LUCE Lectures

Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 38

Summary Lectures Land Use Change and Ecosystems

Lecture 1: Introduction to Land System Science

There are different types of land use, and urbanization (type of land use) is a big part of this.
Furthermore, we are not good at taking care of the land that we have on earth.

We want to use land for different type of reasons, such as:


- Living space
- Infrastructure
- Biodiversity conservation
- Recreation
- Food security
- Economic activities
- Climate mitigation

Challenges: we have limited resources (limited land) and we want to use that land for all these use
cases as depicted above.

Humans use a lot of land:


 Humans use directly affects around 72% of the 130 mKm2 global ice-free land surface, leaving
23% unused.

In 1700: there was a little used. By humans (almost 10%) and more than 45% was still wild or semi-
natural (also 45%).
 Thus the last 300 years a lot have change because of the human influence on the environment.

Deforestation and reforestation:


- 8000 years prior to Neolithic agriculture, Europe was mostly a wooded continent.
- Then there was a lot of deforestation. The forest has been progressively fragmentated
- From the beginning of 1900 to 200s we see a pattern of reforestation. Before that a lot of
grazing areas, nowadays these areas are turning into forests.

Forest transition theory

Because of social economic development we see a pattern of deforestation that leads than to
reforestation. We can put every country on the part of the curve and as you move across this curve you
find a transition. Europe has past the stage a long time ago and are now in the post-transition.
This transition is due to: Agricultural intensification, New technologies, Urbanization &
Industrialization, Environmental Awareness, Policies & regulations, and Globalization.

You can categorize all the countries where they are on the forest transition phase:
1. Pre-transition: Suriname, Guyana
2. Early-transition: Peru, Central Africa, Kenia..
3. Late-transition: Brazil, Argentina, Australia
4. Post-transition: Europe, North America, Russia, China

Land cover change vs. Population growth


There is a strong relationship between forest area and population: obvious, the more people we have
the more people we must feed. So as the population increases, the forest area will decrease.

Population growth:
- Expect population growth to continue to 10 billion people in 2100.
- Especially in Africa. Asia will also increase but expect to decrease after 2050.
We have reached a peak of agricultural land use (in 2000) and is now falling. This is due to a
reduction in global pasture.
- Does not mean that it has peaked everywhere. In tropical regions, it continues to increase.
- Reached peak, but we produce more; this is due to better productivity.
How did we increase the productivity?
- Green revolution
- New Technologies
- Fertilizers use (nitrogen): due to this we can produce so much food.

Global Agricultural Productivity Index:


The metric Total Factor Productivity (TFP): is a ratio that measures changes in how efficiently
agricultural inputs are transformed into outputs (we want this to be very high). It is somewhat a
measure for efficiency.
- In 2022 we are producing all the food we need (in theory). However, this does not mean all
the produced food reaches all the regions in the world.  It is not even distributed.
- The required TFP rate (thus to feed everyone) is bigger than the current rate of TFP growth..
- The current rate of low-income TFP is far lower. If we could increase the TFP rate of the low-
income countries, we can come closer to closing this gap.

It is important to understand global land use patterns (global trade flows)


- A lot of countries consuming palm oil are not the same as the land that that produces it.
- The palm oil export and the import (or the consumption is skewed).
- There is a lot of deforestation for the production of different products.

Example soybean: are they used for food, feed or fuel?


- Little direct use for human food, little for direct animal food.
 most is for processed animal feed, biofuels and vegetable oil.
Biofuels: it is a good thing that biofuels replacing fossil fuels. Because it reduces GHG emissions and
reduces the running out of limited resources. But now we use the food for our fuels…

The world population is increasing, and there will be a less percentage that is in extreme poverty
when you compare it to 1820. That the world will be in less poverty has implications on land use.
 More consumption of meat and dairy.
 China is an example of this theory. Due to the better social economics of this country. It completely
changed the diet of the population. Chinese population consumes a lot more meat.
This shows if everyone in the world adopted the average diet of a given country vs. GDP per capita.
The countries with the lowest GDP per capita use the lowest land and the highest use the most land.
Highest Greenhouse gas emissions per kilogram food: Beef, Lamb, Prawns, Cheese.
Highest Annual deforestation: Cattle, Oilseeds, Forestry logging, Cereals …

Accelerated modern human induced species lost.


- The rate of distinction is far higher than what we would expect without the human influence.
The Global Land Program (GLP)
Came up with 8 hot topics (land use processes) in the land use science.

Topic 1: Urban-rural interaction


Share of people living in urban areas:
- In Japan and Netherlands above 90%
- In Ethiopia lower than 30%.

The rural population is expected to flatten, and the urban population is expected to increase. Thus the
increase in population will live in urban areas and not in in rural areas.
 There will be an increasing demand for housing in urban areas.

What are the effects when there is a higher demand for urban areas?
- People tend to move to the hotspots, thus to the cities. The big cities will grow more and
more.  Land use change, but also more economic change.
- More slums: group of individuals living under the same roof and lacking one or more from
water, sanitation, sufficient living area, and security. There are a lot of slums in Central
Africa.

Topic 2: land-climate-interaction
Land systems drive climate change.
- To stop climate change we must cut greenhouse gas emissions and use land to draw down
carbon dioxide from the atmosphere.
- The food system produces around a quarter of global greenhouse gas emissions.  solutions:
better farming practices, halting deforestation, healthier diets, and stop food waste.
Effects of land use change on the climate:
- Precipitation pattern: due to deforestation, there is a higher albedo and reduced precipitation
rates.
- Tropical tree cover: major source of CO2 emissions.

