Unraveling the urban forest: spatial configurations and landscape effects on Plecturocebus oenanthe occupancy patterns in Moyobamba, Peru

Abstract Urban expansion causes habitat fragmentation and forces wildlife to adapt to altered environments, often compromising their persistence. In Moyobamba (Peru), populations of the critically endangered endemic primate, the San Martín titi monkey (Plecturocebus oenanthe), survive in urban ravines. Understanding how ravine attributes and spatial configurations affect occupancy is essential for conservation efforts. We surveyed 45 ravines using systematic playback and direct observations and analyzed occupancy using generalized linear models (GLMs, binomial distribution) that incorporated ravine attributes, vegetation indices (NDVI), and graph-based connectivity metrics. Three approaches were evaluated: (i) occupancy explained only by ravine attributes, (ii) connectivity through any route across the urban matrix, and (iii) restricted connectivity along the shortest dispersal paths, with model selection based on the Akaike Information Criterion (AIC). Occupancy was best explained by the third approach, highlighting the combined relevance of ravine areas and restricted connectivity. Ravines larger than 2.85 ha had ≥ 50% probability of occupancy, and those 11.45 ha guaranteed animal presence in the study area. The maximum spatial autocorrelation at 70 m identified a critical dispersal threshold in the urban landscape. NDVI heterogeneity had a weak negative effect, suggesting that P. oenanthe tolerates moderate disturbance but avoids highly heterogeneous habitats with high disturbances. Several occupied ravines (e.g. 31 and 43) are highly vulnerable because of their small size and isolation, whereas others present potential for colonization if connectivity is restored. Conservation actions should prioritize maintaining large ravines, enhancing functional connectivity, and integrating ravines into urban planning to ensure the long-term survival of P. oenanthe in the Moyobamba Region.