The role of remote sensing in species distribution models: a review

Species distribution models (SDMs) are invaluable for delineating ecological niches and assessing habitat suitability, facilitating the projection of species distributions across spatial and temporal dimensions. This capability is crucial for conservation planning, habitat management and understanding the impacts of climate change. Remote sensing has emerged as a superior alternative to traditional field surveys in developing SDMs, offering cost-effective, repetitive data collection over comprehensive spatial and temporal scales. Despite the rapid advancements in remote sensing technologies and analytical methods, the specific contributions of remote sensing to SDMs historically, and the potential pathways for its integration with SDMs remain ambiguous. Therefore, our study has set forth two objectives: firstly, to conduct a thorough review of remote sensing’s role in SDMs, focusing on environmental predictors, response variables, scalability and validation; secondly, to outline prospective research trajectories for remote sensing within SDMs. Our findings reveal that remote sensing offers a plethora of environmental predictors for SDMs, encompassing climate, topography, land cover and use, spectral metrics and biogeochemical cycles. A variety of remote sensing techniques, including random forest, deep learning and linear unmixing, facilitate the derivation of SDM response variables and the development of species distribution models across diverse scales. Furthermore, remote sensing enables the validation of SDMs through its mapping outputs.