Figure 1.
Terrain map of Mpala Research Centre showing north-south rainfall gradient and schematic of the experimental design.
Three blocks, each containing one 100×100-m replicate of each treatment, are situated at each circled location on the map; 20 km separates the northern (low-rainfall) and southern (high-rainfall) sites. Red lines indicate dirt roads. (A) Total-exclusion plots exclude all large herbivores; (B) mesoherbivore-exclusion plots exclude all herbivores larger than warthogs; (C) megaherbivore-exclusion plots exclude elephants and giraffes only; (D) open plots are accessible to all species.
Figure 2.
Patterns of dung deposition by the eight most common LMH species, arranged in order of increasing body mass.
Data are the mean number of dung groups per plot across eight surveys from April 2009 to November 2011.
Figure 3.
Mean monthly rainfall (A) within years and (B) across years at each of the three sites.
The asterisks next to years 2008 and 2012 in the top panel indicate that data were available only for 3 and 5 months, respectively.
Table 1.
Surface-soil attributes showing significant variation across treatments and/or sites.
Figure 4.
Relationship between two metrics of productivity and rainfall in the six months prior to productivity measurement.
(A) Mean peak understory biomass (grasses and forbs) in two 64-m2 grids located within total-exclusion plots at each site. (B) Mean NDVI in each plot, calculated from Quickbird satellite imagery. Rainfall was squared in regression analyses to better fit the data.
Figure 5.
Trends in (A) extent of bare ground and (B) total understory vegetation pin hits across sites and treatments.
Data are out of 490 total pin drops per plot per survey.
Table 2.
Understory plant species showing significant variation across treatments.
Figure 6.
Non-metric multidimensional scaling plots illustrating patterns of community similarity across sites and treatments for (A) understory and (B) overstory plants.
Figure 7.
Rate of growth in height (A, C, E) and canopy breadth (B, D) for the three dominant Acacia species across sites and treatments.
Figure 8.
Temporal dynamics in the minimum number known alive of all small mammals in total-exclusion (filled markers) and open plots (open markers) at (A) low-rainfall, (B) intermediate-rainfall, and (C) high-rainfall sites.