Managing Nonlinear Ecosystem Services

Research Objective
A great deal of attention has been directed at ecosystem services and their potential to improve natural resource management (Daily et al. 1997, MA 2005, NRC 2005). By understanding how human populations benefit from ecosystems, we can more strategically protect valuable ecosystem functions. Yet despite the burst of enthusiasm for research on ecosystem services, the significant spatial and temporal variation in ecosystems themselves and the services they provide have prevented the successful integration of ecologic and economic systems into a coherent framework for decision-making. For example, the capacity of wetlands to treat sewage, mitigate floods, or purify water, often depends on their proximity to human populations as well as seasonal fluctuations. Researchers agree that ecosystem services fluctuate across space and time, and yet surprisingly little research has examined the management implications of non-linear ecosystem service provision (Costanza et al. 2001, Koch et al. 2009). As a result, researchers studying ecosystem services still confront a fundamental challenge: How to integrate highly variable and non-linear ecosystem services into decision making.

To answer this question we propose a two step approach. First we must adequately describe the variability in ecosystem service provision within our area of interest. Second, once equipped with a more realistic understanding of ecosystem service provision across space and time, we will then test the impact of various management decisions on the watershed and observe how the value of ecosystem services varies non-linearly or with sensitivity to particular thresholds. As a case study, we will examine the ecosystem services related to drinking water supplies in the Llobregat watershed. Thresholds and non-linearity have important implications for how ecosystem services can be exploited to meet drinking water objectives. This research should contribute to our understanding of non-linear ecosystem service provision, as well as provide a specific example of integrating non-linear ecosystem services into decision-making.

We begin with a provocative hypothesis: recent changes in water treatment technology have helped realign environmental and economic interests in the watershed so that these goals have become mutually reinforcing. Furthermore, understanding the spatial and seasonal fluctuation in ecosystem service provision will be essential to accurately exploit ecosystem service provision. The Llobregat watershed near Barcelona, Spain provides an ideal location in which to study non-linear ecosystem service provision due to the natural variability in Mediterranean climates and the explicit and measurable links between river water quality, ecosystem services and treatment costs.

Ecosystem Services

Ecosystem services are the benefits that people derive from ecosystems (MA 2005). It is the natural capital that we rely on for food production, climate stabilization, pollination and drinking water. Research on ecosystem services has exploded as academics and practitioners alike search for more explicit connections between ecosystems and human well being. And while much has been written about integrating ecosystem services into decision making, there are few examples to draw from. The noteworthy exception is the case of New York City’s drinking water supply, which proponents of ecosystem services repeatedly cite as a success adoption of the ecosystem services framework. There, water managers found that investments in watershed protection in the Catskill Mountains provided equivalent water quality at only a fraction of the cost of building a new filtration plant (Chichilnisky & Heal 1998, NRC 2000). The city saved millions of dollars through investments in upstream land conservation that maintained high drinking water quality at the source. The innovative management of New York City’s drinking water provides a concrete example of how policy makers can use knowledge about ecosystem services to direct future investments. Over a decade later, however, few additional examples have surfaced. Some have begun to question whether the case of New York City’s drinking water was an anomaly rather than a replicable strategy (McCauley 2006). Pressure is building to move ecosystem services from theory to practice (Daily et al. 2009). In particular, we need cases that show how resource managers can successfully integrate ecosystem services into decision-making despite their non-linear qualities (Koch et al. 2009).

Researchers have only recently attempted to understand the stochastic qualities of ecosystem services. Aburto-Oropeza et al. (2008) studied the spatial and temporal fluctuations of fish catch as they related to mangrove conditions, while Koch et al. (2009) studied the irregular protection of coastal property provided by marine vegetation. The paucity of studies on non-linear ecosystem service provision has generated calls for improving our understanding of how ecosystem services fluctuate across space and time (Kremen and Ostfeld 2005, Koch et al. 2009). Studying ecosystem services under these more realistic assumptions will improve our ability to operationalize this attractive conceptual framework for resource management. Furthermore, our focus on meeting regulatory objectives allows us to sidestep controversial questions pertaining to valuation and instead concentrate on how ecosystem services can make tangible contributions to meeting water management objectives.

