Determining where animals are and if they are persisting across protected landscapes is necessary to implement appropriate management and conservation actions. For long-lived animals and those with boom and bust life histories, perspective across time contributes to discerning temporal trends in occupancy and persistence, and potentially in identifying mechanisms affecting those parameters. Long-term data are particularly useful in protected areas to quantify indicators of change that may be less obvious or occur more slowly. We used thirty-four years of amphibian data specific to Rocky Mountain National Park (RMNP) in a Bayesian occupancy modeling framework to estimate changes in colonization, persistence, and occupancy of  Pseudacris maculata, Lithobates sylvaticus, and Ambystoma mavortium. We explored the effects of climate, landscape change, and visitor use as mechanisms behind observed changes in RMNP. Our results indicate that visitor use has a negative effect on initial occupancy and persistence for all three species, and that higher NDVI values (a proxy for habitat complexity) are associated with higher persistence and colonization probabilities. These results provide a way forward where mitigation efforts can target identified drivers of decline.