This post was originally intended to be about my frustration with the ecosystem services concept. In trying to articulate and understand this frustration, I’ve gone through a range of thoughts, which I will explain here. I am getting a bit paranoid that I always seem to come back to issues of scale in my research, and I seem to have done it here again. But I hope it makes some sense, and is more than just an incoherent rambling.
I will start with why I like the ecosystem service concept. I am an interdisciplinary researcher studying natural resource management. The ecosystem service concept is a clear framework for connecting the social world to the physical world. It makes explicit the links between a component of an ecosystem and the various things that it is valued for by people. It seems simple; pollinators are valued because they pollinate crops and other plants, and we like this because we eat, we like pretty meadows, etc. We can then follow this on to further services supported by the pretty meadows, such as recreation and the existence values humans ascribe to such meadows. Being able to follow these chains is useful in understanding the socio-ecological system in any given location. It is also useful for explaining to people how environmental change might directly affect them by impacting on the things they value.
I do share frustrations with other researchers over the grouping together of services and benefits, and the different stages of service (and benefit) in the ecosystem service concept. For this reason, I try to use the idea of intermediary services, final services and benefits. Whereby pollinators pollinating is an intermediary service, the crop is the final service, and the profit from that crop is the benefit to the producer. The consumer may also experience benefit through having food to eat, and preferably at a lower price than they would have been willing to pay. These groupings get long, and interconnect with each other. So the pollinators could lead to multiple benefits, but also could be created through multiple earlier services. Then we are more within a cascade model (e.g. Haines-Young and Potschin, 2010), whereby there is structure (e.g. habitat), process, function, service and benefit.
I find that my main frustrations are introduced when we start to use the concept for practical management. We start to think about how we can increase the number of pollinators. But then we need to recognise that such actions have a trade-off; for example increasing wildflower meadows to support bees may decrease the crop production space, or the habitat for another animal, which then negatively impacts upon another ecosystem service, or multiple ecosystem services in a complex web whereby we need to trade-off goals and priorities (see e.g. Bennet et al., 2009; Raudsepp-Hearne et al., 2010 and others). Some researchers have started to ‘bundle’ ecosystem services to simplify understandings of such trade-offs. Bundles comprise of services that usually appear together and are influenced similarly, such that actions that are beneficial to one service in the group will be beneficial to others, but possibly act negatively on another group. Indeed, a benefit of the ecosystem services concept is that we have a framework for thinking about trade-offs. However, for management purposes, we really lack the knowledge of what actions done in what quantity have which impacts (positive and negative) over which ecosystem services.
While thinking about various actions that could manage ecosystem services, I started to think about ecosystem services within a systems thinking framework. I borrowed the figure below from Donella Meadows’ essay ‘Leverage Points: Places to Intervene in a System’. The idea being that where there is a discrepancy between how we want a system to look, and how it really looks, we can target either the inflows or the outflows from that system in order to remove the discrepancy. Meadows outlines leverage points as being points to intervene in a system, with changing parameters as shallower (and less effective) points, and changing goals as deeper (and more effective). If we use the ecosystem service concept within this framework, we could put pollinators in the central box. Then we can define the goal (e.g. to produce a given amount of oil seed rape). Then we find leverage points to target either the inflow (births, in-migration) or outflow (death through habitat loss, disease) of pollinators in the landscape such that the discrepancy reduction becomes the practical problem.
Currently, according to the way the ecosystem services concept is being operationalized, we are seeking to understand how to target inflow and/or outflow. Most systems are complex, such that this individual component is connected with many others. And often, relationships between components work differently across space. Thus if we are to manage by ecosystem services, we need to model relationships for all locations where there may be variation. And this is being done; we are characterising benefits, understanding how changes in the system affect them. In doing so, I feel somewhat as though we are distracting ourselves by creating ever more complex physical constructs that require even more detailed physical understandings, and ever more complex chains of structures, processes, services and benefits. Great – it is interesting, and should be pursued in the interests of knowledge. But in the end, we are left with very prescriptive sets of measures that can be applied in very specific circumstances locations, depending on what goals we want to achieve.
And to me, it is this questions of ‘what goals?’ and ‘who decides?’ that are my fundamental concerns with the ecosystem services concept. The way the ecosystem service concept is currently being enacted encourages us to work backwards. We are picking a small number of services, and defining goals for each, or for small groups by making decisions on trade-offs. But we aren’t looking at the overall collective system. We are defining the interventions for small components of the system before defining the overall goals. In doing so, we aren’t allowing ourselves to target the deepest, most effective leverage points. We should be asking questions around what we want to manage the system for. Do we want to optimise certain services? Or balance all services? Do we have a particular goal for a resilient system? If we had a goal, we could start to really think about what the discrepancy is, and how to intervene; knowledge could be targeted towards it.
I wonder if we need to start by considering scales of a nested system. If we have started out at the most detailed scale with individual ecosystem services, the next scale up might be biodiversity as the system that incorporates the individual services. This way, the services included within the biodiversity system and their goals influence the working definition of biodiversity. Alternatively, or at same time, by setting goals around biodiversity, we could follow these back to figure out what goals to set for individual ecosystem services. I’m not sure biodiversity is the right grouping concept at this scale – perhaps others have thoughts?! We also get to consider the larger scale system that ‘biodiversity’ (or whatever we settle on) is a part of. Perhaps that system is one of sustainable development (or perhaps I’ve skipped some scales), in which biodiversity might be a sub-system, alongside public health, economic growth, education, etc. Again, we get to define goals for this system, but also see that the sub-systems provide operational definitions for the system goals through their own goals.
So in short, I think I have ended up with my frustration with ecosystem services being that they isolate components of an ecosystem from its broader, interlinked, multi-scale ecosystems. And I have yet to be able to use it to manage anything.