Patterns and predictors of ungulate space use across an isolated Miombo woodland reserve

Abstract

AbstractUnderstanding ecological‐ and management‐related predictors of mammal space use within protected areas is critical for management planning. This is particularly true in small, fenced and isolated reserves where we hypothesised that management‐related activities will influence large herbivore space use more than ecological characteristics. We used camera trap data to assess the space use patterns for 18 ungulate species in the small and isolated Majete Wildlife Reserve (Majete) in the understudied Miombo woodland ecoregion of southern Malawi. In the 2018 dry season, the 691 km2 reserve was systematically surveyed for ungulate presence at 140 camera trap locations. Over a period of 5456 camera days, the survey yielded 11 078 independent detections of 18 ungulate species and three predators. Using a single‐species occupancy modelling framework, the probability of space use of ungulates was assessed in relation to five management (fire exposure, fire frequency, water availability, distance from fence and road) and five ecological (visibility, grass biomass, vegetation type, terrain ruggedness and predator abundance) space use covariates, while accounting for imperfect detection. Top‐ranked models contained multiple covariates for 15 of the 16 species modelled, with only nyala's (Tragelaphus angasii) space use best predicted by vegetation type only. Distance to water, vegetation type, visibility and fire frequency were predictors having strong influences on six or more species each. More management‐related covariates reflected in top models, but ecological covariates had more meaningful effect sizes making us reject the hypothesis. Importantly though, distance from the roads and fence were also identified as prominent predictors. Notably, black rhinoceros' (Diceros bicornis) probability of space use increased with distance from the reserve boundary. Beyond informing habitat management for ungulates in Majete, these results can form the basis for understanding species‐specific space use patterns in other small reserves with similar characteristics and threats.