The on-going declines in biodiversity caused by global and local environmental changes call for improved monitoring and conservation schemes. Remote-sensing (RS) of earth surface stands at the forefront to tackle this challenge, by providing data at different spatial and temporal resolutions that can be related with a wide range of environmental variables. Spatiotemporal dynamics of ecosystems and vegetation functioning (depicting several facets of matter and energy fluxes) can affect habitat suitability and therefore the persistence of species and the patterns of biodiversity. In this study we analysed habitat suitability and species diversity patterns by combining Species Distribution Models (SDMs) with multi-temporal RS-based variables of vegetation primary productivity, seasonality and phenology calculated from MODIS products and also from MODIS/Landsat data fusion using the StarFM algorithm. Predictors’ related to structural variables of landscape composition and configuration were compared to functional variables of vegetation dynamics calculated from RS NDVI time-series in a Multi-model Inference (MMI) framework, allowing to assess the relative predictive importance of each set of variables. Multi-annual RS data was used to explore post-fire alterations in biodiversity and short-term changes in habitat suitability dynamics. Overall, MMI results showed a good support for vegetation functioning variables (derived from RS data) in some cases exceeding the model performance of structural landscape variables. In addition, multi-annual RS data were capable of improving habitat suitability models, evaluating short-term changes and assessing post-fire variations in biodiversity. We argue that coupling SDMs with RS functional indicators can provide early-warnings of changes affecting habitat suitability well before assessments based on structural indicators. Possible applications of this methodology range from the improvement of biodiversity monitoring schemes to the design of more effective conservation strategies by explicitly considering the spatiotemporal dynamics of ecosystems.