Feasibility of developing ecological indicators

After the 2004 Paris symposium that resulted in the special issue of ICES 2005. The following conclusion about ecosystem indicators was drawn and published in Cury and Christensen (2005). (The below are a copy-paste from Perry et al (2010))

• defining and implementing indicators is achievable with present knowledge, data, and frameworks.
• no single indicator describes all aspect of ecosystem dynamics, a suite of indicators is needed (covering different data, groups and processes)
• environmental and low trophic-level indicators capture environmental change and bottom-up effects
  • global effects of environmental changes (eg regime shifts) on higher trophic levels are not well captured by most indicators (at least individually, suites can elucidate these impacts)
• High trophic-level indicators (eg birds, marine mammals) summarise changes in fish communities
  • top-down effects can be quantified using trophodynamic indicators
• size-based indicators are promising for characterising fish community dynamics in a context of over-exploitation
• ecosystem-based indicators are conservative
  • they only show if the ecosystem is strongly affected, so trends and rapid changes must be evaluated by research and/or management
• interpretation of indicators requires scientific expertise because of potential error and bias in their analysis
• some indicators are better used for surveillance than for prediction. Regime shifts illustrate a situation where surveillance indicators may be useful
• in an ecosystem approach to fisheries management, the objective is not to find the best indicator but rather a relevant suite of indicators with known properties
• a strong feedback between scientific expertise ad management is necessary to improve indicators and their practical use.

Proposed ways to communicate the ecosystem status from indicators in an understandable way.

Need a meaningful, accurate and simple representation of the ecosystem indicators

Polar-area pie charts

Pie charts are a good option because they are simple and multi-variate summaries of the ecosystem (Shin et al 2010).

Each pie piece corresponds to a selected indicator

Polar-area pie charts proposed by Shin et al (2010):

• science is still unable to determine reference values for ecosystem indicators
• instead the axis of each indicator displayed is bounded by the minimum (centre) and maximum (outer boundary) values observed for a particular indicator in each ecosystem
• in this orientation the centre of the plot is associated with worse conditions and the outer with better ones
• depending on the indicators may have to use the inverse in order to orient the worse/best scenario to the be same as the others.
• the boundaries are not meant to illustrate the optimum or target values, instead they are there to scale the indicators in the graphic representation.
• in this way the same relative boundaries are used across all ecosystems to facilitate comparison.
• in this paper the minimum value of each pie piece/indicator observed was given a value of 0.1 in a scale of 0-1. This as used to differentiate between minimum value (0.1) and no indicator (0).
• indicator values are only representative of the state of the ecosystem if they are monotonic (preferably linear) across a range of observed values

In order to not have misleading interpretations of the pie diagrams:

1. 'as multiple indicators were handled and represented, their selection was constrained so as to obtain a final set with a good balance between different ecological features deemed important when evaluating the ecosystem effects of fishing.' Each indicator was weighted the same in evaluating the state of an ecosystem (Shin et al 2010).
2. As multiple ecosystem are looked at it is necessary to adapt the same normalised scale for each indicator. As a consequence of this all pies had to be interpreted as relative rather than absolute (Shin et al 2010).

Kite diagrams

Although kite diagrams are visually appealing they can be misleading unless interpreted correctly (Shin et al 2010).

For example the order that the axis are oriented can modify the perception of the ecosystem (Shin et al 2010).

The total are delimited by the values of the indicators differs depending approximately on the axes because each triangular area between the two branches depends on two indicators rather than just one (Shin et al 2010).

Frequency distribution plots

Can be used to evaluate the distribution of actual indicators across ecosystems.

Help to evaluate if pie diagram are good for comparative purposes.

References

Cury, P.M. and Christensen, V. (2005) Quantitative ecosystem indicators for fisheries management. ICES Journal of Marine Science 62: 307-310.

Perry, R.I., Livingston, P., and Fulton, E.A. (2010) Ecosystem indicators. In G. Jamieson, P. Livingston, and C.I. Zhang (eds) Report of Working group 19 on Ecosystem-based management Science and its Application to the North Pacific. PICES Science Report no. 37, Sydney, Canada, 83-89.

Shin, Y-J., Bundy, A., Shannon, L.J., Simier, M., Coll, M., Fulton, E.A., Link, J.S., Jouffre, D., Ojaveer, H., Mackinson, S., Heymans, J.J. and Raid, T. (2010) Can simple be useful and reliable? Using ecological indicators to represent and compare the states of marine ecosystems. ICES Journal of Marine Science, 67, 717-731.

Citation

Please cite this page as:

<>

Page created by:

Last modified on:

Versions: