There is a growing interest to understand the static and dynamic components of population ranges. In general, the frequently used environmental forecasting and evaluating methods of occurrences like niche-based statistical processes are based on the static evaluation of the causative environmental variables. These techniques do not consider that natural populations of species form the systems of complex, connected networks. The aim of this study was to suggest a possible solution to this methodological problem. The proposed variable pattern comparison tool (Spatial pattern identification (SPI) for ecological modelling) provides an opportunity of deep examination of spatial connections between environmental variables and occurrence data in GIS models. The idea of the developed method is, that the network characteristic of the primary point-like occurrence data provides statistically evaluable new and valuable information about the nature and reasons for the interconnections of populations. In technical sense, the approach is based on which the key variables of the models can be identified, thus establishing the targeted variable selection and possible solutions for model reduction.•Exploring the relationships between variables of a GIS model.•Static and pattern similarity-based comparison of the model variables.•Identification of key variables of the model and model reduction.•The network allows the understanding intra- and interspecific population connections.
Keywords: Key variable identification; Pattern similarity; Spatial comparison.
© 2021 The Author(s). Published by Elsevier B.V.