist various market and non-market based valuation techniques to generate a
lower bound for this value. However, this will only be possible for a subset of
ecosystem services for which ecological production functions can be described.
7.5.7 Recommendations for testing, refining and validating

7.81. More testing is required of suitable spatial scales for biodiversity accounting.
This should be supported with further testing of the modelling and other approaches
for generating spatially explicit information on biodiversity via various downscaling
and upscaling approaches. A range of options has been presented in this section.
Protocols for validation and calibration of these approaches should also be explored.
7.82. Selecting the appropriate scale has significant implications for the
aggregation of biodiversity information. Thus, further research and testing of methods
to aggregate ecosystem and species data and condition indicators across ecosystem
units is required. This should consider the implications of ecotones (as areas of high
biodiversity on ecosystem borders) and the diversity among different ecosystem types
across larger spatial areas.
7.83. The asset accounts for biodiversity proposed in the SEEA EEA allow for
causes of addition and reduction in the stocks of species diversity to be recorded.
There are obvious benefits to establishing such a clear causal relationship. However,
completing these entries would require additional data collection and may often be
difficult to complete in a balanced manner. The possibilities for undertaking this
work would benefit from testing in a specific case study, possibly via linkages to land
ownership or land use. At this stage, it is recommended that countries focus on the
development of time series of biodiversity reflected as a sequence of opening and
closing positions.
7.84. Biodiversity is considered as an indicator of condition in the ecosystem
condition account. Improvements and reductions in condition are also recorded in the
condition account. However, there exist multiple drivers of biodiversity loss and so a
supplementary account for drivers of change in ecosystem condition could be a
possibility for testing. This would also provide a suitable structure for capturing
factors such as habitat fragmentation and invasive species.
7.85. The link between biodiversity and ecosystem service delivery is complex.
There will often be time lags between changes in biodiversity and changes to the
supply of ecosystem services. Furthermore, capturing information on the importance
of biodiversity to ecosystem functional redundancy and resilience is challenging due
to non-linear and threshold effects. Given the importance of biodiversity to ecosystem
functioning and sustaining ecosystem service provision, measurement of ecosystem
functional redundancy and resilience is a key issue to be addressed in the ecosystem
accounting framework. Further research is required in this regard, potentially via
research on the capacity of ecosystems to deliver services.
7.6 Other thematic accounts and data on drivers of ecosystem change
90
7.86. As noted in the introduction to this chapter, a wide range of data will need to
be integrated in the compilation of ecosystem accounts. Data on land cover, water,
carbon and biodiversity are likely to be relevant across many ecosystem types. Other
data areas, for which accounting frameworks have been developed in some cases,
include:
 Timber resources (accounting described in the SEEA Central
Framework)
 Fish and other aquatic resources (accounting described in the SEEA
Central Framework)
 Other biological resources including livestock, orchards, plantations,
wild animals (accounting described in the SEEA Central Framework)
 Soil resources (accounting described in the SEEA Central Framework
although much further development is required)
 Nutrient flows and balances for nitrogen and phosphorous (accounting
described in the SEEA Agriculture, Forestry and Fisheries (FAO,
forthcoming) and in OECD/Eurostat manuals (e.g. Eurostat and OECD,
2013))
 GHG emissions and residual flows (e.g. solid waste, wastewater)
(accounting described in the SEEA Central Framework)
 Data on production and use of outputs from agricultural, forestry and
fisheries activity (accounting described in the SEEA Agriculture,
Forestry and Fisheries (FAO, forthcoming))
 Data on tourism and recreation (some coverage of accounting in Tourism
Satellite Accounts) (UN et al., 2010)
 Population data.
7.87. In other contexts some of these data are considered indicators of “drivers” of
changes in ecosystem condition and the supply of ecosystem services. That is, many
of these types of data point to the changing extent of human interaction with the
environment. Information on drivers is likely to be of particular relevance in (i)
understanding changes in condition for specific ecosystems; (ii) developing
appropriate assumptions about future flows of ecosystem services; (iii) assessing
ecosystem capacity; and (iv) valuing ecosystem assets.
7.88. Particular note is made here on the relevance of accounting for GHG
emissions and other residual flows such as solid waste. These flows are not
ecosystem services within the ecosystem accounting model but given the potential
negative impact of these flows on environmental condition there may be significant
interest in how residual flows may be incorporated. In practice, the most
straightforward first step would be presenting information on flows of emissions and
residual flows by type of economic unit by spatial area where possible, alongside
information on changes in environmental condition for the same areas. Subsequently,
analysis may be able to determine linkages between changes in condition and the
residual flows and the associated economic units.
7.89. A more complete integration of residual flows into the ecosystem accounting
model would require a more complete understanding of dependencies between
ecosystems. In particular, it would require incorporation of the atmosphere as a type
of spatial “area”, whose condition is affected by economic activity, including for
example, forest fires. Where there is a decline in condition then it would be possible,
within the ecosystem accounting model, to assess the effects on flows of ecosystem
services and other environmental services, such as the provision of clean air space for
air transport. The extension described here will however, require much further
consideration.
91
7.90. It is likely that, in order to generate the data at the appropriate spatial scale
for ecosystem accounting, some scaling and modelling of the information covered by
the accounts listed above will be required.
7.91. Further, particularly for the measurement of ecosystem services, it will be
necessary to use models of ecosystem processes to estimate the relevant flows. These
models will require additional data, usually of a scientific and ecological nature. Over
time, as the accounts develop, it is likely to be possible to investigate the alignment
and consistency between the scientific data and the socio-economic data, particularly
as it pertains to specific spatial areas or ecosystems. In this sense, the ecosystem
accounting model provides both a rationale and a platform for data integration.
92
8. Valuation in ecosystem accounting

The estimation of monetary values for ecosystem services and ecosystem assets can be
undertaken for a variety of purposes. It is essential that the purpose of valuation is well
understood.
In ecosystem accounting, the primary purpose of valuation is the integration of ecosystem
accounting information with information in the standard national accounts. For this purpose
the valuation used in ecosystem accounting needs to be consistent with the valuation concept
used in the national accounts.
The purpose of integration leads to the application of the valuation concept of exchange
values – i.e. those values that reflect the price at which ecosystem services and ecosystem
assets would be exchanged between buyer and seller if a market existed.
The use of the exchange value concept implies that some valuation techniques commonly
used in the valuation of ecosystem services are not appropriate. However, quite a number of
techniques are relevant or may be adapted for use in accounting.
The focus on exchange values is not intended to suggest that valuation for other purposes is
not appropriate. For example, valuation that takes into account changes in welfare is central to
much economic policy and analysis. However, such values should not be used directly in
accounting.
In ecosystem accounting, the valuation of ecosystem services is the starting point for the
valuation of ecosystem assets. A clear distinction should be made between these to objects of
valuation.
Further testing on valuation methods is required, especially in the context of ecosystem
accounting.
Research is needed to understand the connection between exchange and welfare based
valuations and the extent to which different valuation techniques may be adapted to estimate
exchange values.
8.1 Introduction
8.1. The issue of valuation can complicate the discussion of ecosystem and
natural capital accounting. This occurs for many reasons. For some, the concerns
about valuation relate to the implication that a “dollar value” is placed on all
environmental assets and services and that this is both inappropriate and misleading.
For others, the measurement concerns are too great and the environment is considered
too complex to consider that useful measures in monetary terms might be compiled.
Finally, there are differences concerning the purposes, concepts and techniques in
relation to monetary valuation.
8.2. As in SEEA EEA chapter 5, the ambition in this chapter is to provide a
possible pathway through these various issues, such that the discussion on valuation
in the context of ecosystem accounting can be undertaken in as an informed way as
possible.
8.3. One general conclusion is that valuation in monetary terms requires careful
consideration of the purpose of the valuation. Alternative purposes include
93
accounting purposes and the assessment of changes in welfare between alternative
scenarios. Once the purpose is defined, the appropriate valuation concept can be
selected and from this, the relevant valuation method and technique can be applied.
Often the discussion in environmental valuation moves directly to discussion of
method and technique without recognising that different purposes for valuation, and
hence different concepts may be relevant.
8.4. A fundamental aspect of valuation in an accounting context is that the first
step is the valuation of individual ecosystem services. In general, this will require
finding an appropriate price to apply to an imputed exchange of ecosystem services
between a given ecosystem asset (e.g. a forest) and an economic unit or individual
(e.g. a forester).
8.5. Valuing ecosystem assets requires considering the future flows of ecosystem
services that are expected to be supplied by the ecosystem asset. Generally, this will
mean that a basket of ecosystem services needs to be assumed and priced, with the
value of the ecosystem asset then equal to the net present value of the future flows of
expected ecosystem services. In general, information on the current uses of the
ecosystem and the current basket of services supplied provides the starting point for
establishing the expected flow of ecosystem services. Recognising the steps that are
required to move from the valuation of ecosystem services to the valuation of
ecosystem assets is important in making decisions about the implementation of
8.6. This chapter is structured in the following way. In section 8.2 the main
valuation principles for ecosystem accounting are outlined drawing out the key points
from the material presented in SEEA EEA chapter 5. In section 8.3 the key
challenges in valuation are described. Section 8.4 considers relevant data and source
materials. The final section provides a summary of recommendations in relation to
valuation based on current practice and knowledge and a summary of the key issues
requiring further research. Issues related to the valuation of ecosystem assets and the
valuation of ecosystem degradation are discussed in Chapter 9.
8.2 Valuation principles for ecosystem accounting

8.7. SEEA EEA recognises that the term valuation can mean different things. For
accountants and economists, valuation is almost always used in the context of placing
a monetary price (dollar value) on assets, goods or services. In other contexts,
valuation may refer to a more general notion of recognising significance or
importance. In SEEA EEA, the focus is on valuation in monetary terms but this is not
to discount the role or importance of other concepts of value. (A useful introduction
to a way in which non-monetary valuation may be conducted is described in Maynard
et al, 2014)
8.8. Monetary valuation in the SEEA EEA is applied to the valuation of
ecosystem services and the valuation of ecosystem assets. There is a direct
connection made between these two distinct targets of valuation whereby the value of
ecosystem assets at any point in time, usually the date to which the balance sheet
relates, is equal to the NPV of the future flows of ecosystem services that are
expected to occur. The application of the NPV technique (explained at length in the
SEEA Central Framework Chapter 5) is required since there are no markets that exist
in the buying and selling of ecosystem assets in such a way that the value of all
ecosystem services is captured.
94
8.9. From a practical perspective, the need to apply NPV techniques to value
ecosystem assets implies that the valuation of ecosystem assets cannot be determined
directly. Instead, the asset value relies on the estimation of the value of ecosystem
services. Thus, in an accounting context, the valuation of ecosystem services and the
valuation of ecosystem assets are distinct but related tasks.
8.10. The relevant valuation concept for ecosystem accounting is exchange values.
If there were observable markets in individual ecosystem services, an exchange value
would reflect the prices paid by consumers of ecosystem services to the relevant
producers (i.e. the ecosystem assets). Since transactions with ecosystems are not
observable, these exchange values must be estimated using one of a variety of
valuation techniques.
8.11. Some ecosystem services can be reasonably closely connected to activities in
markets. This is generally the case for provisioning services where ecosystem
services that contribute to the production of food, fibre, fuel and energy can be valued
using prices for the relevant benefits. Here a close connection can be made to the
values used in the SNA to estimate production and consumption. One description of
these types of ecosystem services is that they their values are “near-market”
(Nordhaus, 2005).
8.12. On the other hand, some ecosystem services contribute to benefits that are
not closely connected to existing markets – so called “far-market” services. Often
these are ecosystem services that may be considered to provide public goods – the
contribution of ecosystems to flood protection is one example. In these cases,
determining valuation techniques is more complicated.
8.13. Some techniques used to value ecosystem services do not reflect only the
value of the exchange but also incorporate the welfare effects that can arise to the
consumer of the ecosystem service. For example, the value of water abstracted from a
river might be increased if one also incorporated the positive effect that consuming
water had on health and subsequently labour productivity.
8.14. While values that incorporate welfare effects may be very useful for
assessing differences between available choices, these welfare values are not of direct
applicability in accounting contexts. Consequently, in the selection of non-market
valuation techniques, if the objective is ecosystem accounting, then techniques must
be found that estimate only the exchange value. At the same time, it is possible within
the same measurement framework, to estimate prices based on either exchange or
welfare valuation concepts. They should be considered alternative rather than
competing valuations.
8.3 Relevant data and source materials

