3.1 Introduction

Ecosystem services are the benefits people obtain from ecosystems and are co-produced by the interactions between ecosystems and societies. Since the Millennium Ecosystem Assessment (MA 2005) governments have embedded ecosystem services and natural capital in explicit policy targets. Globally, for example, the Parties to the Convention on Biological Diversity (CBD; www.cbd.int) have committed to ‘enhancing the benefits to all from biodiversity and ecosystem services’. The CBD Aichi Target 14 is of particular relevance to ecosystem services: ‘By 2020, ecosystems that provide essential services, including services related to water, and contribute to health, livelihoods and well-being, are restored and safeguarded, taking into account the needs of women, indigenous and local communities, and the poor and vulnerable’. Beyond the conservation sector, interest in ecosystem services is increasingly aimed at the development of policies at national and global scales (Griggs et al. 2013). Regionally, the European Union Biodiversity Strategy to 2020, for example, aimed to halt the degradation of ecosystem services, and to map and assess the state of ecosystems and their services in their national territories by 2014 (Maes et al. 2016). This study also aimed to assess the economic value of such services, and promote the integration of these values into accounting and reporting systems at EU and national levels by 2020. Non-EU governments of nations such as Australia, Canada and Mexico are also incorporating ecosystem services and natural capital into national accounts.

At a national and sub-national scale, ecosystem services can be an effective tool for informing decisions about the use and management of the planet’s resources, especially when trade-offs and synergies need to be taken into account. Without this information, decisions that determine the fate of terrestrial, coastal, and marine systems and the benefits they provide, are made in the dark, with little understanding of the ecosystem services outcomes (benefits and costs) of any given decision or its consequences for the different stakeholders depending on these services.

While many observations and datasets are available to measure progress towards global, regional, and national goals for ecosystem services, and to ensure effective decision-making for sustainable human use of the planet’s resources (Egoh et al. 2012), their coverage is patchy, incomplete and inconsistent. The challenge is to find meaningful and robust indicators to quantify ecosystem services, measure changes in demand and supply and predict future scenarios. At present, most governments are not effectively measuring or monitoring ecosystem services. This chapter addresses the basic requirements for collecting information on ecosystem services.

3.2 Biodiversity and Ecosystem Services

Biodiversity is related to ecosystem services through a variety of mechanisms operating at different spatial scales (Fig. 3.1) (Mace et al. 2012). Biodiversity regulates the state, the rates and in many cases the stability of ecosystem processes fundamental to most ecosystem services (Cardinale et al. 2012). Components of biodiversity are also directly harvested to meet people’s material needs, and are also valued by societies for their non-tangible contributions to well-being, for example to psychological health, people’s identity and the asset it can be for future generations. Fundamentally, biodiversity provides the evolutionary building blocks of life on Earth and therefore provides important adaptive capacity through its continued ability to support desired ecosystem services and processes in the face of often rapidly changing selective pressures (Mace et al. 2014).

Fig. 3.1
figure 1

Biodiversity is linked to ecosystem services in three different ways: (i) as a regulator of the ecosystem functions that lead to the supply of provisioning, regulating or supporting services, (ii) as a provisioning service, (iii) as something that is appreciated in itself rather than for the benefits obtained from it. Selected examples are used to illustrate these linkages. Source Modified from Mace et al. (2012), Reyers et al. (2012)

Due to the complexity of the links between biodiversity and ecosystem services, as well as the important role played by other non-biophysical inputs into the goods and benefits we obtain from ecosystems (Díaz et al. 2015), monitoring biodiversity alone is not sufficient to understand the status and trends of the services it provides. In fact, monitoring annual changes in the state of ecosystems and determining trends in ecosystem services, can contribute to our understanding of changes in biodiversity and inform on the underlying dynamics of the complex interactions between societies and ecosystems.

3.3 Key Ecosystem Service Concepts

Societies are embedded within ecosystems, depending on and influencing the ecosystem services they produce. The characteristics of ecosystems, such as species composition, tree cover or growth conditions, modulate the type and magnitude of ecosystem services that can flow to societies. Management regimes, technologies, as well as tenure and access arrangements modulate the ways by which ecosystem services are produced and benefit societies. In other words, ecosystem services result from the interactions between ecosystems and societies, which together form a social-ecological system.

