3. Mapping and assessment of ecosystem services
To download the whole theory content (PDF)
Since the European Commission stated in the Action 5 of the Biodiversity Strategy to 2020, that Member States “…will map and assess the state of ecosystems and their services in their national territory…”, there has been a growing need to value the provision of ecosystem services, but also to map, in a spatially explicit manner, the provision and demand of ecosystem services at a wide variety of scales, from transnational to local.
A question may be asked then, why is there a need to map ecosystem services? Firstly, the processes that lead to the production of ecosystem services are of a spatial nature (Fig. 1.3). The ecosystems functions and processes that are responsible for the production of ecosystem services vary greatly in time and space and are scale dependent. Moreover, the drivers of change which affect and modify ecosystems functions and processes show a strong spatial variation: Land use patterns, fragmentation of the land or agriculture intensification, just to name a few.
Therefore, ecosystem services maps are much needed in order to describe and assess the production of ecosystem services as a function of ecosystem processes, patterns of land use, climate and environmental variation (Maes et al., 2013).
The supply of ecosystem services is a complex process and it is often the case when different ecosystem services are interrelated. Synergies and trade-offs within different ecosystem services, and between ecosystem services and biodiversity are common. In some cases the production of a certain ecosystem service will be increased at the expense of another service, or the increase in the production of one service causes the increase in another service (bundles and synergies). Only if the ecosystem services are mapped and their spatial distribution is known, we will be able to disentangle this complex system.
As explained in Chapters 1 and 2, the ecosystems services framework has two interrelated dimensions: supply and demand. The demand for ecosystem services is defined as “ecosystem goods and services currently consumed or used in a particular area over a given time period” (Burkhard et al., 2014). This demand can change over space and time, and may be independent of the actual supply. Once again, maps of the supply and demand of ecosystem services are needed in order to assess and quantify the flows of benefits from ecosystem service supply areas to near and distant human populations.
Ultimately, the visualization of ecosystem services as supply and demand maps can be used in a wide array of processes by decision makers, e.g. Land use planning, Environmental Impact Assessment or landscape management.
3.1. A framework for modelling ecosystem services
An essential first step before the quantification and mapping of ecosystem services is the definition of a modelling framework. These decision frameworks vary in terms of data required, scale, drivers and knowledge required and therefore the model choice will be driven by our project’s characteristics. Kienast and Helfenstein (2016) compiled a classification of ecosystem services models:
- Process based models
- Empirical models
- Tiered approaches
- Indicator-based assessments
- Landscape models
Kienast and Helfenstein (2016) also propose a 6 point framework to describe ecosystem services models. This 6 point framework should also work as a guide for choosing the right model given the project requirements:
Variable (used) knowledge: Refers to the level of knowledge available about the ecosystem services under study, from very basic, narrative-based or experience-based to process-oriented and analytical knowledge.
Spatial scale: The scale of the ecosystem service assessment may vary from local or municipal level to global level and will be a main driver of the type of data required for the ecosystem services assessment.
Temporal scale: Similarly, the temporal scale of the ecosystem services assessment will directly influence the results and the data needs. The temporal scale may vary from months to decades or centuries.
Available (used) data: Data availability and data characteristics (spatial and thematic scales) will drive the choice of models for ecosystem services assessments. For example, if high spatial and thematic resolution data are available, then more complex process-based model could be used.
Stakeholder involvement: Refers to the degree to which we want to open the ecosystem services assessment to the wider public. For example, if stakeholder involvement is a key requirement in our project, we may need to use bottom-up and participatory assessment tools.
Output: The output of an ecosystem services assessment may be qualitative or quantitative and is directly related with the data needs and the choice of model. Quantitative outputs usually require detailed data and mathematical models, whereas qualitative outputs may need expert opinion assessment and qualitative scales.
An essential step in the implementation of the ecosystem services framework is the biophysical quantification of the ecosystem services. Most of the ecosystem services under the provisioning category can be directly quantified. However, the measurement of regulating, supporting and cultural services is more complex and therefore indicators or proxy data are needed (Egoh et al., 2012). As defined by Wiggering and Müller (2004) ”indicators generally are variables that provide aggregated information on certain phenomena”. Robust biophysical indicators are required not only to evaluate ecosystem services, but also to assess the change of ecosystem services provision over time. In an attempt to structurize the quantification of ecosystem services and the choice of indicators, the DPSIR framework (Drivers, Pressures, State, Impact, Response) (Fig. 3.1) has been widely adopted (Müller &Buckhard, 2012).
According to the DPSIR framework, political decisions, production systems and societal developments (drivers) generate pressures in environmental systems. These pressures eventually lead to changes in the state of environmental systems. Consequently, impacts on human and natural systems may lead to changes in the provision of ecosystem goods and services. Finally, societies try to minimize these impacts or adapt to them through response strategies.
The DPSIR framework also captures the connexions between the environmental state (ecosystems and biodiversity) and the human systems. Following this framework, ecosystem services indicators should capture cause-effect relations between pressures, states and impacts.
The role of scale should also be taken into account in the choice of indicators for ecosystem services. The scale (temporal or spatial dimension) of ecological patterns and processes that lead to the provision of ecosystem services should be assessed before an adequate indicator is chosen (Postchin and Haines-Young, 2016). Most provisioning services can be assessed at multiple scales, whereas certain regulation services (e.g. local climate regulation or flood protection) depend strongly on the local or regional context.
Given a particular project, the choice of indicators will mainly be driven by:
- Scope of the study and selection of ecosystem services to be assessed
- Scale of the study
- Data availability
Several guidelines and indicator sets have been proposed at a wide variety of scales. We provide just a few examples in this chapter:
- Mapping and Assessment of Ecosystems and their Services (Indicators for ecosystem assessments under Action 5 of the EU Biodiversity Strategy to 2020) (Maes et al., 2013): The second MAES report presents a wide selection of ecosystem services indicators aimed at the European and Member State’s level, based on the CICES classification.
- Indicators for mapping ecosystem services: A review (JRC scientific and policy reports) (Egoh et al., 2012): A review of spatial information and indicators for mapping and modelling ecosystem services at global, continental and national level.
- A European assessment of the provision of ecosystem services (JRC scientific and policy reports) (Maes et al., 2011): A set of indicators is provided, based on spatial data available at the European scale.
3.3. Methodologies for assessment and mapping of ecosystem services
In an attempt to group and classify all the available methodologies for mapping and assessing ecosystem services, four main approaches may be distinguished:
- Biophysical methods
- Socio-cultural methods
- Economic methods
- Expert-based quantification.
3.3.1. Biophysical methods
Biophysical methodologies are the most widespread approach to map and assess both the supply and the actual use and demand of ecosystem services. A biophysical quantification is the measurement of ES in biophysical units (e.g. quantities of water infiltrated in an aquifer, volume of timber produced in a forest or amount of carbon stored in the soil). Therefore, biophysical methods rely strongly in indicators, proxies and biophysical models. Indicators and biophysical models allow not only to quantify ecosystem services but also to assess the conditions of the ecosystems in terms of structure and function.
In order to guide the biophysical evaluation of ES, we need to answer two questions:
- What do we measure?
- How do we measure?
What to measure?
When the set of ES relevant to our project has been selected, ES indicators must be chosen to assess and monitor the state and provision of ES (see section 3.3). The choice of an indicator depends on multiple factors such as the purpose of the analysis, the audience, spatial and temporal scales and data availability. An important aspect to consider when choosing indicators is whether they will be used to measure stock (potential to deliver ES), or flow (the actual use or realisation of the service). Flow indicators are usually expressed by unit of time. As an example, the grass produced in meadows can be measured as harvested hay (ES flow) in t/ha/year. However, the total amount of standing biomass may not be harvested and can be expressed as t/ha. If the stock is harvested, stock becomes flow (Burkhard and Maes, 2017).