Land use can reduce and absorb the effects of climate change:
- Planting trees will help to store and absorb the carbon from the atmosphere. Land use
practices, such as trees can be act as a net-sink for carbon: they take more carbon in than they
release.
- Planting trees is going to help if it is planted at the right place: applied at grassy biomes such
as savannas can be catastrophic. As it can severely compromise ecosystems services,
including hydrology and soil nutrient cycles, and reduce biodiversity.
 there is also a trade-off: we need to plant trees, but we also need to produce food.

Topic 3: Land Management


Land degradation: the reduction or loss of the biological or economic productivity of land resulting
from a combination of pressures, including land use and management practices.
 we are not doing a great job and taking care of our land.

Agricultural intensification: process of increasing the use of capital, and inputs (fertilizers,
machinery) relative to land area.
- Lot of land spared because of agricultural intensification.
- Increasing productivity comes at the cost of using more machinery, fertilizers, and pesticides.
- Netherlands is one of the top produces, but also uses a lot of fertilizers (meststof).
 67% of the global nitrous exodie emissions are due to agriculture…
Excess nitrogen: difference between nutrient inputs (from fertilizers, manure) and the amount
harvested in crop material. This represents nitrogen that is lost to the environment and can create
ecological imbalances in ecosystems and water bodies. In the Netherlands this is a lot! We use far
more fertilizer than we should…

Greenhouse gas emissions: 1/3 of global greenhouse gas emissions come from food systems. With
the most coming from beef, dark chocolate, lamb and coffee.

Tillage: is the agricultural preparation of soil by mechanical agitation of various types, such as
digging, stirring and overturning.

Conservation agriculture/ no-till management has a lot of benefits such as: better water holding
capacity, less carbon dioxide released, less soil erosion and more.
 lot of farmers do not go for this, because of the less output productivity.

Topic 4: Trade-offs services and biodiversity

Trade-offs: if we loose a bit of biodiversity, then we will gain production and the other way around.
Design cities, we are losing biodiversity.
 Synergies (win-win) are scarce in biodiversity, trade-offs are dominant.
There can be deforestation to create “cultural landscapes”. So trade-off of loss of biodiversity, to gain
a service of values for us.

Topic 5: Telecoupling
Telecoupling: is a strategy that comprehensively analyzes both the socioeconomic and environmental
impacts over long distances.
Topic 6: Land Governance
It refers to set of processes, rules and institutions through which decisions about land use
management, and ownerships are made and implemented.
 it includes the mechanisms by which land is allocated, accessed, and controlled within a society.
 Land governance is a crucial aspect of sustainable development, as it influences various social,
economic, and environmental factors.
 For example: land use planning, policies, regulations. It regulates everything.

1% of farms operate 70% of worlds’ farmland. That is why we need land governance.
There are trade-offs again. Because that 1% is probably very efficient and has a high productivity.

Land grabbing: big companies or countries that buy or lease land to other countries. For the purpose
of security, in terms of food security, we can lease land in other countries to produce the food.
 Congo is most affected by land grabs, but also Indonesia, Cameroon, or Ukraine.
Land use planning is very complex..

Land policies & Interventions

Every policy has different effects and which is the best one comes down to its effectiveness. These
different policies can have different effectiveness and it also depends on the place. There is not silver
bullet on which policy is the most effective.

Topic 7: Land Use and Conflict


With monocultures, the productivity is high: if you do not have monocultures, you might need a lot
more land for the same productivity.
What is fair to look at who emit the most emissions?
- We need lower-income countries to help us with the net zero by 2050. Otherwise, we would
never get the 1.5-degree Celsius target, but is that fair…
 as richest 1% account for more carbon emissions than poorest 66%.
Week 1: Online Lecture
Land use and Food Systems

Biomes: represent ecosystems like rain forests or savannas.  these are natural biomes.
Anthropogenic biomes of the world: more than 75% of land is influenced by humans. Such as
croplands.

Croplands and Rangelands (agriculture) represent a huge portion of the available land  38%.
 it is the largest biome of land of earth today.

 the increase of pasture and cropland comes at the expanse of forest and natural areas.

 the cropland area has not changed much in the last 40 years, but there is a lot of crop production.
 This is due the increase in yield (productivity). This increase in yield is due to the use of
fertilizers. Which has increased 6 times as much compared to 1960.

Take-home message 1: agriculture has transformed our planet, not only by expanding the area, but
also intensively using this area. It is taking up so much of our planet.

Agriculture is not only for food production also for: biofuels or clothing.
Biofuels: represent 1% of global agricultural area. So, the focus is food production.
 however, biofuels do have an increasing rate, so it still matters.

What effects do this big agricultural footprint have on our planet.


 planetary boundaries: are 9 boundaries that have been defined by scientist. If we exceed these
planetary boundaries, we as humans cannot live on this planet anymore. It will disturb the balance,
and for some of the planetary boundaries we are already in the critical zones.
Planetary boundaries that we will discuss and where agriculture has a big effect on:
1. Land system change: we have lost of 1.8 billion hectares of natural areas. in the last years.
So that is 1/3 of the area in the last 5000 years and agriculture is responsible of 80% of the
deforestation. Thus, majority of land system change is due to agriculture (80%).
2. Biosphere integrity: state of biodiversity on our planet.

