Distribution models of ecosystem types
New paper out now: composite landscape predictors improve distribution models of ecosystem types. Trond Simensen, Peter Horvath, Julien Vollering, Lars Erikstad, Rune Halvorsen, and Anders Bryn are the authors behind this effort to improve distribution models of ecosystem types.
This landscape from Skjækerfjella, Trøndelag, contains several of the modelled ecosystem types.
Distribution modelling is one of several cost-effective ways of making maps of natural variation. Distribution modelling methods have been applied mostly to single species, but modelling of ecosystem types produces results that are more directly relevant for management and decision‐making.
As a part of the NiN project, the authors have made complete predicted distribution maps of nine ecosystem types throughout Norway. Most of the nine ecosystem types could be predicted relatively reliably.
They found that the current climate alone is not enough to predict the current distribution of ecosystems. Composite landscape gradients can successfully be used to improve modelling of ecosystem types.
Several models were made per ecosystem type, using different sets of predictor variables. Predictor variables included both single environmental variables and so-called composite landscape gradients that represent complex environmental gradients. The climatic variables alone performed poorly compared to when the composite landscape gradients were included.
This paper thus contributes to improving our knowledge about the world around us, and will be useful for managers and for ecosystem conservation efforts.