8.15. The SEEA EEA Chapter 5 suggests a logic in the valuation process such that
the first step is to determine the purpose of valuation, with ecosystem accounting
being one among a number of purposes. Based on the purpose, the appropriate
valuation concept can be determined. For ecosystem accounting, the exchange value
concept is appropriate. Finally, knowing the concept, a choice can be made between
various valuation techniques such that the exchange value concept can be consistently
applied across different ecosystem services.
8.16. A number of valuation techniques have been considered appropriate for
measuring exchange values although further discussion on this topic is required as it
has generally not been a focus on the ecosystem services valuation literature. The
95
SEEA EEA Chapter 5 outlines a number of the approaches and an updated summary
of valuation techniques is provided in Table 8.1.
8.17. In terms of implementation, valuation exercises generally require, in the first
instance, estimation of physical flows of ecosystem services. These flows are then
multiplied by a relevant price in order to estimate the value of the flows. Information
on physical flows of ecosystem services is thus of direct relevance.
8.18. In terms of estimating prices, usually it is necessary to find studies that have
estimated a price for the relevant ecosystem service in a particular ecosystem type.
There are a number of databases that hold relevant studies, including the Ecosystem
Services Valuation Database (ESVD) that has built on the original work of the TEEB
study, the Environmental Valuation Reference Inventory (EVRI) database, and the
Ecosystem Valuation Toolkit by Earth Economics. A useful link to these and other
valuation databases is on the Ecosystem Services Partnership website (see
http://www.fsd.nl/esp/80136/5/0/50).
8.19. Additional support for applying valuation in national accounting contexts can
be found in materials from the UNEP Ecosystem Services Economics unit, the
materials developed as part of the TEEB study, work being undertaken within the
World Bank WAVES project and the discussion of valuation in the context of the
IPBES.
8.20. It is noted however, that generally these materials are not explicit about the
valuation concept being applied. Hence, it is often unclear as to whether the
approaches and recommendations are suitable for ecosystem accounting purposes in
terms of measuring exchange values. In particular, since much of the work on
environmental valuation has been led by environmental economists, the focus is often
on the valuation of externalities (and associated shadow prices) or the estimation of
welfare effects between alternative scenarios. More work is required to understand
further how these approaches can support valuation in an accounting context.
Nonetheless, in conjunction with the discussions in SEEA EEA Chapter 5, these
materials should provide a reasonable base for investigating the valuation of
ecosystem services at national level.
Table 8.1 Summary of valuation methods and their use in ecosystem accounting
Valuation method Description Comments Suitability for
ecosystem accounting
Unit resource rent Prices determined by Estimates will be In principle this
deducting costs of affected by the method is appropriate
labour, produced assets property rights and but consideration of
and intermediate inputs market structures market structures is
from market price of surrounding required.
outputs (benefits). production. For
example, open access
fisheries and markets
for water supply often
generate low or zero
rents.
Production function, Prices obtained by In principle analogous Appropriate provided
cost function and profit determining the to resource rent but the market based price
function methods contribution of the generally focused on being decomposed
ecosystem to a market the valuation of refers to a product
based price using an regulating services. rather than an asset –
assumed production, May be difficult to e.g. value of housing
cost or profit function. estimate the functions. services rather than the
value of a house.
96
Payment for Ecosystem Prices are obtained Estimates will be Possibly appropriate
Services (PES) from markets for affected by the type of depending on the
schemes specific regulating market structures put in nature of the market
services (e.g. in place for each PES (see structures.
relation to carbon SEEA EEA 5.88-94)
sequestration)
Hedonic pricing Prices are estimated by Very data intensive Appropriate in
decomposing the value approach and principle. Heavily used
of an asset (e.g. a house separating out the in the pricing of
block including the effects of different computers in the
dwelling and the land) characteristics may be national accounts.
into its characteristics difficult, unless there
and pricing each are large sample sizes.
characteristic through
regression analysis
Replacement cost Prices reflect the This method requires Appropriate under the
estimated cost of an understanding of the assumptions (i) that the
replacing a specific ecosystem function estimation of the costs
ecosystem services underpinning the reflects the ecosystem
using produced assets supply of the service services being lost; (ii)
and associated inputs. and an ability to find a that it is a least-cost
comparable “produced” treatment; and (iii) that
method of supplying it would be expected
the same service. that society would
replace the service if it
was removed.
(Assumption (iii) may
be tested using stated
preference methods.)
Damage costs avoided Prices are estimated in May be challenging to Appropriate under the
terms of the value of determine the value of assumptions (i) that the
production losses or the contribution/impact estimation of the
damages that would of an individual damage costs reflects
occur if the ecosystem ecosystem service. the specific ecosystem
services were reduced services being lost; (ii)
or lost due to that the services
ecosystem changes continued to be
(e.g. as a result of demanded; and (iii)
pollution of that the estimated
waterways). damage costs are lower
than potential costs of
abatement or
replacement.
Averting behaviour Prices are estimated Requires an Likely inappropriate
based on individuals understanding of since it relies on
willingness to pay for individual preferences individuals being aware
improved or avoided and may be difficult to of the impacts arising
health outcomes. link the activity of the from environmental
individual to a specific changes.
ecosystem service.
Restoration cost Refers to the estimated The main issue here is Inappropriate since it
cost to restore an that the costs relate to a does not determine a
ecosystem asset to an basket of ecosystem price for an individual
earlier, benchmark services rather than a ecosystem service.
condition. specific one. More
Should be clearly often used as a means
distinguished from the to estimate ecosystem
replacement cost degradation but there
method. are issues in its
application in this
97
context also.
Travel cost Estimates reflect the Key challenge here is Possibly appropriate
price that consumers determining the actual depending on the actual
are willing to pay in contribution of the estimation techniques
relation to visits to ecosystem to the total and whether the
recreational sites. estimated willingness approach provides an
to pay. There are also exchange value, i.e.
many applications of excludes consumer
this method with surplus.
varying assumptions
and techniques being
used with a common
objective of estimating
consumer surplus.
Finally, some travel
cost methods include a
value of time taken by
the household which
would be considered
outside the scope of the
production boundary
used for accounting
purposes.
Stated preference Prices reflect These approaches are Inappropriate since
willingness to pay from generally used to does not measure
either contingent estimate consumer exchange values
valuation studies or surplus and welfare
choice modelling. effects. Within the
range of techniques
used there can be
potential biases that
should be taken into
account.
Marginal values from Prices are estimated by This method can use Appropriate since
revealed demand utilising an appropriate demand functions aims to directly
functions demand function and estimated through measure exchange
setting the price as a travel cost, state values. However, the
point on that function preference, or averting creation of meaningful
using (i) observed behaviour methods. demand functions and
behaviour to reflect The use of supply estimating hypothetical
supply (e.g. visits to functions has been markets may be
parks) or (ii) modelling termed the simulation challenging.
a supply function. exchange method
(Campos & Caparros,
2011)
8.4 Key challenges in valuation

8.21. There is a wide range of challenges in valuation. The following section
describes those that may be most commonly confronted.
8.22. The target of valuation. In the SEEA EEA, the ecosystem accounting model
(see chapter 2) has a clear distinction between ecosystem services and the benefits to
which they contribute. Particularly for provisioning services, it is not uncommon for
the market price of the extracted good (e.g. fish caught or timber harvested) to be
considered equivalent to the price of the ecosystem service. In fact, the market price
reflects the value of the benefit and estimating the appropriate price for the associated
98
ecosystem service must deduct the costs of extraction and harvest, thus leaving a
residual that reflects the ecosystem contribution.
8.23. By way of example, in the case of timber harvest there will commonly be a
price for the logged timber – perhaps a roadside price. This price should be sufficient
to cover the costs of felling (labour, fuel, equipment, etc.). It should also cover any
payments that are made to the owner of the forest for the right to harvest the timber.
These are commonly referred to as stumpage prices. The price of the ecosystem
services in this case is not the roadside price after felling, but the stumpage price,
equivalent to the residual after deducting the costs of extraction. In effect the forest
owner, in setting a stumpage price, is setting a price on the ecosystem services.
8.24. Unfortunately, in some cases, this residual may be very small or negative (for
example, in the case of abstracted water or open access fishing). Consequently, the
implied price of the ecosystem service is very low, zero or negative. A number of
different cases can be identified. For example in the case of water the resource rent is
often near zero as there is commonly no competitive market for distributed water. In
the case of open access fishing the lack of defined property rights is the key driver. In
recreational hunting the costs are often higher than the potential sale price of the
game meat but this will reflect the recreational value of the activity.
8.25. Depending on the situation different valuation approaches may be possible,
for example using replacement costs in the case of water (Remme et al. 2014) or
hunting costs in the case of recreational hunting (Remme et al. 2014). Most
problematic is determining an approach in the case of common pool / open access
resources. Note that the benefits produced in these instances (e.g. fish or water) still
have market prices, but the ecosystem services are implicitly valued at near zero. A
clear resolution of this matter is required since the value obtained using residual or
resource rent techniques do not seem to not reflect the broadening of the production
boundary that underpins the ecosystem accounting approach.
8.26. A second aspect concerning the target of valuation is the distinction between
the valuation of ecosystem services and the valuation of ecosystem assets (and the
related issue of valuing ecosystem degradation). Within ecosystem accounting, the
valuation of ecosystem assets reflects the overall value of a given spatial area and is
estimated by aggregating the net present value of all relevant ecosystem services.
These issues are discussed in Chapter 9.
8.27. Valuation of intermediate services. The focus of valuation in the SEEA EEA,
and in the majority of other studies, is on final ecosystem services. This focus
supports understanding the interactions with beneficiaries (economic units including
households). However, if the valuations of final ecosystem services are attributed to
specific ecosystems it may imply that ecosystem assets that supply final ecosystem
services have a particularly high value, relative to ecosystem assets that do not supply
final services directly. Where there are important dependencies between ecosystem
assets in the supply of a final ecosystem service, ignoring the value of intermediate
services may provide misleading information on the relative importance of certain
ecosystem assets. The description and measurement of ecological production
functions is an important pathway in understanding these connections.
8.28. Consistency in the use of valuation concepts and techniques. For ecosystem
accounting, since the ultimate objective in valuation is the integration of data with the
standard national accounts, it is essential to use a valuation concept that is consistent
with the accounts. SEEA EEA describes the appropriate concept as exchange values,
i.e. the prices that arise at the time of exchange between buyer and seller. If exchange
values are not used to estimate the value of ecosystem services, then there will be no
consistent integration with values in the standard national accounts.
99
8.29. It is quite reasonable and at times necessary to compile estimates using
alternative valuation concepts. Thus, welfare valuations are highly relevant when
comparing alternative scenarios. However, these valuations should not be directly
compared with the value of other assets from the national accounts since the
underlying valuation concept is different.
8.30. The use of a consistent valuation concept does not imply that the same
estimation technique must be applied in all circumstances. Indeed, a variety of
different techniques are likely to be required to cover the range of situations and the
different types of ecosystem services.
8.31. Scaling and aggregation. Often, studies on the valuation of ecosystem
services are undertaken for specific ecosystem services in specific ecosystems. A
significant challenge from an ecosystem accounting perspective is therefore
translating these “point” estimates into information that can be applied at broader
scales. This challenge is generally considered under the banner of “benefit transfer”.
A range of techniques have been developed some of which are considered more
refined and appropriate than others.
8.32. Valuation of regulating services. For most provisioning services there is a
connection to market values of benefits that can provide a base for measurement. This
is also true for some – but by no means all - cultural services (such as those relating
to economic activity in tourism and recreation). However, in the area of regulating
services such connections to marketed benefits are unusual. Indeed, for regulating
services it can be difficult to appropriately define and measure the actual physical
flow of the service because often the service is simply part of ongoing ecosystem
processes rather than a function of direct human activity – for example, air filtration
and carbon sequestration.
8.33. The measurement of non-use values. An important part of the value of
ecosystems from a societal perspective can lie in the non-use values that, in principle,
are captured in various cultural services provided by ecosystem assets. These values
include existence values (based on the utility derived from knowing that an
ecosystem exists); altruistic values (based on the utility derived from knowing that
someone else is benefiting from the ecosystem) and bequest values (based on the
utility derived from knowing that the ecosystem may be used by future generations).
At this point, there are relatively few studies in this area of valuation from the
perspective of ecosystem services. Further, there is an open question of the extent to
which non-use values can be captured within an exchange value concept.
8.34. The valuation of ecosystem assets with respect to land. In estimating the
value of ecosystem assets at exchange values, one important consideration is the
value of land that is commonly traded in markets – including, for example,
agricultural land. Depending on the circumstance, values of land will incorporate the
value of some ecosystem services. However, they are unlikely to capture the value of
all of the ecosystem services particularly those that are of a public good nature.
Further, market based land values will incorporate elements of value that are not
dependent on ecosystems, such as the prospects for property development or the
capitalisation of farm subsidies. Consequently, when considering the integration of
ecosystem asset valuations into existing national accounts balance sheets, some
adjustments will be required to ensure there is no double counting or gaps in
valuation for the estimation of total net wealth.
8.35. The valuation of biodiversity and resilience. Biodiversity and resilience are
considered in SEEA EEA more as characteristics of ecosystem assets and not as
ecosystem services. Consequently, they are not separately valued using the general
approach outlined here and the relative contribution of biodiversity and resilience is
100
unlikely to be identifiable. Further consideration on how these aspects of ecosystem
may be valued is required.
8.36. Uncertainty in measurement. While there is always uncertainty in
measurement, the valuation of ecosystem services tends to bring together a number of
uncertainties into one place. SEEA EEA (section 5.6.4) explains these uncertainties in
more depth; here they are simply listed: (i) uncertainty related to the measurement of
ecosystem services and ecosystem assets in physical terms; (ii) uncertainty in the
valuation of ecosystem services and assets; (iii) uncertainty related to the dynamics of
ecosystems and changes in flows of ecosystem services; and (iv) uncertainty
regarding future prices and values of ecosystem services.
8.5 Recommendations
8.37. There remains a substantial amount of work to be conducted to advance
valuation in the context of ecosystem accounting. At one level there is a need to
continue the discussion about the role of valuation both in general terms and with
respect to accounting. The main challenge is to provide the appropriate context for
the discussion since commonly there are many misunderstandings of the relevant
points. A key issue is understanding the different purposes of valuation and the types
of questions that may, or may not, be supported using information in monetary terms.
8.38. At a second level, there is a need for understanding and explaining the
concept of exchange values for accounting purposes and the development, or
adjustment of, valuation techniques to support the estimation of this valuation
concept. A possible path forward on this is to distinguish better between the relevant
valuation techniques as to: (i) when the ecosystem services can be relatively easily
linked to existing market prices and (ii) when the ecosystem services relate to public
goods. These two principal approaches involve different challenges.
8.39. In the case of ecosystem services that can be relatively easily linked to
market prices, an important part of the information required for valuing the service
may already be in the national accounts. This may be the case for provisioning
services or tourism. For these services, valuation serves to specify the contribution of
the ecosystem to the related benefits included in the national accounts in monetary
terms. Following Table 8.1, for such services a unit resource rent-based valuation
approach may in many cases be appropriate. The SNA (2008) provides detailed
guidance on how intermediate inputs, labour and fixed capital should be costed.
8.40. The SEEA-EEA approach involves the combination of tabular and mapped
information. Producing maps for ecosystem services that are valued based on
information that is in the national accounts generally involves spatial allocation. In
some cases this is straightforward, as in the case of forests providing the opportunity
to harvest timber to a logging company. In other cases, some modelling of spatial
interactions between ecosystem users and the ecosystem is required, as in the case of
allocating the resource rent generated in the tourism sector to (natural) ecosystems.
8.41. In the case of public services, including most regulating services, that are not
captured in the national accounts spatial, physical models for the ecosystem services
involved provide the basis for valuation. Significant uncertainty pertains to both the
physical models (as discussed in Chapters 5 and 6) and the unit values for the
ecosystem services. Different regulating services require different valuation methods.
Replacement costs methods can be applied, based on the least-cost alternative, if it
can reasonably be assumed that the service would indeed be replaced if lost. This
method is relevant, for instance, for the flood protection service of coastal or riparian
101
ecosystems in densely populated areas. In case it cannot be assumed that the service
would be replaced, an avoided damage cost method may be appropriate (see Table
8.1). Hedonic pricing is another valuation method with which there is ample
experience. It can be used to value for example the amenity service, as in the case of
eliciting the incremental value of houses with a view or close to a greenspace.
8.42. Other valuation methods have as yet been less frequently applied, but offer
the potential to broaden the pallet of valuation methods available for SEEA-EEA. For
instance, the Simulated Exchange Value approach and the Travel Cost Method can be
used to reveal demand curves for tourism and recreation which would allow a more
comprehensive inclusion of tourism and recreation in the ecosystem accounts. Well-
functioning Payment Schemes for Ecosystem Services may indicate partial market
equilibrium demand and supply for ecosystem services, and associated market prices,
but it needs to be examined under what conditions the prices paid for ecosystem
services in a PES truly reflect exchange values.
8.43. In general, there is a need for further efforts to estimate exchange values of
ecosystem services in practice, for a basket of ecosystem services, at a broad, macro
scale. There are some examples of work heading in this direction (see for example
Remme et al., 2014; Sumarga and Hein, 2013) but more testing is required. In some
cases (hedonic pricing, replacement costs, avoided damage costs) there is ample
experience in the environmental economics literature that can be built upon, in other
cases, e.g. simulated exchange values and using the travel cost method and prices
from PES schemes in the context of accounting, there is a need for further research
before such valuation approaches can become standardized.
102
9. Integrating ecosystem accounting with standard economic data

