Data from long‐term monitoring programs, such as the US Fish and Wildlife Service (USFWS) line distance sampling (LDS) program for Mojave desert tortoises (Gopherus agassizii), are increasingly being used in new ways to elucidate trends in population dynamics. We used the USFWS LDS data in a novel way to generate range‐wide predictions of occupancy, colonization, and local extinction rates from 2001 to 2018. We developed a dynamic occupancy model to answer fundamental questions posed by Bureau of Land Management personnel regarding how G. agassizii are distributed across the landscape over space and time. We transformed the LDS data into detection/nondetection data and constructed a Bayesian dynamic occupancy model using several time‐varying (e.g., temperature, precipitation, normalized difference vegetation index, fire, and a proxy for invasive grasses) and static covariates (e.g., soil properties, topography, distance to roads, distance to urban areas) hypothesized to influence G. agassizii occupancy dynamics. We estimated that over the entire time series (2001–2018) the probability of G. agassizii occupancy is declining in over one quarter (26%) of the range, largely in the northeastern part of the range, but that from 2011 to 2018, 77% of the range has a declining trend. Drawing on these model outputs, we developed an interactive, web‐based tool for exploring trends in dynamic occupancy across the species range, allowing users to focus on areas of management interest or concern.