The final HS map can be improved, for instance by reclassifying or smoothing it.


Once you have got the perfect HS model and map, you can further process it in several ways:
  • You may reclassify it to show only a few classes (e.g. “unsuitable”, “marginal habitat”, “core habitat”). Indeed, the 100 HS classes produced by the algorithms provide a misleading precision and a coarser scale would better represent what the model really mean (look at the area-adjusted-frequency cross-validation to decide which classes are significant). Moreover, displaying less classes may improve the interpretability of your map, as well as making it better suited for a black-and-white publication.
  • Conversely, you may want to smooth the HS map to remove some stair steps that may occur in some cases. Smoothing will probably not improve the scientific value of your map, but may enhance its visual appeal.
  • Finally, you might be asked to provide a Boolean map displaying good versus bad habitat. The menu Post-production/Booleanise HS map will provide you with a few tools to guide you in the process of choosing the right threshold (note that the area-adjusted-frequency cross-validation could be used too).



  • Hirzel, A. H., G. Le Lay, V. Helfer, C. Randin, and A. Guisan. 2006. Evaluating the ability of habitat suitability models to predict species presences. Ecological Modelling 199:142-152.