Full integration of ecosystem accounting information with the standard national accounts
comprises many steps and is the end point of measurement of ecosystem condition and
ecosystem services in physical and monetary terms.
Four broad types of integration can be described: combined presentations, extended supply
and use tables, institutional sector accounts and balance sheets. Each of these types of
integration provides information suited to answering different policy and analytical questions.
Combined presentations bring together information on ecosystems and the economy without
requiring the estimation of ecosystem services and assets in monetary terms. For example,
information of physical flows of ecosystem services to agriculture can be compared to
agricultural value added from the national accounts.
Extended supply and use tables support the analysis of extended supply chains and the
integration of ecological production functions.
Institutional sector accounts provide the means by which standard aggregates of income and
production can be adjusted for ecosystem degradation – i.e. the cost of using up ecosystem
capital.
Balance sheets provide the framework for extended measures of wealth incorporating the
value of the full range of ecosystem services; standard economic accounts only incorporate
values related to provisioning services.
Aside from combined presentations, the other types of integration require ecosystem data to
be estimated in monetary terms. Thus the measurement challenges outlined in chapter 8
apply.
In addition, there are some specific measurement issues concerning the valuation and
attribution of ecosystem degradation and the application of net present value techniques in the
measurement of ecosystem assets.
There are other approaches to the integration of ecosystem and economic data, including
wealth accounting and full cost accounting. These have a similar intent to the SEEA based
approaches but apply some different measurement concepts and boundaries.
9.1 Introduction
9.1. The integration of ecosystem accounting information with standard economic
data is a key driver for work within the context of the SEEA. This reflects that the
SEEA has been developed as a system that extends and complements the standard
economic accounts of the SNA. Indeed, for some, the prime ambition of developing
the SEEA is deriving adjusted measures of national income and economic activity
that take into account environmental information, for example in the form of
depletion or degradation adjusted measures of GDP.
9.2. The reality that emerges from the development and testing of the SEEA EEA
is that calculating adjustments to national income for ecosystem degradation cannot
be regarded as straightforward or direct. Indeed, what has emerged in recent years is
the need to consider a series of issues as outlined in the SEEA EEA and in these
Technical Recommendations. These issues concern spatial units, scaling and
103
aggregation, ecosystem services, ecosystem condition, ecosystem capacity and
valuation.
9.3. As a result, while a theoretical framework for integrated accounting of
ecosystems and economic activity is largely in place, its implementation represents
the end point of a series of compilation steps (described in section 9.2) and also
requires a range of assumptions on the nature of the required valuation and
integration. Compilers should recognise that some of these accounting matters remain
the subject of ongoing discussion.
9.4. While the ambition to complete a full integration of ecosystem accounting
information continues, it is important to recognise that there are various means by
which ecosystem accounting data can be combined with economic data. Section 9.3
describes the use of combined presentations that are valuable in this context.
9.5. This chapter builds on the text provided in SEEA EEA Chapter 6 and
summarises some of the key points in integrating ecosystem accounting data with
standard economic data.
9.2 Steps required for full integration with the national accounts
9.6. Historically, the approach to integrating ecosystem related information with
the national accounts has moved directly to the question of the valuation of
degradation and the appropriate recording and allocation of degradation in the
accounts. This is characteristic of the approaches outlined by national accountants
(see for example, Harrison 1993 and Vanoli 1995). However, the question of exactly
how the integration should be undertaken has never been fully resolved.
9.7. As explained in SEEA EEA and also in recent literature (e.g. Edens and
Hein, 2013; Obst et al, 2015) the emergence of the concept of ecosystem services has
allowed a reconceptualization of the integration with the national accounts. It is this
new basis for integration that is inherent in the SEEA EEA.
9.8. Through the concept of ecosystem services the following (generalised) steps
toward full integration emerge. The precise ordering of these steps will vary in
practice, and iteration between the steps is to be expected.
i. Delineate the relevant spatial areas to create mutually exclusive
ecosystem assets
ii. Identify and measure the supply of ecosystem services from each
ecosystem asset and determine the relevant beneficiaries
iii. Measure the condition of each ecosystem asset
iv. Assess the future flows of ecosystem services from each ecosystem
asset based on consideration of the current and future condition and
capacity of ecosystem assets
v. Estimate the monetary value of all ecosystem services
vi. Estimate the net present value of the future flows of each ecosystem
service and aggregate to provide a point in time estimate of the
monetary value of each ecosystem asset
vii. Estimate the change in net present value over an accounting period
and determine the monetary value of ecosystem degradation
viii. Integrate values of the production and consumption of ecosystem
services, the value of ecosystem degradation and the value of
ecosystem assets into the standard economic accounts.
104
9.9. It is clear from this list, which itself is somewhat stylised, that the full
integration of ecosystem accounting information into the standard national accounts
(step viii) is not straightforward. At the same time, maintaining a longer-term
objective of integration gives a clear purpose and rationale for the selection and
structuring of the ecosystem information that is required in the early phases. Further,
the information organised in the early phases is likely to be of direct usefulness for
decision making and monitoring in its own right. Consequently, while the objective
of full integration may be challenging, it plays an important part in providing
direction for ecosystem accounting.
9.10. A significant challenge in working through these eight steps, is the
requirement for aggregation across ecosystem services and ecosystem assets.
Aggregation requires a range of assumptions about the relationships between
different ecosystem services and different ecosystem assets. In particular, there is
often an implicit assumption, that separate estimates for different services and assets
can be summed. The reality is that such a summation will tend to abstract, to some
degree, from the inherent complexity of the underlying ecosystem functions and
processes. (In the same way as the national accounts is an abstraction of the
underlying economic system.) The question for compilers and analysts is whether the
abstraction that is represented in ecosystem accounts is appropriate in terms of
making better informed decision on the use and management of ecosystems.
9.11. The measurement issues relating to the initial steps outlined above have been
described in earlier chapters in these Technical Recommendations. This chapter
discusses measurement issues related to steps vi to viii. It is important to recognise
that the content of this chapter is largely in the realm of ongoing research and at this
stage full integration of ecosystem accounts with the standard national accounts is
likely to be a medium to longer term objective at national level.
9.3 The role of combined presentations

9.12. A more immediate means of combining the information from ecosystem
accounting with the standard national accounts is by means of combined
presentations.
9.13. Combined presentations are described in the SEEA Central Framework
Chapter 6. In essence, they are tables that support the presentation of information
from a variety of sources in a manner that facilitates comparison between economic
and environmental data. This is achieved by use of common classifications and
accounting principles.
9.14. Two examples with respect to ecosystem accounting are (i) the provision of
information for specific ecosystem assets (spatial areas) regarding changes in
condition of the asset combined with information on the expenditure on
environmental protection on those assets; and (ii) information on the flows of
ecosystem services generated by an ecosystem asset combined with information on
economic activity associated with that asset. Examples in this second case would be
showing data on flows of ecosystem services from a forest alongside data on
employment in the forestry industry or comparison of agriculture related ecosystem
services to agricultural income. Such comparison may give an insight into the relative
significance of ecosystem flows to various beneficiaries.
9.15. Specifically in relation to ecosystem degradation, a common link is made to
potential restoration costs. That is, for a given decline in ecosystem condition what
would be the costs involved in restoring the ecosystem to the original condition.
105
Information on restoration costs is likely to be of particular relevance in the
management of ecosystems and in understanding the degree of investment in
ecosystems that might be needed to maintain or improve condition.
9.16. Over time, as information is gathered on the actual expenditure on restoring
ecosystem assets, this may be complemented with information on flows of ecosystem
services, and a more complete picture of the relationships between ecosystem
condition and ecosystem services should emerge. Indeed, one of the key roles of the
ecosystem accounting model is to facilitate the organisation of information of this
type and thus support more detailed analysis in the future.
9.17. SEEA EEA Chapter 6 provides some additional comments in relation to
combined presentations. The key point is that there is considerable flexibility in the
design of combined presentations. While they do not represent a full integration of
information in accounting terms, they may support a more informed discussion of the
relationship between ecosystems and economic activity. Further, they may help
underpin the presentation of indicators for monitoring trends in ecosystem related
outcomes.
9.4 Integrated ecosystem-economic accounting structures

9.18. Chapter 4 introduced three types of integrated accounts in the context of the
broad suite of ecosystem accounts. In this section, those three types of accounts –
extended ecosystem supply and use tables in monetary terms, the full sequence of
institutional sector accounts and balance sheets are described in more detail.
9.4.1 Extended supply and use tables (SUT)

