A quest for the Holy Grail? Balancing precision and practicality in ecosystem services metrics.

Ecosystems are highly complex. Building tools that allow us to quantify that complexity with certainty and that produce understandable and useable results is a real and pressing challenge in the ecosystem services arena.

A good metric almost always includes credibility, usability, and cost-effectiveness as criteria.  To build a credible metric you use the best science to accurately account for how an ecological system functions. Usability and cost-effectiveness is achieved with a metric that is straightforward to use and can be applied with limited resources and time. 

Have we launched ourselves on a quest for the Holy Grail?  The perfect metric is virtually unattainable given these criteria.  An overly simplistic approach will not provide us with an accurate picture of how an ecosystem is functioning.  But a metric that is highly complex requires significant time and resources to implement and is unlikely to be used.

How do we balance the need for both precision and practicality when it comes to metrics?  Can we argue that while current measurement systems are imperfect, they are better than nothing? Is commitment to long-term monitoring and adaptive management a sufficient strategy to mitigate against the risks of incomplete or imprecise measurement systems? What’s “good enough” when it comes to metrics?

Hundreds of metrics have been developed.  It’s now time to re-balance our efforts and focus on robust testing of existing approaches and implementing adaptive management frameworks that will tell us if we’re moving in the right direction to meet conservation goals.

Author bio: 
Nicole Robinson Maness is a contractor with the Willamette Partnership working on metrics habitat metrics for ecosystem services markets and is co-author of “Measuring Up: Synchronizing Biodiversity Measurement Systems for Markets and Other Incentive Programs”, a report recently released by the USDA Office of Environmental Markets (http://willamettepartnership.org/measuring-up). She is also a faculty research assistant at Oregon State University working on projects investigating policy issues related to private landowner involvement with voluntary carbon markets.


Sara Vickerman

Nicole is correct that a good

Nicole is correct that a good metric must balance precision and practicality or it will not be effectively applied to help us make better decisions about conservation. Adding to the challenge of developing useful metrics is the scale problem. Metrics that work at the site scale are not necessarily useful for measuring landscape scale effects.  Ideally, a nested system that accommodates information neeeds at multiple scales would be available to apply across jurisdictional and programmatic boundaries. Agencies and the private sector need to work together to develop consistent measures for ecological integrity across multiple scales. Given shrinking resources for this work, it is important to pool resources and expertise to finance and apply these tools.    

Tim Gieseke

Moving in the right direction

Moving in the right direction is a good starting point for usable metrics.  I agree that many have reached a summit in the search for the perfect metric, only to envision the range of issues ahead of them.  I assume that the search for the scientific secrets of a teaspoon of soil decades ago must have brought the same euphoria and then realization that no matter how intensive our study is of the soil world, there will always be thousands of organisms interacting with millions of constituents generating billions of metabolic reactions that we will never grasp.  So we moved on and we work with the soil world and understand its parameters a little more each year.  As l look across a landscape I see a lot of teaspoons of soil and the dynamic world that lives atop it.  Our metrics must capture those parameters and how our interactions may impact them and we need to value those management strategies to some degree.  If we like, Monday morning we could begin to compile some basic metrics for soil, water, air, plants and animals and using GIS, spreadsheets and the cloud, we could demonstrate how this new economic system with ecological metrics work.  The only thing stopping us is us and our need to find the holy grail of metrics.


As a retired oceanographer

As a retired oceanographer from the Fisheries Lab in Woods Hole, Ma. who was a member of the New England Fishery Management Council's (NEFMC) Habitat Plan Development Team, I would question the underlying assumption that ecosystem services metrics developed by academics will be that useful in developing more sustainable fishing practices and better conservation of essential fish habitat (EFH).  For a number of years I supported the ecosystem indicators endeavor launched by the Gulf of Maine Council for the Marine Environment and headed a committee that examined potential options for fisheries and aquaculture.  State managers and regulators had more interest in socioeconomic indicators (direct, indirect and induced benefits in counties/states from commercual/recreational fishing expenditures in their jurisdictions) than ecosystem service indicators.  Developing useful ecosystem services metrics to support fisheries management and aquaculture that will be acceptd by the management entities and their clients is very challenging.

In more recent times there has been a dialog on developing ecosystem approaches to fisheries management (EAFM) as tools to supplement traditional fisheries agency bottom trawl surveys and stock assessments that develop fishing quotas in sector management approaches.  The EAFM methods are more likely to be applied for strategic fisheries management concepts than be utilized in operational management decisions made by the fishery management councils (FMCs).  Ecosystem services metrics may constrain EAFM concepts to encourage more sustainable fisheries management and reduce the direct/indirect effects of fisheries on other marine wildlife; marine biodiversity and EFH.  

