Sustainable Wind Energy Planning Through Ecosystem Service Impact

Renewable and Sustainable Energy Reviews 178 (2023) 113252
Contents lists available at ScienceDirect
Renewable and Sustainable Energy Reviews

journal homepage: www.elsevier.com/locate/rser
Sustainable wind energy planning through ecosystem service impact

valuation and exergy: A study case in south-central Chile
Yenisleidy Martínez-Martínez a, Jo Dewulf b, **, Mauricio Aguayo c, Yannay Casas-Ledón d, e, *
a
Faculty of Engineering, University of Concepción, Víctor Lamas 1290, Concepción, Biobío Region, Chile
b
Research Group Sustainable Systems Engineering (STEN), Ghent University, Coupure Links 653, 9000, Gent, Belgium
c
Territorial Planning and Urban Systems Department, Faculty of Environmental Sciences-EULA Center, University of Concepción, Víctor Lamas 1290, Concepción, Biobío
Region, Chile
d
Environmental Engineering Department, Faculty of Environmental Sciences-EULA Center, University of Concepción, Víctor Lamas 1290, Concepción, Biobío Region,
Chile
e
Unidad de Desarrollo Tecnológico (UDT), University of Concepción, Avenida Cordillera 2634, Parque Industrial Coronel, 4191996, Coronel, Chile
A R T I C L E I N F O A B S T R A C T
Keywords: Wind energy has great potential to drive Chile’s energy sector towards a more sustainable model; however, it
Wind energy could induce landscape changes impacting ecosystem services. Accordingly, this study aimed to identify the best
Suitability analysis areas for wind energy development in Biobío and Ñuble regions, by integrating metrics from suitability analysis,
Ecosystem services
ecosystem service valuation and exergy. The trade-offs between ecosystem services and wind electricity gener
Exergy indicator
ation were assessed through biophysical indicators. Then, a new indicator based on ecosystem services impacts
Impact assessment
and exergy concept was developed to determine the most sustainable sites for wind energy development. Results
showed that deploying wind energy could have the highest adverse effects on erosion regulation services, with an
increase of 40%–50% in soil loss for Ñuble and Biobío. Moreover, the crop and timber provision were also
reduced by 10%, while water-related services were favored by an increment of 10% in water savings. The
proposed indicator took negative values in all suitable zones under analysis (− 25 to − 307 GJex/GWh), indi
cating that ecosystem services losses exceeded gains. Results suggest that seven zones are potential sites for wind
energy development based on their impacts which were up to − 39 GJex/GWh. This work provides evidence for
promoting sustainable wind energy planning by locating the projects in zones with the lowest detrimental
ecosystem services impacts.
1. Introduction development [2]. To minimize potential conflicts and maximize bene