Graph show the extinction rates of animals. These values are 100 times higher than the
background rate.
What are the drivers of biodiversity loss: habitat loss, pollution (pesticide and fertilizer),
illegal hunting, overexploitation (fish), climate change, invasive species.
 Agriculture is a major driver on habitat loss, pollution and climate change.
3. Biogeochemical flows: this includes the nitrogen problem. The amount of reactive nitrogen is
a problem and this nitrogen has increased 4 times then the natural range.
 is due the Haber Bosch process. This Haber Bosch process has been used for the
construction of the fertilizers and this a major driver of the increase of reactive nitrogen.
 86% of additional Nitrogen is due to agriculture.
4. Freshwater Use:
Irrigated land: using more water on agricultural land. This increased a lot (doubled last 40
years). This increase is mostly for agriculture. 70% of extra water usage is due to agriculture.
5. Climate Change: Agriculture is the sector which produces 24% of the greenhouse gas
emissions.
- Methane from rice production and from cows
- Carbon emissions from fertilizers and deforestation.
 the deforestation have the biggest impact on the emissions as it comes from burning of
biomass and soil. Methane from the live-stock production has second biggest impact. Third
impact is from the nitrogen from fertilizers.

Take-home message 2: Agriculture is the most important driver of many of our environmental
problems. It is behind a lot of planetary boundaries. (and have already crossed the critical state).

The Food System


Type of diet varies a lot in various areas of the world:
- Average global food supply is 2940 kcal per person per day  more than we need.
- Problem is that it is not evenly distributed.
- Richest countries are eating a lot more kcal than the poorer countries.
 In orange: if everyone consumed like them, we do not have enough land.
We have huge inequality in access to nutrition. Also have a huge inequality in contribution to
environmental impacts.

Water footprint has a similar pattern: the water that is required for the production. Beef requires 9
time more water than wheat does…

The richer we get the more protein and the more calories per person. So, more demand for meat due to
population growth and dietary changes.
 in the future we will use more and more agriculture to meet everyone demand. We have already
pushed the planetary boundaries. If we continue on like this, we need to change the way we do the
agriculture.

Take home message 3: All diets are not equal, in terms of nutrition that they provide. In terms of
kcal, but also in terms of the environmental impacts of a diet.

Solutions
1. We need to halt deforestation: expansion of agriculture and deforestation is one of the major
reasons of greenhouse gasses emissions
2. We need to close the yield gaps: so we need to have more yield achieved. For example, in
Africa, there it has really low yields
3. We need to change our diets: Calories delivered per calorie produced. We need calories used
for human consumption. Som more vegetables.
4. We need to reduce food waste: 1/3 of food is currently wasted. Huge potential increasing the
amount of food we get by reducing this waste.

If we implement the solution above we can stay within the planetary boundaries.
Tutorial 1.1

Land Sharing: A situation where low-yield farming enables biodiversity to be maintained within the
agricultural landscape.
Land Sparing: Where high-yielding agriculture is practiced, requiring a smaller area of land to attain
the same yields and therefore leaving greater areas of natural habitat untouched.

Dietary choices have a direct impact on land use/cover.


Half Index: human appropriation of land for food. How much land area would be used if every person
in the world would adopt some “standard” diet.
 USA 95% of half index: share of pasture area must bigger than cropland area.
 Indian 18% of land index: share of cropland area and pasture area equal.
Thus, Indian diet is characterized by the consumption of less but healthier food, whereas the USA diet
is characterized by the consumption of more but less-nutrional food.
Week 1: Thursday Tutorial

Urbanization: The urban population is growing master and is more than the rural population since
2010.  more and more people are living in the cities.
The fraction of urban is very little if you compare it to the croplands or forests.

Why do we care? As the process of urbanization increased rapidly, thus it needs to be cosidered.
 Urbanization area in the Netherlands has doubled in 19 years time.

This graph represents the rate of change (cumulated change) or net change of each of the 6 land cover
types on a global level. What do we see:
- Largest net increase: cropland, deforestation is predominantly due to the conversion of
cropland.  losing a lot of forests.
- Built up land (urban) is the 2nd largest net increase globally. It is increasing rapidly. Urban
land expansion is a very important land cover type because it is increasing fast.

If you look at the different parts of the world you can see:
- Europe: urban expansion is by far the biggest, and the only one that is positive. The rest has a
negative growth. We are using cropland as we replace it for forests or for urban lands.
- China/India: cropland increase, but since last few years is going down. Urban increases a lot.
Remember: if one land type increases another one must increase.
- The urban land type are everywhere increases (thus they are a permanent increase)
- Middle east/ southeast Asia: cropland increases.
The graphs show per area the relative importance of one land type and not on a global scale.
Because the scales are not the same, you can not compare them directly by amounts.

Different type of changes


You can see that percentage of all land does not change.
The percentages (or relatives) of built-up area does change.

Different type of changes:


1. Densification (or intensity in population ): everything is much higher. But you can not see this
on a satellite. Think of skyscrapers.
2. Expansion: this would be the land cover change what would you see from a satellite. So the
increase of area.

Decomposition of urban land: the urban land per person in the US is much higher than that of China.

BPC: build-up land per person.


- In Africa there is a lot of population growth but the BPC is increasing less fast than the
population.
- In Europe for example: the population stays the same but the BPC is increasing a lot faster.
This gets you to the expansion phase.

Urban cities get denser, small and medium cities get less denser. The built-up land get denser in large
cities.
 these patterns also differ a lot per country.

Drivers underlying urbanization:


- Regionally we are going do a transition: we do not have a lot of new babies. The main reason
of population growth is that we are getting older.
- Migration
- Wealth: larger contributor to increase in built-up land. Villas use a lot of land per person.
- Land per person: next to houses, but also offices, schools etc.
- Economic development: huge effect on urbanization.

Drivers urban land consumption;


What are the drivers of the urban land consumptions?
- Population density
- GDP
- But also, young age dependency, large families mean less urban land per person.
- Gini?
- Higher educated society leads also to higher urban land consumption.
 thus, not only GDP and people are the drivers, but there are also more drivers that have effect on
the urban land consumption.

To reduce the urban foot print you should avoid the development of countries. This is of course
inevitable and it comes thus with trade-offs.

