Spatial prediction of Laminaria hyperborea at the Norwegian Skagerrak coast

The overall aim of the study was to investigate to what extent the distribution of L. hyperborea in Skagerrak has changed during the last 10-15 years. To do this, we first needed to test to what extent available data, i.e. data from the National monitoring program (KYO) are representative enough and could be used to for predicting spatial changes in the L. hyperborea distribution. Three models were compared. The first model, referred to as the “Full model”, included registrations from the National program for mapping of prioritised nature types, and 2004 data from KYO. The second model, called KYO 2004, used only recordings from KYO 2004. The third model using KYO data from 1995 was also compared to the other two. If the predictions from the KYO 2004 model were approximately similar to the predictions of the Full model, we would assume that comparing KYO models from different years would give information about temporal differences in the distribution of L. hyperborea. The results showed that there were generally larger differences in the prediction between the full model and the model including KYO data from 2004, than between the two KYO models including data from 1995 and 2004. The KYO models underestimated the distribution of kelp compared to the Full model. The data used in KYO models did not cover the whole exposure gradient which is the main reason for the less accurate predictions. While time series data from KYO represent a very important tool for monitoring community changes in the Skagerrak, they are not designed to analyse spatial changes in the distribution of species. To do so, specific monitoring programs are needed.