Four types of ecosystem services can be distinguished (MA 2005), though we focus only on three of them in this chapter. Provisioning services are the goods that can be extracted and consumed from ecosystems and are often valued in markets: for example, water, food, wood and biofuels. Regulating services are the benefits derived from ecosystem processes that modulate the conditions which we experience: such as the regulation of climate, soil fertility or floods. They seldom have markets, and must be valued indirectly. Cultural services are the real but not physical (‘intangible’) benefits that emerge from interactions between humans and ecosystems (Chan et al. 2012), for instance employment, sense of identity, spiritual value, aesthetic value and cognitive development. Some cultural services, such as recreation, do have markets, while others do not. The fourth category, which we do not elaborate on, is supporting services, the fundamental ecosystem processes such as photosynthesis, nutrient cycling and evolution, which permit the delivery of the first three categories, and thus find societal benefit through them.

In order to fully understand ecosystem services, we need to measure and monitor four different components: supply, delivery, contribution to well-being, and value (Tallis et al. 2012). Table 3.1 provides a detailed examination of each of these components across different categories of ecosystem services. The table includes a definition and some popular metrics or indicators used in the quantification and assessment of services. This list is not exhaustive since it does not cover all services or potential indicators, but rather presents a range of different types of services that have been found to be very relevant to societies.

Table 3.1 Examples of provisioning, regulating and cultural ecosystem services, including descriptions, drivers and potential indicators for each of the four components of the ecosystem service (supply, delivery, contribution to well-being and value)

Supply refers to the potential of a social-ecological system to generate a service, typically quantified as a flow (i.e., an amount per unit time). Ecosystem condition (e.g., intact or degraded, stressed or unstressed) and processes (e.g., primary productivity), as well as the way ecosystems are managed, are taken into account when determining supply. This is the component of ecosystem services that has been most commonly measured.

Delivery accounts for how much of the service is actually extracted (e.g., amount of timber harvested), used (e.g., area of avoided flood damage, area that is enjoyed by visitors), and delivered to societies (e.g., spatial location of those benefiting from flood regulation), and how societies have access to these services (e.g., laws rules, norms and restrictions that limit access to a service). Delivery thus depends on the links between ecosystem services supply and people’s location, activities and societal factors determining access to services.

Contribution to well-being accounts for the change in people’s well-being, which results from consuming, using, or having access to the service. Changes in living standards, nutrition status, mortality rates, social conflicts, security in the face of extreme environmental conditions, or happiness partially depend on the delivery of ecosystem services. This component of ecosystem services is the least understood and seldom quantified. One of the issues is that well-being typically has many components and many causes, so it hard to isolate the contributions of a particular service.

Value refers here to the relative importance society attributes to the service. The value of ecosystem services is often accounted in monetary terms, but other ways of establishing the socio-cultural value are potentially equally valid, and may be more appropriate than monetary valuation for some services. For instance, contributions to longevity or perceived quality of life need not be expressed in monetary terms. The monetary value of most provisioning services (e.g., timber) is provided by markets. Where freely-traded markets do not exists (for instance, this is frequently the case for water service), the value can be estimated through a variety of methods, such as the cost of delivering a substitute, or the marginal value addition of the service to other services which do have markets. Valuation approaches, based on willingness to pay, damage costs avoided, travel costs, or hedonic values, have been used to attribute economic value to many regulating and cultural services. Socio-cultural values of ecosystem services to an individual can be assessed through various valuation methods, such as through preference surveys, paired comparisons, and narrative or participatory methods. What is frequently reported is the aggregate societal value resulting from some combination of individual valuations.

These components of ecosystem services feed back into the way social-ecological systems are managed and governed. Supply allows for delivery which allows for contributions to well-being which, in turn, influences value. Ecosystem service contributions to well-being, shape the status of and vision for the well-being of individuals and societies, which directly influences the way formal and informal institutions are designed to modulate interactions with the environment. Value determines which services are fostered, and shape institutions and management interventions, aimed at modifying social-ecological conditions to promote the supply of the desired services at the cost of other services (Díaz et al. 2015).

3.4 Monitoring Ecosystem Services

Ecosystem services can be monitored at multiple spatial scales. For global observation systems, emphasizing the nation state as the focal unit allows for better tracking of progress towards national targets for ecosystem services. In addition, many key global policies, such as the Convention on Biological Diversity (CBD; www.cbd.int), the Sustainable Development Goals (https://sustainabledevelopment.un.org/), and the Commission on Climate Change and Development (www.ccdcommission.org) are governed by mutual agreement of participating nations, requiring monitoring of progress toward global targets. Monitoring, however, can also take place at the local scale, and data can then be aggregated up to the national and global scales, but this is not always a straightforward procedure (Scholes 2009). A multiple scale approach makes it possible for information from one spatial scale to be tested or refined using data produced at other scales. Such comprehensive monitoring at different spatial scales can include national statistics and remote sensing to cover national to global scales, as well as remote sensing and field-based assessments to cover local scales. Models can be developed at all spatial scales.