If we look locally and especially at Europe process of peri-urbanization:

Focus on planning:
Planning you think it affects land cover and land use. More cases that first land cover and then
planning. Often the reverse is case. Plans are adjusted to reality, so land cover change and then the
planning needs to adjust that!

Land use impacts of urbanization


Direct land use cover impacts: urban expansion comes in the expanse of cropland.
Indirect land use cover impacts: The loss of crop land due to urbanization, will be replaced anywhere
else, because the people will not eat less. Thus, cropland will be in place of forest  deforestation.

In areas where croplands are very productive (China) we lose croplands. In areas where croplands are
not very productive (Amazon), we gain croplands.
Week 2: Introduction to impact evaluation (causal inference, ex-post analysis)

Two types of models of LUCC models:


1. Ex-post models:
- Used to: study something that has already happened.
- Example: evaluate the impact of a policy (that has already been implemented). We can go
back and see what the effects are.
- Methods: Statistics/ Econometrics (regression analysis).

2. Ex-ante models:
- Used to study something may happen in the future explore hypothetical scenarios.
- Example: assess the potential impacts of climate change (2050)
- Methods: Simulation models

Examples of popular impact evaluation topics


- Protected areas (and indigenous lands)
- Payments for environmental services (carbon, biodiversity)
- Law enforcement (monitoring systems, fines, embargos)
- Supply chains
- Governance (land tenure)
- Taxes/subsidies/credit access

We can classify these policies into different groups such as enabling, disincentives and incentives.

How to rigorously measure impacts of these policies or interventions.  paper form Allen Blackman.

Gold standard: randomized controlled trial (RCT)


- You start with a population; you randomly sample two groups. One is the treatment group
(exposed to the intervention) and other one is the control group (not exposed to the
intervention). Then you monitor you estimate the impact after monitoring by comparison.
-
- The problem with RCT is that they are difficult to implement. They cost a lot, and sometimes
it is not even possible as you need to have a population that you can split.
But there is an alternative: (Reality) Quasi-experiments.
- Studies that aim to evaluate interventions but do not use randomization.
- They are similar to randomized trials; quasi-experiments aim to demonstrate causality
between an intervention and an outcome.
Example: protected areas and deforestation.
- Conventional methods as in the example in the lecture likely overestimate the effectiveness of
protected areas in Costa Rica by more than 65%.  this is also called location bias. Location
bias is that because of the location there was already less deforestation.
Fairer to come up with a measurement for the effectiveness of protected areas. To create a control to
estimate the average protected area.  called a counterfactual.

But how do we select/create these control areas (counterfactuals) in practice?


Unit of Analysis: individual pixels vs pixel windows.
- Individual pixels: so, each pixel it is looked at if it is forest or not forest.
- Pixel windows: for example, 68% of the pixels is forest and 32% is non-forest.

By looking at the trend of deforestation to get a


counterfactual? The best way to get the
counterfactual is to use controls. Impact of the
REDD+ project (rigorous approach). Because
then you can see that the control is also affected
(it is a confounding factor).
Confounding Factor: variable whose presence
affects the variables being studied so that the
results do not reflect the actual relationship.
The intervention did not cause the reduction!

Confounding factor example: Brazilian


Amazon
- The decrease in deforestation was caused by the confounding factors: by decline in prices of
beef and soy and not per se by the intervention policy.
- Confounding factors: factors that are not the one to explain but do have effect on the outcome.

BACI: Before-after-control- impact quasi experimental design.


It is best we can do! It designs are an effective method to evaluate natural and human-induced
perturbations (or interventions) on ecological (or social) variables when treatment sites cannot be
randomly chosen.

Most used method: difference-in-difference (DiD)

in

First difference is before the intervention starts, second difference is after the intervention.
 with the ATT you can measure the impact.

How to identify control units when you have a quasi-experiment?


Matching analysis:
- You have a population with varying characteristics. With the characteristics you will match
the treatment with the control that has similar characteristics.
- You compare apples to apples  to address the lack of randomization.

Protected areas are the corner stone of government policies. But we learn from fair comparisons that
the impact of protected areas is not that big (between 1-2 to 6%).
 effectiveness depends on local context…
Part 2: Simulations (ex-ante analysis)

Now we will focus on the ex-ante models. Thus the simulation models to study something that may
happen in the future and explore different scenarios.

LUCC models: the workflows.


1. Estimate the rate of LUCC.
2. Identify spatial drivers of LUCC.
3. Apply rate of LUCC to the LUCC suitability map.

How to model the future rate (or amount) of land use change? Estimate the rate of LUCC.
- Take the average between two or more years.
- Trend observed across multiple years.
- Econometric time series forecasts: AR, ARMA< ARIMA
- Transition Matrix.

Partial Computable General Equilibrium Models.


 focused on supply and demand. They describe the allocation of resources in a market economy.
Partial Equilibrium models: consider only one market at a time, ignoring potential interactions across
markets/ different sectors of the economy.

Computable General Equilibrium models: look at the whole economy across different sectors.

Example of a partial equilibrium model GLOBIOM


- Socioeconomic data & forecasts
- Ecological and economic sub-models
- Multiple maps

Rate of change with: Agent-based models (AGM)


 bottom-up type of model.
1. Farmer-agents
2. Land-use cover change decisions
3. Make decisions based on their resources..

 These are models to estimate the rate of LUCC.

How to allocate the future rate of land-use change across the landscape?
Location of change: example.
Step 1. Randomly sample afforestation and non-afforestation pixels.
Step 2. Extract pixel information from multiple explanatory maps.
Step 3. Model afforestation likelihoods based on the explanatory variables (method logistic
regression, random forest or Artificial Neural Network).
So then you have first identified the suitability of LUCC or LUCC risk/likelihoods maps (identify
spatial drivers of LUCC) and then you can apply rate of LUCC to the LUCC suitability map.

