Dr Ismael Núñez-Riboni1, Dr Anna Akimova1
1Thünen-institut Für Seefischerei, Bremerhaven, Germany
Matching environmental and fishery data is a fundamental step when modeling fish distributions and marine habitat suitability. In this study, we show how resolution reduction (downsampling) of both fishery and environmental data improves the performance of habitat modeling. We use spatially resolved abundances throughout 5 decades from 12 functional groups of North Sea fish, as well as bathymetry and bottom temperature. We match environmental and fishery data at 5 spatial scales (1’, 0.2º, 0.5º, 1.0º and 2.0º) by downsampling environmental data and mapping irregular fishery data on the same grid. Through combinations of filter scales, we obtain 9 data-sets per functional group yielding 108 data-sets in total. Afterwards, a realistic habitat model (with mixed effect for spatial autocorrelation) was fit to each data-set. The performance of the habitat model is evaluated with various metrics on training data (noise-to-signal ratio, explained variance and AIC) and on cross-validating data (median Pearson residual and residual auto-correlation). Our results show that downsampling of both environmental and fishery data to scales between 0.2º to 0.5º improves the vast majority of the metrics in comparison to using raw fish and upsampled (interpolated) environmental data. Furthermore, matching data at higher spatial resolutions often yielded unrealistic habitat models. To our knowledge, the present is the most comprehensive study of the effect of scale on matching environmental and fish abundance data and shows that the choice of an appropriate
Ismael was born in Uruguay, studied physics in Mexico and made a PhD in physical oceanography in the Alfred Wegener Institute in Germany. He has worked on climate modelling in the Max Planck Institute of Meteorology and since 2012 he is working in fishery science in the Thünen Institute of Sea Fisheries. He is interested on statistical modelling, in general, and on changes of fish habitat due to climate change, in particular.