Different data sources are best suited to account for different components and spatial scales of ecosystem services (see Table 3.2). Supply is best characterised by data sources that consider the condition of social-ecological systems, for example, from remote sensing and models. Delivery is often based on societal characteristics and can be accounted for from national statistics, field-assessment and models. Contributions to well-being are documented in different ways (mostly field assessments, national statistics and census) and have seldom been explicitly incorporated into models. Economic value can be derived from markets, national statistics or from economic models. Sociocultural value can be obtained from field assessments of preferences, or from the analysis of cultural norms. Different types of value have been incorporated into models.

Table 3.2 Comparison between different ecosystem service data sources

3.5 National Statistics

Census data at national scales are readily available for several ecosystem services. In most cases the census has been conducted at a much more resolved scale (the census district, which may be as small as a neighbourhood). Sometimes such data is available for local analysis, subject to special procedures designed to protect the privacy of individual respondents. The United Nation’s Food and Agriculture Organisation publishes a global database (http://faostat.fao.org/) of the amount produced or extracted (delivery), traded, and the monetary value (value) of several ecosystem services, for example, total production of all commercial crops for countries or regions, export or import quantity of trade crops and their economic value per unit. Other databases, such as that of the World Bank (http://data.worldbank.org) report water withdrawals and water availability to people. Some of the services are monitored in most countries and updated annually (e.g., crops), while others are only available for a small subset of nation states and updated infrequently (~5 years; e.g., water withdrawal). While these statistics provide very relevant information for assessing provisioning ecosystem services, they imperfectly reflect their delivery and economic value. They cannot, for instance, inform on the supply of the services. They further inform only partially on the delivery of the services, as they can only account for the fraction of the food production that enters markets and national statistics. The stronger biases are for economic values, which are the product of markets and incentives, and do not necessarily account for the marginal contribution of ecosystems to food production through primary productivity, water for irrigation, soil fertility, pollination, or pest regulation, relative to those contributed by society. Also, these values do not include the negative impacts of agricultural intensification and expansion, nor that of industrial fisheries, on biodiversity conservation and the degradation of supporting and regulating ecosystem services. The societal costs of intensive agriculture or fisheries are not accounted for either.

Data accuracy in national statistics is quite variable and is dependent on national monitoring infrastructure (human and technical capacity), relative importance of informal activities (e.g., subsistence production or unreported extraction cannot be accounted for), and governmental policies on transparent reporting. Temporal data gaps are common for many countries and are often filled using a variety of techniques, including interpolation, models or expert judgement, which all have well-documented biases. In all cases, uncertainty analyses are needed to quantify and help improve reliability of existing data.

3.6 Remote Sensing

Remote sensing (see Chap. 8) consists of data collection ‘at a distance’: from sensors on the ground, in the water, on aircraft, or in space. Remote sensing of ecosystem services relies on hybrid methods, that use models to combine in situ information (collected either by humans or machines) with that collected at coarser spatial scales (e.g., climate, landform, social or economic variables).

Remote sensing has not been used directly to measure ecosystem services, yet in combination with other data sources it can contribute to the assessment of many ecosystem services (e.g., water quantity and quality, erosion prevention, moderation of extreme events; Horning et al. 2010). These data sources can either contribute to assessing the potential supply of ecosystem services or to assess the social-ecological drivers that influence the supply, delivery, contribution to well-being, and value of ecosystem services (Andrew et al. 2015).

Products from multiple frequencies within the range of visible and near-infrared bands contribute to vegetation indices, such as greenness measures like the Normalized Difference Vegetation Index (NDVI) that indicates plant vigour. Such information can be used as one of several data sources to assess crop delivery (through potential productivity of known plant/crop species), carbon stocks and carbon uptake, fisheries (through ocean productivity), water quality (through changes in water colour), and land use change (a driver). High-resolution data can inform on small-scale ecological features, such as individual trees. Information on roads, fields and habitat patches can be used to provide information on drivers of many ecosystem services. Products from radar devices provide high-resolution information for topography, vegetation and water cover, and potentially on the aboveground biomass. These can contribute to assessing land use change, crops, or water cover (superficial water bodies) over a targeted region. Products based on Laser Imaging Detection and Ranging (LIDAR) devices provide high resolution information on above-ground carbon stocks, water (water surface elevation, and in combination with bathymetry, the volume of freshwater bodies), and ecosystem structure, that can be used to model a range of provisioning, regulating and cultural services. High resolution images (with individual pixels of around 1 m2) are increasingly available from commercial satellites and can be used to refine information for particular locations. The cost is currently high, but may still be cost-effective if compared with manual mapping on the ground, and is being driven down by the advent of unmanned autonomous vehicles or ‘drones’ (e.g., see www.conservationdrones.org) equipped with cameras.

