Constructing a hybrid species distribution model from standard large-scale distribution data

Abstract

Species range shifts under climate change have predominantly been projected by models correlating species observations with climatic conditions. However, geographic range shifting may depend on biotic factors such as demography, dispersal and species interactions. Recently suggested hybrid models include these factors. However, parameterization of hybrid models suffers from lack of detailed ecological data across many taxa. Further, it is methodologically unclear how to upscale ecological information from scales relevant to ecological processes to the coarser resolution of species distribution data (often 100 km(2) or even 2500 km(2)). We tackle these problems by developing a novel modelling and calibration framework, which allows hybrid model calibration from (static) presence-absence data that is available for many species. The framework improves understanding of the influence of biotic processes on range projections and reveals critical sources of uncertainty that limit projection reliability. We demonstrate its performance for the case of the butterfly Titania's Fritillary (Boloria titania).

Keywords

Biotic interaction; Colonization; Extinction; Range projection; Process-based; Dispersal

Published in

Ecological Modelling
2018, volume: 373, pages: 39-52
Publisher: ELSEVIER SCIENCE BV

UKÄ Subject classification

Ecology


Publication identifier

DOI: https://doi.org/10.1016/j.ecolmodel.2018.02.002

Permanent link to this page (URI)

https://res.slu.se/id/publ/94862