How to measure?
When the set of ES has been selected, and appropriate indicators have been chosen to assess the stock and provision of ES, the following step would be the actual quantification of the biophysical stock and flow of ES. Burkhard and Maes (2017) distinguish three general approaches: direct measurements, indirect measurements and ES modelling.
188.8.131.52. Direct measurements of ecosystem services
Direct measurements of an ecosystem service indicator are those derived from observations, monitoring surveys or questionnaires. Examples of direct measurements are: measuring the total amount of grass produced in a grassland (biomass production) or counting the total number and number of species of pollinating insects along a transect in a grassland plot (pollination).
Direct measurements are the most accurate way of quantification, but require a high amount time and resources. Therefore, these type of measurements of ES are appropriate at the site or local level. However, in some cases these indicators have already been measured for different purpose (e.g. crop and timber production statistics) and can be used to assess stock and flow of ES.
184.108.40.206. Indirect measurements of ecosystem services
Indirect measurements also provide a biophysical value, but further interpretations, assumptions or data processing are needed in order to be used as measures of ES.
Data collected through remote sensing techniques is a good example of indirect measurements (e.g. vegetation indices or surface temperature). Most of these products are originally not designed to measure the stock and flow of ecosystem services. However, if the relation between the measured variables and the ecosystem functions and processes are know, ES values can be derived. For example, erosion protection is strongly related with the presence, volume and type of vegetation, which can be derived from vegetation indices such as NDVI (Normalized Difference Vegetation Index).
The use of landcover or habitat maps for ES stock and flow assessments can be considered a form of indirect measures. The most common approach is to generate an average value of each ES per land cover type (e.g. the average value of biomass produced in Estonian coastal meadows is 3050 kg/ha of dry biomass). The ES stock or flow values are averaged from either scientific literature sources or fieldworks. These values can be further linked to landcover units in a map in order to make the analysis spatially explicit.
Indirect measurements are usually a more resource-efficient strategy to assess the provision of ES. Moreover, earth observation datasets are regularly updated, which allows to assess the rate of change in the stock and flow of ES.
220.127.116.11. Ecosystem services modelling
Models are simulations or representation of an ecological system. When direct and indirect data are unavailable, other ecological and socio-economic data and knowledge can be used as surrogate data to estimate the provision and demand of ecosystem services.
The advantage of using ES models is that the input data can be modified in order to simulate hypothetical scenarios of land management, landcover change, climate change, etc. in order to predict possible impacts on the provision of ES.
3.3.2. Socio-cultural methods
Socio-cultural methods generally aim at assessing human preferences for ecosystem services, leaving aside monetary valuations. Values and perceptions of both demand and supply of ecosystem services are commonly assessed and mapped through a wide array of methods based on eliciting social needs and preferences. It is important to make a clear distinction between socio-cultural methods and socio-cultural ecosystem services. Socio-cultural methods are used to quantify and map the three categories of ecosystem services: Provisioning, regulating and cultural. There are several methodologies available, here we highlight three: Preference assessment, PPGIS and time-use assessment.
Preference assessment: Preference assessments aim at assessing values, perceptions knowledge, supply, use and demand of ecosystem services through “traditional” socio-cultural data collection techniques: (ecosystem services) rankings, questionnaires, preference and rating assessments or free listing exercises.
Participatory Mapping and Assessment (PPGIS): PPGIS methodologies allow end users to utilize very basic GIS capabilities, usually through an online platform. In the context of ecosystem services, PPGIS allow to assess the spatial distribution of ecosystem services based on local knowledge, preferences or perceptions. PPGIS approaches are integrative and spatially explicit, therefore allowing for spatial comparisons between supply and demand. Trough PPGIS tools, users are commonly able to mark point or area in a map and answer a questionnaire about the perceived supply or demand of one or more ecosystem services.
Time-use assessment: Time use assessment utilize time as a proxy for assessing the value of certain ecosystem services by directly asking people how much time they would be willing to invest to change the quantity or quality of a given ecosystem service. Similarly to willingness to pay approaches, time-use assessments are based on hypothetical scenarios for willingness to invest time.
