Potentials of ecosystem service accounting at multiple scales
Zurlini G.*, Jones K.B.^, LiB.-L.**, Petrosillo I*
*Landscape Ecology Laboratory, Dept. of Biological and Environmental Sciences and Technologies, University of Salento, Lecce, Italy.
^US Geological Survey, Reston, Virginia, USA.
**Ecological Complexity and Modeling Laboratory, Dept. of Botany and Plant Sciences, University of California at Riverside, Riverside, CA, USA
Why valuating ecosystem services at multiple scales?
To acknowledge the value of natural capital, one of society’s important assets, a better understanding of the patterns and spatial scales at which ecosystem services operate is essential to developing ecosystem services valuation (ESV) to support landscape-level conservation and land management plans.
Ecosystem services (ESs) are benefits that can be provided by human-modified and natural systems (Costanza et al. 1997; MEA 2003). The structure and function of such systems represent a stock of natural capital providing flows of goods (such as food, timber, forage, and non-traditional forest products) and services (such as soil formation and nutrient/carbon storage, erosion control, water filtration, pollution assimilation, and recreation) that are valued by society. These goods and services constitute a large share of our social and economic welfare, and are vital to the Earth’s life-support system (Costanza et al. 1997) (Table 1). Biodiversity is a critical part in providing ecosystem services in that most are due to the component populations, species, functional groups (guilds), food webs, habitat types or mosaics of habitats and land uses that collectively produce them, i.e., the ecosystem service providers (the ESPs) (Kremen 2005, Feld et al. 2009).
One of the greatest challenges facing humanity involves the understanding of distinct scales of environmental change and human response (Folke et al. 2005), and how the interplay of those scales has evolved historically, which is of paramount importance to get a better insight into how specific geographic areas are interrelated (Costanza et al. 2007).A clear understanding of scale can lead to development of more effective environmental policies and management. It can tell the local land owners and jurisdictions whether or not they can protect and enhance ESs locally (through their actions alone) or whether they must rely on regional, national, or global environmental policies to achieve and sustain certain levels of ESs. For example, reduction of hypoxic or anoxic conditions in estuaries and near-shore marine environments requires reduction of nutrient inputs coming off of entire river basins. As such, a comprehensive understanding of scale relations across entire river basins is needed to improve and maintain estuarine and near-shore habitats.
Human activities have become so extensive that they have increased the interconnectivity of our global social-ecological system. This increased interconnectivity has resulted in a global scale alteration of some of our most important ESs (Cronin 1996; Vitousek et al. 1997). Resulting changes in ESs are caused by multiple interacting direct drivers (MEA 2005) such as land cover change, global changes in atmosphere and climate from industrial emissions, water use and availability due to irrigation, alien invasive species expansion, and local changes in species populations due to harvesting, habitat destruction and landscape degradation. Direct drivers operate at multiple scales and, in turn are controlled by indirect drivers (e.g., economic, demographic, or cultural changes) (MEA 2003) operating more diffusely, by altering one or more direct drivers. An example of how climate variabilty/change is magnified by impervious surface created through local or county level land use change (e.g., transition to urban and developed from agricultural lands) is given by Jennings and Jarnagin (2002). Land-cover change has been identified as one of the most important drivers of change in ecosystems and their associated services, representing a primary human effect on natural systems and underlying fragmentation and habitat loss, which are the greatest threats to biodiversity (Vitousek et al. 1997; MEA 2005).
At the landscape scale, the dynamic spatial configuration resulting from human appropriation and management of regional landscapes can have a variety of ecological effects over a wide range of spatial scales. A direct effect is the alteration of ecological processes at local scales through the modification of land cover. For example, converting forest to agriculture land cover alters soil biophysical and chemical properties and associated animal and microbial communities, and agricultural practices such as crop rotation alter the frequency of these disturbances. The spatial configurations of land cover in a region also affect ecological patterns and processes. This is one of the fundamental underlying concepts coming out of the field of landscape ecology that is spatial distribution and pattern affect important ecological flows (e.g., nutrients/materials, biota, energy, water) that sustain certain levels of ESs and their associated benefits. New land cover types can be juxtaposed and shifted within increasingly fragmented remnant native land cover types, and changes in the structure of the landscape can alter (in a negative way) nutrient transport and transformation (e.g. Peterjohn and Correll, 1984), species persistence and biodiversity (e.g. Tilman et al., 2002; Benton et al., 2003), and nurture invasive species (e.g. Fox and Fox, 1986; With, 2004).
Based on the last decade of research, we are beginning to understand the complex ways in which humans have affected and have been affected by natural systems of the Earth (Steffen et al. 2004, Diamond 2005, MEA 2005). Much work has been oriented to describe and categorize ESs, identify methods for economic valuation, mapping the supply and demand for services, assessing threats, and estimating economic values (Costanza et al. 1997; Daily 1997; MEA 2003; MEA 2005).