3.7 Field-Based Estimations

Field-based estimations contribute to local or site-based monitoring and assessment, as well as to validation of models and remotely sensed data products. Ultimately, field-based estimations are a principal source of new data on the supply, delivery, contributions to well-being and value for all services. Some services, such as the flow of water in rivers, are routinely monitored by in-field devices, and new technologies such as eddy covariance are extending the range of in situ observations of services such as carbon sequestration.

Conducting primary data collection can be costly, time consuming and technically specialised, and the methods and information from different data sources need to be standardized. Toolkits are emerging to deal with these issues, and promote standardized rapid assessments at the site scale. Such toolkits provide guidance on the steps to be followed, the kind of data to be gathered and the methods suggested to gather or model quantitative data at this scale that can then be used in an assessment under a range of contexts. Assessments incorporate local knowledge, basic local data collection and other data sources to create fine scale, locally-relevant assessments of multiple ecosystem services.

Two of these toolkits have been particularly useful (Table 3.3). The Toolkit for Ecosystem Service Site-based Assessments (TESSA; Peh et al. 2014) was developed to assist site-scale users with limited capacity and resources, to develop simple estimates of ecosystem services. The Natura toolkit was developed for assessing the socio-economic benefits associated with the ecosystem services of 200 conserved or protected sites in Europe (Kettunen et al. 2009).

Table 3.3 Examples of toolkits available to assess ecosystem services and their advantages and disadvantages

3.8 Community Monitoring of Ecosystem Services

Ecosystem services that are locally relevant can be monitored by local stakeholders, such as land owners and consumers (see Chap. 9 on Citizen Science). Several studies have shown that local communities without conventional scientific training have successfully collected accurate data on a wide range of ecosystem services such as forest carbon storage and sequestration, water quantity and quality, and their links to well-being (Hein et al. 2006; Dinerstein et al. 2013).

Involving communities in data generation enables year-round, low cost generation of local data (plot to landscape level) and wide spatial coverage. It provides information for local-level decision-making for ecosystem service management, and it can also generate employment, enthusiasm, and personal investment in ecosystem service based initiatives. Additionally, it can better incorporate traditional ecological knowledge and help maintain cultural heritage, identity, and values. Community involvement in monitoring can increase local interest and investment in the maintenance of ecosystems and the services they provide.

Information generated by locally-based monitoring systems, however, can be influenced by power struggles and incentives surrounding the monitored resource and validation mechanisms need to be implemented.

Numerous data collection and management tools have been developed in the last 5–10 years to facilitate gathering, storage, and sharing of data by communities.

3.9 Models

Numerical models, understood here as practical tools that predict how ecosystem services change through time and space, are increasingly being used to support decision-making. These models are often developed when data availability is scarce, when spatially explicit information is needed, and in order to assess trade-offs among services under alternative future management scenarios.

A wide variety of approaches have been used for building and applying such models. Five of the more commonly used modelling platforms are described here (Table 3.4).

Table 3.4 Comparative table of the ecosystem services models described in this chapter and their individual advantages and disadvantages
  • The Integrated Valuation of Environmental Services and Tradeoffs (InVEST) suite is a free and open-source software tool to help inform and improve natural resource management and investment decisions (Tallis et al. 2013).

  • The Lund-Potsdam-Jena managed Land Dynamic Global Vegetation and Water Balance Model (LPJmL; www.pik-potsdam.de/research/climate-impacts-and-vulnerabilities/models/lpjml) is a tool that was not specifically designed for ecosystem service assessment, but still allows deducing a number of ecosystem services consistently from the same process based model (Bondeau et al. 2007).

  • The ARtificial Intelligence for Ecosystem Services (ARIES; www.ariesonline.org) can be used to model supply, demand (delivery), flow (the link between the areas of supply and those of delivery), depletion (the balance between supply and delivery), and values (differential preferences among stakeholders) of ecosystem services (Bagstad et al. 2013b). A range of tools (www.ariesonline.org/resources/toolkit.html) and models for a range of case studies (www.ariesonline.org/resources.html) is available.

  • The Ecosystem Service Trade-off Analysis (ESTA) was initially developed to inform and evaluate the trade-off between biodiversity and fisheries objectives, and has been applied to an increasing number of case studies with a range of ecosystem services, including offshore wind and wave energy, aquaculture, and ecotourism (White et al. 2012).