Part 3: The validation of the model

You have two models (A and B) and the simulation from that. If we do it from the past.
We have the observed of the map.

Validation metric: Figure of Merit.


Hit: predicted change and it happened.
Miss: did not predicted change but it happened.
False alarm: predicted change, but it did not happen.

 the closer to 1 the

Comparison against a null model (benchmark):


- Null model: no change, for example no deforestation is simulated
- Simulated model B: change.
 Question about this on the exam.
 sometimes the null model performs better than the simulation.
Comparisons across different scales: you need to resample.

Model validation: crisp-set and fuzzy-set theories:


- Crisp-set theory: you compare each pixel (in the simulation) to the same pixel in the
observed. Doesn’t matter that you have deforestation really close by
- Fuzzy-set theory: the neighborhood matters. So, it is almost a hit. But almost a miss too.
Getting a hit very close to the it matters.

Example of fuzzy validation metric: Constanza 1989.


Week 2: Tutorial 2.1
Land use modelling: models can help us with many different questions and the modelling tools we
develop depend on the question we have. It structures our understanding and develop theories.
 use models for scientific purposes, but also for very practical real-world questions.
Field surveys + remote sensing + ethnography.

Model as a predictive/explorative tool:


- Be prepared of the future.
- Test suitability of interventions under different scenarios
Example: where should we delineate the next protected area?
- Simulation as a tool to discuss the future we want (boundary object).
Example: should Amsterdam expand or densify? What would both options look like?
- Assess impacts and avoid conflicts.
- Design optimal solutions.

Ex-ante models: agenda setting, design and implementation.


So , there are broadly 4 different use cases of a model. The question you want to answer determined
the tool you use.

Four purposes of modelling:


1. Agenda setting (ex-ante):
 Explanatory scenarios: start in the present and look at what could happen in different
scenarios in the future.  What will the future be like?
Example: increasing demand scenario of land use types and the effects.
2. Design (ex-ante):
 Target seeking scenarios  How will I get there? Have a target in mind, and look at
different options how do you get to the target.
3. Implementation (ex-ante):
 Policy screening scenarios: what would happen if I did an intervention/policy?
 (ex-ante analysis of policy.
 Example: What are the land use effects of building a new road? Or new conservation
areas?
 More obvious model.
4. Review (Ex-post):
 Retrospective policy evaluation: you look back and look if what you did if that works
(ex-post).  did it the policy work?
 We have protected large parts of the world’s forest? Did it work?
 The tool to use depends on the question then select appropriate method and tool for question.
Models can have different goals/ambitions, and corresponding complexity (three levels of
complexity/ambition):
1. Understand drivers of land use change: static empirical models relating drivers to land use.
2. Identify areas suitable for land use types: land evaluation.
3. Understand potential impacts of land use change and interventions: dynamic land change
model (most complicated one).
 of course we will be the most ambitious.

Characteristics of dynamic land use models?


- Account for different land use types
- Simulate past and/or future dynamics.
- Can answer wat if questions and scnearios.
Land grabs: one big company takes land from someone else.

LUCC modelling: explore/explain the spatial aspects of interactions between individuals or societies
and their natural environment

Land cover: physical or biological coverage of the Earth’s surface. Forests, water bodies, urban areas,
grasslands, etc. Surface features can be observed and are regardless of how land is used or managed.
Land use: human activities that take place on the land and how the land is used for various purposes:
residential, commercial, industrial, agricultural, recreational. Emphasis on human activities and
purposes for which the land is being used.

VON THUNEN MODEL


How market processes could determine how land in different locations would be used. To model how
crops organized themselves spatially according to the trade-off between their value (market price) and
the cost of transporting them to the market.
 simplest in agricultural use
 can also in urbanization patterns: office buildings tend to replace most residential space at the city
center given the high rents.
The assumptions of the von Thünen model:
- Located centrally within an Isolated state: self-sufficient and has no external influences.
- Isolated State is surrounded by an unoccupied wilderness.
- Land of the State is completely flat and has no rivers or mountains to interrupt the terrain.
- Soil quality and climate are consistent throughout the state.
- Farmer in the Isolated State transport their agricultural goods to market via oxcart, across
land, directly to the central city (roads are not used).
- Farmers act to maximize profits.

There are 4 rings (zones) of agricultural activity surrounding the city.


1. Zone 1: Dairying and intensive farming occur in the ring closest to city. Since vegetables,
fruit, milk and other dairy products must get to market quickly, they are produced close to the
city.
2. Zone 2: Timber and firewood would be produced for fuel and building materials in the
second ring closest to the city. Wood was very important but also it is heavy and difficult to
transport.
3. Zone 3: third ring consists of extensive field crops, such as grains for bread. Since grains last
longer than dairy products and are much lighter than wood, thus can be located further from
the city.
4. Ranching: final ring since animals can be raised far from the city because they are self-
transporting. Animals can walk to the central city for sale or butchering.
 beyond the fourth ring lies the unoccupied wilderness, which is too great a distance from the
central city for any type of agricultural product.
Even though Von Thünen was created in a time before factories, highways and even railroads, it is still
an important LUCC model.
Von Thünen: illustration between the trade-off between land cost and transportation costs (closer tot
the city, the price of land increases). However, it is not difficult to adapt the model to include roads
and rivers that might decrease transportation costs to some location.

DYNA CLUE
Dynamic Conversion of Land Use and its Effects at Small Regional Extent
 conversion of land use and its effects framework is devised to simulate LUCC by leveraging
empirically quantified relationships between land use cover and its spatial driving factors.
It is sub-divides into two distinct modules:
1. Non-spatial demand module: calculates the area change for all land use/cover types at the
study-region level.
2. Spatially explicit allocation procedure: these demands are translated into LUCC at different
locations within the study region using a raster-based system.