9.19. Extended supply and use tables (SUT) represent the first account in which
explicit consideration must be given to the boundaries between the current economic
measures and measures of ecosystem services in terms of the structure of the
accounts. The ambition in the augmented SUT is to present the information on the
supply and use of ecosystem services as extensions to the standard national accounts
SUT table.
9.20. Building on the discussion in Chapter 4 concerning ecosystem services
supply and use tables, and as reflected in Table 4.4, there are two key aspects to this
extension. First, recalling that the ecosystem accounting model implies an extension
to the standard production boundary, the set of products within scope of the SUT
table is broader and hence the size of the SUT must increase. This can be done
through the addition of new rows (representing the ecosystem services).
9.21. The requirement here is to ensure that these ecosystem services are
distinguished clearly from the products that are already within the standard SUT – i.e.
the SNA benefits. For these benefits, the relevant ecosystem services represent the
intermediate consumption of the producers of the SNA benefits. For ecosystem
services that contribute to non-SNA benefits, then rows for both the ecosystem
services and the new benefits need to be incorporated.
9.22. Conceptually, it is possible to extend the SUT further to incorporate both
final and intermediate ecosystem services. However, it is recommended that the
extension be limited to final ecosystem services. In part, this reflects that if
intermediate services were also to be added then the complexity of the table would be
increased. However, it is also the case that from an analytical perspective there is
little gain. The focus of the extended SUT is on the link between the economy and
106
ecosystems and this requires only links to final ecosystem services. Put differently,
from a production perspective, the intermediate services would net out in accounting
terms and are, in effect, embodied in the final ecosystem services. The analysis of
intermediate services, and hence flows between ecosystems, may be better analysed
using data from the basic ecosystem services SUT.
9.23. The second key aspect of the extended SUT is that additional columns are
required to take into account the production of ecosystem services – i.e. the
ecosystems are considered additional producing units alongside the current set of
establishments classified by industry (agriculture, manufacturing, etc.). Given that
SUT are generally compiled at national level, it may be sufficient to introduce simply
one additional column to cover the production of all ecosystem services. In this case,
the detail would be covered in the ecosystem services supply and use account.
However, there may be interest in adding columns by type of EU (ensuring
aggregation to national level) or by specific geographical areas within a country.
9.24. A related extension is environmentally-extended input-output tables (EE-
IOT). These tables are regularly compiled, including at regional and world levels, for
the analysis of embodied GHG emissions, water and similar environmental flows. An
introduction to EE-IOT is contained in SEEA Applications and Extensions Chapter 3.
9.25. For EE-IOT, information on environmental flows (e.g. GHG emissions by
industry) is appended to the standard input-output table and then matrix algebra is
used to integrate the data for analytical purposes. What is required is that the
information on environmental flows is classified and structured in the same manner
as for the standard input-output data. The additional information may be in physical
or monetary form even while the standard input-output data remain in monetary form.
Thus, using EE-IOT techniques, it is possible to analyse selected ecosystem services
without developing a full extended SUT.
9.26. For the extended SUT envisioned here, the ecosystem services are fully
integrated within the standard SUT reflecting the extension of the production
boundary. This is an important development.
9.27. An important result of integrating the flows of ecosystem services in
extended SUT is that it is clear how the commonly discussed topic of “double
counting” is managed. Quite commonly, there is concern that integrating ecosystem
services with the national accounts will result in double counting if certain flows are
included. The stylised presentation in Table 9.1 demonstrates that double counting is
avoided, provided that the series of entries, from production through to final use via
the supply chain, are recorded appropriately. The gross basis of recording that is used
in Table 9.1 is by far the most transparent manner in which double counting is dealt
with for accounting purposes.
9.28. Table 9.1 is a stylized supply and use table and is divided into three parts.
Part A reflects a standard recording of timber production, i.e. no ecosystem services,
of timber production for furniture purchased by households. The recording ignores all
other inputs and potentially relevant flows (e.g. labour costs, retail margins).
9.29. Part B extends this recording to include the flow of the provisioning service
of timber from the ecosystem asset (a forest) to the forestry industry. The main effect
is to partition the value added of the forestry industry between the industry and the
ecosystem asset. Note that the overall value added is unchanged (at 80 currency units)
even though total supply has increased due to the inclusion of the production of
ecosystem services. This reflects the increase in the production boundary and
demonstrates how the accounting framework deals with the challenge of double
counting.
107
9.30. Part C introduces a second ecosystem service, air filtration, which is supplied
by the ecosystem asset. Again total production is increased, but in this case value
added also rises because the additional production is not an input to existing, i.e.
SNA, products. The increase in value added is also reflected in increased final
demand of households.
Table 9.1: Integration of final ecosystem services with current national accounts
estimates
Ecosystem Forestry Manufacturing Households TOTAL
asset (Forest) industry industry Final
Demand
PART A
Supply
Logged timber 50 50
Furniture 80 80
Use
Logged timber 50 50
Furniture 80 80
Value added (supply less use) 50 30 80
PART B
Supply
Ecosystem service – growth 30 30
in timber
Logged timber 50 50
Furniture 80 80
Use
in timber
Logged timber 50 50
Furniture 80 80
Value added (supply less use) 30 20 30 80
PART C
Supply
in timber
Ecosystem service – air 15 15
filtration
Logged timber 50 50
Furniture 80 80
Use
in timber
Ecosystem service – air 15 15
filtration
Logged timber 50 50
Furniture 80 80
Value added 45 20 30 95
Source: Obst et al, 2015
108
9.4.2 Integrated ecosystem accounts: full sequence of institutional sector accounts and
balance sheets
9.31. It is also relevant to integrate ecosystem information into the broader
sequence of institutional sector accounts and balance sheets of the SNA. The general
logic and structure of the sequence of accounts is described in detail in the SNA and
is summarised in the SEEA Central Framework, Chapter 6. The focus in these
accounts moves away from information on production and consumption and instead
focuses on the institutional sector level (i.e. corporations, governments, households)
and measures of income, saving, investment and wealth.
9.32. One of the main functions of the sequence of accounts is to demonstrate the
linkages between incomes, investment and balance sheets and, in this context, a key
feature of the standard sequence of accounts is the attribution of consumption of fixed
capital (depreciation) to sectors as a cost against income.
9.33. Chapter 6 of SEEA EEA describes a possible sequence of accounts where
there is integration of information on ecosystem degradation. (Table 4.6 in Chapter 4
of these Technical Recommendations shows the sequence of accounts proposed in the
SEEA EEA.) The SEEA EEA treatment is not definitive however, and there is no
clear resolution of the way in which degradation might be allocated.
9.34. The appropriate allocation of degradation requires judgements on the
attribution of the impacts of economic activity on the environment. These impacts
may occur in areas well away from the source of the impact, may occur in time
periods well after the impact occurred, and may be unknown to the relevant units. In
addition, it is not necessarily clear in what way the loss of benefits incurred by the
impacted sectors should be related to the income of the sector causing the impact.
These matters have been debated at length in the national accounting community
without any clear resolution. It may well be that, as stated in the SEEA EEA, the
choice of structure for the sequence of sector accounts should depend on the type of
question being posed.
9.35. The significance of developing a sequence of accounts that integrates
ecosystem information is two fold. First, it is in these accounts that the cost of
ecosystem degradation can be attributed to individual sectors and linked, at the same
time, to changes in net wealth. Second, the effect of extending the asset boundary of
the standard national accounts to include various regulating and cultural services
from ecosystems can be seen in an extended balance sheet.
9.36. From an implementation perspective, it should be recognised that the
compilation of a sequence of institutional sector accounts and balance sheets is not
straightforward. It relies on compilation of aggregated data for ecosystem services
and ecosystem assets in monetary terms and hence information from all of the
accounts described elsewhere in these Technical Recommendations will be required
before a sequence of accounts can be compiled. In that sense, the completion of these
accounts should be considered of lower priority in the short term.
9.5 Key integration issues

9.5.1 Measurement of net present value
9.37. The valuation of ecosystem assets, and the estimation of associated changes
in the value of ecosystem assets, requires the derivation of net present values for the
flows of ecosystem services. Setting aside the issues discussed in other chapters on
the measurement of the flows of ecosystem services in physical terms (chapter 5) and
109
the estimation of relevant prices (chapter 8); there remain other measurement
considerations in estimating NPV.
9.38. First, it is necessary to understand the likely asset life of the ecosystem. That
is, an understanding is required of the length of time that ecosystem services will be
supplied into the future. This cannot be done independently for each ecosystem
service. In many cases, the supply of regulating and cultural services will be
competing with the supply of provisioning services. It is likely that the estimated
asset life for provisioning services provides the relevant lower bound for the asset life
assumption. Where an ecosystem asset is being used sustainably, i.e. with no decline
in ecosystem condition, the asset life will be infinite.
9.39. Second, the derivation of NPV requires describing the likely future flows of
ecosystem services. This flow may be affected by many considerations, in particular
ecosystem condition. It may be useful to separately consider future flows in physical
terms and future changes in prices. A common assumption in national accounting for
both of these factors is to assume that the unit resource rent stays constant over the
remaining asset life. This may be considered a default basis for estimation, although it
would be considerably better if some assessment of possible future trade-offs between
different services was taken into account.
9.40. Third, a discount rate is required. The choice of discount rate is an area of
much discussion. Economists are by no means clear on what discount rates might be
appropriate and there are a number of factors to take into consideration. SEEA
Central Framework Annex A5.2 provides a useful summary. Perhaps the key issue
for ecosystem accounting is clearly articulating the intended valuation concept.
Where integration with existing national accounts estimates of income and assets is
required, then an exchange value concept is appropriate. Consistent with this choice,
the use of a market based discount rate is appropriate. Where a societal based
valuation is required, then other considerations are relevant. Generally, social
discount rates are lower than market based discount rates and the resulting NPV
estimates will be higher.
9.41. There is an important link between the choice of approach to estimating
future flows of ecosystem services and the choice of discount rate. Where it is
assumed that the prices of ecosystem services will remain constant over the asset life,
it is necessary to use a discount rate in real terms, i.e. after adjusting for inflation.
Conversely, where the future path of ecosystem services prices is directly estimated
and included in the calculations, then a nominal discount rate should be used.
9.5.2 Allocation of ecosystem degradation to economic units

9.42. One of the most longstanding challenges in developing fully integrated
accounts is the measurement and allocation of ecosystem degradation. The SEEA
Central Framework proposes a means by which the depletion of natural resources can
be incorporated within the standard sequence of accounts of the SNA. This treatment
recognises that the “using up” of natural resources is a cost against the future income
of the extractor, and one that should be attributed to the extractor.
9.43. Depletion due to natural resource extraction is a component of ecosystem
degradation. However, in concept, degradation is broader since it also incorporates
the cost of reducing the future supply of other ecosystem services, such as regulating
and cultural services. Unfortunately, it is commonly the case that the loss in
ecosystem condition and capacity that arises due to economic and human activity
may be difficult to directly attribute to individual economic actors.
110
9.44. A number of alternative approaches to dealing with this issue have been
suggested. Perhaps the most obvious is that the degradation should be attributed to
the economic unit that caused the degradation, presuming that this can be determined.
It may be made difficult due to distance (i.e. impacts are felt in neighbouring
ecosystems) and time (i.e. when the impacts become evident after the activity
occurred). Due to both of these factors, the relevant economic unit (i.e. the unit who
should be shown as bearing the cost) may not be the manager or owner of a particular
ecosystem asset. Further, attributing the overall impacts is complex, since physical
degradation of an ecosystem is likely to impact on the supply of multiple ecosystem
services that are received by various beneficiaries.
9.45. These factors are all quite distinct from the estimation of depreciation (or
consumption of fixed capital) for produced assets. Depreciation can be directly
attributed since there is only one owner/user who receives all of the benefits/services
of the asset (in the generation of output and income).
9.46. Overall, the issue of allocating ecosystem degradation has not been resolved
and remains on the research agenda.
9.5.3 Valuation of ecosystem degradation

9.47. Together with the challenge of attribution, a range of valuation approaches
for ecosystem degradation have been suggested. These are described in SEEA EEA
section 6.3.3. The approach that emerges from the ecosystem service based valuation
approach described here is that the value of ecosystem degradation will be equal to
the change in the net present value of an ecosystem asset, putting aside changes in
value that are not due to economic and human activity.
9.48. A commonly discussed alternative approach to valuing ecosystem
degradation is the use of restoration (maintenance) costs. Such approaches were
suggested in the original 1993 SEEA (UN, 1993). Under this approach, an estimate is
made of the expenditure that would be required to restore an ecosystem to its
condition at the beginning of the accounting period. This line of thinking is
sometimes extended to consider that the accumulated, unpaid restoration costs
represent a liability – an ecological debt.
9.49. Generally, restoration cost approaches are not well accepted by the
environmental economic community (see for example, Barbier, 2013) who prefer to
see degradation valued as the change in the value of the associated benefits and
income flows. In accounting terms as well, recent work suggests that restoration costs
are not equivalent to what is undertaken in estimating depreciation of fixed capital
(Obst and Vardon, 2014; Obst et al., 2015). Importantly, the extension of the
accounting framework to integrate the value of ecosystem services allows a different
perspective on degradation to be supported within the accounting framework.
9.50. Notwithstanding concerns about the use of restoration cost approaches for the
valuation of degradation, the estimation of potential restoration costs can provide
valuable information for policy purposes. Estimation of these costs can provide a
sense of scale to degradation discussion, especially where the discussion revolves
around the resources required to maintain condition. It may also be analytically useful
to compare the estimated restoration costs with the actual expenditures on ecosystem
maintenance. When these costs and expenditures are tracked against actual changes in
ecosystem condition, some useful information for policy purposes seems likely to
emerge.
9.51. A more recent suggestion for the valuation of ecosystem degradation
concerns ecosystem capacity. The suggestion involves estimating the net present
111
value of the future flow of services linked to an ecosystem’s capacity. This is distinct
from the net present value of the flow of services that are expected to occur, which
may be higher or lower than capacity. The difference will depend on difference in the
expected asset life under either scenario and, where the difference in asset life is large
the choice of discount rate may have a significant effect on the results. The change in
the NPV of ecosystem capacity might be a more appropriate estimate of the effect of
the reduction in future income that arises from a decline in ecosystem condition but
further investigation of the national accounting aspects of this approach is required.
9.5.4 Integrating balance sheet valuations

