< back to presentation overview

Details of the presentation
Presentation Oral presentation
Title Opportunities, challenges and limitations of species distribution models: Implications for decision-making in support of biodiversity conservation and sustainable land use

PDF Download   Download PDF (Filesize 5.6 MB)

Short title Critical Issues in Species Distribution Modeling

Author(s) Herkt, K.M.B.(1); Barnikel, G.(1); Fahr, J.(1); Finckh, M.(4); García Marquez, J.R.(2); König, K.(3); Oldeland, J.(4); Penner, J.(5); Rödel. M.-O.(5); Schmidt, M.(3); Sommer, J. H.(2); Wegmann, M.(6)

Presenting author Herkt, K. Matthias B.

Institution(s) (1) Institute of Experimental Ecology, Ulm University, Germany; (2) Nees Institute for Biodiversity of Plants, Rheinische Friedrich-Wilhelms-Universität, Bonn, Germany; (3) Research Institute Senckenberg, Frankfurt am Main, Germany; (4) Biozentrum Klein Flottbek, University of Hamburg, Germany; (5) Museum of Natural History, Berlin, Germany; (6) Department of Remote Sensing, University of Würzburg, German Aerospace Center (DLR), German Remote Sensing Data Center (DFD), Germany

Keywords Species distribution models; biodiversity patterns; model reliability; spatial scaling; sampling bias

Abstract Species distribution models (SDM) are of high value for decision-makers in the field of biodiversity conservation and land use planning. Model reliability, however, depends on the availability and accuracy of species occurrence data, knowledge about the magnitude of sampling biases, the selection of predictor variables, choice of spatial scale (both resolution and extent), the statistical algorithm employed, the incorporation of metapopulation aspects of the modelled organism into the analysis process, comprehensiveness of model validation, and several other factors. Our presentation summarizes the SDM workflow and discusses selected issues in more depth. Principles, challenges and potential solutions are illustrated with case studies from ongoing research projects within the BIOTA network.

We highlight promising future research topics related to species distribution and biodiversity modelling, which BIOTA AFRICA has not addressed yet but would be particularly well prepared to tackle. Based on these findings, we propose to develop an early warning system designed to detect threats to biodiversity. Essential features should be a dynamic link between model predictions and remotely sensed land cover change, and the ability to change the scale of analysis in response to the policy question at hand.

Congress Topic Observation System

Topic No. 1.4
Notes ---

Ref. No. 527