TREE DIVERSITY, USE AND LAND COVER CHANGES IN Mt. OTZI FOREST RESERVE, NORTHERN UGANDA

By ACANAKWO ERIK FRANCIS BSc. (Forestry) (Mak)

A report submitted to the International Leadership and Capacity Building Bursary (ILCB), Royal Geographical Society with IBG as a requirement for the fulfillment of the terms and conditions for the Research Bursary received September 2010

1.0 BACKGROUND Forests in Africa have been relied upon as a source of livelihood to provide both wood and non wood forest products, yet their value is gradually declining due to changes in the land cover (Murphy & Lugo, 1986, Dovie et al., 2002, Nampindo et al., 2005). The changes are largely caused by the people living in and around the forests. In arid and semiarid areas, variations in climate as well as human activities have been the primary factors leading to land and environmental degradation. Trends have been observed to be high in tropical dry forests (Murphy & Lugo, 1986). The dry forests are characterized by slower growth rates with more restricted reproductive episodes which make them susceptible to disturbance (Kalacska et al., 2004). The challenge with biodiversity loss is that it corresponds to the irreversible loss of the potentially valuable genetic libraries and existence values. It also affects the local service of the ecosystem (Myers, 1997). In Uganda where the population rate is growing at a rate of 3.2% according to the 2002 census (UBOS, 2002), a high proportion of the vegetation is lost due to burning, settlement and grazing (Winterbottom & Eilu, 2006). The remaining natural vegetation is limited to protected areas where they have some protection from human activities and disturbances. The Northern part of Uganda however had a collapse of the system of governance due to the over two decade long civil strife that led to the abandoning of the forest reserves and the protected areas (Winterbottom & Eilu 2006). The unmonitored and unregulated utilization of the forest reserves might have then resulted in the degradation of the forest reserves in the North of Uganda.

1.1 Statement of the problem Mt. Otzi Forest Reserve is categorized as a prime conservation forest with some unique species of flora and fauna not known in any other protected areas of Uganda (Davenport et al., 1993; Uganda Forest Nature Conservation Master Plan, 2002). However Mt. Otzi Forest Reserve is located in the Northern part of Uganda that suffered civil unrest for over two decades at the hands of rebel outfits that operated within and out of Uganda, in the neighbouring Southern Sudan. The environment of insecurity led to difficulty for managers of the forest reserve to be effective custodians of the forest resource. The failure to ensure patrols and management of the forest reserve resulted in unregulated exploitation of the forest resources. The effect of the local people’s utilization of the Mt. Otzi Forest Reserve was not known as no studies had been carried out to assess diversity, use and land cover change. There was need therefore to obtain accurate and reliable information to plan for effective management and monitoring of the forest reserve.

1.2 Objectives The general objective of the study was to assess the effect of forest use by the local people on tree species diversity and the resultant land cover change. The specific objectives were; 1. To assess spatial variation in tree diversity, size class distribution and stem density between the Eastern and Western arms of the forest 2. To establish the value and use of the forest reserve to the community 3. To assess land cover changes within and around the forest reserve in Metu Sub-county for the time period between 1986 and 2004 1.3 Hypotheses H0: There is no difference in tree diversity in the two arms of Mt. Otzi Forest Reserve H1: The people in and around the forest depend entirely on the forest resource as a source of livelihood. 1.4 Justification The data generated will help in the development of a work plan and management plan for Mt.Otzi Forest Reserve since none has been made before. Information relating to structural characteristics of the woody species as well as the anthropogenic activities could be used as baseline data to monitor changes of the forest eco-system. 2.0 MATERIALS AND METHODS The study was divided into three parts of forest survey, social survey and remote sensing. Mt. Otzi Forest Reserve is situated in Metu County in the North of Moyo District (Figure 1). The forest is located between 03035’-03049’N and 31047’-31057’E. The relief of the area is characterized by a rugged terrain with an altitudinal range of 760-1667m above sea level and rolling hills along the River Nile, rising in a series of hills and peaks in the North and North-east. The reserve is bounded by the international border with Sudan in the North east and overlooks the confluence of the Achwa River and White Nile as it flows into Sudan. The geology of the area is composed of schist, quartzite and marble which occur in the mountains surrounding Metu County (SOME, 1994). There has been evidence of the presence of Chimpanzees (Pan troglodytes) in Metu Sub-county in addition to other smaller animals. The flora of the area is mainly dominated by perennial grass e.g. Andropogon species, while the tree species are represented mainly by a mixture of Butyrospermum, Combretum and Terminalia in the grass land. Rainfall in Moyo is divided into 2 peaks of April and August-October (UPPA, 2000) with a total of 1266.8mm of rainfall received annually. The dominant agricultural system is the traditional subsistence farming.