  • The Multi-scale Integrated Models of Ecosystem Services (MIMES; www.ebmtools.org/mimes.html) platform is designed to address the magnitude, dynamics, and spatial patterns of ecosystem service values (Altman et al. 2014).

  • Co$ting Nature (www.policysupport.org/costingnature) is a web-based tool for natural capital accounting and analysing the ecosystem services provided by natural environments (i.e., nature’s benefits), identifying the beneficiaries of these services and assessing the impacts of human interventions (Mulligan 2015a).

  • WaterWorld (www.policysupport.org/waterworld) is a web-based tool can be used to understand the hydrological and water resources baseline and water risk factors associated with specific activities under current conditions and under scenarios for land use, land management and climate change (Mulligan 2015b).

3.10 Current Tools to Monitor Ecosystem Services

Ecosystem services can be monitored and assessed at different spatial scales using readily available data sources (Table 3.5). However clear gaps exist, especially when one considers all four components requiring data per ecosystem service (see Table 3.6). We explore progress and gaps per ecosystem service category below.

Table 3.5 Data sources for ecosystem services
Table 3.6 Ecosystem service data sources for different ecosystem services components

Mismatches can occur between data sources and data needs. Some data sources, such as LPJmL models or the older remote sensing data, are only available at low spatial resolution (50 km2 grid cells in the case of LPJmL) and might not be suitable for assessments at landscapes scales. Similarly, assessments of changes in services within very short time frames are incompatible with some data sources that are only available on a yearly basis, as is the case of national statistics, or those that are modelled from data for which data sources are not updated regularly, as is the case of governmental land use and land cover maps in Mexico. The converse situation can also be true: changes in soil carbon or soil fertility within the same land cover type through time could be estimated from repeated remote sensed data, but changes would not be observed given the long time frame over which the processes that regulate them operate.

The data needed for ecosystem service estimation is often the flow of service rather than the particular conditions of the service in one point in time. This is the case of water flowing from a river, or the amount of carbon being taken up by vegetation. The most commonly found approach is for rates to be estimated from differences in the magnitude of the stock which provides or receives the service between two selected dates, as is the case of carbon uptake, most commonly estimated from changes in carbon stocks. Actual flows of ecosystem services, such as in the case of water, can be assessed by some of the models such as ARIES, or by in situ flow measuring devices.

3.11 Provisioning Services

Most provisioning services are already observed at national and local scales in most parts of the world using one or more of the data sources above. National statistics are available (at least partially) for many provisioning services, but are typically blind to subsistence (‘informal’, family consumption, not traded in monitored markets) or illegal operations that can contribute to large proportions of delivery in some countries. Remote sensing data are available for services related to vegetation primary productivity, biomass harvest and water quantity. Field estimations are available for provisioning services (from e.g., TESSA and Natura). Models are available for most provisioning services, from at least one of the four platforms described above.

Observations of supply, that largely depend on biophysical conditions are only available for a few provisioning services. Instead, delivery data sources are commonly reported for services associated with commonly used goods, although only those that are accounted for in statistics. As many provisioning services are commercialised in markets, economic (especially monetary) values are also readily available, but such values do not reflect all the contributions of the ecosystem to these services, nor the consequences. Data on the contributions to well-being are largely missing or in development for most services.

Information on the balance between the demand of the services and the supply, or other estimators of the long-term ability of the ecosystem to sustain the supply of these services are not currently available for most provisioning services.

3.12 Regulating Services

Data on regulating services is increasingly available from national statistics or from remote sensing in conjunction with models, particularly for carbon stocks and uptake (climate regulation). The emphasis has been put on carbon stocks and carbon uptake through primary productivity, which is relatively easily measured and quite relevant to climate change mitigation, while the links to actual carbon dynamics and climate processes is largely absent. Models of regulating services associated with hydrological processes (water quality, erosion regulation), those on the impacts of extreme meteorological events (flood and coastal regulation), as well as those for pest regulation and pollination are increasingly available. Today models are available for most regulating services and most of these models have been developed at landscape and regional scales, but seldom at national scales. Field estimations are available for services (most of which are available from TESSA or Natura, and from a plethora of approaches).

Both supply and delivery of regulating services are accounted for in most models. Data and models for contributions to well-being are absent or in development. Economic values are largely related to avoided costs or marginal contributions to economic activities from regulating services.

Given that regulating services depend on multiple social-ecological processes operating at several spatial and temporal scales, data, models and field estimations of regulating services are necessarily a simplification and, in some cases, they may be an oversimplification which is more misleading than useful.