Valuation approaches remains highly controversial because of technical difficulties in obtaining willingness to pay values, and because they generally assume a linear or monotonic relationship between the ecosystem’s condition and the value of the service provided. This linearity does not hold for either the ecosystem conditions or for the services provided. Ecologists have documented that ecosystems can fluctuate between states where marginal changes have no impact if thresholds have been passed (Holling 1978). Economic systems also have thresholds whereby costs kick in after certain points. Thus assuming linearity in our study of ecosystem services is likely to provide misleading information. A new consensus is emerging that research on ecosystem services must consider the inherent variability and thresholds that are characteristic of both ecological and economic systems (Barbier et al. 2008, Koch et al. 2009).

Research Site: The Llobregat Watershed
The Llobregat Watershed provides an ideal site for examining non-linearity and thresholds in ecosystem service provision because it combines high ecosystem variability, plus human dependency . The Llobregat River flows 145 kilometers from the Pyrenees Mountains to the Mediterranean Sea and provides the city of Barcelona with its drinking water. At the same time Mediterranean rivers experience extreme seasonal fluctuations. Barcelona depends on the highly variable and polluted waters from the Llobregat for industrial, agricultural and domestic uses, and water managers at the two treatment facilities in the lower watershed have been dealing with contaminants dumped into the Llobregat for decades. In particular, mine tailings upstream release sodium chloride (NaCl) into the river. For years, addressing the source of this pollution has been deemed financially prohibitive (ACA 2006). The salts and bromides from the mine tailings react with disinfectants, such as chlorine, to form carcinogenic compounds known as trihalomethanes (THMs). These contaminants have plagued the region’s drinking water supply and generated incompliance with European public health standards for drinking water quality.

To address this challenge, the two major drinking water facilities in the Llobregat recently installed desalination systems. The publically owned water treatment facility managed by Aigües Ter-Llobregat (ATLL) installed a reversible electrodialysis system, while further downstream, the private water company Aigües de Barcelona (AGBAR) installed a reverse osmosis system. Both water treatment plants are regulated by the regional Catalan Water Agency, Agència Catalana de l’Aigua (ACA). Thus in this case, the scale of the management institutions—the water providers and the regulatory agency—closely match the scale of the ecosystem services, allowing for the value of the ecosystem services to be taken into account by decision making authorities. The AGBAR treatment plant also extracts groundwater from a coastal aquifer that is experiencing seawater intrusion, and further compounding the salinity problem in its water supply.

Now treatment managers are confident that they will comply with European public health standards, but at a cost. The new treatment technology consumes vast amounts of energy and is expensive to operate. Now more than ever, the cost of water treatment will depend on water quality (Fig. 1). This creates an explicit link between economic costs and the ecological 

integrity of the river ecosystem. Since various ecosystem functions provide services that potentially can reduce treatment costs, the challenge is to understand the spatial and temporal variations of these ecosystem services so as to take full advantage of their provision. If ecosystem services are integrated into water treatment decisions, it will be possible to meet the same water quality standards at much lower costs because we can rely on the ecosystem services instead of energy intensive treatment. Furthermore, investments in ecosystem management can help restore the ecological integrity of the watershed, in addition to

 potentially generating financial benefits.

Figure 1. Prior to the adoption of desalination 

water treatment, fluctuations in salinity did not

 influence water treatment costs. As of 2009, two treatment plants installed membrane desalination plants. This has significantly increased the cost of treatment. Future fluctuations in salinity will be directly correlated with treatment costs (left panel). This creates a new explicit link between ecosystem conditions and economic costs. It also provides a useful analytical framework with which to study the benefits of ecosystem service

 provision. Depending on the relation between ecosystem variability and treatment costs, the response to changed water quality in the future may be highly non-linear with significant threshold effects (right panel)

As mentioned, this research is inspired by the experience in New York City where water managers saved millions by investing in strategic land use management in the Catskill watershed. While this research follows the spirit of the New York case, it adds three important features that may contribute to more valuable or generalizable results. First, the technological context is vastly different. New York did not have a filtration plant to begin with, and knowledge about ecosystem services was used to avoid the construction of such a system. In contrast, our research examines the management of ecosystem services when water managers are already using the most advanced treatment technology available. Second, we will study ecosystem services in the context of restoration, not conservation. This allows us to quantify the tangible savings derived from improving surface water quality instead of speculating on the avoided costs that resulted from land conservation. Lastly, the technological and environmental conditions at our research site are more representative of urban water challenges globally. Most treatment facilities have already installed the filtration systems that New York avoided. This led to criticisms that the New York model was not replicable. Our research site reflects conditions that water managers are likely to face in the future, whereby poor water sources will be treated with newer and more costly technologies. In this respect, our case study should be more valuable to a wider audience of researchers interested in both in ecosystem services and municipal water treatment.