fits, planners should locate wind projects in areas showing the highest
Wind energy has become a leading option to diversify the energy energy production and the lowest ESs impacts.
sector and reduce carbon emissions [1]. However, wind development Land suitability analysis is widely used in spatial planning contexts,
can induce landscape changes, triggering conflicts with land uses and to identify sites with high resource potential that simultaneously meet a
the provision of ecosystem services (ESs) [2]. For example, siting wind set of techno-economic and environmental criteria. Multiple interna
farms on agroforestry lands can impact crop and timber production and tional studies have successfully conducted onshore and offshore wind
local climate regulation [3]. Clearing vegetation to locate turbine plat farms suitability analysis through a combination of GIS and different
forms and roads can affect erosion control [1]. Wind turbines can induce Multi-criteria Decision-Making (MCDM) methods (Sotiropoulou et al.
bird and bat mortality, plus they have a visual and noise impact on [6], Vavatsikos et al. [7], Tercan [8], Gigović et al. [9], Aydin et al. [10],
nearby residents [1,4,5]. Land use conflicts and ESs trade-offs may affect Jangid et al. [11], Serdar Geng et al. [12], Karipoglu et al. [13]). Spe
public support for this technology, jeopardizing new project cifically for the onshore wind farm, Aydin et al. [10] identified priority
* Corresponding author. Environmental Engineering Department, Faculty of Environmental Sciences-EULA Center, University of Concepción, Víctor Lamas 1290,
Concepción, Biobío Region, Chile.
** Corresponding author.
E-mail addresses: [email protected] (Y. Martínez-Martínez), [email protected] (J. Dewulf), [email protected] (M. Aguayo), [email protected]
(Y. Casas-Ledón).
https://doi.org/10.1016/j.rser.2023.113252
Received 6 September 2022; Received in revised form 17 February 2023; Accepted 6 March 2023
Available online 11 March 2023
1364-0321/© 2023 Elsevier Ltd. All rights reserved.
Y. Martínez-Martínez et al. Renewable and Sustainable Energy Reviews 178 (2023) 113252
sites for wind energy development in Turkey, integrating GIS and a fuzzy in ESs evaluation [17]; nonetheless, it is a controversial issue, and some
decision-making approach. In this study, wind potential and environ ESs cannot be expressed in monetary terms [18]. In this sense, the
mental criteria (noise, bird habitats, safety, and aesthetics) were exergy concept has emerged as a valuable metric for accounting the
considered and mapped as fussy sets. A similar analysis was conducted natural (e.g., net primary productivity) and non-natural resources (e.g.,
by Karipoglu et al. [13], who found two suitable sites for wind energy fertilizer) [19], which represents the maximal amount of work that can
implementation in Develi (Turkey) when the number of environmental be extracted from resources, indicating the resource quality and effi
constraints (wind speed, forests, military regions, civil and military ciency in its use [20,21]. In addition, the exergy reflects the potential
aviation, designated regions, agricultural, water sources, roads and work capacity of the ecosystems, measuring its flow, dynamic and value
ports, fault lines, bird’s migration paths, and energy transmission lines) of ecosystem intrinsic value [22], so it can provide a more holistic
was higher than those included in Aydin et al. [10]. On the other hand, characterization of ecosystems. Therefore, it could offer a new
Vavatsikos et al. [7] have incorporated the economic aspect (e.g., in perspective as ESs aggregation tool because the resources (e.g., biomass,
ternal rate of return, discounted payback period, levelized electricity water, soil) can be expressed in exergy units (MJex).
cost) in GIS-based suitability analysis for evaluating the best candidate Exergy has been widely used to improve thermal processes’ effi
wind energy projects. ciency and lately has gained relevance in environmental and global
Besides, other studies have focused on developing a comprehensive sustainable studies [20]. Mainly, it has been used to assess land use
decision-making framework and improving multicriteria techniques. In change impacts on primary production [23], land use impact in the Life
this sense, Gigović et al. [9] developed a reliable model for best zone Cycle Assessment framework [19,24,25], to evaluate resource use effi
identification for wind farm installations using the province of Vojvo ciency in agricultural activities [26] and other economic areas like the
dina, Serbia, as a case study. The proposed model combined GIS and energy sector [27,28]. Moreover, the exergy principles have been also
MCDA through various multicriteria methods (e.g., decision-making applied for quantifying the ecosystems services values [22]. In this
trial and evaluation laboratory, the analytic network process, and sense, Sheng et al. [22] evaluated the intrinsic value of marine ecosys
multi-attributive border approximation area comparison). In the same tems (river and coastal bay); nonetheless, the study only focused on the
line, Sotiropoulou et al. [6] proposed a new decision-making framework value of each ecosystem can provide, but not considered the impacts of
through the combination of GIS and PROMETHEE II method to improve the human activities on the value of ESs. To our best knowledge, exergy
the selection of wind farm siting. indicators have not been used in spatial energy planning studies.
Most land suitability studies discussed previously consider technical- In the Chilean context, efforts have also been made in spatial energy
economic and environmental attributes regardless of the multicriteria planning. A GIS-based multicriteria evaluation was performed by
techniques used. Regarding ecological attributes, the most common sub- Suuronen et al. [29] to optimize the location of solar facilities in the
criteria considered in the literature are protected areas, water bodies, Atacama Desert. Likewise, MINENERGIA/GIZ [30] assessed the poten
and bird migration paths, among others; however, the ecosystem ser tial for solar, hydro, and wind development at the country level. Several
vices values have not been considered as a criterion. In this sense, few techno-economic and environmental constraints were used in the
studies integrated the ecosystem services evaluation into land suitability GIS-based suitability analysis; however, selecting the best suitable zones
analysis for renewable energy development. For example, Kienast et al. for renewable energy implementation based on ESs evaluation have not
[14] implemented a landscape service mapping approach to identify been considered. The first and exclusive precedent found at the local
low-conflicts sites for energy production from three landscape-sensitive level was reported by Martínez-Martínez et al. [16], who complimented
renewable sources, including wind power. Low-conflicting lands were the GIS-based suitability analysis with expert-derived ESs matrix and
found by overlaying the potential energy layer and six ESs maps. Simi multicriteria methods for identifying wind farm suitable sites. Never
larly, Egli et al. [15] integrated ESs and optimization tools to evaluate theless, the authors restricted their analysis to visualizing the existing
the trade-offs between five services and wind generation in complex supply capacities of ESs that could co-exist with potential wind energy
landscapes. This method identified the low-impact sites with potential sources, excluding the quantitative valuation of ESs changes for future
for wind generation, including an optimal wind turbine system design at regional wind development scenarios.
the lowest environmental costs. A pioneering work in this area for the To the best of our knowledge, the previous studies have not provided
Chilean context, was recently published by Martínez-Martínez et al. a comprehensive methodological framework that integrates the spatial
[16], which integrated GIS-based land suitability analysis with land suitability analysis, ESs evaluation and exergy concept for selecting
expert-derived ESs matrix and multicriteria methods to identify suitable the more suitable site for wind energy development scenarios with less
sites for different renewable energy sources (e.g., wind, solar, biomass) ESs impacts. Therefore, the present study propose new contributions to
in two Chilean regions. The author analyzed the suitability level from the field by i) developing a comprehensive methodological framework
the perspective of the existing supply capacities of ESs as a proxy for that integrates the ESs approach in land suitability analysis for the
potential ESs trade-offs. development and effective planning of wind energy, ii) a novel exergy-
The above-mentioned studies were limited to assessing the trade-offs based indicator for evaluating the ESs impact in a single metric, and iii) a
between some ESs and renewable energy and they did not quantify or validation of the proposed methodological framework and exergy-based
not explicitly report on the ESs flow changes associated with wind en indicator for identifying the best suitable sites for future wind energy
ergy development scenarios. Despite their attempt to integrate the ESs scenarios for Biobío and Ñuble regions, located at the south of Chile.
approach, a conceptual framework to address spatial energy planning
from an ESs valuation perspective is still missing. The ecosystem services 2. Material and methods
concept explicitly connects the natural environment with human well
being, and its integration with land suitability analysis could provide a Fig. 1 shows the methodological steps followed in the work, which
solid basis for decision-making. first includes identifying suitable sites (no technical or territorial con
On the other hand, quantifying multiples ESs indicators is complex straints) available for wind energy development and delimiting sub-
due to various existing indicators, site-specific data availability, and areas with potential. Ecosystem service impacts (ΔESs) on services
metrics. This latter aspect is crucial for decision-making because the were calculated as differences between ESs in scenario two minus
variety of ESs metrics does not readily understand for decision-makers to baseline. ESs indicators were selected to quantify the value of those
find the best alternative in terms of sustainability. Based on that, the services in two scenarios: 1) a baseline scenario representing current
aggregation methods help to simplify the communication for decision- land use conditions and ESs provision and 2) a scenario denoting the
making process, and the ESs metrics could be brought to comparable development of the wind potential of suitable sub-areas. All biophysical
units. In this regard, the monetary valuation method is commonly used indicators were converted into exergy terms by using the resource’s
2
Fig. 1. Methodological scheme (Crop: indicator for food crops (kg/GWh); Timber: indicator for timber production (m3/GWh), WF blue: indicator for water (m3/
GWh); WF gray: indicator for mediation of waste (m3/GWh) and Soil loss: indicator for erosion regulation. Subscripts 1 and 2 represent the estimation of each