9.52. Perhaps the most significant measurement challenge in full integration is the
need to generate balance sheet values for ecosystem assets on a consistent basis with
the items already recorded on the national balance sheet as defined by the SNA. In
the absence of observed market prices for ecosystem assets, the logic of the SEEA is
that the value of the asset would be equal to the net present value of the future flows
of all relevant ecosystem services.
9.53. As with all NPV based approaches, this requires various assumptions (see
SEEA Central Framework Chapter 5) including estimating the future rates of use and
extraction and applying appropriate discount rates. Estimating the future rates of use
of ecosystems must in turn imply an understanding of the likely mix or basket of
ecosystems that will occur in the future, and also the likely impact of supplying this
assumed basket on the future condition of the ecosystem asset.
9.54. It is possible to estimate future service flows using biophysical models that
estimate future conditions according to assumed scenarios (e.g. an extraction profile).
Advances in recording and understanding the links between ecosystem condition and
ecosystem services will, over time, improve these estimates.
9.55. Some parts of the value of ecosystem assets are reflected in the value of
assets currently recorded on national balance sheet. The most obvious example is the
value of agricultural land which must, reasonably, be considered to incorporate the
value of some of the ecosystem services used by farmers in the production of
agricultural outputs. Similar logic would apply in the case of forests.
9.56. However, as explained at some length in SEEA EEA section 6.4.2,
untangling the overlap in valuations is likely to be complex. For example, in the
valuation of agricultural land, part of the market value currently on the balance sheet
will reflect the contribution of ecosystem services. To avoid double counting, a
reconciliation is needed between the total market value and the value of agricultural
land derived summing the NPV of a basket of ecosystem services.
9.57. An intermediate step may be the compilation of an ecosystem monetary asset
account as described in Chapter 4. This account shows the opening stocks, additions
and reductions in stocks and closing stocks for ecosystem assets in a standalone
account. This may then be compared to current SNA based balance sheet entries,
particularly for land and natural resources, as a means of understanding the potential
differences and the significance of the recognition of ecosystem services that are
currently outside the production boundary.
9.58. In comparing the values of ecosystem assets with values currently
incorporated into SNA balance sheets, it is important to recognise the different
underlying scopes of environmental assets. In broad terms, the SNA balance sheets
will have lower values for environmental assets as a result of the SEEA EEA
including the values of additional ecosystem services. At the same time, the SEEA
EEA values of ecosystem assets do not cover all environmental assets – most notably
112
sub-soil mineral and energy resources16. The effects of these two differences on the
total value of environmental assets will vary from country to country.
9.5.5 Application of integrated approaches for individual ecosystem assets

9.59. A final challenge in the area of integrating the accounts arises when applying
the accounting approach at the level of an individual ecosystem asset. Recall that the
valuation of an ecosystem asset is directly related to the basket of final ecosystem
services that are expected to be generated from an asset. At the level of individual
ecosystem assets however, there will be cases where an asset supplies few or no final
ecosystem services (for example, a high mountain forest) but, instead plays a
supporting role in supplying intermediate services to neighbouring ecosystems. In
this situation, an ecosystem asset may be recorded as having zero monetary value and
its value becomes embodied in the value of the neighbouring ecosystems. While at an
aggregate, national level this may not be a significant issue, it is likely to be of
concern if attribution of value is being examined or accounting is being undertaken at
smaller sub-national scales. Resolution of this issue requires the incorporation of
intermediate services into the ecosystem accounting model in a far more explicit
manner.
9.6 Alternative approaches to integration

9.60. If the objective is not full integration with the standard national accounts,
then integration of ecosystem and economic data may be considered in different
ways. Three alternative integrated approaches are summarised here.
9.61. A well-developed approach, usually referred to as wealth accounting, has
developed as a branch of economics since the mid-1970s. Wealth accounting seeks to
aggregate the value of all relevant assets/capitals including produced, natural, human
and social capital. The most prominent work has been completed by the World Bank
(2011) and by UNU-IHDP and UNEP (2014). Their methods vary in the detail but
they are broadly similar approaches.
9.62. In concept, wealth accounting aims to value each form of capital in terms of
its marginal contribution to human welfare (Dasgupta, 2009). By doing so shadow
prices are estimated for each asset type. From a national accounting perspective while
a focus on marginal prices is appropriate, estimation of the contribution to welfare is
different from a focus on exchange value.
9.63. Given the purpose of wealth accounting, the conceptual approach to
integration is quite appropriate. However, in practice, often values for produced
assets from the standard national accounts based on exchange values are combined
with valuations for other capitals based on welfare valuation concepts. Hence, there
may be a lack of alignment between the valuation approaches for different capitals.
For natural capital, it is clear that the use of exchange values for ecosystem services
would not correspond directly to the conceptual requirements of wealth accounting,
although there will be strong connections between the two approaches.
9.64. Another approach to integration builds on the use of restoration costs as a
measure of ecosystem degradation to create ecological liabilities on the national
balance sheet. That is, unpaid restoration costs that arise when an ecosystem declines
in condition are treated as a liability. This approach is described as a possible
extension in the ENCA QSP and has also been suggested for use at the corporate
16
Accounting for these environmental assets is described in the SEEA Central Framework.
113
level by the UK Natural Capital Committee. From a national accounting perspective
there are a number of difficulties with this approach:
 First, there is the question of whether restoration costs are a suitable estimate
of ecosystem degradation. This was discussed earlier in this chapter.
 Second, there is a question of when liabilities should be recognised. If there
is no expectation that the restoration will take place then, at least for
accounting purposes, no liability should be recognised. In effect, recognising
these liabilities is a social or analytical choice rather than an application of
accounting principles.
 Third, if a liability is recognised then, all else being equal, net wealth will
fall. However, since the recognition of the liability reflects the degradation of
an asset there will be both a fall in an asset and an increase in a liability for
the same event. This implies a double counting on the balance sheet. This
issue does not arise in accounting for depreciation where the only balance
sheet change is the fall in the asset value. A solution would be to record the
liability but keep the ecosystem asset value unchanged but this seems
counter-intuitive.
9.65. Overall, while recording ecological debts seems attractive and may be a
useful tool in communicating the extent of ecosystem degradation, it has some
deficiencies in terms of its consistency with national accounting principles.
9.66. The final integrating approach noted here is that of full cost accounting. This
is an accounting approach that has developed in corporate accounting. The intent
behind full cost accounting is to estimate and record the broader costs of a company’s
impacts on the environment as part of their ongoing operating costs. For example, the
costs of GHG emissions and the release of pollutants are common areas of interest.
Such information may be helpful in a range of management situations.
9.67. From an ecosystem accounting perspective a few points can be noted. First,
the approach largely excludes consideration of ecosystem services in terms of
recognising ecosystem services as inputs to the production process. Hence, within the
full cost accounting approach there is no change in the standard production or income
boundaries.
9.68. Second, there is no recognition of ecosystem assets as part of the capital base
of a company and hence no impact on the company’s balance sheet or recording of
ecosystem degradation as a capital cost.
9.69. Third, the incorporation of costs associated with residual flows (emissions,
pollutants, etc.) is not something undertaken directly in ecosystem accounting. In
broad terms a focus on residual flows reflects the valuation of a company’s negative
externalities and externalities are specifically excluded from the national accounts. It
may be that, in fact, the attribution of these costs can be part of a measure of
ecosystem degradation. Further work is required to understand the links between the
valuation of externalities and ecosystem accounting, recognising that the links may be
different for different types of externalities.
9.70. Overall, while full cost accounting does represent a form of integration it is
somewhat different in scope relative to the concepts and intent of ecosystem
accounting.
114
9.7 Recommendations
9.71. From a national accounting perspective, the full integration of ecosystem
accounting information into the standard national accounts and the derivation of
adjusted estimates of GDP and other measures of economic activity represents
somewhat of a required outcome from SEEA. While recognising the need for some
final work on the allocation of degradation, the adoption of ecosystem accounting has
demonstrated that a full integration is conceptually possible.
9.72. At the same time, there are numerous challenges in measurement that must
be worked through. These particularly concern the aggregation of stocks and flows
across ecosystem services and ecosystem assets. These are important challenges to be
confronted through testing and on-going research. This chapter has described some
specific issues concerning integration but the measurement issues raised in other
chapters are equally relevant in working towards a complete integrated accounting
dataset.
9.73. While work continues on developing these full links between ecosystem and
economic data, there are other options that can be pursued. Combined presentations,
as described in SEEA EEA Chapter 6 and here in section 9.3, are important tools for
presenting data that support a comparison and discussion of environmental and
economic issues. It is strongly recommended that countries work towards the
development of combined presentations of data. Work in this area is likely to be
considerably supported by a focus on presenting data at meaningful spatial level.
9.74. As work within a country proceeds on ecosystem accounting, it is important
that any related or similar work on the integration of environmental and economic
data is placed in context. Generally, it is not a question of data being in competition,
but rather it is a case of different data being suited for specific purposes. Explaining
the link between different measurement approaches and different policy questions is
an important role for national statisticians.
115
Annex 1: Clarifications on SEEA EEA
Introduction
These Technical Recommendations build directly on the conceptual framework for ecosystem
accounting described in the SEEA EEA. For the most part, they provide additional
explanation and direction for compilation. However, there are some areas in which a
clarification of the conceptual model is described. This reflects the ongoing discussion and
consideration of ecosystem accounting since the completion of the SEEA EEA in 2013. There
are five areas in which conceptual clarifications are introduced. These areas are described in
this annex.
The treatment of spatial units

The treatment of spatial units for ecosystem accounting has been advanced and clarified in a
number of ways in the Technical Recommendations (see Chapter 3 for details). In the first
instance, while three types of spatial area have been retained, the labels have changed. BSU
remain Basic Spatial Units, but LCEU have been re-labelled Ecosystem Units (EU) and EAU
have been labelled geographical aggregations.
More significantly, the EU is now clearly the core spatial area concept for ecosystem
accounting representing the ecosystem asset that supplies ecosystem services. The Technical
Recommendations clarify that the delineation of EU will, ideally, involve the use of a range
of criteria, including land cover, vegetation type, soil type, hydrology, and potential land
management or use. These criteria can be used to define various EU types. At the highest
level of aggregation EU types should aggregate to the 15 land cover classes of the SEEA
Central Framework thus helping to align datasets and also provide a starting point for
ecosystem accounting work.
Concerning BSU, it has now been clarified that these may be formed in various ways but that
the use of a reference grid, delineated using a common reference system or datum is the most
useful approach. Within a reference grid both raster and vector data can be more readily
combined.
Account labelling and structure

The SEEA EEA included a range of accounts but, on reflection, the structure and naming
conventions were not well developed. In the Technical Recommendations (Chapter 4), there
are three key developments:
i. A distinction has been drawn between ecosystem accounts and thematic accounts.
Ecosystem accounts are those covering specifically stocks and changes in stocks of
ecosystem assets, flows of ecosystem services, and accounts that integrate ecosystem
information with the standard national accounts. Thematic accounts are for specific
topics including land, carbon, water and biodiversity. Data from thematic accounts
may be used in compiling ecosystem accounts and may also provide important
context for the analysis of ecosystem accounting information.
ii. Some of the ecosystem accounts have been relabelled – most notably the ecosystem
monetary asset account which was formerly simply the ecosystem asset account.
iii. In terms of account structure most retain a similar approach as in SEEA EEA. The
exception is the supply and use table for ecosystem services which now has a more
articulated framing that builds on the physical supply and use tables of the SEEA
Central Framework.
116
The measurement of ecosystem services
The focus for ecosystem accounting on final ecosystem services as contributions to the
production of benefits remains unchanged. However, there are two aspects surrounding this
focus that have been clarified in these Technical Recommendations.
First, there is a clearer explanation that the incorporation of final ecosystem services in the
accounting model can be seen as an extension in the production boundary of the SNA. Thus,
in a national accounting context, the integration of final ecosystem services leads to an
expansion in measures of output. Where the ecosystem services contribute to the production
of SNA benefits there is a corresponding increase in intermediate consumption and hence
value added is unaffected.
This expansion of the production boundary has a range of “natural” implications for national
accounting. These include the broadening of measures of income and hence the associated
value of assets that supply the services. Also, it is consistent to extend standard supply and
use tables through the additional of rows reflecting the “new” final ecosystem services.
Additional columns are also added reflecting the ecosystem assets and “new” producing units.
Second, there is a clearer recognition of the potential to record intermediate ecosystem
services reflecting flows between ecosystem assets. In an accounting sense, these flows net
out in the production of final ecosystem services and hence recording them has no impact on
value added. However, recognising that they may be recorded in the system, supports a better
conceptualisation of the connections between ecosystem assets and hence illustrates the
potential of ecosystem accounting to recognise the contributions of possibly remote
ecosystems.
At the same time, there is the practical reality that there are a very large number of potential
intermediate services. Consequently, it is not anticipated that at this stage there should be
significant focus on these flows.
Ecosystem condition
The concept of ecosystem condition remains the same as in the SEEA EEA. However, on
reflection, there was a certain naivety concerning the measurement of condition and hence
some important framing of this issue has been included. This framing introduces the notion of
top-down and bottom-up approaches to measurement, recognises that some indicators of
condition may relate to fixed characteristics as distinct from variable ones, and there is
important clarification on the issue of measuring condition from small to larger scales. On
this last point, a continuum is described from the definition of indicators for individual
characteristics for a single ecosystem type, up to the potential to define comparable indicators
across ecosystem types with multiple characteristics.
A more general point is that it is also recognised more explicitly that the measurement of
condition will depend on the current pattern of land use/management and the associated mix
of ecosystem services. In turn, this is likely to affect the way in which ecosystem units are
delineated.
Ecosystem capacity
In the SEEA EEA, ecosystem capacity was mentioned but not defined. In the discussion
through the development of the SEEA EEA, the relevance of the concept was recognised but
no agreement could be found on how it might be best described in an accounting context.
Since the release of the SEEA EEA, it has become increasingly clear that the concept of
ecosystem capacity that links the concepts of ecosystem condition and ecosystem services is
in fact quite fundamental in an accounting context. Most importantly, the concept of
117
ecosystem capacity can be directly linked to the measurement of ecosystem degradation, itself
a fundamental variable in national accounting.
These Technical Recommendations (Chapter 6) therefore provide a more thorough
description of the concept of ecosystem capacity, propose a definition and outline some
associated measurement thoughts. As yet, no final position has been reached regarding the
definition and measurement of this concept and research is continuing on this topic.
118
Annex 2: Summary of various National Capital Accounting initiatives
a. CBD Ecosystem Natural Capital Accounts: Quick Start Package (ENCA QSP)
(Weber, 2014a)
The ENCA QSP is a detailed technical document aimed at supporting countries in
the implementation of Aichi Biodiversity Target 2 on the integration of
biodiversity values in national accounting systems. Using techniques developed
in a European context and applied in Europe (European Environment Agency,
2011) and in Mauritius (Weber, 2014b), the ENCA QSP gives practical guidance
on establishing detailed spatial datasets on land cover, carbon, water, species
diversity, and various landscape level indicators (e.g. on fragmentation and
ecotones).
The two key strengths of the ENCA QSP are its demonstration of the potential to
integrate large volumes of data at country level, often using global level datasets;
and its demonstration of a national accounting approach to ecosystem
measurement wherein data are scaled up and down as required to provide an
overall picture of change for a country as a whole. The ambition to provide a
broad picture for a country as distinct from a precise estimate for a specific
ecosystem is an important distinction of ecosystem accounting.
The focus of the ENCA QSP is on the measurement of ecosystem extent and
condition. It does indicate a link to the measurement of ecosystem services but
this is done only via an assumption that for a given ecosystem condition there
will be a specific basket of ecosystem services – it is in effect a top down
approach. Ecosystem services are not measured directly in what might be termed
a bottom up approach. A consequence is that the measurement requirements in
ENCA QSP are reduced relative to the SEEA EEA.
With regard to the measurement of ecosystem condition the ENCA QSP proposes
an approach that uses indicators of a limited number of ecosystem characteristics
that are applied to all ecosystem types. This broad approach may seem
inappropriate from an ecological perspective but the intention is to provide a
quick and broad assessment.
The core of the ENCA QSP lies in the measurement of ecosystem condition and
the assessment of ecosystem capability. It does however, articulate a number of
potential extensions. These include recording different economic sectors
accountability for ecosystem degradation (in physical units), the compilation of
an ecological balance sheet and discussion of the recording of ecosystem
restoration costs. There is no valuation of ecosystem services nor valuation of
ecosystem assets as outlined in the SEEA EEA.
Overall, its detailed proposals for the estimation of accounts with national
coverage for land, carbon and water and various high-level indicators concerning
ecosystem function are important contributions and should be of direct support to
compilers of ecosystem accounts as described in the SEEA EEA.
b. World Bank WAVES Designing Pilots for Ecosystem Accounting (World Bank,
2014)
This guidance material provides a summary of the key features of ecosystem
accounting and how a country or region might work towards developing a set of
ecosystem accounts. Its coverage includes discussion on the types of issues that
might benefit from the compilation of ecosystem accounts, the selection of a case
119
study area/site, assessment of the relevant ecosystem services, guidance on the
biophysical mapping and analysis of ecosystem services, and shows an
application of the approach to a study area in Peru.
The focus of the material is on providing appropriate context and criteria / factors
that are relevant for making decisions with respect to ecosystem accounting.
While there is some mention of the measurement of ecosystem condition and
somewhat more discussion on the issue of ecosystem capacity, on the whole the
primary focus of the material concerns ecosystem services. Methods for the
valuation of ecosystem services are mentioned.
This material should provide a useful complement to other materials, such as
those focused on ecosystem condition (ENCA QSP, above) and those focused on
valuation (UNEP Small Island Developing States Guide, below). Indeed, this
presence of complementarity speaks to the breadth of the requirements for
Since the focus of the guidance is on the practical implementation and testing of
ecosystem accounting there are no specific departures from the SEEA EEA
concepts. Of course, the precise manner and methods by which ecosystem
accounts should be compiled remains the object of the testing and in this regard
the WAVES guidance material should usefully complement these Technical
Recommendations as well.
c. UNEP Guidance Manual on Valuation and Accounting of Ecosystem Services for

