Conservation biology faces the challenge of safeguarding the ecosystem functions and ecological processes (the water cycle, nutrients, energy flow, and community dynamics) that sustain the multiple facets of biodiversity. Characterization and evaluation of these processes and functions can be carried out through functional attributes or traits related to the exchanges of matter and energy between vegetation and the atmosphere. Based on this principle, satellite imagery can provide integrative spatiotemporal characterizations of ecosystem functions at local to global scales. Here, we provide a multitemporal dataset at protected-area level that characterizes the spatial patterns and temporal dynamics of ecosystem functioning in the Biosphere Reserve of the Sierra Nevada (Spain), captured through the spectral Enhanced Vegetation Index (EVI, using product MOD13Q1.006 from the MODIS sensor) from 2001 to 2018. The database contains, at the annual scale, a synthetic map of Ecosystem Functional Type (EFT) classes from three Ecosystem Functional Attributes (EFAs): (i) descriptors of annual primary production, (ii) seasonality, and (iii) phenology of carbon gains. It also includes two ecosystem functional-diversity indices derived from the above datasets: (i) EFT richness and (ii) EFT rarity. Finally, it provides interannual summaries for all previously mentioned variables, i.e., their long-term means and interannual variability. The datasets are available at two open-source sites (PANGAEA: https://doi.org/10.1594/PANGAEA.924792; Cazorla et al., 2020a; interannual summaries at http://obsnev.es/apps/efts_SN.html, last access: 17 April 2023). This dataset provides scientists, environmental managers, and the public in general with valuable information on the first characterization of ecosystem functional diversity based on primary production developed in the Sierra Nevada, a biodiversity hotspot in the Mediterranean basin and an exceptional natural laboratory for ecological research within the Long-Term Social-Ecological Research (LTER) network.