The allocation is based upon multiple possibilities. There are four categories that together create a set
of conditions and possibilities for which the model calculates the best solution in an iterative
procedure:
1. Spatial policies and restrictions: indicate areas where land use changes are restricted
through policies or tenure status.
2. land use/cover type conversion settings: 1) conversion elasticities are related to the
reversibility of land use cover. Some use high capital investment and are not easily converted
unless there is sufficient demand. 0 is easy conversion and 1 is irreversible change. 2)
transition sequence: a conversion matrix present to which other land use/cover type the
present land use/cover type can be converted, in which regions it is allowed. How many years
(steps) it takes, and the maximum number of years that a land use cover type can remain the
same.
3. land use/cover requirements (demands): they constrain the simulation by defining the
amount of LUCC required for the study region. All the changes should meet these
requirements.
4. locations characteristics: conversion happen at the highest preference or suitability.
Week 3: Land Use, Biodiversity, and Ecosystem services
Federico Riva

Topic 1: Land Use Change

Land use is increase and it affects biodiversity and thus our ecosystem services. So we need
to properly understand land use changes.
What is the relationship among land use change biodiversity and ecosystem services?

Paper by Elis et al: Biggest forest in Amazon or in the north.


The land use type distributions were a long time the same, but now that there is change in the
distributions and that is due to human activity.

There are four fundamental ways in which landscapes can change:


- Changes in habitat area: changes that can be quantified. Habitat loss/deforestation
- Changes in habitat configuration (fragmentation): same total area of forest but
broken into four patches.
- Changes in habitat composition (quality): different types of land type. From mostly
forest to extensive cropland. Thus, habitat heterogeneity, and the movement to land
use intensity
- Changes in habitat connectivity: Changes in the degree to which a landscape allows
movement (through for example a corridor or steppingstones).
In real landscapes all these aspects change together often in connection. So, through the loss
of forest there is a fragmentation of the forest.
Some examples:
- Habitat loss is not habitat fragmentation: see picture, most of the habitat we have is
fragmentated. But they can go hand in hand.
- In the amazon and Congo there is continuous forest. In Florida there is a fragmented
forest due to human activity. However, in the north of Canada it is natural that there is
fragmentation of forest.
Habitat connectivity: most key areas unprotected.
It is important to differentiate the different types of habitat change, because they differently
affect biodiversity and ecosystem services.

Topic 2: Biodiversity

What is biodiversity? Biodiversity is not one number and you can


look at it at different ways:
- Different species on earth (taxonomic diversity)
- Different evolutionary histories. Thus, the evolutionary
pathways (phylogenetic diversity)
- Different forms and functions of life: so, for example that
very different plants converge now to the same function
(functional diversity)
- Different ways in which species interact (interaction
diversity)
- Different ways in which species interact.
What we do not know:
There is a lot we do not know about biodiversity: we do not know how many species there
are, where they are located and whether they are increasing or decreasing. Only little subset
of the species that live on the planet and know mostly about species in North America and
Europe.

What we do know:
Defaunation in the Anthropocene: facing 6th mass extinction due to human activity.
 the extinction rates for species are 100 to 1000 times higher than what would be expected
without the human influence.
 also been investigated: human activities threaten biodiversity.
How can we manage land use to positively affect biodiversity?
- As habitat loss is the first threat to species: thus, land use change is a big threat!
- Confusing habitat loss and fragmentation implies neglecting habitat protection when
habitat exists in the form of many small patches: small, isolated habitat patches may
not support the same level of biodiversity and ecosystem functioning as larger,
interconnected habitat

Most habitat patches on Earth are small due to human activities. But several small patches
hold more species than large patches. Thus, there is surprising value of species in small patch
of grass. To save nature we can think small..

To protect biodiversity we need to provide as much habitat as possible to species:


1. Protect habitat in every biome and ecoregion
2. Protect as much habitat as possible, including smaller and larger patches
3. Protect habitat to facilitate connectivity.

Topic 3: Ecosystem services

Ecosystem service: “the benfits that humans obtain form ecosystems. The ecosystem
services can be divided into:
- Supporting services: nutrient cycling, primary production
- Provisioning: food, fresh water, wood and fiber
- Regulating: climate regulation, flood regulation, water purification
- Cultural: how we as humans perceive the nature.
 so, ecosystem services are complex but they are fundamental for human societies.
Ecosystem services are fundamental for human societies.

Biodiversity and Ecosystem Services


- Variation of species and within species guarantees a variation in functions and we all
need these functions: need to ensure that all the different things maintain because they
all have different function and they are interconnected with each other.
-
Three key functions for functional diversity:
- Complementary: different life forms typically complement each other: Different
plants with roots at different depths more easily survive drought, bees with different
mouth lengths will pollinate different flowers.
- Stability: biodiversity often enhances stability. With a higher biodiversity it is more
likely that one species or another will buffer undesirable effects of environmental
variation.
- Keystone Species: Some species are key for ecosystem persistence: such as corals
create diverse reef ecosystems; sea otters eat sea urchins. Without the sea otter, which
prevents sea urchins from overbreeding, kelp forests would be severely depleted.
 biodiversity often enhances ecosystem services.

Functional redundancy (duplication) as a risk strategy: different bee species ensure


pollination also if one species doesn’t thrive in a give year.
 also by a study: more species are required to reach the same level of successful pollination
when considering large scales.
Thus biodiversity is good for ecosystem services and we need to sustain biodiversity. But
how?