3.3.3. Economic methods
Economic methodologies for mapping and assessing ecosystem services aim at quantifying the welfare (in monetary terms) that society gains from the use of ecosystem services. The spatial variation of economic values can be assessed through mapping approaches. The economic valuation of ecosystem services is a very complex field and there are publications that deal specifically with this. For a deeper understanding of economic valuations, we recommend: Brander and Crossman (2017). Economic methods for the evaluation of ecosystem services support decision making processes in which several management, project or policy options are considered. Three economic methods have been selected to illustrate the wide collection of methods available: Cost-effectiveness analysis, cost-benefit analysis and multi-criteria analysis.
Cost-effectiveness analysis (CEA): Cost-effective analysis compares alternative options in terms of their costs. The different options considered aim at achieving one specific goal and all costs can be expressed in monetary terms. Cost-effective analysis identifies the option with the lowest cost. In the context of ecosystem service, CEA is a relatively limited approach, since it is often not the case that a single goal for ecosystem services provision can be set.
Cost-benefit analysis(CBA): CBA is often used to asses multiple planning and policy options in which all impacts can be quantified in monetary terms. CBA considers and compares all costs and benefits from the different options being assessed. This approach is applied in the ecosystem by estimating the costs and benefits that different planning and policy options have on the delivery of ecosystem services, but it requires a deep knowledge of ecosystem processes.
Multi-criteria analysis (MCA): MCA is commonly used when not all the costs and benefits of a certain option can be valued in monetary terms. The basic idea behind MCA is to allow the integration of different objectives (or criteria) without assigning monetary values to all of them. MCA is used to establish preferences between different options referencing to a common set of criteria established by a decision making body.
3.3.4. Expert-based quantification of ecosystem services
When other sources are lacking, expert knowledge can provide the information needed for an ES stock, flow and demand assessment. Moreover, when experts from multiple disciplines are engaged in the assessment, a deeper understanding will be gained about the complex interrelations of drivers, pressures, state, impacts and responses in the ES stock, flow and demand system.
In an expert based-assessment, a deliberative process among the experts leads to an agreement on the estimates of ES supply and demand. When biophysical or other forms of data are missing, expert assessment are an efficient way to obtain an approximation of ES values.
Expert-based quantifications are commonly used together with the lookup table approach for mapping ecosystem services (see section 3.5). The combination of these two techniques is a cost-efficient way to obtain reliable maps.
A common technique to quantify the provision of ES in the context of expert-based assessments is the use of relative scores: Experts are asked to value the provision of a certain ES in a relative scoring scale of e.g. 1 to 5.
3.4. Mapping ecosystem services
As explained in section 3.3, the indicators used to quantify ecosystem services vary in scale. Therefore, the mapping resolution at which ecosystem services can be mapped depends on the spatial scale of the biophysical models used to calculate the indicators and the spatial scale at which data is available (Maes et al., 2011).
Similarly, different ecosystem services, related to different biophysical processes, require specific thematic maps in order to precisely capture the spatially explicit character of ecosystem functions. For example, soil related services such as carbon storage in soils or nutrient retention will require a soil map. On the other hand, production-related services such as fodder or timber production will be best captured through a landcover map, a habitats map or a forest types map. In this regard, it is essential to identify what is the service providing unit (SPU) of an ecosystem services map. Burkhard et al. (2014) defines a service providing unit as “spatial units that are the source of an ecosystem service (Syrbe and Walz, 2012). Include the total collection of organisms and their traits required to deliver a given ecosystem service (Vandewalle et al., 2009) as well as abiotic ecosystem components (Syrbe and Walz, 2012). Commensurate with ecosystem service supply (Crossman et al., 2013)”. SPUs should be carefully chosen and should match the scale of their geobiophysical supply origin (Burkhard et al., 2014) in order to avoid spatial mismatches that would lead to misinterpretations and misleading results of the ecosystem services quantification.
Broadly, ES mapping approaches can be classified into 5 categories (Burkhard and Maes, 2017):
- Lookup table: Also known as matrix. Land cover classes are used as proxies for ES provision. Each land cover class is linked to an ES average value (this data is commonly obtained from statistical databases or scientific literature).
- Lookup table with expert-based estimates: Similarly to the lookup tables, landcover classes are linked to ES values that have been previously agreed by a panel of experts (see section 3.4.4).
- Causal relationships: ES are estimated spatially based on known relationships between ES and spatial information. For example, the amount of grass produced in a grassland can be estimated using yield statistics for different regions, soil fertility and slope.
- Extrapolations from primary data: Direct measurements or primary data are collected in field surveys and linked to spatially defined units. ES value are extrapolated from these.
- ES models: A combination of field data of ES, socio-economic data as well as information from literature and statistics can be structured in the form of complex models that predict the provision of ES under different scenarios. This models can be linked to spatial units in order to make spatially explicit predictions or elicit the demand of certain services.
Ecosystem services mapping is a complex process that requires data at a wide variety of scales. Therefore a flexible methodology is need to account for all possible biophysical models, data needs and mapping scales. In a tiered mapping approach (Fig. 3.2), each tier of level adds more mapping complexity, uses more detailed data and requires more expertise:
Tier 1 maps: It is the simplest of the three tiers. In tier one, land cover and landuse data are used to map ecosystem services supply and demand. LULC maps are often combined with vegetation and habitats maps. From these maps, inferences about the relative quantity of services are estimated.
Tier 2 maps: In tier 2, previous LULC and/or vegetation and habitats maps are linked to datasets that reflect the provision of ecosystem services. These datasets could be location-based information, scientific literature or statistics datasets. The linkage between maps and datasets allow for ecosystem services quantifications at different locations and scales. Tier 2 quantifications require basic GIS processing.
Tier 3 maps: The third and most detailed level of mapping involves modelling the biophysical processes responsible for the delivery of ecosystem services. Environmental biotic and abiotic variables are combined in models in order to predict the spatial distribution and quantity of ecosystem services. Tier 3 requires complex GIS processing and in-depth knowledge of the processes being modelled.
3.5. Assessing and mapping the demand
The demand is often an overlooked component of ecosystem service mapping and assessment processes. However, mapping the demand for ecosystem services should be a key aspect of the ecosystem services framework and several important points should be taken into account:
- The provision and demand of ecosystem services often occur at different locations. It is not uncommon that the beneficiaries of ecosystem services are located far away from the actual ecosystem services provision spots. Consequently, the demand for ecosystem services should be specifically quantified and mapped, and flows from supply to demand estimated. The spatial relations between the supply and the demand, as defined by Burkhard et al. (2014) are:
- In situ: Supply and demand happen at the same location.
- Omni-directional: A certain ecosystem service is produced in one location but benefits the surrounding landscape without a directional bias. It is the case of many regulation ecosystem services.
- Directional: There is a clear flow direction from the ecosystem service produced at a certain spot to the area where beneficiaries are located.
- Decoupled: The ecosystem service flows over long distances.
- Supply and demand of ecosystem services may occur at different spatial scales and spatial units responsible for supply and demand are often times not the same. The areas were ecosystem services are used are often not related to ecosystems or geobiopysical units. More commonly, areas where the use of ecosystem services is realized are urban areas and rural settlements.
- The indicators and/or methods used to quantify the supply of a certain ecosystem services are rarely the same methods used to quantify the demand of the same ecosystem service. In most cases, the demand cannot be measured directly; therefore proxies such as the density of population or density of housing are used. In numerous occasions, social methods (see section 3.4.2) are used to measure the demand of ecosystem services, by directly asking the users of the service.
Slides of the lecture “Methods for mapping and assessment of ecosystem services”
Brander, L.M., Crossman, N.D., 2017. Economic quantification. In Burkhard, B. and J. Maes (eds). Mapping Ecosystem Services. Pensoft Publishers Ltd, Sofia.
