1. Cattle and other livestock graze more than a quarter of the world’s terrestrial area and are widely regarded to be drivers of global biodiversity declines. Studies often compare the effects of livestock presence/absence but, to our knowledge, no studies have tested for interactive effects between large wild herbivores and livestock at varying stocking rates on small-bodied wild vertebrates.

2. We investigated the effects of cattle stocking rates (none/moderate/high) on the diversity of wildlife 0.05-1000 kg using camera traps at a long-term exclosure experiment within a semi-arid savanna ecosystem in central Kenya. In addition, by selectively excluding wild ‘mesoherbivores’ (50-1000 kg) and ‘megaherbivores’ (>1000 kg; elephant and giraffe), we tested whether the presence of these two wild herbivore guilds (collectively, ‘larger wild herbivores’) mediates the effect of cattle stocking rate on habitat use and diversity of ‘smaller wildlife’ (mammals ranging between 10-70 cm shoulder height and birds).

3. Our results show that cattle enhance alpha diversity of smaller wildlife (with or without larger wild herbivore presence) and of all wildlife 0.05-1000 kg (with or without megaherbivore presence), by altering vegetation structure. However, for smaller wildlife, this effect is less pronounced in the presence of larger wild herbivores, which also shorten grass. In the absence of cattle, mesoherbivore-accessible sites showed higher alpha diversity of smaller wildlife than sites excluding mesoherbivores.

4. Smaller wildlife habitat use was increased by high-cattle-stocking rates and wild mesoherbivores more in the presence of the other.

5. Synthesis and applications. Our findings imply that grazing, whether by livestock or wildlife, can enhance local savanna wildlife diversity. The biodiversity benefits of localised increases in herbivory are likely to be due to shortened grass and associated visibility improvements (for predator-avoidance/foraging). This suggests that land managers can increase local biodiversity by shortening grass, with wild or domestic herbivores (or both), at least in patches within a taller-grass matrix.

Camera traps deployed in a long-term exclosure experiment for one year (2019-2020).

Camera trap images organised using camtrapR package, with images of the same species at the same location taken 60 minutes appart assumed to be independent.

See README file.