Open Access

Biodiversity is quickly diminishing across the planet, primarily owing to human pressures. Protected areas are an essential tool for conserving biodiversity in response to increasing human pressures. However, their ecological effectiveness is contested and their capacity to resist human pressures differ. This dissertation aimed to assess the ecological effectiveness of different protection levels (from strict to less strictly protected: national park, game reserve, forest reserve, game-controlled area, and unprotected areas) in biodiversity (both mega diverse butterflies and mammals), maintaining habitat connectivity, and reducing anthropogenic threats at the wider landscape in the Katavi-Rukwa Ecosystem of southwestern Tanzania. To achieve this overarching goal, I employed an interdisciplinary approach. First, I analyzed butterfly diversity and community composition patterns across protection levels in the Katavi-Rukwa Ecosystem. I found that species richness and abundance were highest in the game reserves and game-controlled areas, intermediate in the forest reserves, national park and unprotected areas. Species composition differed significantly among protection levels. Landscape heterogeneity, forest cover, and primary productivity influenced species composition. Land-use, burned areas, forest cover, and primary productivity explained the richness of species and functional traits. Game reserves hosted most indicator species. Second, I modelled the spatial distribution of six large mammal target species (buffalo Syncerus caffer, elephant Loxodonta africana, giraffe Giraffa camelopardalis, hartebeest Alcelaphus buselaphus, topi Damaliscus korrigum, and zebra Equus burchellii) across environmental and protection gradients in the Katavi-Rukwa Ecosystem. Based on species-specific density surface models, I found relatively consistent effects of protection level and land-use variables on the spatial distribution of the target mammal species: relative densities were highest in the national park and game reserves, intermediate in forest reserves and game-controlled areas and lowest in un-protected areas. Beyond species-specific environmental predictors for relative densities, our results highlight consistent negative associations between relative densities of the target species and distance to cropland and avoidance of areas in proximity to houses. Third, I examined temporal changes in land-use, population densities and distribution of six large mammal target species across protection levels between 1991 and 2018. During the surveyed period, cropland increased from 3.4 % to 9.6 % on unprotected land and from ≤0.05 % to <1 % on protected land. Wildlife densities of most, but not all target species declined across the entire landscape, yet the onset of the observed wildlife declines occurred several years before the onset of cropland expansion. Across protection levels, wildlife densities occurred at much greater densities in the national park and game reserves and lowest in the forest reserves, game-controlled areas and unprotected areas. Based on logistic regression models, target species preferred the national park over less strictly protection levels and areas distant to cropland. Because these analyses do not support a direct relationship between the timing of land-use change and wildlife population dynamics, other factors may account for the apparent ecosystem-wide decline in wildlife. Fourth, I quantified land-use changes, modelled habitat suitability and connectivity of elephant over time across a large protected area network in southwestern Tanzania. Based on analyses of remotely-sensed data, cropland increased from 7% in 2000 to 13% in 2019, with an average expansion of 634 km2 per year. Based on ensemble models, distance from cropland influenced survey-specific habitat suitability for elephant the most. Despite cropland expansion, the locations of the modelled elephant corridors (n=10) remained similar throughout the survey period. According to ecological knowledge, nine of the modelled corridors were active, whereas one modelled corridor had been inactive since the 1970s. Based on circuit theory, I prioritize three corridors for protected area connectivity. Key indicators of corridor quality varied over time, whereas elephant movement through some corridors appears to have increased over time. Overall, this dissertation underpins differences in ecological effectiveness of protected areas within one ecosystem. It highlights the need to utilize a landscape conservation approach to guide effective conservation across the entire protection gradient. It also suggests the need to enforcing land use plans and having alternative and sustainable forms for generating income from the land without impairing wildlife habitat.