Box 3.1. The Demand for Ecosystem Services at Drinking Water Treatment Facilities in Barcelona

Engagement with drinking water managers in Barcelona, Spain allowed for the identification of ecosystem services relevant for decision-makers. Discussions revealed that treatment costs were particularly sensitive to three water quality parameters: stream temperature, ammonium and conductivity. In particular, high stream temperature increased water treatment costs because of the water treatment technology used and the high concentration of sterilisation products during warm summer months (Valero and Arbós 2010). Understanding the demand for reduced stream temperatures by water treatment managers allowed for the development of a targeted research program focusing on ecosystem structures that would reduce thermal heating in the Llobregat River. It was found that the restoration of riparian forests upstream would be able to recover ecosystem processes, reduce stream temperature in the summer and therefore reduce water treatment costs. After modelling multiple restoration scenarios, nearly half of the investment in riparian river restoration was estimated to be recovered in a 20 year period through a reduction in water treatment costs (Honey-Rosés et al. 2013). Understanding the demand for reduced stream temperatures by water treatment managers allowed for the development of a targeted research program focusing on ecosystem structures that would reduce thermal heating in the Llobregat River.

3.13 Cultural Services

Cultural services present a challenge when it comes to observation and assessment because some of them are not easily disentangled from other ecosystem services, such as provisioning services. For instance many important cultural services are co-produced by the same ecosystem components and human activities that produce material objects for consumption (Chan et al. 2012), such as agricultural landscapes or harvested forests. The different cultural services are highly intertwined, and unlike with provisioning or regulating services, it is not possible to clearly delineate the different components of the services. Cultural services are highly context dependent and thus information on these is often only available and relevant at local scales. This is not true for all cultural services: some are well-defined, discrete and routinely monitored, such as the use of national parks, or the income from nature-based tourism and recreation.

Readily available sources of information on cultural services are very wide ranging. These include local assessments of cultural preferences (for aesthetic views; Bagstad et al. 2013c) (can be obtained from the above toolkits), and databases on use of particular areas or ecosystems for ecotourism at national scales (governmental database). Further sources of information on cultural services are embedded into local artistic expression (e.g., poetry, music) or in social norms that articulate a value or impact of nature on the human condition.

3.14 Observing Multiple Ecosystem Services

Historically, ecosystem management has often focused on delivery of a single service from that ecosystem (often a provisioning service, such as timber or grazing) without recognition that the same ecosystem produces multiple, often interacting services which are also affected by management interventions. This often leads to trade-offs (where one service decreases while the other increases), but can also lead to synergies (where increasing the supply of one services also increases the supply of another). Moving observation systems beyond single services to the full bundle of services (a set of services that tend to co-occur in space or time), to quantify and reflect the synergies (positive interactions) and trade-offs (negative interactions) is a major challenge for current research efforts. Also, an understanding of the interactions among stakeholders that have differential preferences for the traded-off services is needed.

The identification of bundles of services that arise under particular biophysical, management, and societal conditions is particularly relevant. Data needed for these assessments is hindered by the reduced replicability of the same measurements across different social-ecological conditions. It is seldom that they supply exactly the same sets of provisioning, regulating and cultural services, at the same spatial and temporal scale, and measuring the same components (e.g., supply or value). While still patchy, such datasets have been increasingly available in the past few years. Comparisons across studies are nevertheless faced with the lack of interoperability among them.

Additional observations of biodiversity (see other chapters) and multiple ecosystem services at different spatial scales will contribute to a better understanding of their inter-linkages, patterns of interactions across scales and time, and common trade-offs and synergies.

3.15 Using Scenarios in Modelling to Predict Future Ecosystem Services

Scenarios are stories about plausible futures, with the power to capture public attention and inform more sustainable decisions (Henrichs et al. 2010). They can help communicate the outcomes of different choices for societies and ecosystems while at the same time involving stakeholders in a powerful learning process. It is important to consider the explicit goals for the use of scenarios in determining which type of scenario will best address those goals and reach their intended audience. Three main uses of scenarios include: (1) assessing the impact of decisions under consideration, (2) exploring hypothetical but plausible futures, and (3) building consensus around a shared vision for the future (e.g., see IPBES 2016).

Certain characteristics can make scenarios more effective. Scenarios that are relevant to the decision context or stakeholder interests will align with the problems and questions of interest to stakeholders. To be legitimate, the scenario development process should include diverse stakeholder views and beliefs. To be credible, scenario storylines should be developed using scientifically robust methods. To be plausible, scenarios should tell coherent stories that could conceivably happen. Finally, to tell a compelling story, scenarios should be distinct enough from one another that they show contrasting ecosystem service impacts. Iteration of scenarios can greatly enhance many of these characteristics, as they are refined over time to incorporate stakeholder feedback, as well as emerging knowledge, trends and issues.