As a grassroots environmental activist who is involved with the New England Ocean Action Network which is promoting coastal and marine spatial planning (cmsp) by the Northeast Regional Ocean Council (NROC), I feel that the ecosystem services indice developers need to engage with the management entities and their users in filling existing information gaps.  I don't see that having academics develop ecosystem service indices that are not linked to the  National Ocean Policy or Pew/U.S. Ocean Commission report recommendations to be a useful endeavor.  I have more influence as a Sierra Club grassroots environmental activist on tthe managment/regulatory process than such academic research.  Academic scientists certainly have the right to conduct such research and pursue grant funding for their research agenda.  Dr. Les Kaufman (Boston University) and other researchers are examining ecosystem service  modeling in the Gulf of Maine, but they are engaged with the New England Fishery Management Council and NOAA Fisheries EAFM proponents.

Even the Nation Science Foudation requires public outreach efforts on the public relevance of their research grant results.  Having revently attended a public workshop on the public outreach components for NSF grants at the Waquoit Bay National Ecological Research Reserve, I was disappointed at the limited scope of this endeavor.  Thus I decided to respond to the dialog on ecosystem metrics in a non-traditional fashion, since it is out of touch with the reality of natural resource management/regulation from my experience in fisheries management.  There needs to be a better integration of applied and pure science into the natural resource planning, management and regulatory systems, since many existing management paradigms are not well suited to cmsp/adaptive management for complex dynamic socioeconomic/ecological systems.  Developing ecosystem metrics could play a role in this dialog if academic scientists engage with the management entities and their clients.



The scale and precision of a

The scale and precision of a metric must fit the question being asked and the system that is the subject of inquiry. Ecosystems produce services at the scale of "teaspoons of soil," but the services may be delivered to society at scales ranging from local to global. Marine fishery production, for example, may depend on reproductive, recruitment, and predator-prey dynamics at very fine spatial scales, but the services are delivered at the scale of major ocean basins, or even globally if we consider the span of markets.

Thus, broad-based metrics that capture status and trends at the most expansive scales are informative and essential, even though they lack precision. At the same time, metrics that reflect the underlying processes are essential and may or may not have precision, depending on the state of knowledge. Linking the scales, spatially and temporally, is an important challenge for ecologists and economists.

Asking what's important and what's good enough for a given question should not be a quest for the Holy Grail. 

Morgan Robertson

This cannot be approached as

This cannot be approached as if it were merely an ecological problem -- the problem of finding adequate metrics, as Nicole says, balances economic criteria and ecological criteria.  Since we're talking in the language of "services", we're talking about things that are primarily economic in their identity whether or not there's actually a market for them (they might be used for indexical purposes -- green GDP etc -- rather than sold in markets).  So the question is how to render ecological complexity in a form that is as abstract and transportable as a commodity. 

Or, as the NRC put it in their 2005 book on the topic, studies must be "designed so that the output from the ecological models can be used as an input to the economic models".  This hierarchy will rankle some ecologists.

The market consequences of getting too complex, ecologically, with your metrics was put to me once by an EPA regulator in the following way:

"You can define a unit so that you’re going to have flourishing mitigation banking.  You can also define a unit so that, should there ever be one exchanged, it would be environmentally precise.  And those are at potentially different extremes… [You might end up with credits in] 'habitat for middle-aged great blue herons who don’t like shrimp', or something.  Obviously, I can’t imagine even trying to do that."

I consider this to be one of the real unresolvable paradoxes of environmental markets.  If a metric is ecologically precise enough to fully reflect the complexity of ecological systems, it will be so hard-to-measure and so specific to a site that it will tend to describe a commodity that can trade successfully in a market of sufficient thickness and volume to sustain trade.  If a metric is abstract enough to describe a commodity which can trade at volume, it will tend to fail to describe ecological complexity. 

In short, as the ecological sensitivity of a metric increases, the fungibility of the commodity it describes decreases.  Empirically, I saw this as wetland credit markets began to evolve -- the metrics actualy got simpler as pressures for increased volume come to bear.  But with the trend toward functional assessment we're seeing a push to complicate the assessment -- so the paradox above creates a stiff headwind, but not an absolute bar.  Will be interesting to see how far it can go.


Wetlandia -- a blog about Ecosystem Services and some other things.


The issue of scale is

The issue of scale is resolved by calculating ecosystem service flows with a multi-scalar model.  The rest is about pragmatism and practice.  On the pragmatism side, one approach that works is to acknowledge that ecosystem services exist as a result of human demands being met by biophysical processes, usually in concert with human work.  This is a reasonable way of going about things because the values associated with ecosystem services are all human constructs.  The relevant metrics, then, are whatever things you can measure that provide a good, solid, surrogate handle on the determination of ecosystem service availability or scarcity.  