Land use strategies can enhance biodiversity and thus, in turn, ecosystem services.
 Crop heterogeneity enhances the diversity in agricultural regions.
 we need to find win-win situations.

Land sparing/ sharing debate.


Land sparing: fragmentation. If you have more intensive agriculture you will also have more
land for nature conservation. Will give the highest amount of production for agricultural field.
- Allows for focusing agricultural activities on portions of the landscape best suited to
supporting production.
- Similarly designating areas for conservation ensures core habitat for wildlife not
suited for cohabitation with humans is secured.
Land sharing: continuous. Sharing nature and people. You have semi-natural habitat, more
smaller spaces.
- Securing multiple benefits provided by agricultural landscape.
- Recognizes the contribution of conservation efforts to supporting agricultural
production (such as pollination and pest control)
- Manage biodiversity to increase the productivity, value, and resilience of the farming
systems.

Also, when we think about strategies in Europe. If you would protect habitat in Europe,
somewhere will need to produce the food. So, it would go overseas.
Why is this debate so controversial:
- It depends on which values you want to converse. It depends on what the trade-offs
you are willing to make, and it also involves feelings.
- There are a lot of trade-offs. It depends on…
- And so many other factors.
- Social issues.
Part 2: The geography of megatrends affecting European agriculture

STEEP: social technological environmental economic political.

Dutch farms are really entrepreneurial. But will the farms in other parts of Europe evolve in
the same way as the Dutch farms? Probably not.

The paper that he wrote:


1. European agriculture will transform: it will be different in the future
2. European agriculture must transform: to meet the sustainability challenges and targets.
3. Where? Different regions different pressures. How can you map out the pressures the
system is under? How do you measure the pressure to change or to staying the same.
You look at the plans? Pressure can be: that the young people move out of these
places.

Systemic resilience
- You have a current state. And you have a desirable future change: this is where you
want to go because for example of sustainability challenges. And the undesirable
future: this is now where you want to go.
- Right now, the system is stable, and this stable has resilience:
- Unhelpful resilience: keeping the marble form flowing to the desirable futures
through pressures (green arrows)
- Helpful resilience: keeping the marble form flowing to the undesirable futre.

Futures studies: the systematic study of possible, probable and preferable futures. We are
trying to use a methodology here.
 Megatrends: are long term driving forces that are observable now and will likely have a
significant influence in the future. This is an important methodology for futures studies.

Difference between forecasts scenarios and megatrends:


- Forecasting models: certain trends + certain causal mechnisms
- Scenarios: uncertain trends + certain causal mechanisms
- Megatrends and future studies: certain trends + uncertain causal mechanisms

Example future studies/megatrends: what will the population aging (we are certain that we
are becoming older) aging do to our cities or areas?
The megatrends method (systematic approach):
1. Framing:
- what is geography of the long term driving forces pressuring European agriculture, at
the regional scale?
- Where is the agricultural system under pressure to change? Where the pressures
making it more persistent? (think of the marble, current system become more stable,
the marble goes deeper in the pit and it more difficult to move).
2. Orienting: know you will choose your trends.
- Longlisting trends using the STEEP framework
- Shortlisting then the longlist by selection those that are:
 Credible and dynamic: must be mentioned by multiple existing academic
 Spatially heterogenous: across Europe (because of differences in pressures and or
sensitivities to pressures
 Quantifiable: trends need to be put into numbers so you can work with it.

MEGATRENDS
He landed with four mega trends to answer the question: how will farms in Europe change in
2050?
1. Climate change: a warming continent
2. Demography: an ageing continent
3. Ideological shifts
- Productivism: if what you value is large amounts of food production
- Post-productivism: if you additionally also value biodiversity and vital rural
communities than it is post-productivism (about ideology!).
- Environmental action space: a continent running against its environmental ceiling.

3. Scanning:
- Mapping the trends: you use data sets

4. Interpreting:
- What are the likely consequences of the trends?
- Literature review to establish links between a trend and megatrends/pressures.
Megatrends and then the pressures has an effect to:
- Persistence:
- Systematic change
- Marginalization, bad news?

Week 3: Tutorial 1 Agent-Based Model

Individual decisions can lead to emergent patterns (system output).

What is an agent-based model?


- Systems are modelled as collections of unique individuals (agents, and they are all
different
- System dynamics arise from the interactions of individuals with their environment
and with each other.

Building blocks of the system:


- Individuals: heterogenous and adaptive (reactive)
- Environment
- Interaction and processes (for example decision rules)
These building blocks build the emergent patterns as for example population dynamics,
overgrazing, re-/de-forestation.
Some characteristics of the ABM model with the example of wolf-sheep:
- Bottom-up: represent individual processes, rather than system-level equations).
- Heterogeneity: start at different locations with different energy stocks
- Interactions: wolves/sheep interact with each other, sheep interact with grass.
- Emergence: complex macro outcomes from the simple micro processes.
 complexity.
Example 2:
- How do farm cessation, expansion, and diversification, collectively shape landscape
structure?
- Agent = farmer
- Typology: hobby, conventional, diversifier, expansionist
- Decisions can be made: expand farm? Protect landscape elements.
- Outputs: projected landscape protection over time and space.
 you can look at the output how they different respond to these scenarios. This is how land
use is going to evolve over time.

So, the characteristics and capabilities of ABM:


- Bottom-up representation of processes (rather than patterns)
- Relationships and feedbacks between individuals and dynamic environments
- Representation of autonomous actors (e.g. people, families, government)/agents:
 Heterogeneity (wealth, gender, selfishness)
 Interactions (social networks, social-environmental)
 Nuanced decision-making representation: rules (utility maximizing,
satisficing, copy-cat) and preferences (economic, environmental, prestige).