Figure 1: Location of Mt. Otzi Forest Reserve in Uganda and within West Moyo County

2.1 Methods Selection of study area The study was carried out in Mt. Otzi Forest Reserve and in the surrounding areas located within Metu Sub-county. Mt. Otzi Forest Reserve area was divided into 2 parts, the Western and Eastern arms. The Western arm had steep gorges with a highly rugged terrain. The Eastern arm on the other hand has rolling hills rising in a series of hills and peaks. The western arm is not easily accessible due to the rugged terrain, deep and steep gorges, while the Eastern arm is relatively accessible by both animals and humans. 2.1.1 Forest survey Plot layout The study area in the Forest survey was divided into 4km2 grid plots using the Hawth’s tools, an add-in tool to the ArcGIS version 9.2 software. Three sample plots were selected at the gridline intersection and their Universal Transverse Mercator (UTM) coordinates recorded using a Garmin 12XL GPS Unit. 32 grid intersections were selected, each having 3 sample plots, making a total of 96 sample plots for the forest survey. The sample plots were located in the field using topographic maps and a GPS unit. Plots of 50 X 100m were established for the study within the forest reserve. Tree Variables recorded In each of the identified plots, all the trees with a diameter at breast height (DBH), measured at 1.3m, equal to or greater than 10cm that occurred in the plot were enumerated. For trees that forked below 1.3m, each stem that had a DBH equal to or greater than 10cm was considered as an individual tree. DBH was measured using a tree caliper. Recordings of DBH were organized into diameter size classes with a class interval range of 5.5cm. All trees encountered were identified by their local names after examination of their morphological features such as leaves, fruit, bark, exudates and flowers. Samples of leaves, fruit and flowers from each tree species encountered for the first time were collected tagged and pressed, later taken to the herbarium in Makerere University for confirmation of the plant identities by the curator. 2.1.2 Social survey The survey was undertaken to assess the local people’s utilization of the tree species, frequency of exploitation of the forest resources. Data were collected using a combination of questionnaires, administration of one-to-one interviews, Focus Group Discussions and observational methods. 50 respondents in each of the four parishes were selected through a

systematic random sampling of households in the different villages that make up the parish, while groups of 10 people were organized for the Focus Group Discussions. 2.1.3 Assessment of Land Use/Cover change Data source Data for the assessment of the land use/ cover change was obtained from satellite images downloaded from the internet. Mt. Otzi Forest Reserve occurs within the image path 172 and row 052 of the Landsat grid. Two satellite images; one for 1986 and 2004 were obtained from the Global Land Cover Facility (GLCF), an Earth Science Data Interface, University of Maryland link: http://glcf.umiacs.umd.edu. A 1995 land use map for the forest reserve was obtained from the National Forestry Authority (NFA). The shape files of Mt. Otzi Forest Reserve, with the different attributes of the area were obtained from the wildlife Conservation Society (WCS) Uganda, Kampala office (Table 1). Table 1 shows details of data source used for the remote sensing assessment. Table 1: Source of Data for Remote Sensing Data type

Production date

Scale

Source

Landsat TM

1986-01-17

30m

GLCF

Landsat ETM+

2004-11-10

30m

GLCF

Land use/cover map

1995

1:466,792 (View scale)

NFA

1:466,792 (View scale)

WCS

Shape files of the study area

Remote Sensing Data Management The Metu Sub-county vector map was obtained by clipping out Metu Sub-county and its attribute features, producing vector maps of the study area. The downloaded satellite images were then merged to form a false colour composite image. Geo-referencing was done to synchronise the identifiable features on the image with those on the vector map to allow for accurate analysis. The area of study was extracted through the process of square extraction. This helped to make the extent of the image smaller, thus enhancing faster analysis of the image while using the Arc GIS 9.2 software.

The extent of extraction was; Maximum X: 387733.245

Maximum Y: 423889.494

Minimum X: 351827.322

Minimum Y: 396621.070

On-screen digitization was done by editing through creation of new polygons on the 1995 land cover map. The process involved studying the satellite images for 1986 and 2004 and creating polygons around the different vegetation types, identified by the different shades and colours. 2.2 Analysis of Data 2.2.1 Forest survey data analysis The Shannon-Wiener index was used to assess the tree species diversity. Margalef’s index was used to determine the species richness in the different plots of the study area, while Pielou J index of evenness was used to determine the species evenness in the different plots. Species Diversity Richness (SDR) IV software was used to carry out all the above analyses. The independent t-test statistics was used to test if there was a significant difference in the diversity by testing difference in species richness and evenness, stem diversity and basal area in the two arms of the study area. 2.2.2 Social survey data analysis Forest use data from questionnaires, FGDs and one-to-one interviews were analyzed by using descriptive statistics. Chi-square tests were used to determine whether there was a relationship between distance from the forest and the livelihood option pursued. 2.2.3 Remote sensing data analysis The overall land cover changes were analysed by adding the 1986 and the 2004 land cover raster maps. The attribute tables of the resulting images were studied to determine the changes in area for the different land use types.

3.0 RESULTS 3.1 Tree species Diversity A total 11,120 individual trees were enumerated belonging to 125 species, 92 genera and 43 families. 20 families were represented by 1 species each, while the rest were represented by 2-11 species. Mimosaceae had the highest number of species (11 species in 6 genera).

The species accumulation curve in Figure 2 showed an increasing trend by the 95th sample plot revealing that there was still an opportunity of adding new species with an increase in sample size.

Figure 2: Tree species accumulation curve for sampling in the forest survey

At 95% confidence Interval, there was a significant difference (P