Name: Metamodeling
Reference: Miller, P.S., and R.C. Lacy. Metamodels as a tool for risk assessment. Pages 333-351 in: F.R. Westley and P.S. Miller, (2003) eds. Experiments in Consilience: Integrating Social and Scientific Responses to Save Endangered Species. Island Press, Washington, DC.
MetaModel Manager is a platform to implement this approach, maintained by the Species Conservation Toolkit Initiative (SCTI), and available from SCTI's website here.
Conservation planning step(s) when this would be used: As for PVA, this tool may be used to assist threats analysis at the Review Status step, and may also be used to assist the Action Planning and Evaluation steps.
Description of tool use: This is an extension of traditional PVA techniques, in which specialized models of specific biological processes (population demographics, disease, spatial movement, social breeding complexity, etc.) are linked together through a central data handling and communication module. Within a metamodel, the component models run synchronously, and the outputs of each can be inputs to modify other models as they run. The resulting metamodel offers a richer and more realistic depiction of a complex environment-wildlife-human system, thereby facilitating the creation of more effective solutions to species decline.
Experience and expertise required to use the tool: Considerable expertise is required to apply this tool effectively.
Data requirements: This tool is not suitable for use in a data poor situation. Data inaccuracies and gaps will be compounded through the combination of models, in ways that can be difficult to predict.
Cost: MetaModel Manager and some of the component modeling programs such as Outbreak and Vortex, are FREE for non-commercial use. However, the expertise to operate them effectively may be expensive, depending on circumstances.
Strengths and weaknesses, when to use and interpret with caution: Metamodels would be most valuable when processes that are normally studied in isolation interact in dynamic ways – for example, predator-prey or other multi-species systems, population demography coupled to infectious disease, or dispersal dependent on changing configurations of habitat on a landscape undergoing alteration by humans. The overall metamodel, however, will only be as reliable as are the component models used to simulate the individual processes, so any cautions regarding the reliability of PVA models, epidemiological models of infectious disease, habitat models, or other models linked through a metamodel should carry through to concerns regarding the usefulness of higher level meta-model outcomes. Moreover, a metamodel is only useful if the important functional linkages between variables in the distinct component models can be specified.
Case study: Bradshaw, C.J.A., C.R. McMahon, P.S. Miller, R.C. Lacy, M.J. Watts, M.L. Verant, J.P. Pollak, D.A. Fordham, T.A.A. Prowse, and B.A. Brook. 2011. Novel coupling of individual-based epidemiological and demographic models predicts realistic dynamics of tuberculosis in alien buffalo. Journal of Applied Ecology, 49: 268–277.
Prowse, T.A.A., C.N. Johnson, R.C. Lacy, C.J.A. Bradshaw, J.P. Pollak, M.J. Watts, and B.W.
Brook. 2013. No need for disease: testing extinction hypotheses for the thylacine using multi-species metamodels. J Anim Ecol. 2013 Mar;82(2):355-64.
Back to Abruzzi Table 1.