Burkhard, B., Kandziora, M., Hou, Y., Müller, F., 2014. Ecosystem service potentials, flows and demand–concepts for spatial localisation, indication and quantification. Landsc. Online 34, 1–32.
Burkhard, B. and J. Maes (eds), 2017. Mapping Ecosystem Services. Pensoft Publishers Ltd, Sofia
Crossman, N.D.; Burkhard, B.; Nedkov, S.; Willemen, L.; Petz, K.; Palomo, I.; Drakou, E.G.; Martín-Lopez, B.; McPhearson, T.; Boyanova, K.; Alkemade, R.; Egoh, B.; Dunbar, M. Maes, J., 2013. A blueprint for mapping and modelling ecosystem services. Ecosystem Services 4: 4-14.
Egoh, B., Drakou, E.G., Dunbar, M.B., Maes, J., Willemen, L., 2012. Indicators for mapping ecosystem services: a review. Report EUR 25456 EN. Publications Office of the European Union, Luxembourg
Kienast, F., Helfenstein, J., 2016. Modelling ecosystem services. In M. Potschin, R. Haines-Young, R. Fish, & R. K. Turner (Eds.), Routledge handbook of ecosystem services (pp. 144-156). Abingdon: Routledge.
Maes, J., Paracchini, M.L., Zulian, G., 2011. A European Assessment of the Provision of Ecosystem Services: Towards an Atlas of Ecosystem Services. Publications Office of the European Union, Luxembourg, doi:10.2788/63557, p. 81.
Maes, J., Teller, A., Erhard, M., Liquete, C., Braat, L., Berry, P., Egoh, B., Puydarrieux, P., Fiorina, C., Santos, F., Paracchini, M.L., Keune, H., Wittmer, H., Hauck, J., Fiala, I., Verburg, P.H., Condé, S., Schägner, J.P., San Miguel, J., Estreguil, C., Ostermann, O., Barredo, J.I., Pereira, H.M., Stott,A., Laporte,V., Meiner,A., Olah, B., Royo Gelabert, E., Spyropoulou, R., Petersen, J.E., Maguire, C., Zal, N., Achilleos, E., Rubin, A., Ledoux, L., Brown, C., Raes, C., Jacobs, S., Vandewalle, M., Connor, D., Bidoglio, G., 2013. Mapping and Assessment of Ecosystems and their Services. An Analytical Framework for Ecosystem Assessments Under Action 5 of the EU Biodiversity Strategy to 2020. Publications Office of the European Union, Luxembourg, 57 p
Müller, F., Burkhard B., 2012. The indicator side of ecosystem services. Ecosystem Services 1, 26-30.
Potschin, M., Haines-Young, R., 2016. Defining and measuring ecosystem services. In: Potschin, M., Haines-Young, R., Fish, R., Turner, R.K. (Eds.), Routledge Handbook of Ecosystem Services. Routledge, Taylor & Francis Group, London; New York, p. 2016.
Syrbe, R.-U.,Walz U., 2012. Spatial indicators for the assessment of ecosystem services: providing, benefiting and connecting areas and landscape metrics. Ecological Indicators 21, 80–88.
Vandewalle, M., Sykes, M.T., Harrison, P.A., Luck, G.W., Berry, P., Bugter, R., Dawson, T.P., Feld, C.K., Harrington, R., Haslett, J.R., Hering, D., Jones, K.B., Jongamn, R., Lavorel. S., 2009. Review paper on concepts of dynamic ecosystems and their services. The Rubicode Project Rationalising Biodiversity Conservation in Dynamic Ecosystems. http://www.rubicode.net/rubicode/RUBICODE_ Review_on_Ecosystem_Services.pdf (Date: 17.10.2013).
Wiggering, H., Müller, F. (Eds.), 2004. Umweltziele und Indikatoren. Springer, Berlin/Heidelberg/New York, p. 670.