Small Island Developing States (SIDS) (UNEP, 2015)
This manual was prepared in the context of the particular imperatives for SIDS to
manage their development in the context of climate change and recognizing the
particularly strong link between SIDS economies and their natural environment.
The first part (chap. 2 - 4) of the guidance is focused on the measurement and
valuation of ecosystem services and a thorough overview of relevant concepts
and methods is provided with a particular focus on measurement in the context of
SIDS. Step by step guides to the most relevant methods are also provided. The
coverage of this discussion is not solely on valuation for accounting purposes
since there are other reasons for valuation other than accounting (e.g. cost benefit
analysis, program evaluation, etc.).
Chapter 5 describes two aspects of “ecosystem service accounting. The first is a
summary of work in Mauritius that is an application of the methods described
above in the ENCA QSP. In effect this work does not reflect accounting for
ecosystem services but rather accounting for ecosystem condition. The second
aspect outlines some steps to the valuation of ecosystem services for inclusion in
the standard national accounts. The use of a production function approach is
summarized for a small set of provisioning and cultural services.
The guidance does not cover the valuation of regulating services in an accounting
context and while pointing towards the integration of ecosystem services into the
national accounts, it does not discuss the relevant measurement issues or mention
issues such as the valuation of ecosystem degradation.
This Guidance Manual should provide useful information for those wishing to
undertake the valuation of ecosystem services as part of implementation of work
on SEEA EEA however care is needed on the discussion of the integration of
ecosystem services value within the standard national accounts since some of the
important integration issues are not considered.
120
d. Mapping and Assessment of Ecosystems and their Services (MAES)
The European Union’s MAES project is a large measurement project working
towards completion of Action 5 of the European Union Biodiversity Strategy to
2020. The MAES framework encompasses the two key dimensions of
measurement that are also in the SEEA EEA namely ecosystem condition and
ecosystem services. In that sense, the developments in the MAES provide a
relevant example of the types of measurement issues likely to arise in ecosystem
accounting. Indeed, part of the MAES project is the development of a
methodological approach to natural capital accounting.
To date the main output from the MAES project is its report on “Indicators for
ecosystem assessments under Action 5 of the European Union Biodiversity
Strategy to 2020” (Maes, et al. 2013). In this report it documents the
establishment of six pilot studies across Europe and the results from assessing
ecosystem condition and an array of ecosystem services in different ecosystem
types (forests, cropland and grasslands, freshwater, and marine).
The document is useful in highlighting measurement possibilities and challenges
in a summary manner thus providing insights for those aiming to establish
ecosystem accounting projects. Particularly useful are the listings of (and
recommendations regarding) potential indicators for different ecosystem services
across the range of provisioning, regulating and cultural services. Such listings
are particularly useful in trying to understand the type of information that might
be relevant.
In the context of ecosystem accounting the approach taken is particularly
appropriate since it is working form the intent of measuring ecosystems and their
services at a national and pan-European level. This type of broad assessment and
the use of relevant frameworks and classifications is well aligned with the
intentions of ecosystem accounting.
A draft reference document on natural capital accounting has also been released
for consultation (January 2015). Largely it is a description of the various
approaches to natural capital accounting, including the SEEA and includes
discussion of natural capital itself, and the role of natural capital accounting in
policy. The document discusses also the role of valuation, in both monetary and
non-monetary terms. The document does not provide methodological guidance
but is useful in providing background material to SEEA EEA based accounting
exercises.
121
Annex 3: Listing of project research papers
1. Compilation of data, tools and methods (Bordt, 2014)
2. Water and ecosystem accounting (Vardon, 2014a)
3. Carbon and ecosystem accounting (Vardon, 2014b)
4. Linkages between ecosystems asset and service accounts (Hein, 2014a)
5. Guidelines for biophysical modelling and mapping (Hein, 2014b)
6. Land and ecosystem condition and capacity (Bordt, 2015a)
7. Spatial units, scaling and aggregation (Bordt, 2015b)
8. Land accounts and ecosystem extent (Ivanov, 2015)
9. Functional approach to ecosystem accounting (Eigenraam, 2015)
10. Experimental Biodiversity Accounting within the SEEA EEA framework (UNEP-
WCMC, 2015)
122
Annex 4: UK guidance for ecosystem accounts scoping studies
Natural Capital Project Board
Guidance for developing ecosystems accounts based on Broad Habitats
UK Department of Environment, Food and Rural Affairs and Office for National Statistics
June 2014
This note sets out the steps involved in conducting scoping studies and compiling initial
accounts for a particular Broad Habitat, based on our experiences so far and the World Bank’s
report on designing ecosystem accounting pilots17.
1. Define extent. Define the different ecosystems/habitats covered within the Broad
Habitat category and assess the available and likely future availability of
measurements of the extent of each habitat
2. Identify key services. Identify the key services these ecosystems provide and their
importance and status by reference to the prioritisation criteria
3. Establish relevant characteristics. Identify what characteristics are key to the
delivery of those services (this might best be done in consultation with experts)
4. Assess data sources. Assess the availability (including expected future availability)
of non-monetary information on those characteristics and those services, and the
degree to which spatially disaggregated data is important for the accounts and its
availability
5. Propose asset account structure. Conclude on the services which should be
included in the initial accounts and hence on the structure of the non-monetary asset
accounts in terms of recording specific habitats separately and the relevant
characteristics for those habitats
6. Propose services account structure(s). Conclude on the units and structure of the
non-monetary services accounts for each of these habitat types
7. Spatially disaggregated accounts. Conclude on the scope for spatially
disaggregated non-monetary asset and services accounts and the process by which
they should be compiled and maintained
8. Assess valuation options. Explore options for the valuation of those services (and
hence the asset value relating to those services)
9. Provide proof of concept. Set out illustrative accounts on the basis of the data
obtained so far and make recommendations about a) how to best fill data gaps b)
when to update and c) how to reconcile with other accounts
10. Unresolved issues. Set out any unresolved (specific or cross-cutting) issues arising
which need further consideration, and report any potential policy applications
identified in the course of the study
11. Resource requirements. Assess the resources and time required to compile the
proposed accounts and resolve outstanding issues
17
https://www.wavespartnership.org/sites/waves/files/documents/PTEC2%20-%20Ecosystem.pdf
123
Annex 5: Key features of a national accounting approach to ecosystem
measurement
Introduction
A5.1 This section explains the key features of a national accounting
approach and why it provides a distinct measurement discipline that works
very effectively towards the mainstreaming of environmental information
into economic measures.
A5.2 To place accounting frameworks in context it is relevant to consider
the information pyramid (Figure A3.1). This pyramid has as its base a full
range of basic statistics and data from various sources including surveys,
censuses, scientific measurement and administrative sources. Generally, these
data will be collected for various purposes with the use of different
measurement scopes, frequencies, definitions and classifications. Each of
these data sources will be relevant to analysis or monitoring of specific
themes.
Figure A3.1 Information pyramid
Indicators
Accounts: assets,
condition, services
Frameworks: measurement,
process, quality
Basic data: environmental, economic,

social statistics
A5.3 The role of accounting frameworks (at the middle levels of the
pyramid) is to integrate these data to provide a single best picture of a
broader concept or set of concepts – for example economic growth or
ecosystem condition. The compiler of accounts must therefore reconcile and
merge data from various sources taking into account differences in scope,
frequency, definition and classification as appropriate.
A5.4 Finally, having integrated the data within a single framework,
indicators can be derived that provide insights into the changes in
composition, changes in relationships between stocks and flows, and other
features taking advantage of the underlying relationships in the accounts
between stocks and flows, between capital and labour, between production
and consumption, etc. Indicators such as GDP, national saving, national
124
wealth, terms of trade and multi-factor productivity all emerge from the one
national accounts framework.
A5.5 The following sub-sections focus on the approach that national
accountants take to providing the single best picture.
Key features of a national accounting approach