indicator in scenario 1 and 2, respectively. Meanwhile, ΔESs broadly denotes ecosystem service impacts, while ΔCrop, ΔTimber, ΔWFblue, ΔWFgray and ΔSoil loss
indicate more specifically the impact on each service. BOS refers to the new balance of service indicator (GJex/GWh)).
standard exergy content. Further, they were aggregated into two groups both regions used to conform the former Biobío Region; however, from
depending on whether their impact represented a benefit or damage. September 2018, they separated and Ñuble was consolidated as an in
The best suitable zones were selected by calculating the balance be dependent region. Despite that, this entire geographical area has been
tween gains and losses (BOS indicator) and applying ESs damage- selected due to its significant contribution to the national electricity
minimizing approach. matrix, including its untapped potential in terms of renewable energy
resources, to promote the development of a more sustainable energy
2.1. Study area sector. In this sense, by 2021, the Biobio region represented the second
region (5590 MW) with highest gross installed generation capacity in
The study area considered in this work is the Biobío and Ñuble re the Chilean electricity matrix, following Antofagasta (8074 MW) [31].
gions, both located in South-Central Chile at 36◦ 00′ S-38◦ 29′ S (Fig. 2) Regarding renewable energy sources, Biobío region features a high wind
and covering a surface of 37068 km2 approximately. Administratively, energy potential [16,30], represented by wind profiles ranging from 2 to
Fig. 2. Study area.
3
12 m/s at 100 m high. Despite, the net installed capacity for wind energy proposed ESs indicators attempted to correlate the effect of land use
in the Biobío region increased by January 2022 to 176.0 MW, this value changes on ESs between these defined scenarios of wind energy devel
is below the feasible technical potential which is projected to 3225.4 opment. For instance, if potential suitable sites for wind energy occupy
MW in the next few years (https://www.sea.gob.cl/). In addition, the croplands, then the potential decrease of crop productivity caused by
current energy policy toward a carbon neutrality target in Chile by 2050 agricultural lands reduction could represent the impact on service sup
encourages wind and solar investment [32]; therefore, it is expected an ply values.
accelerated wind energy development at regional scale, for which could The ESs selection was based on previous studies reported by Martí
be essential the development and adoption of new spatial planning nez-Martínez et al. [16], who identified priority ESs provided by existing
strategies to locate those projects with the lowest potential environ land cover types in Biobío and Ñuble regions (SM Table S2). From these
mental impact. priority ESs, only those sensitive to wind energy development were
In other aspects, like climate, this region characterizes by a Medi selected. For a more precise selection of indicators for each ESs, the
terranean climate with a transition to temperate wet and rainy in the CICES v5.1 [35] classification system was used and complemented with
south direction. The mean temperature and monthly precipitation in the a literature review [1,3,4,36,37]. In this study, cultural ESs were not
winter-summer period fluctuate from 10 to 17 ◦ C to 20–250 mm, included because it is complex to quantify them in terms of exergy owing
respectively. The vegetation in the area by 2015 was dominated by their intangible and non-material nature. Several of the ESs included in
forest plantations (34.2% of total surface), native forests (24.3% of the regulation and maintenance categories listed in Table S2 were not
total), and cropland (19.3% of total), which has a significant influence considered in this study, due to the lack of data for these regions. Finally,
on the agro-forestry sector being a very important economic activity in three provisioning services and two regulation and maintenance services
the region. indicators were chosen (Table 1).
All indicators listed in Table 1 are calculated for two scenarios: 1) a
2.2. Suitable areas for wind energy development and potential energy scenario depicting current conditions (baseline scenario) and 2) a sce
output nario where the full wind potential of suitable areas is developed. The
difference between the two scenarios (scenario 2 minus 1) equals the
This work was based on the wind suitability map developed by potential impact (ΔESs) as described in eq. (1).
Martínez-Martínez et al. [16], who constructed by excluding from a
ΔESs = ESsscenario 2 − ESsscenario 1 (1)
wind resource layer (wind speed, m/s) five layers depicting technical
and territorial constraints (altitude >2000 masl, slope >15◦ , distance to
2.3.1. Crop indicator
urban/industrial centers <1 km, distance to rivers/water bodies/wet
Crop indicator (eq. (2)) was used to quantify the total amount of food
lands <300 m, protected areas). The base suitability map was improved
crops that could be harvested in suitable zones for wind power gener
by adding a constraint about a restricted distance of 700 m [30] to wind
ation (kg/GWh, annual basis).
turbines of projects currently installed or under evaluation. The wind
power project portfolio (Supplementary Material SM, Table S1) avail mj
∑17
Yieldi • Surfaceij
able at the Environmental Evaluation Service’s website (https://sea.gob. Cropj = = i=1 (2)
Ej Ej
cl/sea/que-es-seia) was analyzed. A layer containing the spatial location
of all wind turbines (928 turbines) and the surface occupied by per where mj is the mass of food crops harvested (kg/yr) and Ej is the wind
manent works were produced in ArcGIS, and a buffer of 700 m was energy output (GWh/yr) for the j-sub-area, Yieldi is the yield of the i-
generated. This constraint layer was then subtracted from the base cultivated crop (kg/ha), Surfaceij is the surface harvested of the i-culti
suitability map. Lastly, sub-areas were delimited by considering a min vated crop (ha) in the j-zone. Regional statistics about yields and har
imum continuous area (i.e., no restrictions in between) of 1 ha based on vested surfaces were obtained from the Bureau of Agrarian Studies and
the minimum space occupied by existing projects. Policies [38] and values used correspond to 11-year averages (2010/11
Potential energy outputs (Ej, GWh/yr) for each suitable zone were to 2020/21). Seventeen crop types, including grains, legumes, vegeta
simulated using the Wind Explorer platform (http://eolico.minenergia. bles, and industrial crops were considered in calculations.
cl/potencia). For the simulation, a Vestas V126-3.3 MW wind turbine
array spaced 390 m apart were assumed because this corresponded to 2.3.2. Timber indicator
the mean distance between turbines and net power per turbine averaged Timber indicator (eq. (3)) accounted for the total biomass produced
for forty-one projects of the wind power project portfolio in the regions. in suitable sites for wind energy (m3/GWh, annual basis).
∑6
2.3. Impacts on ecosystem services: selection of indicators and Vj
Timberj = = i=1
MAIi • Surfaceij
(3)
quantification Ej Ej
The impact quantification of ESs was carried out following the where Vj is the volume of biomass from forestlands in the j-zone (m3/yr),
Ecosystem Services Assessment (ESA) framework proposed by Rosenthal MAIi is the mean annual increment in commercial volume for the i-forest
et al. [33] and Burkhard et al. [34], which involves the following steps: specie (m3/ha.yr), Surfaceij is the surface occupied by i-specie in the j-
i) problem identification and scoping, ii) data collection, iii) building of zone (ha). Three forest species of native forest (Roble-Rauli-Coihue,
scenarios, iv) analysis of ESs and their indicators, v) integration of the Sclerophyll, and Evergreen) and exotic plantations (Pinus radiata, Euca
results and vi) results communication. Accordingly, the scenarios defi lyptus globulus, and other species) were detected. Data about the mean
nition and the qualitative and quantitative methods for evaluating the annual growth for all forest species was obtained from the Ministry of
ESs supply are required. Environment [39] (SM, Table S3).
In this study, two scenarios were proposed for evaluating the change
in specific ESs supply caused by wind energy development. The first 2.3.3. WFblue indicator
scenario represents the current land-use conditions (baseline scenario), WFblue indicator (eq. (4)) is based on Hoekstra et al. (2011)’s blue
and a second scenario involves the land use changes associated with water footprint indicator (BWF) [40] and it was used to assess the vol
wind farms installation. ume of irrigation water consumed in suitable zones for wind develop
The land-use change was considered the main stressor in the cause- ment (m3/GWh, annual basis). This indicator was calculated following
effect chain in ESA, as it directly affects the biophysical structures and the methods proposed by Jaramillo (2017) [41] in Chile, i.e., the
processes of ecosystems (hence, service provision) [18]. Based on that, amount of water needed to cover the deficit between crop
4
Table 1
Potential ecosystem services impacts related to wind development in the study area and proposed indicators for quantification. Ecosystem services classification and
description according to the Common International Classification of Ecosystem Services (CICES v5.1) [35]. Potential impacts and proposed indicators were retrieved
from the literature [1,3,4,36,37].
Section Division Group Class Potential impact Indicator
Provisioning Biomass Cultivated terrestrial Cultivated terrestrial plants Agriculture co-exists with wind parks, but the Crop (kg/GWh)
plants for nutrition, grown for nutritional transformation and further occupation of
materials, or energy purposes croplands by wind farm infrastructure can impact
food crop production.
Fibres and other materials Wind farms can co-exist with forest lands, but Timber (m3/
from cultivated plants, (…) timber provision could be impacted by the GWh)
for direct use or processing clearing of forest vegetation near wind parks.
Water Surface water used for Surface water used as a The usage of irrigation water could decrease Water footprint
nutrition, materials, or material (for non-drinking because of the occupation of agricultural land by blue (WFblue,
energy purposes) wind infrastructure. m3/GWh)
Regulation & Regulation of physical, Regulation of baseline Control of erosion rates Vegetation removal to develop wind farms can Soil loss (Mg/
Maintenance chemical, and biological flows and extreme events aggravate soil loss due to erosion processes. GWh)
conditions
Transformation of Mediation of waste, Dilution by freshwater and A reduction in agricultural activities could reduce Water footprint
biochemical or physical toxics and other marine ecosystems pollutant loads from fertilizers reaching water gray (WF gray,
inputs to ecosystems nuisances by non-living bodies. m3/GWh)
processes
evapotranspiration and effective precipitation during the growing sea minus the nitrogen uptake by harvested crops.
son of the crop. ∑
17
∑12 ( ) β• (Appli − Yieldi •Nyld i )

