Till startsida
To content Read more about how we use cookies on gu.se

How accurate is your predicted map?

When you have obtained the final map of predicted species distribution or abundance, some care should be taken when interpreting the results:

High sampling density produce more reliable maps
The figure below shows the predicted distribution of the red alga Furcellaria lumbricalis (red zones) along the Lithuanian coast, using four modelling methods. Evident differences occur between methods in areas with low number of biologial samples (upper parts) compared to the areas more densely covered by samples (lower parts). Thus, predictions are quite confident in the latter case, whereas no strict conclusions should be drawn in the areas with relatively low number of samples.

Interpretation of predicted map Predicted distribution of red algae along the Lithuanian coast using four modelling methods. Results are more consistent in areas with high number of biological samples (lower part of maps), compared to in areas with low sample sizes. Methods: GAM (Generalized additive models), MARS (Multivariate adaptive regression splines), RF (Random forest and MaxEnt (Maximum entropy modelling).

Accumulation of measurement errors
Usually, the environmental data used for the prediction are obtained from different sources (e.g. geological map, field data) with their own measurement errors. When using these data as GIS-layers for the prediction of species distribution, the final habitat map will accumulate in itself all these errors and this is called error propagation. So, if the final map error is not indicated, it must be evaluated from the description of data before interpreting the map. Otherwise it is very dangerous to draw any conclusion from map without information, how it was composed and how reliable it is. 

© University of Gothenburg, Sweden Box 100, S-405 30 Gothenburg
Phone +46 31-786 0000, About the website

The University of Gothenburg uses cookies to provide you with the best possible user experience. By continuing on this website, you approve of our use of cookies.  What are cookies?