A5.6 For those not familiar with the way in which national accountants
work through measurement issues there are two key aspects that should be
understood. First, national accounting approaches generally always
commence using data from multiple sources that has already been collected.
National accounting is therefore not focused on defining survey questions,
determining sample sizes, collecting and processing data, etc. These
important tasks are assumed to be completed by experts in specific subject
matter areas, relevant methodologists and those in charge of administrative
data. Ideally, there would be a close relationship between the national
accounts compiler and those collecting the data but this can take time to
evolve and in any event the national accountant will always remain one step
removed from the source data.
A5.7 Second, in part as a result of not collecting data, but largely as a
result of the underpinning conceptual framework, national accountants work
“from the outside in”. National accounting is not a “bottom up” measurement
approach whereby aggregates are formed by summing available data. Rather,
most effort goes into ensuring that the estimates that are compiled
appropriately reflect the target concept, for example, economic growth or
fixed capital formation or household consumption. Generally, it will be the
case that no single data source can fully encapsulate a single concept and
hence the role of the national accountant is to meld, integrate and otherwise
combine data from multiple sources to estimate the concept as best as
possible.
A5.8 Further, on this same point. It is not sufficient to obtain the best
estimate of each concept in isolation. Rather the measurement of each
concept must be considered in the context of the measurement of other
concepts following national accounts identities. Thus, for example, total
supply and total use of each product must align. Ultimately it is the ambition
to produce, at a single point in time, the single best picture, of the concepts in
scope of the national accounts framework. This cannot be achieved by
relying on a bottom up strategy where the micro builds neatly to the macro.
Instead, a top down or working from the outside in approach must be applied.
A5.9 Building on these two key aspects there are some related national
accounting compilation principles that should be recognised.
i. The maintenance of time series is fundamental. Perhaps the most important
principle is that in creating the “single best picture” it is not sufficient for
each data point to stand alone in time. Hence changes over time must be
considered as part of the picture. Often national accounts time series extend
for over 30 or 40 years and there are few if any data sources that are
maintained consistently over these time frames. Indeed, generally data
sources will improve their methods and coverage over time. Consequently, a
key role in national accounts is linking information from different sources
and over time, and hence various methods may be applied to consistently
measure the same concept.
125
ii. Prices, quantities (volumes) and values are all relevant. While the vast bulk
of the national accounts framework is presented in terms of relationships in
value terms (i.e. in terms of the actual monetary amounts transacted); the
most significant proportion of resources on compiling national accounts are
targeted at decomposing the changes in value between changes in prices or
changes in underlying volumes. Generally, most analysis of the national
accounts, e.g. growth rates, productivity, investment, is conducted in volume
terms (i.e. after removing price effects). Again the single best picture
ambition requires balancing these different perspectives at a component and
aggregate level.
iii. The need for revisions. Without a time constraint on the integration of data
and the release of results it is likely that national accounts would never be
completed. Given their scope, there is always new information that might be
considered or new methods that might be adopted to refine the single best
picture. National accounting thus works by ensuring the release at regular
intervals of the best picture in the knowledge that it will be revised in due
course when additional information comes to hand.
iv. Accounting is iterative. Fundamentally, the process of integrating data for
accounting is not a single, one-off process. Each time a set of accounts is
compiled different integration issues will arise and will generally only be
resolved through attempting integration, understanding the reasons for
imbalances, and implementing possible solutions. Gradually, a single best
picture emerges. Ideally, resolving these integration issues is a task that
involves both accountants and data supplying areas. Such joint resolution is
an important aspect in mainstreaming different data as part of an overall
picture.
A5.10 One overall consequence of a national accounting approach to
compilation is that comparability between different estimates is not assessed
primarily on the basis of method. In the first instance, comparability is based
on the extent to which different estimates accurately reflect the target
concept. Indeed, since each national accountant will be faced with the
integration of different source data, a focus on comparability of methods is
likely not a helpful starting point although it must be accepted that not all
methods will produce estimates of equal quality.
A5.11 One benefit of a focus on concepts is that countries will tend to focus
their resources on measuring those aspects within the accounting framework
that are of most relevance to them. For example, in a country in which
agriculture is a dominant activity, resources should be allocated to
measurement of this activity. In a different economic structure, for example a
country with a large finance sector, the balance of resources and the choice of
data and methods will and should be different. Since economic structures
changes over time, methods will also need to adapt. The development of
services statistics and associated measurement methods over the past 25 years
is a good example of this evolution in compilation approaches even as the
underlying concepts remain stable.
Applying the national accounting approach to ecosystem accounting

A5.12 In most cases, including for the datasets that underpin ecosystem
accounting, the ambition is to generate databases pertaining to a single theme
or topic and to provide the best estimates based on the selected methods and
resources available. While this may well and should involve comparison with
126
other datasets as part of editing the dataset, it generally does not involve full
integration and reconciliation with other datasets.
A5.13 A national accountant, on the other hand, is not compiling such a
dataset but rather is seeking to undertake the integration. In many respects
this is a role that must, at some point, be undertaken by a data user, analyst or
decision maker. That is, at some point interpretations and judgements are
needed concerning data from different sources that may suggest different
trends. Within the scope of macro-economic analysis, national accountants
make such judgements about relative data quality using the rigour of the
national accounting framework. The alternative would be a situation where
each economic analyst made their own judgements possibly using varying
definitions of economic aggregates and measurement scope.
A5.14 The application of a national accounting approach within ecosystem
accounting extends this national accounting compilation approach to
biophysical and scientific data. That is, within ecosystem accounting the
ambition is to integrate the various sources of information on ecosystem
condition, ecosystem services, economic production and consumption, to
present the single best picture.
A5.15 One consequence is that for ecosystem accounting it is necessary but
not sufficient to have data for a particular ecosystem type or for a selected set
of ecosystem services. Rather, effort must be made to obtain information that
permits assessment of the whole area of interest and full scope of supply of
ecosystem services. Certainly it would be relevant to place most resources
into measuring those ecosystems and their services that are considered most
relevant and significant, but this should not detract from the ambition to
measure the whole.
A5.16 In putting national accounts based estimates together it means that
data that may be regarded as of good quality are adjusted to ensure an
integrated picture. As well, since the emphasis is on the measurement of a
defined framework, some data sources may not be used, whatever their
quality, since they are not defined following the required concepts.
A5.17 While these statements are somewhat stark, in practice, a national
accounts approach is very reluctant to ignore any information. Rather, efforts
are generally made to examine all relevant data and, where necessary, make
adjustments to concepts to permit integration.
A5.18 In the area of ecosystem accounting, work is ongoing to define the
final integrated framework. In this context, there remains considerable scope
for an active dialogue between those managing the underlying data sets and
those designing the ecosystem accounting framework. This dialogue is
essential for the generation of high quality information.
Principles and tools of national accounting

A5.19 The focus here is on the main principles and tools that national
accountants apply to ensure coherence in the integration of data from
multiple sources. The following paragraphs present a brief description of the
relevant principles. An extensive discussion of the principles is contained in
the SNA 2008 and an extended overview is provided in SEEA Central
Framework.
A5.20 Accounting identities. The accounting system relies on a number of
identities – that is, expressions of relationships between different variables.
127
There are two relationships of particular importance in ecosystem accounting.
First, there is the supply and use identity in which the supply of a product (or,
in this case, an ecosystem service) must balance with the use of that same
product. This identity applies in both physical and monetary terms. Often
information on the supply and use of a product will be from multiple sources
and hence this identity provides a means by which data can be reconciled.
A5.21 Second, there is the relationship between balance sheets and changes
in assets. This identity is that the opening stock plus additions to stock less
reductions in stock must equal the closing stock. Again, this identity applies
in both physical and monetary terms. Without this identity there would be no
particular reason to ensure that observed changes in ecosystem assets (e.g.
through natural growth or extraction) aligned with the series of point-in-time
estimates of ecosystem condition that underpin the balance sheets.
A5.22 Frequency of recording. In order to provide a single best picture
across multiple data sources it is essential that there is a common reference
point referred to in accounting terms as the accounting period. Generally, it is
recommended that the accounting period used across a set of SEEA based
accounts is one year. This supports alignment with economic data that are
usually compiled on this periodicity. Flows are measured such that all
activity that takes place during the selected accounting period is recorded.
Stocks are measured at the opening and closing dates of the accounting
period.
A5.23 Commonly, different data sources will have different reference
periods and thus adjustments will be required to allow appropriate
integration. For example, flows may cover a date range that is not aligned
with the selected accounting period and/or stock information will relate to a
non-opening or closing period date. Where adjustments are made these
should be made explicit or if no adjustments are made then the implicit
assumptions should be described.
A5.24 For the measurement of some ecosystem characteristics and services
the use of an annual frequency may not be ideal. For example at larger scales
changes in ecosystem extent may only be detectable over periods of three to
five years. In the other direction, measurement of changes in water resources
may require sub-annual data to detect seasonal variation. As appropriate it is
relevant to record and present specific data using these alternative
frequencies such that decision making and analysis can be best supported. At
the same time, a single frequency is required for the integration of all data,
including economic data, and it is for this purpose that annual recording is
proposed. This frequency also ensures a regular presentation of ecosystem
accounting data to decision makers and supports the mainstreaming of
environmental information that is a core ambition of the SEEA.
A5.25 In addition to these key principles there are a few common tools and
methods that national accounts apply. These are
A5.26 Benchmarking, interpolation and extrapolation. Among the range of
different data sources there will usually be a particularly high quality source
in terms of coverage and quality. Commonly such a source will provide a
benchmark estimate at a point in time or for a given accounting period. Using
this information as a base, it is then common to use indicators to extrapolate
this information to provide more up to date estimates (a process known as
“nowcasting”) and also to interpolate between benchmarks, for example in
128
cases where the best data are collected every 3 years but annual estimates are
required for accounting purposes. Generally, these techniques are applied to
generate the initial estimates for a particular variable and may be
subsequently adjusted through the balancing and integration process.
A5.27 In some respects these types of benchmarking and
interpolation/extrapolation techniques may be regarded as a form of
modelling. The extent to which this is the case will depend on the
sophistication of the technique that is used. Generally, regressions and the
like are not utilised since maintaining these models across the full extent of a
national accounts framework would be very resource intensive. Further, since
the estimates for an individual time series are eventually integrated within a
series of accounting identities it may be difficult to rationalise the statistical
advantage of applying detailed modelling approaches for individual series.
A5.28 Modelling. Where modelling does become more in evidence is when
there is a clear shortage of data for particular variables – i.e. there are no
direct estimates or benchmarks that can be used to provide a starting point. In
this case, modelling may be required. An example in standard national
accounts is the estimation of consumption of fixed capital (depreciation)
which are commonly derived using the so-called perpetual inventory model
(PIM) that requires estimates of capital formation and assumptions regarding
asset lives and depreciation rates.
A5.29 In the context of ecosystem accounting, the spatial detail required is
likely to considerably increase the need for modelling and this will be new
ground for many national accountants. Chapter 5 of the EEA TG considers
the role of biophysical modelling in ecosystem accounting and the general
issue of spatial imputation where information estimated in one location is
applied in other locations. Such modelling and imputation may be relevant in
the measurement of ecosystem extent, ecosystem condition and ecosystem
services. While these may not be traditional “sources” of information for
national accounts type work, there is no particular reason that such modelled
data cannot be directly incorporated. It remains the task of the accountant to
integrate all available data as best as possible. At the same time, a balance
must be found concerning the proportion of data that are modelled within the
overall dataset. Excessive reliance on modelled rather than directly collected
data may raise questions about the accuracy of the information.
A5.30 A general issue that crosses all of the discussion through this section
is that of data quality. Unlike many of the source data that feed into the
national accounts it is not usually possible to give a precise estimate of
common measures of data quality such as standard errors. The melding and
synthesis of multiple data sources makes this task relatively intractable. In the
same context it is challenging to measure the significance of the application
of accounting principles on data quality. While clearly these principles lead
to coherence in the final data – it is often unclear how much adjustment
might have been required in order for the coherence to be enforced.
A5.31 Ultimately it will often be the case that accounts are considered of a
relatively good quality if the picture that they present is broadly considered a
reasonably accurate one. This may emerge from consideration of
i. How well the accounts reflect and incorporate data that are considered to
be of high quality.
129
ii. Commentary by accountants as to the extent of adjustment required
(noting that in a number of situations accounts may be left unbalanced
and the size of the discrepancy may be a measure of quality).
iii. The size of revisions to the estimates where a consistent pattern of large
revisions to initial estimates either up or down would give an indication
as to the relative quality of the source and methods.
iv. The usefulness of the data from the accounts to users. At the end of the
day if the data from the accounts do not support meaningful decision
making or analysis then the quality of the accounts must be questioned.
A5.32 A final area concerns the treatment of uncertainty in accounting
contexts. SEEA EEA Chapter 5 provides an overview of several areas of
uncertainty that may affect information used in ecosystem accounting. By its
nature, accounting aims to provide a single best picture and in this context it
would seem to ignore issues of uncertainty. Three points should be noted.
First, to the extent that the inputs into an accounting exercise are subject to
uncertainty then this should be taken into consideration in the compilation of
the accounts themselves. Ideally, degrees of concern about the data would be
the subject of description in the reporting of accounting outputs. The same
holds true for any assumptions that are applied in the construction of
accounting estimates – for example in terms of estimating future flows of
ecosystem services in net present value calculations.
A5.33 Second, while not generally undertaken, it would be plausible to
consider publishing some ecosystem accounting aggregates within sensitivity
bounds. The challenge of course is to ensure that a balance in the accounting
identities would be meaningfully maintained but with further consideration of
how uncertainty can be usefully reflected within an accounting context would
be welcome.
A5.34 Third, accounting does not provide a model for forecasting future
changes in systems. The national accounts organise information about the
composition and changes in economic activity but do not purport to provide
future estimates of economic growth. Economic models, generally using time
series of national accounts data, perform this role.
A5.35 In the same way, ecosystem accounting is not designed to provide a
model of how the ecosystem behaves that can be used to forecast ecological
outcomes. It records, ex post, measures of changes in ecosystem condition
and flows of ecosystem services. How this information might be combined to
support estimates of future flows or changes in condition is a separate issue
and likely subject to considerable uncertainties. This distinction between
creating a structured set of information and modelling future states is often
not made in scientific discourse and usually forgotten by economists.
However, it is fundamental to understanding the role that accounting may be
able to play in supporting the mainstreaming of environmental information
into decision making.
A5.36 The inappropriateness of the national accounts as a forecasting model
must be distinguished from the use of future data in the derivation of some
national accounting estimates. A particular example is the use of information
on future flows of services in the measurement of ecosystem capacity and
ecosystem asset net present values. While it is true that net present values
require information on future flows, ideally this information should be
obtained from specific data sources, models and expert opinion. Where such
inputs are not available, national accountants will commonly make
130
assumptions about the future flows (usually based on past history) such that a
net present value can be estimated. However, this is quite different from
concluding that the national accounts framework provides a model that can
be used for forecasting.
131
References
Ajani, J and Comisari, C. (2014). Towards a Comprehensive and Fully Integrated Stock and
Flow Framework for Carbon Accounting in Australia. Australian National University,
Canberra.
Aslaksen, I., Framstad, E., Garnåsjordet, P. A., & Lillegård, M. (2012). The Norwegian
nature index: Expert evaluations in precautionary approaches to biodiversity policy, Norsk
Geografisk Tidsskrift–Norwegian Journal of Geography, 66, 257- 271.
Australian Bureau of Statistics (2015). Information paper: An Experimental Ecosystem

Account for the Great Barrier Reef Region. ABS catalogue number 4680.0.55.001, ABS,
Canberra.
Bagstad et al (2014). From theoretical to actual ecosystem services: mapping beneficiaries

and spatial flows in ecosystem services assessments. Ecological Services, 19:64
Banzhaf S, Boyd J (2012) The architecture and measurement of an ecosystem services index.
Sustainability 4:430-461
Barbier E (2013) Wealth accounting, ecological capital and ecosystem services. Environment
and Development Economics 18(2):133–161
Campos and Caparros (2011)
Certain G, Skarpaas O (2010). Nature Index: Statistical Framework and Implementation for
Norway. NINA Report, 542.
CBD (2003) Interlinkages between biological diversity and climate change. CBD Technical
Series No 10. Secretariat of Convention on Biological Diversity, Montreal.
CBD (2010). The strategic plan for biodiversity 2011–2020 and the Aichi biodiversity targets.
UNEP/CBD/COP/DEC/ X/2. United Nations Environmental Program, New York.
Cosier P, McDonald J (2010). A common currency for building environmental (ecosystem)

accounts, Meeting of the London Group on. Environmental Accounting. Santiago, 25 – 28
October 2010 2010.
Dasgupta P S (2009) The Welfare Economic Theory of Green National Accounts.

