Assessing the effects of zero abundance data on habitat preference modelling using a genetic Takagi-Sugeno fuzzy model

In this paper, the effects of zero abundance data on fish habitat modelling using a genetic Takagi-Sugeno fuzzy system were assessed with specific focus on habitat preference curves (HPCs) and model performance. Three independent data sets were prepared from a series of fish habitat surveys conducted in an agricultural canal in Japan. To quantify the effects of zero abundance data, two kinds of data (full abundance data and presence-only abundance data) were used, from which a fuzzy habitat preference model (FHPM) were developed. As a result, the HPCs obtained using presence-only abundance data resulted in similar HPCs across the different data sets used, while those obtained using full abundance data differed by the data sets. Because the model performance with regard to generalization ability was higher, the present study concluded that the use of presence-only abundance data can better capture the habitat preference of the target species.