There is a growing demand for answers to questions like: How much of a watershed’s (catchment) area must be forest and where are the most important areas to maintain or restore in order to provide clean water for downstream communities? How should patches of different land uses/land covers be distributed within a rural landscape to provide flood regulation or pollination and pest control services for crops? Up to what distances might adjacent land uses affect the capacity of natural habitat to provide pest control and pollination services? Answers to these questions will help determine how different land uses and set-asides (protected designations) should be spatially distributed in the landscape in order to protect and manage ESs.
Ecosystem services valuation (ESV) is the process of assessing the contributions of ESs to certain sustainable scales, fair distribution, and efficient allocation of space and resources(Liu et al. 2010),like having an appropriate mix of land-use types across scales such that ESs can be maintained.There is also the need to move from general estimates provided by Costanza et al. (1997) to more specific statements at regional and local scales. To effectively manage changes in the flow of goods and service we need to incorporate ESV into resource management decisions. However, moving from general statements about the tremendous benefits nature provides to people to credible, quantitative estimates of ESs has proven difficult (MEA 2005).
Among the crucial issues for obtaining reliable, quantitative estimates of ESs are those related to scales and interaction between services. Goods and services that humans obtain from ecosystems may span different scales in space and time, and can interact with one another in complex, often unpredictable ways. Improvements in the understanding of the patterns at multiple spatial and temporal scales at which ecosystem services operate are likely to provide more realistic ES values and also to improve ecosystem-based management practices.
Scales of operation of the services
Most ESs are broadly classified as operating on local, regional, global or multiple scales, and different providers of the same ES may operate across a range of spatial and temporal scales (Costanza, 2008; Ricketts et al., 2008). Services can be localized, e.g., fruit from a single tree, or derived from a relatively large area, e.g., flood control by wetlands or climate regulation through carbon sequestration (Costanza et al. 1997; Daily 1997; Costanza 2008). ES providers are landscape- or habitat-wide or ecological community attributes, and can often be characterized by the component populations, species, functional groups (guilds), food webs or habitat types that collectively produce the services (Kremen 2005) (Table 2). They can operate at different scales. So, for example, pollinators or native predators providing pest control on crops generally operate at a local scale, whereas forests contribute to climate regulation at local (shading), regional (rainfall patterns and albedo) and global (carbon sequestration) scales.
Costanza (2008) has recently classified the spatial characteristics of 17 ESs listed in Costanza et al. (1997) based not only on the scale of operation of the service, but also on the spatial proximity of service delivery to human beneficiaries (Table 3). Services like carbon sequestration, which is an intermediate input to climate regulation, is classified as global non-proximal. This is because the atmosphere is generally well-mixed and removing carbon dioxide (or other greenhouse gases) at any location is equivalent to removing it anywhere else. However, at local or even regional scale, in certain areas and during certain periods of the year, conditions of stratified thermal inversion may occur in the lower atmospheric layers resulting in respiratory distress and infections in people.
Local proximal services, on the other hand, are dependent on the spatial proximity of the ecosystem to the human beneficiaries. Use values will be determined by local patterns of land use, human population settlements and proximity to beneficiaries. The recreational function of a landscape or ecosystem, for example, is not only defined by the land cover of a specific location (e.g. natural area) but depends also on accessibility properties (e.g. distance to roads) and the characteristics of the surrounding landscape. Also pollination requires that the ecosystem with pollinators be proximal to the land being pollinated.
Directional flow-related services are dependent on the flow from upstream to down-stream as is the case for water supply and water regulation. This kind of service is the basis for ecosystem service payments established in 1996 in Costa Rica to induce landowners to provide ecosystem services. In this case, downstream users pay upstream landowners to maintain forest cover on their land to benefit from protection of upstream ecosystem services (MEA 2005). Other examples of this kind of service can be found where, like in Italy, the economic value of non-marketed services provided by forests for watershed protection and slope stability against landslides, can be very high (MEA 2005).
The effects of land-use intensity on local biodiversity and ecological functioning depend on spatial scales much larger than a single field or land use/land cover. This demands a landscape perspective, which takes into account the spatial arrangement of surrounding land-use types at multiple scales. To be characterized adequately, pattern–process relationships must be assessed at the multiple scales (Turner et al. 2001, Levin, 1992) relevant to the inherent structure (or rate) of the system (or process) being studied (Holling 1992), or the scale of perception of the organism(s) (Wiens 1989; Wiens and Milne 1989) providing the service (ESPs). In this respect, spatially explicit estimates of ES efficiency across landscapes at multiple scales that might inform land-use and management decisions are still lacking (Balmford et al. 2002; MEA 2005) (Table 4).
Non-linearity
The valuation process of ecosystem services (ESV) to more accurately represent their value needs to incorporate the non-linear properties of these services over space and time (Kock et al. 2009). In many field settings, like rural landscapes, there is a nested set of cycles, each occurring over its own range of scales (Holling et al. 1996). Some cycles occur annually, some take around a decade, and still others may take a century. Photosynthetic activity, for example, is a fundamental process necessary for the production of other ecosystem services like carbon fixation, oxygen and primary production. Photosynthetic activity can be highly variable over time and space, but it is often assumed to change linearly, i.e. at a steady, unvarying rate for each habitat considered. When it is measured by green index (NDVI) from satellite data it shows rather distinct and characteristic seasonal inter-annual periodicities for different broad land-use/land-cover categories corroborating well-known vegetation changes given by seasonal variations in climate and water regime as well agricultural practices. For example, in Apulia (south Italy) mean NDVI paths of land-use/land-cover show the greatest separation during the hot and dry summers owing to drought of semi-arid grasslands, and water regulation of forests (Zaccarelli et al. 2008).
