Millennia of human activity around Lake Victoria have fragmented and degraded the once-extensive moist broadleaf forest. Despite legal protection through a network of forest reserves, pressure on these forests increases, both from illegal timber and charcoal production and from clearing forests for agriculture and housing. Forest biodiversity is known to be severely affected by habitat disturbance (both past and ongoing), habitat fragmentation, and land-use intensity in the surrounding landscape. The relative balance of these influences and their importance to different species groups is under-recorded, particularly in the tropics. We focussed particularly on the last aspect - the land use of the 'matrix' landscape surrounding remaining habitat patches. The matrix can influence movement between patches, edge effects within forest patches, and it can contain resources, such as food.
The team assessed the impact of forest fragmentation and matrix quality on diverse species groups: butterflies and moths, grasshoppers, birds, monkeys, bats, trees, climbers, herbs and ferns. The project also included a legal component looking at reasons for forest clearance and the impact of legal frameworks for ownership.
During the project, the research team collected data on a range of species groups as listed above. We have investigated how quickly plants recover from disturbance, and how fragment size affects the diversity of different species groups. We have recorded both structural and compositional changes of the tree community in small fragments around the city of Kampala – urbanisation and the attendant need for housing, food and recreational facilities mean that around half of the fragments have been lost since 1990.
The structural change studies show that the above ground biomass (and therefore stored carbon) has also dropped significantly in the more disturbed forests. Within the same set of fragments, the red-tailed monkey population was resurveyed and compared with 1995 data. In the remaining fragments, the group sizes were significantly smaller in 2012 than in 1995 and the groups contained fewer juveniles.
Reserved or protected forests should be expected to fare better than unprotected fragments. However, within reserves in southern Central Uganda, we have recorded consistent losses of moth diversity over 20-40 years in three very different forests. Zika, Mpanga and Mabira are all legally protected, but very different in size and surroundings; all three have lost moth diversity.
Butterflies we were sampled in different-aged forest areas, and we used the known habitat preferences of the butterflies to understand how butterfly species distribute across different forest types, and what processes drive these patterns. We found that secondary forest has lower species diversity and a greater proportion of generalist species than undisturbed forest.
Different aspects of fragmentation (isolation, patch area, and matrix cover) appear to have a cumulative effect on bird communities. In fragments far from the main forest and surrounded by sugar cane, bird abundance is lower and the proportion of particular guilds (especially ground-nesting insectivores) is lower. There is no change in species richness, suggesting that with fragmentation, forest specialists are replaced by generalist species.
People living in around these fragments lack suitable incentives to protect the forests, and we found that they were demotivated by this lack of support. However, the fragments themselves play a role even for rare and red-listed species. The existence of such fragments was found to depend less than expected on the security of land ownership – forest owners with clearly defined boundaries, good relationships with neighbouring landowners, and clear aims for their forest were more likely to protected their forest land and not sell it.