Risk perception and tolerance shape variation in agricultural use for a transboundary elephant population.

To conserve wide-ranging species in human-modified landscapes, it is essential to understand how animals selectively use or avoid cultivated areas. Use of agriculture leads to human-wildlife conflict, but evidence suggests that individuals may differ in their tendency to be involved in conflict. This is particularly relevant to wild elephant populations. We analysed GPS data of 66 free-ranging elephants in the Serengeti-Mara ecosystem to quantify their use of agriculture. We then examined factors influencing the level of agricultural use, individual change in use across years and differences in activity budgets associated with use. Using clustering methods, our data grouped into four agricultural use tactics: rare (<0.6% time in agriculture; 26% of population), sporadic (0.6%-3.8%; 34%), seasonal (3.9%-12.8%; 31%) and habitual (>12.8%; 9%). Sporadic and seasonal individuals represented two-thirds (67%) of recorded GPS fixes in agriculture, compared to 32% from habitual individuals. Increased agricultural use was associated with higher daily distance travelled and larger home range size, but not with age or sex. Individual tactic change was prevalent and the habitual tactic was maintained in consecutive years by only five elephants. Across tactics, individuals switched from diurnal to nocturnal activity during agricultural use, interpreted as representing similar risk perception of cultivated areas. Conversely, tactic choice appeared to be associated with differences in risk tolerance between individuals. Together, our results suggest that elephants are balancing the costs and benefits of crop usage at both fine (e.g. crop raid events) and long (e.g. yearly tactic change) temporal scales. The high proportion of sporadic and seasonal tactics also highlights the importance of mitigation strategies that address conflict arising from many animals, rather than targeted management of habitual crop raiders. Our approach can be applied to other species and systems to characterize individual variation in human resource use and inform mitigations for human-wildlife coexistence.

Potential impacts of climate change on geographical distribution of three primary vectors of African Trypanosomiasis in Tanzania's Maasai Steppe: G. m. morsitans, G. pallidipes and G. swynnertoni.

In the Maasai Steppe, public health and economy are threatened by African Trypanosomiasis, a debilitating and fatal disease to livestock (African Animal Trypanosomiasis -AAT) and humans (Human African Trypanosomiasis-HAT), if not treated. The tsetse fly is the primary vector for both HAT and AAT and climate is an important predictor of their occurrence and the parasites they carry. While understanding tsetse fly distribution is essential for informing vector and disease control strategies, existing distribution maps are old and were based on coarse spatial resolution data, consequently, inaccurately representing vector and disease dynamics necessary to design and implement fit-for-purpose mitigation strategies. Also, the assertion that climate change is altering tsetse fly distribution in Tanzania lacks empirical evidence. Despite tsetse flies posing public health risks and economic hardship, no study has modelled their distributions at a scale needed for local planning. This study used MaxEnt species distribution modelling (SDM) and ecological niche modeling tools to predict potential distribution of three tsetse fly species in Tanzania's Maasai Steppe from current climate information, and project their distributions to midcentury climatic conditions under representative concentration pathways (RCP) 4.5 scenarios. Current climate results predicted that G. m. morsitans, G. pallidipes and G swynnertoni cover 19,225 km2, 7,113 km2 and 32,335 km2 and future prediction indicated that by the year 2050, the habitable area may decrease by up to 23.13%, 12.9% and 22.8% of current habitable area, respectively. This information can serve as a useful predictor of potential HAT and AAT hotspots and inform surveillance strategies. Distribution maps generated by this study can be useful in guiding tsetse fly control managers, and health, livestock and wildlife officers when setting surveys and surveillance programs. The maps can also inform protected area managers of potential encroachment into the protected areas (PAs) due to shrinkage of tsetse fly habitats outside PAs.