BWFj max 0, ETckj − Pefkj • 0.001 • Surf GWFj
i=1
WFblue j = = k=1 (4) WFgray j = = Nmax − Nnat

(8)
Ej Ej Ej Ej
where ETckj is the crop evapotranspiration (mm/month), Pefkj is the where β is the leaching-runoff fraction (dimensionless), assumed to be
effective precipitation for the k-month in the j-zone (mm/month), Surf is 0.1 according to Osorio U. (2013) [47]. Appli is the nitrogen load applied
the area (m2), and the term 0.001⋅Surf is a conversion factor from mm to according to the fertilization formula of the i-cultivated crop (kg/yr).
m3. ETc denotes crop water demand, and it is the sum of the evaporated The term Yieldi is the crop yield (kg/yr), Nyld i represents nitrogen con
water from the soil plus the crop’s transpiration. This parameter was tent in the crop yield (fraction), Nmax is the maximum allowed con
assessed according to the FAO’s Irrigation and Drainage Paper No. 56 centration of nitrogen (kg/m3) and Nnat is the natural concentration of
[42] as follows in eq. (5). nitrogen in the water body (kg/m3). The fertilizer application formula
∑
17 for harvested crops was obtained from ODEPA (2010) [48], the nitrogen
ETcj = EToj • Kci (5) fraction in crop yield was obtained from Soil and Water Assessment Tool
i=1 (SWAT+) database, Nmax equal to 0.0006 kg/m3 for those zones in
Biobío (D.S. Nº 9/2015 MMA [49]) and equal to 0.01 kg/m3 for zones in
where EToj and Kci are the reference evapotranspiration (mm/month) Ñuble [47]. Natural nitrogen concentrations were assumed to be zero.
and the crop coefficient for the i-crop (dimensionless). Kc values were
obtained from the literature [42,43] for each crop development stage 2.3.5. Soil loss indicator
(Kcinitial, Kcdevelopment, Kcfinal), and ETo was estimated using the modi The Soil Loss indicator (eq. (9)) assesses potential soil losses in
fied Hargreaves (MH) method [44]. Hargreaves equation required suitable zones for wind generation (Mg/GWh, annual basis). In this case,
extraterrestrial radiation (MJ/m2.day) and 20-year average climate data the Universal Soil Loss Equation (USLE) [50] was applied to predict
from the Center for Climate and Resilience Research (CR)2 (https: average rates of soil erosion according to rainfall patterns, soil types,
//www.cr2.cl/datos-productos-grillados/). Monthly values were esti topographic features, crop systems, and management practices of the
mated as the sum of daily calculations. Meanwhile, Pef was calculated study area.
using the USDA Soil Conservation Service method like in Ganesh et al.
[45] and total monthly precipitations (Pj in mm/month) according to Soil lossj =
Aj Rj • Kj • LSj • Cl • Pj
= (9)
eqs. (6) and (7). Ej Ej
( )
Pj 125 − 0.2 • Pj where Aj is the computed soil loss for the j-zone (Mg/ha.yr) and Rj
Pefj = for P ≤ 250 mm (6)
125 represents the rainfall erosivity factor (MJ.mm/ha.h.yr). Kj and LSj are
the soil erodibility factor (Mg.ha.h/MJ.ha.mm) and the slope-steepness
Pefj = 125 + 0.1 • Pj for P > 250 mm (7) factor (dimensionless). While Cl is the cover and management factor for
the l-land cover type (dimensionless) and Pj is the support practice factor
2.3.4. WFgray indicator (dimensionless). R-factor was calculated using the regression equation
The WFgray indicator is based on Hoekstra et al. (2011)’s gray water proposed by Bonilla and Vidal [51] for Central Chile, as indicated in eq.
footprint indicator (GWF) [40], quantifying the volume of freshwater (10).
needed to assimilate nitrogen pollution from fertilizer application in
zones with potential for wind energy (m3/GWh, annual basis), as Rj = 0.028 • Pj 1.534 (10)
expressed in expression 8.
The freshwater volume was calculated as the ratio of nitrogen where Pj is the annual precipitation (mm/yr), and K-factor was retrieved
loading reaching the recipient body divided by the difference between from Bonilla and Johnson [52]. The LS factor was computed following
water quality standards for nitrogen minus its natural concentration in the method reported by Balabathina et al. [53] (eq. (11)).
the receiving body [40,46]. The nitrogen loading reaching water body
constitutes a fraction of the difference between total nitrogen applied
5
( )m ( )n
Flow Accumulation • Cell Size sin S benefits obtained minus the damages caused by landscape trans
LSj = (m + 1) • • (11)
22.13 0.0896 formation for generating wind energy (see eq. (16) and Fig. 3).
BOSj = Gainj − Lossj (16)
where exponents m = 0.5 and n = 1.3. Flow accumulation was generated
in ArcGIS from a digital elevation model using a multiple flow direction
where: BOSj < 0, then losses exceed benefits.
algorithm. Cell size refers to the pixel size (30 × 30 m), and S is the slope
BOSj = 0, then gains equal to the losses.
(radians). Meanwhile, C-factor was estimated for each l-land cover type
BOSj > 0, then the benefits exceed the losses.
using the NDVI-based approach [54], as described in eq. (12), which
Finally, suitable zones were ranked from best-to-worst an ESs
requires satellite imagery and a land cover database.
damage-minimizing approach was applied to select the best areas to
( )
NDVI develop wind energy projects: i.e., areas with the highest BOSj values are
Cl = exp − α • (12) the best options to develop wind energy.
(β − NDVI)
where α = 2, β = 1, and NDVI is the Normalised Difference Vegetation 2.5. Data processing
Index (dimensionless). NDVI was calculated using the RED and NIR
bands of the Landsat 8 OLI/TIRS C2 L2 Science Products (https://eart The quantification and mapping of each ESs indicator was conducted
hexplorer.usgs.gov/). P-factor was assumed to be one because there is in a GIS environment, and it was complemented with spreadsheets and
no information regarding soil conservation practices in the region. Python programming depending on the input data formats, time and
computing capacity requirements of each model. The primary input data
mainly comes from cartographic documents, Landsat collections, cli
2.4. Exergy-based indicator for selecting the best areas to develop wind matic recordings, and regional statistical reports. The datasets were
energy usually arranged as raster format in ArcGIS 10.8 software at 30 m spatial
resolution.
The ESs-exergy principle could represents the potential work ca More precisely, crop and timber indicators, dependent on regional
pacity of the ecosystems, indicating the fraction of ecosystem’s that agroforestry statistics data (e.g., crop yield and the surface area occu
cannot be used by human well-being and/or by the ecosystem itself. pied), were calculated via spreadsheets. Since results were not geore
Based on this principle, a novel Balance of Service (BOS) indicator was ferenced at the regional scale, they were mapped in ArcGIS 10.8
developed here to provide a single metric for the quantification of the software using a land cover map and a look-up table approach [14]. The
ESs impacts associated to wind energy in suitable sites, by integrating land cover maps used for the study area correspond to the last update
the traditional ESs assessment with the exergy principle. For this, all ESs provided by CONAF in 2015 (https://sit.conaf.cl/). Crop and timber
impacts were expressed in exergy units (GJex/GWh) and grouped into a indicators were mapped by adding their corresponding values to the
single metric called BOS, representing the ecosystem degradation. land cover map’s attribute table, assuming equal distribution across
Exergy is defined as maximus amount of work that can be extracted cropland and forest land use types, respectively.
from resources in relation to a reference environment, which is defined The blue and gray water footprints integrate different input data
by temperature of To = 298.15 K, pressure P = 1.01 bar and the atmo such as land cover maps, climatic data (e.g., precipitations, tempera
sphere composition [55]. The exergetic value estimated for each ESs ture), agricultural statistics and specific parameters (e.g., Kc, rainfall
flows is only determined by the chemical exergy component because the erosivity, soil erodibility factor, cover and management factor). The pre-
physical, kinetic and potential interactions concerning the reference processing of parameters such as Kc, applied nutrient load, and nitrogen
state are negligible [27]. uptake, derived mainly from agricultural statistics and tabulated data,
The chemical exergy values for freshwater, biomass and soil re was carried out via spreadsheets. Subsequently, both indicators were
sources were determined according to eq. (13). estimated in ArcGIS using spatial analyst tools (e.g., raster calculator),
providing the spatial distribution of blue and gray water footprints maps
ΔESsex = Ex0i • ΔESs (13)
in raster format.
The assessment of soil loss indicator was carried out entirely in
where ΔEssex is the potential ESs impact in exergy units (Gjex/GWh),
ArcGIS using spatial analyst tools such as hydrology tools, slope, and
Ex0i is the chemical exergy content of i-resource linked to the service
raster calculator, generating a raster map with annual soil losses as
(ΔCrop and ΔTimber = biomass; ΔWFblue and ΔWFgray = water; ΔSoil
output.
loss = soil). Standard chemical exergy values were obtained from
Meanwhile, the annual electricity generation (GWh) in the suitable
literature: freshwater = 5e-11 GJex/m3 water [27], soil = 0.02161985
zones was estimated by creating a point grid of 390 m away in ArcGIS,
GJex/kg soil [56].
considering that each point corresponds to a wind turbine. Then, po
For crop and timber provisioning, the amounts obtained from eqs.
tential electricity generation at each point/turbine was simulated
(2) and (3) were converted in carbon-based (kgC/kg dry matter) metric,
(http://eolico.minenergia.cl/potencia) according to existing wind speed
assuming a carbon content per dry matter factor of 50% [57]. The
values. This analysis required the use of Python programming and web
biomass-to-exergy conversion factor (0.0429 GJex/kgC) provided by
scraping techniques. The point map was then converted to raster using
Alvarenga et al. [58], was used for converting biomass in exergy value.
ArcGIS tools.
Then, all ΔEssex (absolute values) for each j-zone were aggregated
Finally, zonal statistics tools were used in ArcGIS to calculate the
into two main groups according to whether they represent a benefit
total value of each ESs indicator and annual electricity generation per
(gain) for the community or damage (loss). “Gain” (eq. (14)) grouped
suitable zone and scenarios, which were then exported to a spreadsheet.
WF blue and WF gray because their reduction implies water savings
There, the ratio between ESs indicators and electricity generation per
(beneficial), while “Loss” (eq. (15)) grouped Crop, Timber and Soil loss
suitable sites was computed, subsequently, the difference in ESs values
because the reduction of food crops and timber and the increase of soil
between scenarios (scenario 2 minus scenario 1), obtaining at the end
losses implies a damage.
⃒ ⃒ ⃒ ⃒ the potential ESs gains or losses per suitable site (see supplementary
Gainj = ⃒ΔWFblue ex j ⃒ + ⃒ΔWFgray ex j ⃒ (14) material, Table S4).
⃒ ⃒ ⃒ ⃒ ⃒ ⃒
Lossj = ⃒ΔCropex j ⃒ + ⃒ΔTimberex j ⃒ + ⃒ΔSoil lossex j ⃒ (15)