Translating scenarios to decision-support tools requires that storylines be made spatially-explicit, with each scenario corresponding to a map of land cover, or coastal or marine habitats and uses that feed into the biophysical and/or economic models underlying ecosystem service assessment. Converting scenario storylines into maps can be accomplished by asking stakeholders to simply draw maps for each scenario; more analytical methods of forecasting where change is most likely to occur on the landscape or seascape are based on past trends; rule-based approaches define which areas are likely to be most suitable for particular uses or activities. Models of future supply, delivery, value and benefit of ecosystem services into alternative scenarios are increasingly being developed.

All the modelling platforms described above may be used to predict ecosystem services under different future scenarios for land/sea use and management patterns. Different models have been built to be differentially sensitive to alternative future issues. For instance, the LPJmL, is highly sensitive to climate change, which is particularly helpful when looking for mid- to long-term effects.

3.16 Linking Ecosystem Service Observations to Decision-Making

Monitoring for ecosystem services to support decision-making is greatly enhanced with early involvement of the actual stakeholders involved in the decisions. One key advantage to examining ecosystem services with a stakeholder driven agenda includes the easy identification of key services recognised and preferred by societies, as well as the identification of indicators that are most meaningful to them. Stakeholders can also participate in community-based or citizen science-based monitoring of ecosystem services. Successfully integrating decision-makers in the assessment and valuation of services also allows for speedier adoption of the ecosystem services framework in practice, and the use of ecosystem service data into actual decision-making.

Emphasis has increasingly been put on the use of ecosystem service indicators towards agreed upon policy goals. That is the case of indicators that can inform on progress towards the Aichi Targets and more recently progress towards the Sustainable Development Goals. The challenge is to identify those indicators that are most relevant to measuring progress towards the goal, while at the same time being supported by actually available data, conceptual understanding and credibility.

Monitoring for ecosystem services at local to national and global scales needs to take into account how preferences and ecosystem services can change in space and time. Services that are most relevant at national to global scales could be monitored systematically, while locally relevant services could be assessed within particular locations.

Box 3.2. Monitoring Ecosystem Services for Coastal Planning in Belize

The coast of Belize includes hundreds of kilometres of mangrove forests, extensive seagrass beds, and the largest unbroken reef in the Western Hemisphere. 800,000 tourists visit the area for its renowned snorkelling and diving sites. Tourism, as well as commercial, recreational, and subsistence fisheries, contribute to income and livelihoods, but at the same time threaten the very ecosystems that make these activities possible. Efforts to put the Belize Barrier Reef on the United Nations Educational, Scientific and Cultural Organization’s list of World Heritage Sites in Danger and the creation of a visionary legislation in 1998 calling for cross-sector, ecosystem-based management of coastal and marine ecosystems were insufficient to halt degradation. In 2010 The Natural Capital Project (www.naturalcapitalproject.org) partnered with the Coastal Management Authority and Institute to use ecosystem-service approaches and models to design a spatial plan (Arkema et al. 2015). Interactions with a range of stakeholders and government agencies led to the identification of different categories of human activities, a zoning scheme, and three alternative future scenarios. The supply and economic value of lobster fisheries, tourism, coastal protection and habitat (to support fisheries) were modelled for current and future scenarios using InVEST. Data sources included: (i) field assessments of lobster catch and revenue; (ii) high resolution land use cover maps developed from remote sensed data, (iii) model of lobster migration, (iv) current visitation data obtained from social media (e.g., flickr). Risk under alternative scenarios for individual services as well as trade-offs among services across zones were assessed using additional spatial data on human activities and habitats, as well as information from the peer- reviewed and grey literature on the expected impacts of human activities on the services and the habitats. The most desirable future scenario was identified and further refined to increase expected delivery of almost all services in all regions into 2025. The results from this future scenario were incorporated into the Coastal Zone Management plan for Belize in 2012. It was refined through further stakeholder involvement and expert review during 2013 and led to changes in national legislation such as the creation of marine reserves and the revocation of offshore drilling contracts issued earlier by the government of Belize.

3.17 Creating a Network for Observing and Managing Ecosystem Services

The ultimate goal of many efforts to monitor ecosystem services is to inform decision-makers and policy to ensure the long-term supply of services and the flow of benefits to societies. While progress has been made on the quantification and mapping of services, less attention has been given to the needs of decision-makers and resource users from local to global scales. Meaningful engagement with resource users and policy makers should occur early, explicitly and formally when monitoring services (Menzel and Teng 2010).

