Decomposition is important for nutrient cycling and the dynamics of soil organic matter. The factors that influence local decomposition rates in savannas dominated by Macrotermes mounds remain uncertain. Here, we experimentally assessed the effects of macro-and micro-detritivores, active and inactive mounds, and vegetation cover on wood decomposition rates for eight common woody plant species in Lake Mburo National Park, in Uganda. Five pairs of Macrotermes mounds, one active and one inactive per pair, were selected. Each mound provided two sample locations, one, the most shaded (with canopy cover), and one, the most open (without canopy cover) edge of mound. In addition, for each mound pair, one additional sample location was located off-mound, in an open level area between the mounds. After one, three, and 12 months, protected (wrapped in 1-mm mesh fiber-glass excluding macrodetritivores) and unprotected wood samples from each location were retrieved, brushed clean, oven-dried, and weighed. After 12 months, mean percentage mass loss was four times higher for unprotected than protected wood samples across all species located on mound sites (when decomposition in shaded and open microhabitats was combined). Mean percentage mass loss across all species combined was 1.2 times higher on active than inactive mounds. Across all mounds, decomposition was on average 1.1 times more rapid in the shaded than open mound parts. These differences were more pronounced on inactive mounds (1.3 times more rapid in the shaded than open parts). Percentage mass loss was markedly lower off-mound (12.6 AE 0.8%) than on active (25.9 AE 1.5%) or inactive mounds (19.7 AE 1.2%). Proportional mass loss for unprotected wood decreased with increasing wood density, but proportional mass loss of protected wood samples was not detectably influenced by wood density. Our study highlights the strong and locally contingent influence of termite mounds, termite activity, vegetation, and their interactions on wood decomposition rates within a savanna landscape. Furthermore, variation in per-species wood decomposition rates, including the negative correlation with wood density, depends on accessibility to macrodetritivores.

Seed removal can influence plant community dynamics, composition, and resulting vegetation characteristics. In the African savanna, termites and large herbivores influence vegetation in various ways, likely including indirect effects on seed predators and secondary dispersers. However, the intensity and variation of seed removal rates in African savannas has seldom been studied. We experimentally investigated whether termites and large herbivores were important factors in the mechanisms contributing to observed patterns in tree species composition on and off mounds, in Lake Mburo National Park, Uganda. Within fenced (excluding large herbivores) and unfenced termite mound and adjacent savanna plots, we placed seeds of nine native tree species within small open "cages," accessed by all animals, roofed cages that only allowed access to small vertebrates and invertebrates, and closed cages that permitted access by smaller invertebrates only (5 mm wire mesh). We found that mean seed removal rate was high (up to 87.3% per 3 d). Mound habitats experienced significantly higher removal rates than off-mound habitats. The mean removal rate of native seeds from closed cages was 11.1% per 3 d compared with 19.4% and 23.3% removed per 3 d in the roofed and open cages, respectively. Smaller seeds experienced higher removal rates than larger seeds. Large herbivore exclusion on mounds reduced native seed removal rates by a mean of 8.8% in the open cages, but increased removal rates by 1.7% in the open cages when off-mound habitats were fenced. While removal rates from open cages were higher on active mounds (30.9%) than on inactive mounds (26.7%), the removal rates from closed cages were lower on active vs. inactive mounds (6.1% vs. 11.6%, respectively). Thus, we conclude that large herbivores and Macrotermes mounds influence seed removal rates, though these effects appear indirect.

Questions Ungulates affect plant community structure and composition. Vegetation response to these effects are variable. Wild large herbivore populations are declining globally, but how tree communities respond to this change is not clear. We experimentally examined how tree communities respond to changes in ungulate abundance in a heterogeneous landscape. Location Lake Mburo National Park, Uganda. Methods We recorded tree species in nine replicate sites each with four treatment plots: fenced off‐mound (excluding ungulates), unfenced off‐mound, fenced and unfenced on‐mound. Each species was assessed for fruit type, leafing strategy, spinescence and bark thickness. We compared tree communities on‐ and off‐mound, with and without ungulates using PERMANOVA, and the effects of habitat, fencing and time on stem density and traits using generalized linear mixed effects model. Results Stem density increased by 88% off‐mound and 138% on‐mound (p = 0.005) with fencing, between 2006 and 2015. Whether tree communities occurred on‐ or off‐mound determined species composition, but fencing had little effect. Tree traits were not markedly altered by fencing on‐mound. Off‐mound, fencing was associated with a 38% increase in the proportion of fleshy‐fruited tree stems (p < 0.001), 18% decline in armed tree stems (p = 0.035) and a 44% reduction in mean bark thickness (p = 0.001). Conclusion Our study highlights the important role mounds play in maintaining tree community composition with declining ungulate abundance. While ungulates influence tree communities off‐mounds they have little effect on tree composition and traits of mound‐borne trees. Thus, Macrotermes mounds support distinct tree communities that are robust to exclusion of ungulates.