Assessing the uncertainty arising from standard land‐cover mapping procedures when modelling species distributions

Abstract Aim Habitat loss and degradation is one of the main threats to biodiversity worldwide. Ecological modelling community has been increasingly incorporating habitat changes in species distribution models (SDMs). However, effect that uncertainty arising from standard procedures land‐use/land‐cover (LULC) mapping can have on SDMs overlooked. Remote sensing offers a great source information assess at different spatiotemporal scales. despite variety satellite imagery preprocessing supervised classification currently available, their effects remain largely unknown. Location Baixa Limia–Serra do Xurés Natural Park, NW Spain. Methods We developed for 27 bird with 4 algorithms using 12 sets LULC predictors derived combination 3 image levels algorithms. Results The level, method algorithm had significant impact SDM performance spatial temporal predictions, considerable variation between species. Even if processing transferability were lower than themselves, predictions overall strongly affected by these procedures, sometimes overriding Main conclusions argue when predicting dynamics should receive same level attention as variability due types climate simulations often used ecological research. also highlight need report accuracy satellite‐based maps each class since some classes be key importance certain applications.