Archive for the ‘Rubicode’ Category

Biodiversity indicators: 10 common mistakes

Wednesday, October 28th, 2009

In a paper published in 2003 in the Journal of Environmental Management, Lee Failing and Robin Gregory list 10 common mistakes made in designing biodiversity indicators for forest management. The paper is a worthy read for anyone dealing with issues of monitoring or decisions concerning land-use or ecosystem management.

According to the authors, indicators can have three uses: tracking performance (for results-based management), discriminating alternative hypotheses (for scientific exploration), discriminating alternative policies or management options.

In their paper, they focus on the latter. They list ten common mistakes made in developing and using biodiversity indicators aimed at providing guidance to policy makers or forest managers who must decide on landscape or forest management policies and plans. Deciding whether or not to allow a specific project to go forward requires a different suite of indicators than assessing whether or not the project was a success.

They provide a nice example to illustrate their point:

When we go to the doctors and ask “what is my risk of heart disease”, we do not expect the answer to be framed as a percentage of the target daily donut intake”. (…) Eating fewer donuts may be part of a sensible management strategy but it does not answer the question “am I healthy?” A report of two dozen indicators may be an important part of the the analysis process, but it is also not an acceptable answer to the question (…). Doctors it seems understand the need to take a complex thing, break it down into a relatively small number of indicators, and provide a summary judgement about the status of our health or the probability of recovery associated with alternative treatments.

The 10 mistakes:

  • 1. Failing to define end-points – Is the aim to preserve ecosystem services or scenic value, to prevent the loss of a particular set of species or the intrinsic values and rights of all species.
  • 2. Mixing means and ends – Appropriate performance indicators should focus on the desired goals, not on whether “actions” were taken. Guideline are no substitute to goals.
  • 3. Ignoring the management context – Outside a specific context, “biodiversity” has no meaning – The context must thus be specified.
  • 4. Making lists instead of indicators
  • 5. Avoiding importance weights for individual indicators – Unfortunately, stating that “everything is important” doesn’t work in practice.
  • 6. Avoiding summary indicators or indices because they are considered overly simple
  • 7. Failing to link indicators to decisions
  • 8. Confusing value judgements with technical judgements
  • 9. Substituting data collection for critical thinking – If no data is available, then the authors suggest using established methods for gathering and synthesizing qualitative expert judgements.
  • 10. Oversimplifying: Ignoring spatial and temporal trade-offs – In giving examples for mistake 10, Failing and Gregory mention the importance of taking into account spatial and temporal trade-offs in designing policies aimed at no-net-loss of biodiversity. Temporary and /or local losses could provide – or be made to – provide gains at a broader scale or on the longer term. The same point is made by Kerry ten Kate in an EM podcast on making biodiversity offsets work (mp3).
  • Research priorities according to Rubicode

    Wednesday, September 30th, 2009

    The RUBICODE project on “Rationalising biodiversity conservation in dynamic ecosystems” has published its latest and final newsletter.

    Among other products, the project has prepared an interesting report on research priorities for ecosystem services (pdf here). The report lists 11 priorities, which are commented briefly below.

  • Quantify the role of biodiversity, including uncharismatic and speciose groups of organisms such as invertebrates, lower plants and fungi, in ecosystem function and service provision.

    This is a major goal of ecosystem service science when it comes to informing decision-makers concerning the management of biodiversity. It is somewhat overarching.

  • Develop trait-based approaches to ecosystem service assessment which include: (i) improved knowledge of trait-based multi-trophic linkages within ecosystems; (ii) trait based thresholds for the provision of services; and (iii) trait-based indicators to assess and define quantitatively service provision at multiple scales.

    Trait-based approaches, also labelled functional diversity approaches, are now well developped for linking biodiversity, ecosystem properties and ecosystem services, in particular for plant diversity. As well as being grounded in theory, they offer useful indicators that are often require less expertise than the identification of species and the estimation of their abundance. Expanding these approaches to address multi-trophic linkages, thresholds and multiple scale is the next step.

  • Develop improved methods for the integrated assessment of ecosystem services at different spatial and temporal scales, including methods for: (i) investigating interactions between the demand and supply of multiple ecosystem services; (ii) upscaling and downscaling; and (iii) integrating valuation processes and results in impact assessments and models.

    This point relates particularly well with policy and management issues. As the incorporation of ecosystem services into the decision-making process of land management marches forward such questions will most likely gain in practicality and lose their esoteric touch – which is probably a good thing.

  • Identify thresholds in the relationships between biodiversity, ecosystem functioning, ecosystem services and human well-being to identify points beyond which the level of ecosystem service delivery changes dramatically and perhaps irreversibly. Thresholds again…
  • Identify and quantify the impact of direct and indirect socio-economic and environmental drivers on ecosystem services, and develop tools to design and evaluate policy options for ecosystem service management under uncertain futures.

    As well as uncertainty in the future changes in direct and indirect drivers, ecosystem service science should also assist decision makers in incorporating uncertainty in the scientific knowledge itself. Given the limitations and contextuality of what we know of the dynamics of biodiversity and ecosystems, dealing with uncertainty should be central to the design and evaluation of policy options as well as concrete – on the ground – actions.

  • Improve understanding of the role of the cultural, economic and policy contexts in ecosystem service assessment, particularly in the choice of: (i) metrics, valuation and appraisal methods; (ii) stakeholder involvement; (iii) required levels of precision; and (iv) policy instruments and decision support tools.

    As well as understanding who and what determines the choice of metrics and who is called upon to value ecosystem services, the conceptual bases for these choices should be made explicit (and possibly challenged).

  • Develop an improved classification for ecosystem services and values, which includes values of flows of ecosystem services and stocks of ecosystem assets and allows for the distinction between final and intermediate services.

    Double accounting in the definition of ecosystem service is a recurrent problem, in particular among ecologists. The issue will be discussed in a later post. Meanwhile, you can read the excellent paper by Boyd & Banzhaf (2006).

  • Enhance the usefulness of value, price and cost estimates for ecosystem services by: (i) improving database coverage, quality, depth and access; (ii) filling key gaps in valuation evidence; (iii) investigating replication, validity and transfer of functional assumptions and values estimates; and (iv) developing agreed protocols for comparing and transferring value estimates.

    These priorities are central to the incorporation of ecosystem services in mainstream decision making concerning land-use and natural resource management.

  • Develop tools, methods and decision-support systems to assist the multi-level governance of ecosystem services. What does that mean?
  • Quantify the role of multifunctional land management and landscape patterns on the provision of ecosystem services and develop options to conserve biodiversity and maintain ecosystem integrity outside protected areas.

    The goal of making land management more biodiversity- and ecosystem service-friendly is a very notable trend in public policies for agriculture and forestry (at least in the developed world). Providing quantitative data on ecosystem service provision by alternative policy options is necessary for designing precise incentives schemes that actually serve the goal of ecosystem service provision (rather than the goal of satisfying a particular political base…). Such incentive schemes are an essential component of the policy mix for biodiversity and ecosystem service enhancement.

  • Develop tools and methods to promote the uptake of business opportunities associated with the sustainable management of ecosystem service delivery.

    The development of biodiversity and ecosystem service offsets in recent years is a step in this direction. Hopefully, ecosystem service scientists will embrace this trend and strengthen its scientific basis.

  • Did you notice how long the list is?