Application of ABM (and models in general)


- Complicatedness: the structure is big there / lot of variables
- Complexity: quite simple structure but has a very complex behavior

AGM is about understanding this kind of complexity.


Relationship between complicatedness and complexity. If you add to much variables it looses
it complexity.

What level of detail is most useful? Two different perspectives?


- Kiss: keep it simple, stupid
- KIDS: keep it descriptive, stupid.
- But if you combine them. WE have different complicatedness for different purposes.
So kiss for the wolf-sheep and the Indonesia-land use for the KIDS one.

How is ABM useful for modelling land systems?

At least for three reasons:


1. Complex Human Behavior / non-economic components of decision making:
Not all land managers are profit maximisers. In Europe someone makes different decisions in
land than someone in Africa or South America
 ABM can formalize a) different decision-making procedures and b) heterogeneity in these
 look at the different implication of those different agents.
How to model decision making?
- Theory-based: rationality (humans are utility maximizers) or theory of planned
behavior (decisions follow from intentions).
- Rule based: more decision tree structures.
- Probabilistic: chances
 parameterize decision making by theory, interview with stakeholder, choice experiments,
survey data or expert judgement.
2. Human-environment feedbacks:
People will decide if they follow the rules and there is a two-way feedback system with the
humans and the trees (the environment).
How does rule adherence affect forest outcomes? This is how we explicitly model the
feedback with the environment. If nobody follows rules and forest is depleted and if everyone
follows the rules the forest survives.
 This is how we can explicitly model the feedbacks with the environment.

3. Participatory modeling
- Agent based models are more intuitively understood by stakeholders.
 participatory simulation of land use change in Vietnam
Why do participatory modeling?
- Mutual learning and knowledge democratization
- Increase policy relevance and impact!

Summary
Useful features of ABM for understanding land use change:
1. Nuanced behavioral representation  non-economic components of decision-making.
2. Interactions and feedback  human environment feedback.
3. Bottom-up description of processes  useful for participatory modelling.
Key concepts:
1. Different modeling purposes
2. Complicatedness vs complexity
Week 4: Lecture 1 Towards Sustainable Land Use

What is sustainable land use?


- Illegal mining is not sustainable.
- However, it is not always straightforward to see what is sustainable?
- Is Land sparing or is land sharing sustainable?  no one can answer this.

Agroforestry: land use management in which combinations of trees or shrubs are grown
around or among crops or pastureland.
Benefits of agroforestry?
- Creates microclimates.
- Provides habitat for pollinator.
- Increases crop yield, can be debatable.

Conventional farming vs. Agroforestry?


 If you are sharing trees with crops. They will be competing against each other and the
lower yields you will find.

Organic Farming: agricultural system that uses fertilizers of organic origin such as compost
manure, green manure, and bone meal and places emphasis on techniques such as crop
rotation and companion planting. Promotes biodiversity and is generally more
environmentally friendly.

Conventional farming vs. Organic farming


 organic farming better for the environment, but the yields are probably lower. Also, a lot
of yield variability.
 conventional farmer has a bigger (negative) impact on the environment.

We can produce food in all different regions  driver for land grabs.

Land abandonment in Europe.


- There is a trend of reforestation.  explained by the Forest Transition Theory.
- Europe is in the post-transition phase, we have already had the economic
development, result the production is now at other places (overseas).

There is a trend of land abandonment in Europe: the trade-offs.


- Cultural heritage and Agricultural production vs Ecological restoration/ Wilderness.
- Agrobiodiversity vs Come-back of megafauna.
- The more forests we have: increased wildfire risk/ alien plant invasion vs carbon
sequestration (storing and capturing).

What is sustainable land use?


No single “sustainable land-use”/
Land use decisions always comes with trade-offs:
- Between the services provided (food, biodiversity, regulation)
- Between stakeholders/beneficiaries
- Between locations: local vs. global
- Between values: monetary, social or physical decision?
- Between the short-term and the long-term
What can we do for more sustainable land use?
1. Reducing land use inefficiencies:
- We have over-consumption, this is an inefficiency we can reduce.
- Waste can be reduced.
- The animal product distribution.
2. Sustainable diets: changing diets would lead to a significant change in the demand
for the land.  so, it is a way to reduce the pressures on the land.

It is all related to different preferences and different values.


Promoting more sustainable practices:
- is the role of consumers,
- role of companies, and
- the role governments.

Some ideas that are sustainable practices:


- Certification: such as sustainable forest management (FSC)
 Study shows that there is no effect of the certification (FSC) on deforestation…
zx
 Small-scale agriculture
- Zero-deforestation commitments (ZDCs): voluntary sustainability initiative that
companies adopt to signal their intention to reduce or eliminate deforestation
associated with commodities that they produce, trade, or sell.
 Why do they follow this, because government pushes them or that the consumers
make them.
 Effect could be great in theory, but in practice it is more complicated.
 limitations: voluntary commitment, not all forests are falling under the ZDC,
mapping and definition of the forests, displacement of other activities (spillage or
leakage), forest to rangeland.
- Payments for Environmental Services (PES): REDD+
 or the carbon credits.
Centralized REDD+: national REDD+ programs. ,subnational REDD_ jurisdictions,
laws and regulations, part of national GHG emission inventories.
Decentralized REDD+: voluntary activities, project-based REDD+, usually
unregulated by governments, carbon offset certification schemes.
- Taxes and subsidies
- Behavioral changes… Nudging: influencing people to make the right decision,
without really changing a lot, so to be subtle motivated.  can be a powerful way to
influence land use.
- Carbon payments could have a possible impact on agriculture.
- Biodiversity credits: offsets are not that ineffective. Idea/theory is great, but still in
practice not that effective.

You might also like