Environmental & Resource Economics 42(1):3–38
Driver A. (2012). National Spatial Biodiversity Assessment 2011: An Assessment of South

Africa's Biodiversity and Ecosystems. Synthesis Report. Capetown, South Africa: South
African National Biodiversity Institute and Department of Environmental Affairs.
Edens B, Hein L (2013) Towards a consistent approach for ecosystem accounting. Ecological
Economics 90:41–52
European Commission (2011) The EU Biodiversity Strategy to 2020, Publications office of

the European Union, Luxembourg
132
European Commission, International Monetary Fund, Organisation for Economic Co-
operation and Development, United Nations, World Bank (2009) System of National
Accounts 2008. United Nations, New York
Eurostat and OECD (2013). Nutrient budgets: Methodology and handbook. Luxembourg.
FAO (2015). Global Forest Resources Assessment, 2015: How are the World’s Forests
Changing? Rome. Available at: http://www.fao.org/3/a-i4793e.pdf
FAO and United Nations (forthcoming) SEEA Agriculture, Forestry and Fisheries.
FAO and Global Land Cover Network (2009). Land Cover Classification System v.3 (or
Land Cover Meta Language): design criteria. Rome.
Haines-Young R, Potschin M (2013) Common International Classification of Ecosystem

Services (CICES), Consultation on version 4, August-December 2012. EEA Framework
Contract No EEA/IEA/09/003
Harrison A (1993) The Draft Handbook and the UNSTAT Framework: Comments. In: Lutz E
(ed) Toward Improved Accounting for the Environment. World Bank, Washington D.C.
Hein L, et al. (2015) Progress and challenges in the development of ecosystem accounting as
a tool to analyse ecosystem capital. Current Opinion in Environmental Sustainability. 14: 86-
92.
International Integrated Reporting Council (2013), “The international framework”, IIRC,

available at: www.theiirc.org
IUCN (2014) The IUCN Red List of Threatened Species. Version 2014.1.
http://www.iucnredlist.org.
Keith, D.A., Rodríguez, J.P., Rodríguez-Clark, K.M., Nicholson, E., Aapala, K., Alonso, A.,
Bachman, S., Basset, A., Barrow, E.G., Benson, J.S., Bishop, M.J., Bonifacio, R., Brooks,
T.M., Burgman, M.A., Comer, P., Comín, F.A., Essl, F., Faber-Langendoen, D., Fairweather,
P.G., Holdaway, R.J., Jennings, M., Kingsford, R.T., Lester, R.E., Mac Nally, R., McCarthy,
M.A., Moat, J., Oliveira-Miranda, M.A., Pisanu, P., Poulin, B., Regan, T..J., Riecken, U.,
Spalding, M.D. & Zambrano-Martinez, A. (2013) Scientific foundations for an IUCN Red
List of ecosystems. PLoS ONE 8(5): e62111.
Landers, D. H., & Nahlik, A. M. (2012). A national ecosystem services classification system
for final ecosystem goods and services (NESCS), EPA/600/R-12/XXX, US Environmental
Protection Agency, Office of Research and Development, Washington, D.C.
Landers, D. H., & Nahlik, A. M. (2013). Final ecosystem goods and services – Classification
system. US Environmental Protection Agency, Office of Research and Development,
Washington, D.C.
Maes, J., Teller, A., Erhard, M., Liquete, C., Braat, L., Berry, P. and Bidoglio, G. (2013),
Mapping and Assessment of Ecosystems and their Services, An Analytical Framework for
Ecosystem Assessments under Action 5 of the EU Biodiversity Strategy to 2020, Publications
133
Office of the European Union, Luxembourg.
Maynard S, James D, Davidson A (2014) Determining the value of multiple ecosystem

services in terms of community wellbeing: Who should be the valuing agent? Ecological
Economics
Millennium Ecosystem Assessment (2005) Volume I. Ecosystems and human well-being:

Current state and trends. Island Press, Washington, D.C.
Nordhaus W D (2005) Principles of national accounting for non-market accounts. Revised

January 5, 2005. Paper presented to the Conference on Research in Income and Wealth
(CRIW): A New Architecture for the US National Accounts, 16-17 April 2004.
Obst C, Vardon M. (2014) Recording environmental assets in the national accounts. Oxford
Review of Economic Policy. Vol 30, 126-144.
Obst C, Hein L, Edens B (2015). National accounting and the valuation of ecosystem assets
and their services. Environmental and Resource Economics. DOI 10.1007/s10640-015-9921-
1
Philips SJ, Anderson RP, Schepire RE. (2006) Maximum entropy modelling of species
geographic distributions. Ecological modelling, 190: 231-259
Remme, R P, Schröter M, Hein L (2014). Developing spatial biophysical accounting for

multiple ecosystem services. Ecosystem Services 10, 6-18
Schröter M, Barton D N, Remme R P, Hein L (2014) Accounting for capacity and flow of
ecosystem services: A conceptual model and a case study for Telemark, Norway. Ecological
Indicators 36:539-551
Statistics Canada (2013). Human Activity and the Environment: Measuring Ecosystem Goods
and Services 2013. 16-201-XWE. Ottawa: Government of Canada.
Sumarga E, Hein L (2013) Mapping ecosystem services for landscape planning, the case of
Central Kalimantan. Environmental management. 54: 84-97.
TEEB (2010) The economics of ecosystems and biodiversity, Mainstreaming the economics
of nature. A synthesis of the approach, conclusions and recommendations of TEEB.
Routledge, Oxford, UK
UK National Ecosystem Assessment (2011) The UK National Ecosystem Assessment:

Synthesis of the key findings. UNEP-WCMC, Cambridge
United Nations (1993) Handbook of National Accounting: Integrated Environmental and

Economic Accounting, Interim version. Studies in methods, Series F, No. 61. United Nations,
New York.
United Nations (2008a). International Standard Industrial Classification of All Economic

Activities (ISIC), Rev.4. Statistical Papers, Series M, No. 4/Rev.4. Sales No. E.08.XVII.25.
United Nations (2008b). Central Product Classification (CPC) Ver.2.
134
United Nations (2012a). International Recommendations for Water Statistics. Statistical
Papers, Series M, No.91. Sales No. 10.XVII.15.
United Nations (2012b). SEEA-Water: System of Environmental-Economic Accounting for

Water. Statistical Papers, Series F, No.100. Sales No. E11.XVII.12.
United Nations (2013), “System of environmental-economic accounting 2012 –

implementation guide”, Draft presented to the London Expert Group on Environmental-
Economic Accounting, November 2013.
United Nations (2014) Open Working Group proposal for Sustainable Development Goals.
Document A/68/970.
United Nations (2015). Framework for the Development of Environment Statistics. New
York.
United Nations, European Commission, International Monetary Fund, Organisation for

Economic Co-operation and Development, World Bank (2003) Handbook of national
accounting: Integrated environmental and economic accounting, 2003. Studies in methods,
Series F, No 61. United Nations, New York.
United Nations, World Tourism Organisation, European Commission and OECD (2010).
Tourism Satellite Account: Recommended Methodological Framework 2008. Studies in
Methods Series F No 80/ rev1. United Nations, New York.
United Nations, European Commission, Food and Agricultural Organization of the United
Nations, Organisation for Economic Co-operation and Development, The World Bank (2013)
System of Environmental-Economic Accounting 2012 – Applications and Extensions. White
cover edition. United Nations, New York
Nations, International Monetary Fund, Organisation for Economic Co-operation and
Development, The World Bank (2014a) System of Environmental-Economic Accounting
2012 – Central Framework. United Nations, New York
Nations, Organisation for Economic Co-operation and Development, The World Bank
(2014b) System of Environmental-Economic Accounting 2012 – Experimental Ecosystem
Accounting. United Nations, New York
UNDP (2014) Human Development Report. UNDP, New York.
UNEP (2015) Guidance Manual on Valuation and Accounting of Ecosystem Services for
Small Island Developing States, UNEP, Nairobi.
UNU-IHDP and UNEP (2014) Inclusive Wealth Report 2014. Measuring Progress Toward
Sustainability. Cambridge University Press, Cambridge, UK
Vanoli A (1995) Reflections on Environmental Accounting Issues. Review of Income and

Wealth 41(2):113-137
135
Victorian Department of Sustainability and Environment (2013). Victorian experimental
ecosystem accounts, Victorian Government, East Melbourne, Australia.
Weber, J.-L. (2011). An experimental framework for ecosystem capital accounting in Europe,
EEA Technical Report No 13/2011, ISSN 1725-2237, European Environment Agency,
Copenhagen, Denmark.
Weber J-L (2012) Recording Ecological Debts in the National Accounts: Possibilities open by
the development of ecosystem capital accounts. Paper submitted to the ISEE2012
Conference, Rio de Janeiro.
Weber J-L (2014a). Ecosystem Natural Capital Accounts: A quick start package. Technical
series No. 77. Secretariat of the CBD. Montreal
Weber J-L (2014b). Mauritius case study of experimental ecosystem natural capital accounts:
Methodology and preliminary results. Indian Ocean Commission, Mauritius.
World Bank (2011) The Changing Wealth of Nations. World Bank, Washington D.C.
World Bank (2014). Designing Pilots for Ecosystem Accounting, WAVES, World Bank,
Washington.
World Wildlife Fund (2014) Living Planet Report. WWF/ESA.
136

SEEA EEA Tech Rec Consultation Draft 8.1 Dec2015 Final

Uploaded by

Copyright:

Available Formats

SEEA EEA Tech Rec Consultation Draft 8.1 Dec2015 Final

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

SEEA EEA Tech Rec Consultation Draft 8.1 Dec2015 Final

Uploaded by

Copyright:

Available Formats

SEEA Experimental Ecosystem Accounting:

Consultation Draft – December 2015

Prepared as part of the joint UNEP / UNSD / CBD project on Advancing

2. Main aspects of ecosystem accounting 8

3. Spatial infrastructure and definition of units for ecosystem accounting 17

4. The ecosystem accounts 29

5. Accounting for flows of ecosystem services 47

6. Accounting for ecosystem assets 61

8. Valuation in ecosystem accounting 92

BSU basic spatial unit

1.1 Definition and role of ecosystem accounting

1.2 Scope and purpose of the SEEA EEA Technical Recommendations

1.2.2 Connection to other materials

1.2.4 The scope of the Technical Recommendations

1.3.2 Corporate initiatives

1.4 Structure of Technical Recommendations

Key points in this chapter

2.2 The SEEA EEA model and key accounting principles

Figure 2.1 Ecosystem accounting model

Source: SEEA EEA Figure 2.2, UN et al 2014b

2.2.2 Assets and services

2.3 The steps in compiling ecosystem accounts

b. Steps in monetary terms

2.3.2 Summary of compilation steps

2.3.3 Key considerations in compiling ecosystem accounts

2.3.4 What constitutes ecosystem accounting?

Key points in this chapter

3.2.1 Ecosystem Units (EU)

Table 3.1 Land cover classes

3.2.2 Basic spatial units (BSU)

3.2.3 Geographical aggregations

Figure 3.1 Relationships between spatial areas for ecosystem accounting

3.38. In practice, a balance must be struck between providing a highly detailed

3.5.2 Issues requiring ongoing research

Key points in this chapter

Integrated accounts* Combined presentations

4.3 From a SEEA perspective, one ambition is to integrate information on

4.3 Ecosystem condition accounts

4.22 Underpinning these accounts will be information from a variety of sources. In

4.4 Ecosystem services supply and use table

4.4.2. The ecosystem services supply table

4.4.3 The ecosystem services use table

4.5 Ecosystem monetary asset account

Table 4.5 Ecosystem monetary asset account (currency units)

Table 4.6 Accounting entries for depletion and degradation

Ecosystem assets Degradation - reappraisals of

4.68 However, within this accounting construct, further consideration of exactly

4.6 Integrating ecosystem accounting information with standard national accounts

Int. consumption – Products 0 0 0 0 0

Gross value added 120 110 230 230 230

Less Consumption of fixed capital 10 10 10 10

Less Compensation of employees – 50 50 50 50

Allocation and use of income

Less Final consumption – Products 200 200 200 200

Key points in this chapter

5.14. Distinguishing final and intermediate ecosystem services: The distinction

5.24. The treatment of ecosystem disservices: Ecosystem disservices arise in cases

5.3 The classification of ecosystem services

5.4 The role and use of biophysical modelling