The BOS indicator, expressed in terms of GJex/GWh, reflects the
6
Fig. 3. Simplified representation of the exergy balance of ecosystem services. The green bar represents services gains and the red bar denotes services losses.
4. Results and discussion delimited in the study regions, as shown in Fig. 4, occupying a total
surface of 279156.2 ha, which is approximately the 7.5% of the total
4.1. Characterization of suitable zones for wind development regional area. Ñuble Region holds 140335.7 ha suitable distributed in 28
zones, while Biobío Region features 138820.5 ha (1.1% lower than
Forty zones with an aptitude to develop wind energy projects were Ñuble’s suitable area) merged in 12 zones.
Fig. 4. Characterization of zones suitable for wind energy deployment. The table shows the mean annual electricity output per wind turbine in GWh, the percentage
that cropland, native forest and forest plantations land use types represent of the total surface of each zone and of the total regional surface.
7
In terms of wind potential, simulations indicated a potential for removed in case of fully developing wind energy at suitable zones. The
electricity generation averaged over all suitable sites of 7.9 GWh and Ñuble Region is expected to be the most affected in terms of loss of
9.0 GWh per turbine in Ñuble and Biobío, respectively. The higher wind agricultural lands and native forest, with the transformation and later
potential corresponded to zones in Biobío as they have better wind occupation of 8481.6 ha (65% of total cropland losses) and 530.6 ha
profiles: e.g., Biobío zones portrayed an average wind speed of 6.2 m/s (53.6% of total native forest losses) by permanent works, respectively. In
at 100 m high, while mean wind velocity in Ñuble was 8% lower. contrast, the Biobío Region gathers the highest reductions of exotic
However, wind potential in some zones could be under or over- plantations with 7504.6 ha (68.8% of total plantations losses).
estimated as a single model of 3.3 MW was used during all simula
tions. In this sense, wind and topographic features in some areas could 4.2. Potential impacts on ecosystem services
require other turbine models with different power specifications, finally
leading to higher or lower electricity generation per zone. Fig. 5 illustrates the potential impacts on ESs of developing wind
Compared to local studies [30], our estimations are higher as we energy in suitable areas identified in Biobío and Ñuble regions, using
assume a higher power wind turbine model for the simulation. In five ecological indicators disaggregated by provisioning and regulation
addition, the suitable areas identified could be overestimated because and maintenance services. For each indicator, results are expressed per
the existing road networks, railways, or areas occupied by other projects unit of electricity (GWh) that could be potentially generated in each
were not considered as constraints; accordingly, the wind energy suit suitable site. Impacts comprise the value of indicators calculated in
able surface could diminish the effective area. scenario 2 (developing wind energy in suitable zones) minus scenario 1
On the other hand, different land use and cover types exist in suitable (baseline scenario).
areas, primarily valuable lands with high potential to supply provi Wind energy development in suitable sites reduces all ESs indicators
sioning services, particularly crop and timber. In this sense, around values compared to their baseline state, except for the soil loss indicator.
133229 ha of cropland, 10134 ha of native forest and 110734 ha of Also, there is a marked variation in ESs values per suitable zones and
exotic forest plantations were quantified in both regions. Ñuble Region regions. This may be influenced, firstly, by land cover types prevailing in
concentrates the highest percentages of areas on agricultural lands each zone and thus mostly intervened. In addition, spatial variations in
(64.9%) and native forests (53.2%). In contrast, zones in Biobío have the climatic variable (precipitation, temperature, radiation or wind speed),
greatest surface of exotic plantations (68.9%). Among all suitable areas, topographic features (slope, altitude) and environmental regulations (D.
zone number 32 showed the highest amount of croplands (20025 ha) S. Nº 9/2015 MMA) directly affect the value of each ESs indicator.
and forest resources (native forest 2128 ha, plantations 27719 ha), while More specifically, crop production in the baseline scenario ranges
site number 11 had the lowest share of croplands (82.5 ha) and forest from 271 to 9057 kg/GWh for the two regions. Suitable areas in Ñuble
resources (170.8 ha). Significant differences between suitable areas in Region can produce an annual crop average of 3647 kg per GWh.
terms of land use and cover were observed, depending on where the Meanwhile, in Biobío Region, the mean production resulted in 45.7%
highest potential was found and the size of the available area. lower than in Ñuble due to the lower availability of cultivated land and
Wind energy development is compatible with the existing land use the higher potential energy outputs. Despite the compatibility between
and cover types. This means that only a small percentage of vegetation agriculture and wind energy, locating wind farms in suitable zones could
(10% of current existences) equivalent to the surface of permanent reduce by 10% on average food crop production caused by agricultural
works (e.g., wind turbine platforms, roads, power lines) should be land transformation. This could endanger food provision, raise food
Fig. 5. Potential impacts on ecosystem services from developing wind energy in suitable areas (punctual values in SM Table S4). Axes represent the value of
ecosystem service plotted for each suitable zones in Ñuble (1–28) and Biobío (29–40) regions.
8
prices and negatively damage local economies highly dependent on vegetation, and thus its protective effect against erosion, for locating the
agriculture [59]. wind farm infrastructure. In this sense, soil erosion has been a relevant
Regarding timber provisioning services, annual timber availability environmental issue in the study area, especially after the Natural Re
currently goes from 0.17 to 31 m3/GWh, including native forests and sources Information Center (CIREN) in 2007 registered 1.18 million
exotic plantations. Generally, zones in Biobío Region have an average hectares of eroded surfaces, the 31.9% of the regional area [64]. High
timber estimation of 19 m3/GWh, a 47% higher than Ñuble. Developing erosion rates may cause soil quality losses and agricultural productivity
the wind potential of suitable sites could lead in each region to a 10% affectations [50], which could eventually backfire the restoration of
decrease in forest lands, implying in average a 10% reduction in timber converted croplands after the decommissioning stage of the wind park.
supply as biomass from annual tree growth becomes unavailable. This is Also, the rise of erosion issues could lead to larger sediment depositions
a detrimental outcome given the great importance of forestry resources in water bodies affecting freshwater quality.
to the regional and national economy. Forest lands are compatible with Besides, it is necessary to highlight that all ESs indicators were
wind energy, but trees are also clear-cut around neighboring areas of calculated using regional data (e.g., agroforestry statistics) georefer
permanent installations to avoid disruptions of circulating wind flows enced according to the 2015 land use map (last update to date). The
[60]. As forest resources are protected by law in Chile (Law above makes ESs valuation sensitive to land use changes. In this context,
20.283/2008 [61], D.L. 701/1974 [62]), any surface of forest vegetation and regardless of assuming that land coverage does not undergo sig
removed during the development of wind farms must be recovered in nificant changes in the short term, to obtain more accurate results the
another location. However, native forests have long growth periods land cover layer should be updated in further studies. Additionally, a
making native resources cut down difficult to restore. In the case of uniform spatial distribution of agroforest statistics was assumed due to
forest plantations, it takes almost as long as the lifespan of wind farms the lack of more specific spatial information. Thus, crop and timber
before the replanted plantation is ready for commercial use (pines: production could be misrepresented in some areas as they are site-
22–24 years, eucalyptus: 12–14 years [63]). In either case, removing the dependent (e.g., not all crop types are grown in all counties in the re
native or exotic forest can have an additional impact on services like gion or have the same yield). Likewise, site-specific data should be used
local climate regulation as forests are generally considered carbon sinks. to replace NDVI-based C-factor in USLE equation aiming to obtain better
Besides, the WFblue indicator indicated crop water requirements up results.
to 92 m3/GWh in suitable zones for the baseline scenario. Areas
belonging to Ñuble Region displayed the highest consumptive water 4.3. Selection of best areas to develop wind energy: ESs-exergy-based
volumes with an average of 38 m3/GWh, while Biobío areas had a assessment
smaller mean value of 8 m3/GWh. This pattern could be related to higher
average crop production in suitable sites of Ñuble (3647 kg/GWh) Fig. 6 shows the aggregated impacts on ESs expressed in GJex/GWh
compared to Biobío (1979 kg/GWh). In addition, Biobío Region zones of wind energy development on suitable areas.
showed average precipitations of 1183.1 mm per year, a 13% higher The highest negative impacts on ESs belong to zones number 24 in
than Ñuble, suggesting that most of crop water needs in Biobío would be Ñuble with − 34 GJex/GWh, followed by zones 33, 30 and 31 (− 33
covered by precipitation. On the other hand, developing wind energy in GJex/GWh and − 31 GJex/GWh, respectively) in Biobío. Conversely,
suitable sites could imply an average 12% reduction in irrigation water zones 36 and 37 in Biobío, including the number 22 in Ñuble (290 GJex/
use due to agricultural land transformation and consequent crop pro GWh, 276 GJex/GWh and 227 GJex/GWh), respectively, had the higher
duction loss. Nevertheless, in this situation, a reduction is not a synonym positive impacts. From all indicators, the increase in soil losses due to
for damaging impacts on people’s wellbeing as it depends on the erosion processes contributed the most (75%) to calculated total im
meaning of reduction. In this context, the observed decrease in blue pacts, which may be explained because the mass of soil that is annually
water footprint implies a benefit (gain) for people because that volume lost exceeds 7 to 14 times the mass of timber and crops lost, respectively.
of freshwater will be available for other uses (e.g., domestic and in In the same line, the exergy content of the soil is higher than those of the
dustrial use, hydropower, recreational). other resources.