A network for monitoring ecosystem services is necessary to synergise work done by multiple partners, taking advantage of others’ insights, increasing consistency, and reducing duplication of efforts. Creating such a network for monitoring ecosystem services at local to global scales will require significant effort from stakeholders from the research, policy and practice communities across the globe. National monitoring systems could create mechanisms by which local stakeholders can provide input and feed into the national system. City and regional governments may help facilitate the engagement with local stakeholders, and help assess the status of services at local scales. Stakeholder participation in monitoring activities will vary widely depending on many factors including local relevance of the services they are monitoring, and whether incentives are provided.

Local scale monitoring could dovetail into existing ecosystem services research which may have very different objectives but could contribute to an observation network. Examples of such on-going efforts include: the already existing networks associated with ARIES, and MIMES the Ecosystem Service Partnership (www.es-partnership.org/esp), the International Long-Term Ecological Research Network (www.ilternet.edu), the Natural Capital Project (www.naturalcapitalproject.org), the Program for Ecosystem Change and Society (PECS; www.pecs-science.org), the Sub-Global Assessment Network (www.unep-wcmc.org/sga-network_770.html), the Tropical Ecology Assessment and Monitoring Network (www.teamnetwork.org), the ESCom Scotland (http://escomscotland.wordpress.com/) and Vital Signs (http://vitalsigns.org/).

One major challenge to date is that multi-scale cross-site comparisons are only possible if comparable approaches and indicators are used. To date a wide diversity of approaches and indicators complicate such comparisons. Great emphasis has been given over the last decade to the development of new metrics, tools and approaches, which has fostered creative solutions. Yet, standard procedures will eventually need to be identified and practical examples be provided to operationalise the ecosystem services concept (e.g., OPERAs; www.operas-project.eu/).

Efforts through the Group on Earth Observations Biodiversity Observation Network (GEO BON; www.geobon.org), to further develop and communicate standards and protocols for the collection of new ecosystem services observations to enhance comparability across scales and data sources, are on-going. Ecosystem Service tools are being incorporated into GEO BON developed toolkits, namely BON-in-a-Box.

Automated, remotely sensed Earth observations will increasingly be used in the future to assess ecosystem services as well as the drivers that modify their supply and delivery. Changes in environmental and socio-economic features are more available than ever with the new sensors, such as those in the Sentinel fleet. The critical issue is integration of the data in ways that make it readily usable for ecosystem service assessments (Cord et al. 2015).

3.18 Monitoring to Support Policy Design

Ecosystem services monitoring can be directly linked to on-going assessments that support policy design. Timely information from monitoring ecosystem services can be useful to the Intergovernmental Platform on Biodiversity and Ecosystem Services (IPBES; www.ipbes.net) that aims to strengthen the science policy interface for biodiversity and ecosystem services for the conservation and sustainable use of biodiversity, long-term human well-being and sustainable development. IPBES is aiming to establish strategic partnerships, such as with monitoring programmes, to assist in the delivery of its work programme.

Similarly, National governments are also signatories to Multilateral Environmental Agreements. In most cases (for instance the CBD), these rely on technical and scientific bodies to assess progress towards implementation of agreed decisions. National progress reports and assessment of needs towards achieving targets rely on monitoring ecosystem services.

Agreements and commitments across different scales (national to global) on biodiversity and ecosystem services would benefit greatly from the extension and linking of various observing networks, which can promote the collection, access, packaging and communication of data. This often will require engagement with existing mechanisms such as the assessments to be performed by IPBES, CBD and individual nations.

3.19 Conclusions

Monitoring ecosystem services is vital for informing policy (or decision-making) to protect human well-being and the natural systems upon which it relies at different scales. While ecosystem services are linked to biodiversity, the social factors involved in their supply, delivery and value to human well-being implies that they cannot be predicted from biodiversity monitoring initiatives alone. Here we emphasise that monitoring systems for ecosystem services must take into account provisioning, regulating and cultural services as well as their components of supply, delivery, contribution to well-being and value. A wide variety of data sources is available and relevant to ecosystem services monitoring, including national statistics, field-based assessments, remote sensing and models. Their elaboration will help ensure monitoring at relevant (and where necessary multiple) scales of interest.

Outputs from monitoring a range of ecosystem services and their components at different spatial scales can actively support decision-making. Analyses of multiple services and biodiversity can inform decision-makers such as land managers as to trade-offs and synergies among them. Modelling and exploring future scenarios of ecosystem services can then clarify the impacts of alternative policies on such trade-offs and synergies.

Monitoring our life support systems and using this information in decision-making across all scales will be central to our endeavours to transform to more sustainable and equitable futures.