The Matrix

Fern and frog Foto: Jenny Reïniö

Millennia of human activity in the fertile region around Lake Victoria have fragmented and degraded the once-extensive moist broadleaf forest. Despite legal protection by gazetting a network of forest reserves, pressure on these forests continues, both from illegal timber and charcoal production and by clearing forests to allow sugar cane and oil palm plantations. Plans to degazette part of Mabira forest were recently dropped after demonstrations and objections from the World Bank. The forests are species rich; Mabira forest, the largest remaining forest fragment has about 300 species of both birds and trees and an endemic primate. This species richness, and the forests' proximity to the capital, Kampala, has allowed the initiation of ecotourism projects. If entrance fees are set at an appropriate level, the income should exceed the costs of conservation.

Species richness in Ugandan forests fits the widely reported power-law species area relation. This relationship can been used to predict that as the size of a forest decreases, the number of species supported by that forest will fall, and hence that many of the Ugandan forest species will become extinct as their habitat is further reduced and degraded. Despite the widespread usage of the SAR in conservation biology, the relationship between species richness and area is confounded by a number of factors.

• Large forests often have heterogeneous habitats containing diverse communities and for this reason alone can be expected to have higher species richness than small forests with a single vegetation type.
• Unequal sampling effort can also bias the results as greater sampling effort will uncover more species; if large species-rich forests have attracted more taxonomists they will have a larger species inventory than under-sampled forests.
• Species richness may not yet have reached equilibrium after a past change in size (i.e., “extinction debt”).

These confounding factors can affect the slope of, and the scatter around, the relationship. If the SAR is to be used as a predictive tool, it needs to be demonstrated that there is a relationship between spatial extent and species richness independent of history, habitat heterogeneity and sampling effort in Ugandan forests.

The SAR treats each forest individually. This makes the analysis simple, but may not be ecologically reasonable. Dispersal links the individual forests together into a metacommunity, where regional extinction-recolonisation dynamics can help to maintain species richness in both the individual fragments and the entire landscape. Metacommunity dynamics depend on the ability of individuals to disperse between forest fragments, so the properties of the matrix, the non-forest area, are important. We predict that where the matrix is conducive for dispersal, the metacommunity will be more tightly linked, and species richness will be higher than in regions where the matrix is not conducive for dispersal.

There are three types of matrix around the moist broadleaf forest fragments in Uganda: subsistence agroforestry; plantation agriculture; and, for forest fragments on islands in Lake Victoria, water. We predict that the structurally-diverse agroforestry matrix will be more conducive for dispersal than the structurally-simple plantation agriculture or water, and that this difference will be detectable in the species richness of different forests.

Metacommunity dynamics depend on the ability of individuals to disperse between forest fragments, so the properties of the matrix, the non-forest area, are important (Vandermeer and Carvajal 2001). We predict that where the matrix is conducive for dispersal, the metacommunity will be more tightly linked, and species richness will be higher than in regions where the matrix is not conducive for dispersal. There are three types of matrix around the moist broadleaf forest fragments in Uganda: subsistence agroforestry; plantation agriculture; and, for forest fragments on islands in Lake Victoria, water. We predict that the structurally-diverse agroforestry matrix will be more conducive for dispersal than the structurally-simple plantation agriculture or water, and that this difference will be detectable in the species richness of different forests.

Species richness of several indicator groups for biodiversity and high conservation value in Uganda's moist broadleaf forests.

We will collect data on tree, bird, mammal, and butterfly species richness using standard field sampling methods for each of these groups. Fragments of different size and situated in three different matrix landscapes will be studied; agroforestry, plantation agriculture and islands on Lake Victoria. Existing data on species richness in the forests will be collated for comparison with the new data. Data on the history of the forests will be acquired by analysing early remote sensing data (e.g. the CORONA spy satellite) and from historical data and documentary evidence from various sources.

 Data analyses.

We will test if there is a significant SAR for each taxonomic group and if there is a difference in the gradient of the SAR between forests surrounded by the different matrix types. The concern of scientists not to release research data can be a constraint for those who could use these data in the development of conservation policy and implementation. To achieve the full benefits of the project, we will deposit data generated by this project in appropriate international data centres as soon as possible after publication.

 Protection and conservation.

One way to protect the biodiversity threatened by human activity is to give special protection to certain areas, as a National Park or other form of protection. One can also adopt rules that forbid certain kinds of activity in specified areas or certain kinds of use of land. The legal questions will then be to interpret and implement the rules. Another possibility is to try to develop a property regime that gives the stakeholders an interest in protecting the biodiversity themselves. The basic question then is if and how different legal rules, system of ownership and property-structure influence the effect of agricultural activities on biodiversity.

In the well known article on “The Tragedy of the Commons” Hardin argues that common resources tend to be overexploited because stakeholders lack incentives to manage the resources sustainably. One possible solution to this is to establish individual ownership. Others have argued that a management system can be established in commonly held areas and that this will prevent overexploitation more effectively than individual ownership.

It has also been argued that the only way to bring about a sustainable development is to re-establish old custom-based systems based on long experience of how different activities influence the environment and the biodiversity.

It is of great interest to find examples where one can try to verify whether ownership structure influences the protection or destruction of biodiversity in an area. It would be valuable to combine the legal research with research on biodiversity made by natural scientists.

In this task, two Masters students will analyse the current property system of the area, try to evaluate the effects the property system has on the biodiversity based on the results achieved by the natural scientists, and discuss alternative systems and custom-based systems that might have been found in the area.

Project partners and roles:

• Ecological and Environmental Change Research Group (EECRG), Department of Biology, University of Bergen. The project leaders are Vigdis Vandvik and Richard Telford who will have responsibility for data management, analyses and project integration, and supervise the biology Masters students.
• Makerere University. Josephine Esaete will lead the project from the Ugandan side, and have main responsibility for taxonomic group selection, site selection, data acquisition, coordination of field work, and facilitation of information transfer. Two PhD students are working on the project: Perpetra Akite and Collins Bulafu, based at Makerere University,
• Ugandan National Environment Management Authority, NEMA. Ms Mary Kitutu will assist with logistical support, and will be responsible for communicating of implications of research to Ugandan policy makers.
• Faculty of Law, University of Bergen. Ernst Nordtveit will have responsibility for investigating the legal framework for biodiversity in forest fragments and the matrix. He will supervise the law Masters students.

MSc theses:

Therese Kronstad: The value of forest matrix habitats for conservation: Butterfly distribution on a land-use gradient from mature forest to small-scale agriculture in Mabira Forest Reserve, Uganda
Kristoffer Hauge: Bat activity and community composition in contrasting agricultural landscapes and the adjacent Budongo forest reserve, Uganda
Ingrid Sandberg: Management of Public Forests, Uganda
Helge Haugland:

Forest fragment in a tea plantation