A similar pattern is obtained for WFgray indicator, whose reduction The impacts on water-related ESs had no appreciable contribution
constitutes a benefit for people because it implies a lower appropriation because of the very low exergy content of freshwater compared to
of the river’s capacity to assimilate pollutant loadings reaching them, biomass and soil resources. Nevertheless, it does not mean that water
increasing the quality and availability of freshwater for other activities. does not play an essential role and function in the provisioning and
In this case, an average 10% reduction was quantified because of regulation and maintenance services.
expanding wind energy towards suitable sites identified for both re Converting biophysical indicators to exergy units (comparable units)
gions, linked to the reduction of agricultural production and thus, lower allowed for accounting for total impacts and identifying the suitable
fertilizer application during the sowing stage. Additional results showed sites with the greatest ESs changes. However, the most suitable locations
that 7.46–1728 m3/GWh of freshwater is required to assimilate for wind development cannot be decided in terms of total impacts alone,
nitrogen-based pollution from current agricultural practices in suitable as they only denote a change in final conditions compared to a baseline
zones. Particularly, the average water needed in zones of the Biobío scenario and say nothing about the impact’s nature (beneficial or
Region to dilute nitrogen concentrations until permitted levels is be damaging). Accordingly, the BOS indicator, which is based on potential
tween seven to eight times the water required in Ñuble (886 and 113 gains (benefits) minus losses, was calculated to determine the best
m3/GWh, respectively). This performance justifies as the secondary suitable sites to deploy wind energy. Fig. 7 depicts the suitable sites for
water quality standard for the Biobío River Basin (D.S. Nº 9/2015 MMA wind energy ranked from best to worst according to calculated values of
[49]) sets very low maximum nitrogen concentration levels (0.6 mg/L); the BOS indicator.
in contrast, the limit values used for Ñuble Region basins are much BOS indicator took negative values in all suitable zones under
higher (10 mg/L). analysis (− 25 to − 307 GJex/GWh), implying that losses exceeded gains,
Meanwhile, soil loss levels up to 42 Mg/GWh were observed in the and thus, damages incurred with the loss of crop, timber and topsoil
baseline scenario. Zones belonging to Ñuble Region averaged potential surpassed the benefits of freshwater savings. Sites showing the highest
soil losses of 12 Mg/GWh, while Biobío Region showed average values of BOS values (− 25>− 307 GJex/GWh) were considered more sustainable
8 Mg/GWh. This indicator demonstrated that wind energy development options than others with lower values to deploy the existing wind po
in suitable sites could negatively impact erosion regulation services, as tential, as they showed smaller harmful ESs impacts. More specifically,
soil losses increase by 40% and 50% in Ñuble and Biobío, respectively, zones number 39, 31, 30 and 29 are among the most sustainable sites for
compared to the baseline scenario. This may be due to the removal of wind energy in Biobío as they had the larger BOS values (− 25.1 GJex/
9
Fig. 6. Aggregated ecosystem service impacts caused by developing wind energy in suitable areas. Bars depict the contribution of indicators to the total impact.
Positive/negative values denote an increase/decrease in final values compared to the baseline, respectively.
GWh, − 36.9 GJex/GWh, − 39.3 GJex/GWh and − 39.4 GJex/GWh, “conventional” land suitability analysis results of Martínez-Martínez
respectively). Meanwhile, zones number 1, 2 and 3 (− 28.9 GJex/GWh, et al. (just permitted lands without technical and environmental con
− 37.9 GJex/GWh and − 38.2 GJex/GWh, respectively) are included straints). Similarly, our results were 30.5% more focused than those
among the top-ranked sites in Ñuble (red boxes in Fig. 7). The maximum obtained by them after including the ESs approach in suitability analysis
theoretical electricity potential (35 TWh/yr) encompassed in the above to determine the best sites for wind development.
areas, according to simulations, has the potential to attenuate the elec The methodology developed in this work, especially the BOS indi
tricity generation coming from coal (25.8 TWh generated by 2021 [65]), cator, have the following strengths when compared to the preceding
becoming a very promising alternative for decarbonization of the na studies: (1) the methodological framework is not exclusive for wind
tional electricity matrix and for Chile to reach the carbon neutrality energy or the study area, instead it can be applied to other renewable
target by 2050. energy sources and study cases (national or international). In such cases,
On the other hand, it is worth noting that there is no explicit criteria the specific parameters that characterize each renewable energy source
or BOS threshold value for selecting the best sites for wind energy (e.g., technical and environmental constraints) and the goals and scopes
development. Consequently, the identification of the most sustainable of each study should be adjusted. (2) The BOS indicator is quite sensitive
areas is not necessarily limited to the zones mentioned above. Besides, it to site-specific land use type, land use changes, land use intensity and
is important to remark that wind generation is also considered a pro management (e.g., biomass yield), climatic conditions (e.g., tempera
visioning service according to the CICES classification; however, the ture, precipitations), socio-economic and cultural conditions and envi
proposed BOS indicator does not consider the energy generated as a ronmental standards, which allows capturing and monitoring spatial
“Gain”. This is because this indicator was constructed to reflect how human interventions at different scale (e.g., regional, country). Based on
much other ESs are affected to supply a GWh of wind energy. On the that, it is expected that BOS values vary significantly between regions
contrary, if the BOS indicator were modified to express the impact per and countries. Additionally, a key strength of the BOS indicator is its
unit of surface area (GJex/ha.yr), then wind generation could be dynamic nature, as the gain-loss concept it embodies can be applied
included as an additional “Gain” parameter. For this case, the new regardless of the ESs valuation method and metrics used, without losing
exergy balance still shows negative values (− 13.9 to − 182.8 GJex/ha. its essence. This allows to include other ESs valuation methods (e.g.,
yr), indicating that soil, crop and timber losses (especially soil loss) economic valuation) to solve any limitation or undesired result affecting
exceeded the joint benefits of water savings and wind generation. Still, the decision-making process. On top of that, the proposed ESs-exergy
the most feasible zones match the results shown in Fig. 7. based BOS indicator provide relevant information for a better under
On the other hand, different approaches have been developed in the standing of dynamic and complexity relationships between ecosystems
literature to include ESs in the spatial planning of renewable energies and human activities at the local level, emphasizing in the valuable part
[14,15]. Nevertheless, the decision-making context, scope of the of ecosystem capacity (quality of resources) that could not be use effi
assessment, ESs class type, number of ESs assessed and variety of valu ciently by socio-economics and ecosystem systems caused by wind en
ation methods limited comparison analysis between studies. In the ergy development [19]. Therefore, the BOS provide scientific evidence
particular case of Biobío and Ñuble regions as study case, Martí for adopting strategies and policies towards a more sustainable man
nez-Martínez et al. [16] attempted to operationalize the ESs concept to agement of the ecosystems focusing on to reduce ESs disturbances.
improve the spatial planning of wind energy. A different approach was On the other hand, it should be mentioned that there are different
applied, which did not include biophysical quantification of ESs or the definitions and classification systems of ESs, which might entail the use
balance of impacts to select the most sustainable sites to deploy wind of other ESs names and indicators during BOS assessment in similar
energy; instead, a MCDM technique was used to integrate maps studies (according to the classification system chosen). In this study,
depicting the supply capacity of ESs. It is worth highlighting that similar CICES V5.1 classification system was used since its four-level hierar
constraints and threshold values to ours were used to define suitable chical structure (e.g., section, division, group and class) enable a more
sites. By comparing the results of both studies (see Fig. 8), it can be detailed ESs description and the indicator selection process could be
observed that the methodology developed in the current work signifi more accurate. Also, it allows that studies at different topics and spatial
cantly reduced the total surface area identified as suitable for wind en resolutions can more easily compared and avoid double counting during
ergy development, obtaining more accurate results and avoiding future service assessments [66].
service conflicts. For example, this work increased by 44% the accuracy
in delimiting suitable areas for wind energy compared to the
10
Fig. 7. Suitable zones for wind energy ranked from best to worst according to calculated BOS indicator values in GJex/GWh (top places within red boxes).
4.4. Final remarks Accordingly, the water-related ESs could be economically more
relevant than soil erosion (the opposite happens with exergy valua
The proposed BOS indicator successfully integrates the ESs approach tion) due to their importance for domestic and economic activities (e.
and thermodynamic laws allowing to discern where it would be most g., agriculture). The ESs valuation from economic perspectives could
sustainable to develop a wind project in the study area. However, the not be comparable with BOS-exergy indicator due to intrinsic limi
methodology and derived results have limitations that need to be tation of each method; nevertheless, both methods have in common
addressed: the system approach in nature. Therefore, a balance of ESs gain and
losses in monetary terms could be an interesting approach for eval
i. Exergy concept allowed to bring all indicators to comparable units to uating ESs.
account the gains minus losses balance. However, as gains grouped ii. A conservative ESs trade-offs analysis was performed since only five
water-related indicators, very low values of benefits compared to provisioning and regulation and maintenance services were assessed.
losses were observed (BOS<0) because of the low standard exergy Nevertheless, there could be a greater number of ESs being provided
content of freshwater compared to biomass and soil. This finding in suitable areas, including cultural services, which are among the