Both spatial and temporal scales matter, however, and recognizing the values realized on different time scales can be helpful. For example, on a short time scale of days to months, the most useful ecosystem service metrics are natural-human pairings:  for example, between measurements of key determinants of ecosystem service delivery on the biophysical side, with measurements of corresponding market or other forms of emergent social behavior on the human side.  Take fisheries, for example.  It is possible to model the determinants of the rate of provisioning of fish biomass that is available to a fishery, after which what is most interesting is the process by which this potential is converted into monetary and/or other values (e.g. cultural, aesthetic, etc.).  However, there is are also other ecosystem services beside food provisioning whose values are created on a longer time scale, such as bioregulation (keeping the system going, maintaining resilience), and existence (e.g. compliance with the Endangered Species Act).  

If both spatial and temporal scaling are dealt with, and the coupling between human and natural players is explicitly defined (the science of doing this is, after all, a major dimension of "coupled human and natural systems"), then many of the semantic problems that haunt the modeling of ecosystem service dynamics aren't so scary any more.  Then, the model itself, is what becomes scary...or even more so, the modelers!

By way of example, interested readers will want to look into applications developed using the modeling platforms Atlantis, Aries, InVEST, and especially, MIMES.  It is a young and exciting field.  There'll be plenty to argue about, but meanwhile, useful work can be done as well.


For something more abstract

For something more abstract like coastal protection or biodiversity on a coral reef, how does one quantify that? Tangible ecosystem services are easy, but managers (well most managers) are hard pressed to identify with abstract intangible services like the two mentioned about. And ideas?


Tim Gieseke

Morgan - I appreciate your

Morgan - I appreciate your comment - "I consider this to be one of the real unresolvable paradoxes of environmental markets. If a metric is ecologically precise enough to fully reflect the complexity of ecological systems, it will be so hard-to-measure and so specific to a site that it will tend to describe a commodity that can trade successfully in a market of sufficient thickness and volume to sustain trade. If a metric is abstract enough to describe a commodity which can trade at volume, it will tend to fail to describe ecological complexity."

But to give an out of the box perspective:

The reason ecological markets are emerging is that our current economic metrics (unit and price) do not fully reflect the complexity of our economic systems.  I recognize that preserving a failing system (chock full of externalities)  because it is simple to operate may be popular and comfortable, but it does not remove the economic fact that it is chock full of externalities.

Morgan Robertson

Tim -- I totally agree!  I've

Tim -- I totally agree!  I've been watching the "acre v. function" argument for a long time, and I certainly see the disadvantages of the acre.  The acre in no way reflects what is actually important about protecting ecosystems, I agree.  My only point is that in pursuing a better metric, we're running afoul of the very dynamics that allow markets to operate -- that is, markets only work when the commodity is abstract, generalized, and fungible.  As we pursue finer and finer ecological distinctions in defining "services" -- and this is precisely the job we set ecological scientists to doing in an ecosystem services context! -- we create potential commodities that are less and less able to circulate in a market.

In lectures I often use the concept of a loaf of bread or a gallon of gas.  These are excellent commodities because although each individual loaf or gallon may differ in a thousand small ways, those differences are successfully ignored when it comes to defining the unit.  This is because as a society we've successfully adopted "the pound" and "the gallon" as uncontroversial ways of measuring value -- of course, there was a moment in history where these measures were very new and controversial, but that moment is long over.

Acres, for all their flaws, are like "pounds" or "gallons".  And I maintain that there will be a headwind against the success of more complicated metrics.  Some of this headwind will come from scientists themselves who find the process of creating durable and uncontroversial metrics to be impossible, and some will come from the market because who wants to buy bread in "units of nutritional service" instead of "pounds"?


Wetlandia -- a blog about ecosystem services and other things.

Tim Gieseke

Since Natural Capital

Since Natural Capital spontaneously produces ecoservices in relation to the its state of integrity, its willing-to-supply is innate.  Physical, man-made capital produces goods and services in relation to humans' willingness-to-supply.  If an ecosystem is in tip-top form it produces an abundance of ecoservices.  If a factory is in tip-top form it only produces an abundance of products when someone decides it should.  The point of this is that in ecosystem service markets, one can value the management of natural capital and receive a quantity of ecoservices without the need to value the specific ecoservices.    Developing a market system from this one-step back, greatly simplifies the process and allows for symbiotic demand to function.  In discussing this point, I think economists have a more difficult time grasping/accepting this economic structure than ecologists.  Economic transitions provide a linear, zero-sum process that is uncomplicated.  Ironically, applying this simple process to the natural economic system, makes it  very complicated.  Seek out simplicity as any system that is not, will not work.  In reference to jahsonb's comment, the challenge is to identify the natural capital that humans can manage.  In terrestrial ecosystems it is land, in oceans, fish stocks become pronounced.  We are not going to develop this ecological dimension to the economy all at once and developing a foundation to understand natural capital will remain a priority prior to playing the market and paying for ecosystem services.  The greatest leverage we can use in ecological system is symbiosis, yet most ecoservice markets shy away from this advantage.


Scientifically derived

Scientifically derived quantifications, empirically derived, predictable and verifiable are all anyone wants to hang their hats on these days.