Another example is coastal protection provided by wave attenuation by some seagrasses that may be at its maximum during summer, when plants are reproducing, at medium levels in spring and fall, and non-existent during winter, when density and biomass are low (Kock et al 2009). Yet, in the Mediterranean, sandy beach erosion can be still mitigated during winter by the stranding of seagrass leaves (Müller et al. 2008).
Non-linearity may occur also because of the interaction between services.
Synergism and trade-off among services
ESs may interact with one another in complex and unpredictable ways, and knowledge of the interactions among ESs is necessary for making sound decisions about how society manages the services. As to directional flow-related services, impounding streams for hydroelectric power, for example, may have negative consequences on downstream food provisioning by fisheries. In other words, a synergism occurs when ESs interact with one another in a multiplicative or exponential way. Synergistic interactions can have positive and negative effects, and pose a major challenge to the management of ESs because the strength and direction of such interactions remains virtually unknown (Sala et al. 2005).
Trade-offs, in contrast, occur when the provision of one ES is reduced as a consequence of increased use of another ES. Agricultural production shows an inverse relationship with water quality and quantity, as we increase agricultural production, the quality of water and the quantity available tend to decrease (MEA 2005). Thus, use of nutrients and pesticides to increase agricultural production can lead to critical declines in water quality that can often propagate down-stream.Trade-offs seem inevitable in many circumstances and will be critical for determining the outcome of environmental decisions. In some cases, a trade-off may be the consequence of an explicit choice; but in others, trade-offs arise without premeditation or awareness that they are taking place. Managers must clearly identify temporal trade-offs to allow policy-makers to understand the long-term effects of preferring one ecosystem service over another. This is crucial for any planning activity.Decisions are made to maintain provisioning services in the present, often at the expense of provisioning services in the future, as many managers are rewarded for short-term success (MEA 2005). This is one of the most fundamental temporal cross-scale challenge we are facing.
Ecosystem services in social-ecological systems
Humans may interact with ecosystems at various scales as individuals or as representatives of organizations responding to environmental changes (e.g., climate, drought, desertification) through multiple pathways. Their responses may in turn alter feedbacks between climate, ecological and social systems, producing a complex web of multidirectional connections in time and space (Costanza et al. 2007). Highly interlinked and connected systems transfershocks through the system faster and more completely than systemsthat are more modular and disconnected.The most important consequence for our globally interconnected social-ecological system is that the dynamics of our economic system(s) are no longer determined by economics alone, but by the dynamics of the total environmental (economic and ecological) system (Limburg et al. 2002).
Anthropogenic disturbances such as changes in land use are determined by the social component of social-ecological landscapes. Such component is made by groups of people, organized in a hierarchy at different levels (e.g., household, village, county, province, region, and nation), that result in a distinct set of spatial patterns and scales influencing fundamental ecological processes (e.g., flows of materials/nutrients, water, biota, and energy). These patterns and resulting processes formed by humans in turn affect the quality and amount of ESs that a particular socio-ecological landscape can provide. Any given land use system in the hierarchy is likely to overlap multiple ownership and jurisdictional boundaries. Within this “panarchy” (Gunderson and Holling, 2002), the participants can have differing views as to which set of ESs is desirable at each level of the panarchy. For example, scale determines different values of wetland services in the Netherlandsfor stakeholders atdifferent jurisdictional levels (Hein et al. 2006). In this example, at the municipal scale, interests refer to recreation,reed cutting and fisheries, whereas at the provincial scale, main concerns are recreation, but also nature conservation. At the national level, nature conservation is by far the mostimportant service.
Because of historical land-use legacies, decision hierarchies of social systems can often be intertwined with the natural hierarchies, scales and frequencies that may emerge at the ecosystem or landscape level (Gunderson and Holling 2002). As humans act as a keystone species (sensu O’Neill and Kahn, 2000), the characteristic scales of particularphenomena like anthropogenic changes are deemed to entrain andconstrain ecological processes that produce services, and to berelated to the scales of human interactions with the biophysicalenvironment (Holling, 1992). If the patterns or scales of humanland use change, then the structure anddynamics of the system as a wholecan change accordingly, leading to transitions between alternativephases,when the integral structure of the systems is changed (Kay, 2000; Li, 2000; Li 2002). In other words, people can structure landscapes leading to a set of different ecological flows that result in different levels (and types) of ESs. Alternatively, natural biophysical conditions can structure what humans do on the landscape, for example, in mountainous regions where agriculture occurs in the river bottoms or valleys but not on the mountains.