could lead to a misunderstanding about the relevance of water- most conflicting services with wind energy development [1,69].
related ESs. Because of that, other methods to aggregate indicators Cultural services were not included since they are complex to inte
(e.g., emergy and monetary value) should be applied. For instance, grate in exergy-based indicator due to their subjective and
monetary valuation could give an insight of the most valuable ESs non-material nature. Therefore, a combination of GIS and partici
and where protection is economically most relevant [67,68]. patory methods (Delphi and multicriteria methods) could be a valid
11
Fig. 8. Comparison between best suitable sites for wind development in Biobío and Ñuble regions obtained by Martínez-Martínez et al. [16] (figure A and B) and
those identified in this research (figure C). A: suitable locations after a simple land suitability analysis (permitted lands after removing technical and environmental
constrained sites), B: suitable location after a land suitability analysis combined with ecosystem service approach (aggregated supply capacity of ecosystem services
used as a decision-making factor to select best locations), C: current work depicting suitable locations after a land suitability analysis combined with ecosystem
service valuation (BOS indicator used as decision-making factor to define best locations).
option to map cultural services and include them in the spatial of the total regional land area (279156.2 ha). It was also demonstrated
planning of wind energy. In this sense, Analytic Hierarchy Process that wind development might entail a trade-offs situation between wind
method have been widely applied in ecosystems services assessment generation and service provision in suitable lands. For example, it was
due to easy and success applicability [70], which could provide the observed that crop and timber production could decrease by 10% while
weighing factor to cultural services and integrated into land suit soil losses can increase by 40–50% in average, all of which is damaging
ability analysis. In such case, changes are expected in those sites for locals. Also, up to 10% reduction in blue and gray water footprint can
identified as most suitable for wind development when cultural be achieved, indicating beneficial water savings. By applying the BOS
services are incorporated. indicator, previous beneficial and damaging impacts were balanced to
decide where would be more favorable to promote wind development.
5. Conclusions The results revealed that, in all cases, losses in terms of ESs impacts
exceeded gains as BOS values were less than zero (BOS<0). The most
This study applied a novel methodology that integrates land suit sustainable locations for wind deployment were zones 39, 31, 30, 29
ability analysis with ESs approach and exergy analysis to identify the (Biobío) and 1, 2, 3 (Ñuble) as they depicted the highest BOS values,
most sustainable locations to deploy wind energy in Biobío and Ñuble thus, the lowest detrimental impacts. These results constitute valuable
regions. By combining the ES approach with traditional land suitability, information for regional authorities to promote a more focused energy
the trade-offs and synergies that may occur between wind electricity sector development.
generation and ESs in suitable lands were readily assessed. These results The proposed framework can be replicable to other renewable en
were used to construct a novel BOS indicator, which improves the se ergy sources considering different spatial scales (e.g., regional, national,
lection process of the most sustainable locations for renewable energy and international) by adjusting the specific parameters that characterize
development. The BOS indicator relied on a balancing gains and losses each energy source (e.g., technical constraints) and the goals and scopes
concept, where gains comprise the aggregation of beneficial impacts of each study. Therefore, the proposed framework could provide prac
resulting from landscape transformations; meanwhile, losses involve the tical guidelines for national and international land planners and
sum of all prejudicial effects. The exergy principles provide the oppor decision-makers to support a more sustainable and efficient spatial en
tunity to compare and unify the biotic resources with abiotic ones in a ergy planning addressed toward sites with the lowest impacts on ESs.
single indicator, representing the valuable part of ESs that are gained or
lost and could not use efficiently by socioeconomic and ecological Credit author statement
systems.
To our best knowledge, the BOS indicator constitutes a novelty in Yenisleidy Martínez-Martínez: Conceptualization, Investigation,
spatial planning studies since other attempts to include ESs in spatial Methodology, Formal analysis, Writing-Original draft preparation,
energy planning have been reported in the literature, but not from an Funding acquisition. Mauricio Aguayo: Methodology, Supervision. Jo
integrated exergy gain-loss balanced perspective. Moreover, the results Dewulf: Supervision.: Conceptualization, Writing- Reviewing and Edit
demonstrated an improvement during selection of the most sustainable ing, Funding acquisition, Supervision.
locations for wind energy development compared with traditional land
suitability analysis.
Declaration of competing interest
After applying this methodology, high-suitability zones for wind
energy development were identified, which accounted for around 7.5%
The authors declare that they have no known competing financial
12
interests or personal relationships that could have appeared to influence [17] Sullivan S, Hannis M. Mathematics maybe, but not money: on balance sheets,
numbers and nature in ecological accounting, Accounting, Audit. Account. J 2017;
the work reported in this paper.
30:1459–80. https://doi.org/10.1108/AAAJ-06-2017-2963.
[18] De Luca Peña LV, Taelman SE, Préat N, Boone L, Van Der Biest K, Custódio M,
Data availability Hernandez Lucas S, Everaert G, Dewulf J. Towards a comprehensive sustainability
methodology to assess anthropogenic impacts on ecosystems : review of the
integration of life cycle assessment , environmental risk assessment and ecosystem
Data will be made available on request. services assessment. Sci Total Environ 2022;808:1–17. https://doi.org/10.1016/j.
scitotenv.2021.152125.
Acknowledgments [19] Alvarenga RAF, Dewulf J, Van Langenhove H. A new natural resource balance
indicator for terrestrial biomass production systems. Ecol Indicat 2013;32:140–6.
https://doi.org/10.1016/j.ecolind.2013.03.029.
Authors would like to thank the funding provided by the Agencia [20] Romero JC, Linares P. Exergy as a global energy sustainability indicator. A review
Nacional de Investigación y Desarrollo (National Research and Devel of the state of the art. Renew Sustain Energy Rev 2014;33:427–42. https://doi.org/
10.1016/j.rser.2014.02.012.
opment Agency, ANID) grant ANID-PFCHA/Beca de Doctorado Nacional [21] Bösch ME, Hellweg S, Huijbregts MA, Frischknecht R. Applying cumulative exergy
21180191, ANID/CONICYT FONDECYT grant number 11170302, and demand (CExD) indicators to the ecoinvent database. Int J LCA 2007;12:181–90.
projects ANID CENTROS BASALES ACE210012 and ANID/FONDAP/ https://doi.org/10.1065/lca2006.11.282.
[22] Sheng H-X, Xu H, Zhang L, Chen W. Ecosystem intrinsic value and its application in
15130015. decision-making for sustainable development. J Nat Conserv 2019;49:27–36.
https://doi.org/10.1016/j.jnc.2019.01.008.
Appendix A. Supplementary data [23] Martínez Martínez Y, Goecke Coll D, Aguayo M, Casas-Ledón Y. Effects of
landcover changes on net primary production (NPP)-based exergy in south-central
of Chile. Appl Geogr 2019;113. https://doi.org/10.1016/j.apgeog.2019.102101.
Supplementary data to this article can be found online at https://doi. [24] Alvarenga RAF, Dewulf J, Van Langenhove H, Huijbregts MAJ. Exergy-based
org/10.1016/j.rser.2023.113252. accounting for land as a natural resource in life cycle assessment. Int J Life Cycle
Assess 2013;18:939–47. https://doi.org/10.1007/s11367-013-0555-7.
[25] Alvarenga RAF, Erb KH, Haberl H, Soares SR, van Zelm R, Dewulf J. Global land
References use impacts on biomass production—a spatial-differentiated resource-related life
cycle impact assessment method. Int J Life Cycle Assess 2015;20:440–50. https://
[1] Dai K, Bergot A, Liang C, Xiang W-N, Huang Z. Environmental issues associated doi.org/10.1007/s11367-014-0843-x.
with wind energy - a review. Renew Energy 2015;75:911–21. https://doi.org/ [26] Zhang B, Jin P, Qiao H, Hayat T, Alsaedi A, Ahmad B. Exergy analysis of Chinese
10.1016/j.renene.2014.10.074. agriculture. Ecol Indicat 2017:1–13. https://doi.org/10.1016/j.
[2] Vallejos-Romero A, Cordoves-Sánchez M, Jacobi P, Aledo A. In transitions we ecolind.2017.08.054.
trust? Understanding citizen, business, and public sector opposition to wind energy [27] Rodríguez-Merchan V, Ulloa-Tesser C, Casas-Ledón Y. Evaluation of the water-
and hydropower in Chile. Energy Res Social Sci 2020;67. https://doi.org/10.1016/ energy-land nexus (WELN) using exergy-based indicators: the Chilean electricity
j.erss.2020.101508. system case. Energies 2020;13. https://doi.org/10.3390/en13010042.
[3] Wang S, Wang S, Smith P. Quantifying impacts of onshore wind farms on [28] Fan Y, Wu X, Wu X, Li C, Yang Q, Hayat T, Alsaedi A, Wang P, Chen G. A unified
ecosystem services at local and global scales. Renew Sustain Energy Rev 2015;52: ecological assessment of a solar concentrating plant based on an integrated
1424–8. https://doi.org/10.1016/j.rser.2015.08.019. approach joining cosmic exergy analysis with ecological indicators. Renew Sustain
[4] Enevoldsen P. Onshore wind energy in Northern European forests: reviewing the Energy Rev 2020;129:109934. https://doi.org/10.1016/j.rser.2020.109934.
risks. Renew Sustain Energy Rev 2016;60:1251–62. https://doi.org/10.1016/j. [29] Suuronen A, Lensu A, Kuitunen M, Andrade-Alvear R, Celis NG, Miranda M,
rser.2016.02.027. Perez M, Kukkonen JVK. Optimization of photovoltaic solar power plant locations
[5] Torres Sibille A del C, Cloquell-Ballester V-A, Darton R. Development and in northern Chile. Environ Earth Sci 2017;76. https://doi.org/10.1007/s12665-
validation of a multicriteria indicator for the assessment of objective aesthetic 017-7170-z.
impact of wind farms. Renew Sustain Energy Rev 2009;13:40–66. https://doi.org/ [30] MINENERGIA/GIZ. Energías Renovables en Chile. El potential eólico, solar e
10.1016/j.rser.2007.05.002. hidroeléctrico de Arica a Chiloé., Proyecto Estrategia de Expansión de las Energías
[6] Sotiropoulou KF, Vavatsikos AP. Onshore wind farms GIS-Assisted suitability Renovables en los Sistemas Eléctricos Interconectados (MINENERGIA/GIZ),
analysis using PROMETHEE II. Energy Pol 2021;158:112531. https://doi.org/ Santiago de Chile, Chile. 2014. http://www.minenergia.gob.cl/archivos_bajar/Est
10.1016/j.enpol.2021.112531. udios/Potencial_ER_en_Chile_AC.pdf.
[7] Vavatsikos AP, Arvanitidou A, Petsas D. Wind farm investments portfolio formation [31] CNE. Anuario estadístico de Energía 2021. https://www.cne.cl/wp-content/uploa
using GIS-based suitability analysis and simulation procedures. J Environ Manag ds/2022/07/AnuarioEstadisticoEnergia2021.pdf; 2021.
2019;252:109670. https://doi.org/10.1016/j.jenvman.2019.109670. [32] de Energía Ministerio, Planificación Energética de Largo Plazo (PELP). Periodo
[8] Tercan E. Land suitability assessment for wind farms through best-worst method 2023-2027. 2021. Informe preliminar, https://energia.gob.cl/sites/default/file
and GIS in Balıkesir province of Turkey. Sustain Energy Technol Assessments 2021; s/documentos/pelp2023-2027_informe_preliminar.pdf.
47:101491. https://doi.org/10.1016/j.seta.2021.101491. [33] Rosenthal A, Verutes G, McKenzie E, Arkema KK, Bhagabati N, Bremer LL,
[9] Gigović L, Pamučar D, Božanić D, Ljubojević S. Application of the GIS-DANP- Olwero N, Vogl AL. Process matters: a framework for conducting decision-relevant
MABAC multi-criteria model for selecting the location of wind farms: a case study assessments of ecosystem services. Int. J. Biodivers. Sci. Ecosyst. Serv. Manag.
of Vojvodina, Serbia. Renew Energy 2017;103:501–21. https://doi.org/10.1016/j. 2015;11:190–204. https://doi.org/10.1080/21513732.2014.966149.
renene.2016.11.057. [34] Burkhard B, Santos-Martin F, Nedkov S, Maes J. An operational framework for
[10] Aydin NY, Kentel E, Duzgun S. GIS-based environmental assessment of wind energy integrated Mapping and Assessment of Ecosystems and their Services (MAES). One
systems for spatial planning: a case study from Western Turkey. Renew Sustain Ecosyst 2018;3:e22831. https://doi.org/10.3897/oneeco.3.e22831.
Energy Rev 2010;14:364–73. https://doi.org/10.1016/j.rser.2009.07.023. [35] Haines-Young R, Potschin MB. Common international classification of ecosystem
[11] Jangid J, Bera AK, Joseph M, Singh V, Singh TP, Pradhan BK, Das S. Potential zones services (CICES) V5.1. Guidance on the application of the revised structure. 2018.
identification for harvesting wind energy resources in desert region of India – a www.cices.eu. [Accessed 26 November 2019]. accessed.
multi criteria evaluation approach using remote sensing and GIS. Renew Sustain [36] Burkhard B, Maes J. Mapping ecosystem services. Bulgaria: Pensoft Publishers;
Energy Rev 2016;65:1–10. https://doi.org/10.1016/j.rser.2016.06.078. 2017. https://doi.org/10.3897/ab.e12837.
[12] Genç MS, Karipoğlu F, Koca K, Azgin ŞT. Suitable site selection for offshore wind [37] Burkhard B, Kroll F, Nedkov S, Müller F. Mapping ecosystem service supply,
farms in Turkey’s seas: GIS-MCDM based approach. Earth Sci. Informatics 2021;14: demand and budgets. Ecol Indicat 2012;21:17–29. https://doi.org/10.1016/j.
1213–25. https://doi.org/10.1007/s12145-021-00632-3. ecolind.2011.06.019.
[13] Karipoğlu F, Genç MS, Koca K. Determination of the most appropriate site selection [38] ODEPA. Cultivos anuales regionales, Santiago de Chile. 2021. https://www.odepa.
of wind power plants based Geographic Information System and Multi-Criteria gob.cl/estadisticas-del-sector/estadisticas-productivas.
Decision-Making approach in develi, Turkey. Int. J. Sustain. Energy Plan. Manag. [39] MMA Segundo. Informe del Inventario Nacional de Gases de Efecto Invernadero de
2021;30:97–114. https://doi.org/10.5278/ijsepm.6242. Chile serie 1990-2013. Chile: Santiago; 2017.
[14] Kienast F, Huber N, Hergert R, Bolliger J, Moran LS, Hersperger AM. Conflicts [40] Hoekstra AY, Chapagain AK, Aldaya MM, Mekonnen MM. The water footprint
between decentralized renewable electricity production and landscape services – a assessment manual. Setting the global standard. Earthscan; 2011. https://doi.org/
spatially-explicit quantitative assessment for Switzerland. Renew Sustain Energy 10.1080/0969160x.2011.593864.
Rev 2017;67:397–407. https://doi.org/10.1016/j.rser.2016.09.045. [41] Jaramillo C. Aplicación de la Metodología de Contabilidad de Huella Hídrica
[15] Egli T, Bolliger J, Kienast F. Evaluating ecosystem service trade-offs with wind Directa a 15 Regiones de Chile. 2017. https://escenarioshidricos.cl/wp-conten
electricity production in Switzerland. Renew Sustain Energy Rev 2017;67:863–75. t/uploads/2020/06/huella-hidrica.pdf.
https://doi.org/10.1016/j.rser.2016.09.074. [42] Allen RG, Pereira LS, Raes D, Smith M. Evapotranspiración del cultivo. Guías para
[16] Martínez-Martínez Y, Dewulf J, Casas-Ledón Y. GIS-based site suitability analysis la determinación de los requerimientos de agua de los cultivos. Roma: FAO; 2006.
and ecosystem services approach for supporting renewable energy development in [43] DGA-MOP. Estimación de la demanda actual, proyecciones futuras y
south-central Chile. Renew Energy 2022;182:363–76. https://doi.org/10.1016/j. caracterización de la calidad de los recursos hídricos en Chile. 2017.
renene.2021.10.008.
13
[44] Droogers P, Allen RG. Estimating reference evapotranspiration under inaccurate impacts on ecosystems. Sci Total Environ 2016;550:143–56. https://doi.org/
data conditions. Irrigat Drain Syst 2002;16:33–45. https://doi.org/10.1023/A: 10.1016/j.scitotenv.2016.01.055.
1015508322413. [59] Burgess PJ, Rivas Casado M, Gavu J, Mead A, Cockerill T, Lord R, Van Der Horst D,
[45] Ganesh Babu R, Ravi Babu G, Hema Kumar HV. Estimation of crop water Howard DC. A framework for reviewing the trade-offs between, renewable energy,
requirement, effective rainfall and irrigation water requirement for vegetable crops food, feed and wood production at a local level. Renew Sustain Energy Rev 2012;
using CROPWAT. Int J Agric Eng 2015;8:15–20. https://doi.org/10.15740/has/ 16:129–42. https://doi.org/10.1016/j.rser.2011.07.142.
ijae/8.1/15-20. [60] Enevoldsen P. A socio-technical framework for examining the consequences of
[46] Muratoglu A. Grey water footprint of agricultural production: an assessment based deforestation: a case study of wind project development in Northern Europe.
on nitrogen surplus and high-resolution leaching runoff fractions in Turkey. Sci Energy Pol 2018;115:138–47. https://doi.org/10.1016/j.enpol.2018.01.007.
Total Environ 2020;742:140553. https://doi.org/10.1016/j. [61] de Agricultura Ministerio. LEY 20.283 Sobre Recuperación del Bosque Nativo y
scitotenv.2020.140553. Fomento Forestal, Biblioteca del Congreso Nacional de Chile, República de Chile.
[47] A. (ed), Osorio U. Determinación de la huella del agua y estrategias de manejo de 2008. https://www.leychile.cl/N?i=274894&f=2008-07-30&p=.
recursos hídricos, La Serena, Chile. 2013. https://www.pj.gov.py/ebook/monogra [62] de Agricultura Ministerio. Decreto Ley 701 Fija Régimen Legal de los Terrenos
fias/nacional/internacional-público/Fernando-Costantini-Agua-virtual-y-Huella Forestales o Preferentemente Aptos Para la Forestación, y Establece Normas de
-Hidrica.pdf. http://bibliotecadigital.ciren.cl//handle/123456789/14673. Fomento Sobre la Materia, Biblioteca del Congreso Nacional de Chile, República de
http://biblioteca.inia.cl/link.cgi/Catalogo/Actas/38988.act. Chile. 1974. http://bcn.cl/2et24. [Accessed 14 October 2020]. accessed.
[48] ODEPA. Estudio de Diagnóstico de Mercado y Estudio de la Cadena de [63] Barros S. Evolución de las plantaciones forestales en Chile. Forestación y
Comercialización de Fertilizantes en Chile. Santiago de Chile; 2010. https://www. reforestación, Cienc. e Investig. For. INFOR. 2018;24:89–114. https://bibliotecadig
odepa.gob.cl/files_mf/1369773301Estudio_Cadena_Comercializacion_Fertilizantes. ital.infor.cl/handle/20.500.12220/28235.
pdf. [64] CIREN. Determinación de la erosión actual y potential de los suelos de Chile: región
[49] D.S. No 9/2015 MMA. ESTABLECE normas SECUNDARIAS de CALIDAD ambiental del Bío-Bío Síntesis de Resultados - diciembre 2010. CIREN, Santiago de Chile;
para la protección de las aguas CONTINENTALES SUPERFICIALES de la cuenca del 2010. http://bibliotecadigital.ciren.cl//handle/123456789/2041. [Accessed 13
río biobío. https://www.leychile.cl/N?i=1084403&f=2015-11-27&p=; 2015. October 2020]. accessed.
[50] Wischmeier WH, Smith DD. Predicting rainfall erosion losses- A guide to [65] CNE. Generación bruta SEN. https://www.cne.cl/normativas/electrica/consu
conservation planning. Handbook N: United States Department of Agriculture lta-publica/electricidad/; 2022.
(USDA); 1978. [66] Babí Almenar J, Rugani B, Geneletti D, Brewer T. Integration of ecosystem services
[51] Bonilla CA, Vidal KL. Rainfall erosivity in Central Chile. J Hydrol 2011;410: into a conceptual spatial planning framework based on a landscape ecology
126–33. https://doi.org/10.1016/j.jhydrol.2011.09.022. perspective. Landsc Ecol 2018;33:2047–59. https://doi.org/10.1007/s10980-018-
[52] Bonilla CA, Johnson OI. Soil erodibility mapping and its correlation with soil 0727-8.
properties in Central Chile. Geoderma 2012;189–190:116–23. https://doi.org/ [67] De Groot R, Fisher B, Christie M, Aronson J, Braat L, Gowdy J, Haines-Young R,
10.1016/j.geoderma.2012.05.005. Maltby E, Neuville A, Polasky S, Portela R, Ring I. Integrating the ecological and
[53] Balabathina VN, Raju RP, Mulualem W, Tadele G. Estimation of soil loss using economic dimensions in biodiversity and ecosystem service valuation. 2010.
remote sensing and GIS-based universal soil loss equation in northern catchment of [68] de Groot R, Brander L, van der Ploeg S, Costanza R, Bernard F, Braat L, Christie M,
Lake Tana Sub-basin, Upper Blue Nile Basin, Northwest Ethiopia. Environ. Syst. Crossman N, Ghermandi A, Hein L, Hussain S, Kumar P, McVittie A, Portela R,
Res. 2020;9:1–32. https://doi.org/10.1186/s40068-020-00203-3. Rodriguez LC, ten Brink P, van Beukering P. Global estimates of the value of
[54] Van Der Knijff JM, Jones RJ, Montanarella L. Soil erosion risk assessment in Italy. ecosystems and their services in monetary units. Ecosyst Serv 2012;1:50–61.
1999. https://doi.org/10.1016/J.ECOSER.2012.07.005.
[55] Szargut JT. Exergy method: technical and ecological applications. first ed. 2005. [69] Tsoutsos T, Tsouchlaraki A, Tsiropoulos M, Serpetsidakis M. Visual impact
[56] Valero A, Palacino B, Ascaso S, Valero A. Exergy assessment of topsoil fertility. Ecol evaluation of a wind park in a Greek island. Appl Energy 2009;86:546–53. https://
Model 2022;464. https://doi.org/10.1016/j.ecolmodel.2021.109802. doi.org/10.1016/j.apenergy.2008.08.013.
[57] INFOR. Los recursos forestales en Chile. Inventario continuo de bosques nativos y [70] Colak HE, Memisoglu T, Gercek Y. Optimal site selection for solar photovoltaic
actualización de plantaciones forestales; 2016. (PV) power plants using GIS and AHP: a case study of Malatya Province, Turkey.
[58] Taelman SE, Schaubroeck T, De Meester S, Boone L, Dewulf J. Accounting for land Renew Energy 2020;149:565–76. https://doi.org/10.1016/j.renene.2019.12.078.
use in life cycle assessment: the value of NPP as a proxy indicator to assess land use
14

Sustainable Wind Energy Planning Through Ecosystem Service Impact

Uploaded by

Copyright:

Available Formats

Sustainable Wind Energy Planning Through Ecosystem Service Impact

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

Sustainable Wind Energy Planning Through Ecosystem Service Impact

Uploaded by

Copyright:

Available Formats

Renewable and Sustainable Energy Reviews 178 (2023) 113252

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews