GROUND COUNTS FOR MEDIUM TO LARGE MAMMALS -LAKE MBURO NATIONAL PARK- 2018

GROUND COUNTS for Medium to Large Mammals in Lake Mburo Conservation Area

Report by F. E. Kisame, F. Wanyama, E. Buhanga and A. Rwetsiba Wildlife Authority 2018 I UGANDA WILDLIFE AUTHORITY

TABLE OF CONTENTS ABSTRACT I 1.0 BACKGROUND 1.1. CONSERVATION HISTORY OF LAKE MBURO II 1.1.1. Pre 1935: The Nkore kingdom ii 1.1.2. 1935 to 1964: Controlled Hunting Areas ii 1.1.3. 1964 to 1986: Lake Mburo Game Reserve ii 1.1.4. Lake Mburo National Park ii 1.1.5. 1987- Today: Lake Mburo National Park 1 1.2. MAMMAL SPECIES THAT HAVE SINCE DISAPPEARED IN LMNP 1 1.3. HUMAN POPULATION 2 1.4. WILD ANIMAL POPULATIONS AND SURVEYS 2 1.5. OBJECTIVES FOR THE SURVEY; 3 2.0 SURVEY AREA AND METHODS 2.1. METHODS 4 2.1.1. Foot -transect survey 4 2.1.1.1. Survey design 4 2.1.1.2. Data collection 4 2.1.2. Road counts 5 2.1.3. count 5 3.0 RESULTS 6 3.1. MAMMAL POPULATION ESTIMATES IN LMNP AND THE RANCHES 6 3.1.1. Estimated Population Combined for LMNP and Ranches 6 3.1.2. Estimated Population in Lake Mburo National Park. 7 3.1.3. Estimated Population in the Ranches 7 3.2. ROAD COUNT ENCOUNTERS 8 3.2.1. Road Verses Transect abundance estimates 9 3.3. HIPPOPOTAMUS SURVEY IN LAKE MBURO WITHIN LMNP 9 3.4. MAMMAL STRUCTURE 10 3.4.1. Mammal structure in Lake Mburo National Park (Block 1) 10 3.4.2. Mammal structure in the ranches (Block 2 and 3 Combined) 11 3.5. DISTRIBUTION 11 3.5.1. Distribution of mammals in LMNP and the ranches outside the park 11 3.5.2. Carcass encounters and their distribution in LMNP and the ranches 14 3.5.2.1. Carcass distribution in LMNP and the Ranches 16 3.5.3. Cattle in Lake Mburo National Park 17 3.5.4. Poaching in the ranches adjacent to Lake Mburo National Park 17 GROUND COUNTS FOR MEDIUM TO LARGE MAMMALS -LAKE MBURO NATIONAL PARK- 2018

5.0 CONCLUSION AND RECOMMENDATIONS 6.0 REFERENCES 22 7.0 APPENDICES 23 7.5.1. Appendix I. Some of the survey team members 23 7.5.2. Appendix II: One of the training sessions before the census 23 7.5.3. Appendix III: Some of the teams practicing use of GPS during the training 24 7.5.4. Appendix IV: Briefing teams at the end of the training 24 FIGURES Figure 1. Ground survey map for LMCA (November, 2016) 4 Figure 2. Species encounter rate during road counts in Lake Mburo National Park. 8 Figure 3. A regression for species encounters on road and transect counts 9 Figure 4. Estimated Hippopotamus population trends in Lake Mburo (2009-2016) 9 Figure 5. Mammal structure in Lake mburo National Park 10 Figure 6. Mammal structure in the ranches (Block 2 and 3 Combined) 11 Figure 7. Distribution of Buffalo, bushbuck, eland and impala in LMNP and ranches 12 Figure 8. Distribution of Bush duiker, baboon, and in LMNP and ranches 13 Figure 9. Distribution of Reedbuck, warthog, and waterbuck in LMNP and ranches 14 Figure 10. Carcass encounters in Lake Mburo National Park and the ranches 15 Figure 11. Carcasses of Zebra, buffalo, impala, waterbuck and warthog in LMNP and the ranches 16 Figure 12. Cattle distribution during the survey in LMNP 17 Figure 13. Poacher arrested with nets and a bushbuck carcass. 17 Figure 14. Species population estimates for 2014 and 2016 in LMNP 18 Figure 15. Cattle in one of the containment areas (Kraal) in LMNP 20 TABLES Table 1: Total population estimates for wild animals in LMNP and the ranches (Nov, 2016) 6 Table 2: Total population estimates for wild animals in LMNP and the ranches (August 2014) 6 Table 3: Population estimates for wild animals in LMNP (November, 2016) 7 Table 4: Population estimates for wild animals in the ranches (November, 2016) 7 Table 5: Species encounters during the road count 8 PLATES Plate 1. Survey team during the boat census in Lake Mburo National Park 5 UGANDA WILDLIFE AUTHORITY

ABSTRACT Lake Mburo National Park is an important biodiversity area with a large number of animals that currently survive in a habitat dominated by the invasive acacia woodland. The survey was to generate information on the population of medium to large mammals as well as their distribution patterns in LMNP and the ranches. This was to allow for further assessment of the effectiveness of current conservation strategies in LMCA. The survey was a ground count carried out towards the end of the dry season in November 2016. The transects used during the survey were designed using programme DISTANCE. The analysis of the results was done using by program DISTANCE also. The survey results showed that the Impala were the dominant mammal species in Lake Mburo landscape with a population of (21,240) individuals followed by zebra (19,589), waterbuck (4,983), warthog (3,083), eland (1,482), buffalo (1,479), bushbuck (1,239) and topi (515) in the order of reducing species population. We recorded 271 Hip- popotamus on lake mburo also. The adult females were more than the adult males in the mburo landscape. The Linear Regression Analysis showed a very strong relationship (R2=80.6%) between the population values for mammals obtained using transects during ground counts and the mammal encounter values obtained through a road count. On the other hand, Lake Mburo National Park had the highest number of mammal carcasses with carcasses of Impala, zebra, waterbuck, buffalo, warthog and hippopotamus being recorded whereas, the only carcass recorded in the ranches was for a zebra in block 2-the ranch. We arrested some individuals suspected to be poachers after they were found with hunting nets and a bushbuck carcass later recorded at police as exhibits. The change in the survey design from a low sampling intensity to a high sampling intensity provided better ground information on species numbers. As a result more species sightings were made than in earlier surveys due to change in the sampling intensity. There is need to encourage conservation management enterprises that improve relations with the communities and minimize human-wildlife conflicts. There is need for translocation of and impala from the ranches to other protected areas such as, Pian Upe and Katonga Wildlife Reserves to increase their potential as well as minimizing the human-wildlife conflicts being caused by the zebras and impalas in the ranches. There is need to construct more water holes/dams that can store water long enough while ensuring adequate water supply at all times within the park. To improve the wildlife habitat, there is need to prioritize invasive species management.

i GROUND COUNTS FOR MEDIUM TO LARGE MAMMALS -LAKE MBURO NATIONAL PARK- 2018

1.0 BACKGROUND 1.1. CONSERVATION HISTORY OF LAKE MBURO Lake Mburo and its surrounding country side form part of an important conservation area. Over the years this area has been through many traumatic changes and the extent of the land under formal protection decreased dramatically. Wildlife and human numbers fluctuated significantly as a result of natural growth, disease, civil war and political unrest and many of the challenges facing the park today arose from issues that occurred during this troubled history.

1.1.1. PRE 1935: THE NKORE KINGDOM Traditionally this entire area, from Tanzania in the south to the Katonga River in the north, was the territory of the Banyankole people and was ruled over by the Omuabe, the king. Access to the Nshaara area around the lakes and swamps was restricted by the king and people were only allowed to live here or graze their cattle here with his special permission. Whilst wildlife was protected, carnivores such as and which attacked livestock were killed. At the turn of the century the area was struck by rinderpest and smallpox and many people, livestock and wildlife died.

1.1.2. 1935 TO 1964: CONTROLLED HUNTING AREAS Throughout the early part of the 20th century the area slowly recovered and wildlife populations increased again. Human settlement escalated and hunting became a popular past time amongst the wealthy and influential. In 1935 in order to-protect the game the protectorate Government declared two adjoining controlled Hunting Areas, lake Mburo and Masha. Although hunting was regulated in these areas, grazing, farming and fishing was allowed to continue as before. In the 1940’s, a second disaster struck with a wave of sleeping sickness and Nagana carried by tsetse fly. Many people left the area with their cattle, leaving behind only a few farmers and fishermen. The tsetse fly eradication programme which followed, drastically reduced game populations.

1.1.3. 1964 TO 1986: LAKE MBURO GAME RESERVE By the 1960’s the tsetse fly eradication programme was complete opening up the area to cattle once again. The consequent increase in use and settlement by farmers and pastoralists prompted the newly independent gov- ernment to rationalize land use. The area was divided into a commercial cattle ranching scheme to the north and the lake Mburo Game Reserve to the south. The Masha Controlled Hunting Area, which was by now well populated, was degazetted. Families living within the reserve were permitted to remain and farm, but pastoralists displaced by the ranching scheme joined the increasing numbers of farmers entering the poorly controlled reserve. In 1969, a large block of land was cut from the reserve to form the Nshaara Dairy Cross Breeding Ranch.

1.1.4. LAKE MBURO NATIONAL PARK In 1983, the Uganda Government, worried by the increased settlement and heavy grazing pressure of the area, created the Lake Mburo National Park, establishing its boundaries as those of the original game reserve. All families living in the area were evicted without compensation and forcefully pushed from their homes. The park opened to the public in 1984; when the government weakened and finally fell, the evicted families returned and joined by others in search of land they drove out the park staff and killed much of the wildlife. By 1986 the entire park had been re-occupied

ii UGANDA WILDLIFE AUTHORITY

1.1.5. 1987- TODAY: LAKE MBURO NATIONAL PARK In 1987 the government saw the need to regain control of the park. Aware of the injustices of the past it consulted the local people and their leaders about the park and following these discussions decided to reduce the park’s area by 60% and handed that land back to the people. Farmers and pastoralists living in the 40% that remained as park were given permission to stay until the government found them alternative land. Fishing on the lake Mburo was controlled but allowed to continue. Under these new conditions wildlife populations were increasing due to the reduced human impact on the area. Park authorities and local communities now work closely together to conserve Uganda’s natural heritage without conflict or confrontation. The park is located entirely in Nyabushozi County which is part of in south western Uganda. It lies between the towns of present Kiruhura District, Mbarara and Lyantonde just to the south of the Mbarara- highway. It borders Rugaaga, Kabingo, Sanga, Nyakashashara, Rubaya, Masha and Kanyaryeru sub-counties. It can be accessed by road about a 3.5 hour drive from Kampala and a left turn at the Nshara gate or right turn at the Sanga gate. Lake Mburo National Park (LMNP) is one of the smallest parks in Uganda with an area of 370km2 lying in a rain shadow between Lake Victoria and the Ruwenzori mountains. The area receives about 800mm of rainfall in two peaks: long rains in February to June and short rains from September to December. The rains are rather erratic and unpredictable but most rain tends to fall in April and November. The average daily temperature is about 27.50c with a minimum of 21.50c and a maximum of 34oc .July and August are the hottest months. The park lies between 1219m and 1828m above sea level. High hills, rocky eroded ridges on the western part where deep valleys support dry forest found nowhere else; the rolling, wooded hills are intersected by wide, flat-bottomed valleys which are seasonally flooded and drain into the swamps and lakes. The Ruizi River flows in a south-easterly direction and forms part of the western boundary of the park. Lake Mburo and its wetland eventually drain into Lake Victoria. 1.2. MAMMAL SPECIES THAT HAVE SINCE DISAPPEARED IN LMNP Lake Mburo National Park is home to many medium sized and large Mammalia species. However, some of the mammal species have not been seen in a long time and these include; Giant forest hogs: Giant forest hogs lived in the forest and bush around the lakes and along the river but were shot to extinction during the tsetse fly eradication programme in the 1950’s and 1960’s. African hunting dog: Loss of the vast open spaces over which the African hunting dog ranged combined with ruthless poisoning and shooting led to their extinction from the park. Elephants: Elephants used to pass through the park area but were cut off by increasing human settlement and have not been seen since the 1940’s. Lions: Lions shot and poisoned by ranchers and pastoralists to protect their livestock, were finally exterminated in the early 1980’s. Recently monitoring records have confirmed the existence of a in the park. Lake Mburo National Park witnessed tremendous changes over the years in terms of conservation. Uganda Wild- life Authority regularly carries out censuses of its protected areas in Uganda aiming at ascertaining how well management efforts are contributing to the well being of the habitats where most of the wildlife is resident. Much as LMCA witnessed many traumatic changes that led to the dramatic reduction of the land under formal protec- tion, these challenges have continued to haunt the park even today. The animal population has increased over the years, but the area under protection has remained the same size. The human population around the park has also increased. This is causing conflicts as the wild animals continue

1 GROUND COUNTS FOR MEDIUM TO LARGE MAMMALS -LAKE MBURO NATIONAL PARK- 2018

to interact with the people within this landscape. A number of developments have taken place around the park that range from settlements to farming and fencing off the ranches. This has complicated wild animals’ move- ment through this landscape, and has created mixed feelings among wildlife conservationist today on how best to manage wildlife in this complex environment; aware that animals do not recognize boundaries. 1.3. HUMAN POPULATION The park is surrounded by a high human population comprising of the Banyankole- a dominant ethnic group (three-quarters of the population) who include Bairu- traditional cultivators and the Bahima- traditional pasto- ralists. Other groups include the Bakiga and Banyarwanda. Cattle keeping households tend to be large as cows remain the main source of livelihoods – a traditional practice of the Bahima before and since the degazettement of the park in 1983. Before, densities of wildlife in this area was higher as hunting pressure on wildlife was reduced due to the traditional belief of shunning game meat, their favored diet being milk, blood and ghee. Low livestock and human populations as a result of sleeping sickness and rinderpests epidemics in 1800s-1900s, reduced burning and increased vegetation led to significant increase in wildlife populations. 1.4. WILD ANIMAL POPULATIONS AND SURVEYS Wildlife surveys/censuses are key to generating useful information about the status in terms of numbers and distribution of wildlife in the entire ecosystem of LMCA as well as the threats to the species habitat. In LMNP the earlier surveys were aerial and were carried out between 1992 and 2004 when a larger part of LMNP was still open. But with time, a transition from grassland to woodland vegetation became clear. At this time the survey protocol also changed from aerial to ground animal counts to clear the discrepancies in mammal population esti- mates. Ground counts are excellent for obtaining data in small to medium sized areas on population structures, on the seasonal pattern of distribution within different vegetation types and condition of the animals that can not be obtained from the aircraft. Ground counts are therefore ideal for detailed studies in small study areas, their use being only limited when ground access is difficult or when the area covered is very large (Norton-Griffiths 1978). For example, during the aerial surveys impala and other large wild mammals were counted. Only data on impala and zebra were reliable as the animals were evenly distributed. while data on duiker, bushbuck, reedbuck, eland and buffalo were less reliable (Averbeck 2001). Dense vegetation and the behaviour of the counted species made it difficult toount c from the air. While bush- bucks, duikers and reedbuck often rest in bushes and thickets, they can be hardly seen from the air, eland and buffalo are gregarious and most of the animals live in herds. As eland and buffaloes were not so common in the Lake Mburo area it was likely that the observer would either count a herd with many animals or no animals. An aerial survey covers only part of the whole area and by extrapolating from the percentage covered to 100% the population densities are either overestimated or underestimated. In order to monitor population trends in the Lake Mburo area regular ground counts are more cost effective and provide more accurate results for different species than aerial surveys in the Lake Mburo area (Averbeck 2001). Aerial surveys tend to underestimate species population in woodland vegetation due to presence of canopy. Aerial surveys undoubtedly grossly underesti- mate population of some species for example, the Bushbuck’s population density because of its preference for cover and its secretive habits. However, East (1999) noted that the Bushbuck’s tendency to remain concealed probably results in significant undercounting in some ground surveys. It was also important to note that aerial surveys are better at recording animals away from roads but less efficient at detecting individuals or small groups of animals, especially of smaller

2 UGANDA WILDLIFE AUTHORITY

mammals. For example, the survey carried out in 2004 tried to compare the two methods but it is advisable to restrict comparisons to those studies that use the same methodology, and refrain from comparing ground and aerial surveys (Caro 1999). Nevertheless, though there were improvements in species populations with ground animal counts in the subsequent ground surveys from 2004, 2006, 2010, 2012 and 2014 questions continued to arise about the populations of certain species such as, the Eland and buffalo. This preempted thoughts to re-de- sign the previous survey protocol for LMNP; from a low sampling intensity (3.5 km spacing between transects) to a high sampling intensity (2km spacing between the transects) in order to bridge the gap. This survey provides the basis for monitoring the wild animals in Lake Mburo National Park over the years. The current survey was carried out at the end of the dry season in November 2016 in line with the UWA Strategic Plan, the Annual Operation Plan and the general survey programme that require a periodic review on species population size, abundance and distribution. The previous ground count was done in August 2014 on medium and large mammals of Lake Mburo National Park and the exercise was solely funded by Uganda Wildlife Authority. 1.5. OBJECTIVES FOR THE SURVEY; i. To generate population estimates for medium to large mammals in LMNP and the ranches. ii. To generate distribution maps for mammal species in LMNP and the ranches. iii. To provide data for monitoring and further assessments for the effectiveness of current conservation strategies (e.g. sport hunting quota setting). iv. To provide information about the species structure in this landscape.

3 GROUND COUNTS FOR MEDIUM TO LARGE MAMMALS -LAKE MBURO NATIONAL PARK- 2018

2.0 SURVEY AREA AND METHODS The survey area comprised the park and the ranches adjacent to the park. The park and the adjacent land were divided into three main blocks where the recent ground count was carried out: (1) Block 1 - Lake Mburo National Park, (2) Block 2 - The government ranches/private ranches and (3) Block 3 - Private ranches (See figure 1).

Figure 1. Ground survey map for LMCA (November, 2016)

2.1. METHODS

2.1.1. FOOT -TRANSECT SURVEY

2.1.1.1. Survey design The Transects were made using the DISTANCE software. The program calculated the start and end coordinates that were uploaded in the GPS units to aid in navigation during the survey. A total of 118 transects each measuring approximately 4.0 Kilometers long and spaced at 2km intervals were covered (Figure 1). This meant that tran- sects are wholly into one block and do not overlap in the separate blocks. It is assumed that at any given time, the wild animals in the conservation area are evenly distributed. But this is not the reality mainly because of irregular dispersion of animals in relation to availability and distribution of the natural resources thus, implying that the temporal and spatial distribution of animals at any one time is not the same. The stratification shall help appro- priate analysis of raw data depending on number of animals sighted in a particular area. (If the spatial distribution is highly divergent, stratification shall be used during analysis)

2.1.1.2. Data collection 13 survey teams, each headed by a competent staff in using Distance sampling walked along a transect each day to collect data. Waypoints were uploaded on to the GPS to ease identification of the start and end points of each transect during census (Figure 1). Each team collected data in the morning hours between 07:00 AM and 11:30 AM. To ease movement of census teams to the starting point of each transect, given the long distances involved 4 UGANDA WILDLIFE AUTHORITY

vehicles were used. While at the starting point of each transect, the census teams walked quietly in a straight line from north to the south direction with the aid of a compass and a GPS. Each individual or a group of mammals sighted were recorded and their perpendicular distances from each tran- sect to the individual or the centre of the animal group was recorded using a range finder. Binoculars were used to provide for clear observations and estimations of species at a distance. Indirect observations such as footprints, droppings and feces were also recorded as opportunistic observations for determination of wild animal absence or presence in the survey area. Data was then recorded on specially designed data sheets.

2.1.2. ROAD COUNTS Road counts were carried out in the morning between 0830hrs and 1140hrs. A four wheel pick up moving at an average speed of 20-25 Km/hr was used. The observers recorded all species sighted on to the datasheets and also marked coordinate references for the locations where animal observations were made using GPS. A dis- tance of 37 km was covered with an area estimate of about 24.35 square kilometers.

2.1.3. HIPPOPOTAMUS COUNT Appropriate information on hippopotamus numbers is best done using a boat. The census was carried out between 07:52 AM and 10:41 AM for hippopotamus, crocodiles and the rare elusive Sitatunga that inhabit lake- shores/ banks and swamp ecosystems with the help of binoculars for sighting at a distance. A motor boat (for total counts) moving slowly to increase chances of observations and to make navigation possible along the lake- shore was used. The observers recorded all species sighted on to the datasheets and also marked coordinate references for the locations where animal observations were made using GPS.

Plate 1. Survey team during the boat census in Lake Mburo National Park

5 GROUND COUNTS FOR MEDIUM TO LARGE MAMMALS -LAKE MBURO NATIONAL PARK- 2018

3.0 RESULTS 3.1. MAMMAL POPULATION ESTIMATES IN LMNP AND THE RANCHES

3.1.1. ESTIMATED POPULATION COMBINED FOR LMNP AND RANCHES The results of the total population estimate for each species in the National park and the ranches is given (Table 1). From the survey results Impala were the dominant mammal species followed by Zebra, waterbuck, warthog, eland, buffalo, bushbuck and topi in the order of reducing species population

SPECIES GLOBAL TOTAL (LMNP AND RANCHES) D POP. EST. 95% CONFIDENCE LCL UCL Buffalo 1.15 1,479 647 3,379 Bushbuck 0.96 1,239 862 1,780 Eland 1,482 246 8,965 Impala 16.60 21,240 14,448 31,227 Topi 1.50 515 238 1,118 Warthog 9.56 3,083 1,843 5,160 Waterbuck 5.13 4,983 2,933 8,466 Zebra 15.30 19,589 11,563 33,175

Table 1: Total population estimates for wild animals in LMNP and the ranches (Nov, 2016) The table below shows population estimates for wild animals in LMNP and ranches (August 2014). The results for 2014 are provided for comparison purposes.

SPECIES DENSITY POPULATION ESTIMATE 95% CONFIDENCE LIMIT LOWER CL UPPER CL Impala 15.3 20,408 10,649 39,108 Waterbuck 1.6 2,166 915 5,127 Zebra 8.9 11,849 6,827 20,565 Eland 0.6 859 114 6,473 Buffalo 0.8 1,077 140 8,287 Warthogs 0.6 805 323 2,005

Table 2: Total population estimates for wild animals in LMNP and the ranches (August 2014)

6 UGANDA WILDLIFE AUTHORITY

3.1.2. ESTIMATED POPULATION IN LAKE MBURO NATIONAL PARK. The results of the estimate for each mammal species counted in the National park is given (Table 3). SPECIES BLOCK 1( NATIONAL PARK) D POP. EST. SE 95% CONFIDENCE LCL UCL Buffalo 3.5 1,206 607 464 3,132 Bushbuck 1.6 591 143 365 954 Eland 4.3 1,484 780 246 8,965 Impala 30.7 10,556 2,396 6,750 16,505 Topi 1.0 344 156 144 822 Warthog 8.9 3,083 808 1,843 5,160 Waterbuck 10.0 3,730 1,210 1,993 6,981 Zebra 35 12,200 4,302 6,150 24,202

Table 3: Population estimates for wild animals in LMNP (November, 2016)

3.1.3. ESTIMATED POPULATION IN THE RANCHES The results of the estimate for each mammal species counted in the ranches is given (Table 4). SPECIES BLOCK 2 (PRIVATE/GOVERNMENT RANCHES) BLOCK 3 (PRIVATE RANCHES) D POP. EST. SE 95% CONFIDENCE D POP. SE 95% CONFIDENCE LCL UCL EST. LCL UCL Buffalo 0.2 116 92 25 538 0.5 157 120 39 633 Bushbuck 0.8 503 131 310 841 0.4 145 51 72 291 Impala 14.0 9,365 2,660 5,382 16,297 4.2 1,320 506 621 2,802 Topi 0.5 311 232 81 1,190 Warthog Waterbuck 2.0 1,253 624 491 3,199 Zebra 9.9 6,224 2,995 2,506 15,459 3.7 1,162 809 321 4,208

Table 4: Population estimates for wild animals in the ranches (November, 2016)

7 GROUND COUNTS FOR MEDIUM TO LARGE MAMMALS -LAKE MBURO NATIONAL PARK- 2018

3.2. ROAD COUNT ENCOUNTERS The results from road count encounters are given (Table 5). 348 Impala individuals were recorded in 2016 co mpared to 142 individuals encountered in 2014 during the road counts.

SPECIES ENCOUNTERS Buffalo 83 Eland 126 Impala 348 Topi 49 Warthog 67 Waterbuck 64 Zebra 256 Bushbuck 3 Table 5: Species encounters during the road count We calculated the number of individual species encountered per kilometer of road covered in Lake Mburo National Park and expressed the value as an abundance index in this case. The result from this is given (Figure 2)

Figure 2. Species encounter rate during road counts in Lake Mburo National Park.

8 UGANDA WILDLIFE AUTHORITY

3.2.1. ROAD VERSES TRANSECT ABUNDANCE ESTIMATES A regression for species encounters (Road count) and Population estimate (Transect counts) is given (Figure 3) showing that a statistically significant relationship exists between the two values. In other words, there is a very strong positive relationship (R2=80.6%) between the species population values obtained during ground counts and the encounter values obtained during road counts.

Figure 3. A regression for species encounters on road and transect counts

3.3. HIPPOPOTAMUS SURVEY IN LAKE MBURO WITHIN LMNP The current estimated hippopotamus abundance in Lake Mburo during the survey was 271 individuals registering a slight increase from the 255 individuals recorded in 2014. The result for the hippopotamus population recorded over the years in LMNP is given (Figure 4).

Figure 4. Estimated Hippopotamus population trends in Lake Mburo (2009-2016)

9 GROUND COUNTS FOR MEDIUM TO LARGE MAMMALS -LAKE MBURO NATIONAL PARK- 2018

3.4. MAMMAL STRUCTURE

3.4.1. MAMMAL STRUCTURE IN LAKE MBURO NATIONAL PARK (BLOCK 1) The following mammals were encountered and their numbers recorded they include; Impala (822), zebra (715), buffalo (254), eland (238), waterbuck (208), baboon (2002), Warthog (152), Vervets (49), topi (29), bushbuck (28), bushduiker (6), oribi (1) and reedbuck (1) in order of reducing numbers. The number of adults (male or female), juvenile and infants that were identified and number of unidentified based on age or sex as observed from the ground mammal counts in Lake Mburo National Park is given (Figure 5). The adult females in the park were more than the adult males.

Figure 5. Mammal structure in Lake mburo National Park

10 UGANDA WILDLIFE AUTHORITY

3.4.2. MAMMAL STRUCTURE IN THE RANCHES (BLOCK 2 AND 3 COMBINED) The number of adults (male or female), juvenile and infants that were identified and number of unidentified based on age or sex as observed from the ground mammal counts in the ranches is given (Figure 6). The adult females were more than the adult males in the ranches also.

Figure 6. Mammal structure in the ranches (Block 2 and 3 Combined)

3.5. DISTRIBUTION

3.5.1. DISTRIBUTION OF MAMMALS IN LMNP AND THE RANCHES OUTSIDE THE PARK Depending on their history and their dispersal capabilities, species can either occupy large areas or be restricted to small regions. The fact that human activities are constantly altering areas of species distribution, there was need to understand the distribution of species for monitoring purposes. The distribution of species as recorded during the survey is given (Figures 7,8 and 9)

11 GROUND COUNTS FOR MEDIUM TO LARGE MAMMALS -LAKE MBURO NATIONAL PARK- 2018

Figure 7. Distribution of Buffalo, bushbuck, eland and impala in LMNP and ranches

12 UGANDA WILDLIFE AUTHORITY

Figure 8. Distribution of Bush duiker, baboon, topi and oribi in LMNP and ranches

13 GROUND COUNTS FOR MEDIUM TO LARGE MAMMALS -LAKE MBURO NATIONAL PARK- 2018

3.5.2. CARCASS ENCOUNTERS AND THEIR DISTRIBUTION IN LMNP AND THE RANCHES The result of mammal carcass encounters during the survey is given. Most of the carcasses for Impala, Warthog, Zebra, Waterbuck, Buffalo and Hippopotamus were found in the park and only a carcass of Zebra was found in block 2 ranch (Figure 10).

Figure 9. Distribution of Reedbuck, warthog, zebra and waterbuck in LMNP and ranches

14 UGANDA WILDLIFE AUTHORITY

Figure 10. Carcass encounters in Lake Mburo National Park and the ranches

15 GROUND COUNTS FOR MEDIUM TO LARGE MAMMALS -LAKE MBURO NATIONAL PARK- 2018

3.5.2.1. Carcass distribution in LMNP and the Ranches The carcasses sighted were recorded and their spatial distribution given in (Figure 11).

Figure 11. Carcasses of Zebra, buffalo, impala, waterbuck and warthog in LMNP and the ranches

16 UGANDA WILDLIFE AUTHORITY

3.5.3. CATTLE IN LAKE MBURO NATIONAL PARK The cattle sighted in the park was recorded and the distribution is given (Figure 12)

Figure 12. Cattle distribution during the survey in LMNP

3.5.4. POACHING IN THE RANCHES ADJACENT TO LAKE MBURO NATIONAL PARK The nets used for poaching and a carcass of a bushbuck confiscated during the census period. Poachers were arrested and brought to sanga police station with the exhibits (see Figure 13.).

Figure 13. Poacher arrested with nets and a bushbuck carcass.

17 GROUND COUNTS FOR MEDIUM TO LARGE MAMMALS -LAKE MBURO NATIONAL PARK- 2018

4.0. DISCUSSION From the survey results Impala were the dominant mammal species followed by Zebra, Waterbuck, Warthog, Eland, Buffalo, Bushbuck and Topi in the order of reducing species population. That is there were more mammals in the park (block 1) followed by block 2-the Government ranches and then block 3-the private ranches in order of reducing numbers. In summary, most of the mammal species were found in the park than in the ranches (Table 2 and 3). Their distribution equally followed the same trend (Figures 7, 8 and 9). Distribution and abundance of wild animals tend to be higher inside than outside the park due to intensive law enforcement inside the park than in the ranches (Rwetsiba and Tumwesigye 2004). In 2014 survey, the topi and bushbuck numbers were so small for the analysis using program Distance and as such the results were discarded. A carcass for topi was also observed during the road count in 2014. Similarly, the current survey recorded sightings for oribi and reedbuck but the numbers were very small for the analysis in program DISTANCE. From the landscape perspective (Figure 14), there was a general increase in species sightings in 2016 when com- pared to 2014 using the same line transect method (Table 2). Secondly, Topi and Bushbucks were recorded in 2016 compared to the 2014 survey where no recording was made (Figure 14). The trend in observation and the variations thereafter, could be attributed to the increase in the sampling intensity as a result of the change in the survey design and the seasonality. However, the seasonality effect was not felt on the ground as the month of november-the survey period was relatively dry. Therefore, as to whether the seasonality had an influence on species sighting is an issue for discussion. Comparing the densities, the 2016 results showed that Impalas and zebra had the highest densities among the species. This implies that impala and zebra were more adaptable to the landscape showing relativley high com- petion for space in the mburo landscape than any other species (Figure 14) hence, the densities of 16.60/km2 and 15.3/km2 respectively.

Figure 14. Species population estimates for 2014 and 2016 in LMNP

We recorded animal carcasses encountered during the survey as opportunistic observations. Most of the car- casses were recorded in the park and the only recorded carcass in the ranches was that of a zebra in block 2

18 UGANDA WILDLIFE AUTHORITY

(Figure 10 and 11). The carcasses could have been part of the carcasses left behind during the drought. Between April and October 2016, LMNP experienced a prolonged drought that led to the death of wild animals in the park. Prior to the survey some of the carcasses had been disposed off. Whether the dry spell had a negative impact on the species population, could not easily be established from the results however, it remains a subject for future research. Block 3 (the private ranch) was heavily settled and the observed mammals were few in number compared to block 3 (Government/private ranch) and (block 1)-the national park. There could be a negative correlation between mammal presence and the settlement patterns observed in the ranches in Block 3. For example, Ogutu et al, 2010 noted that Settlement activities and daytime livestock watering provide a push that keep wildlife away. This could be true for the observed trend in block 3. Poaching is still a challenge in Lake Mburo landscape (Figure 13). Poaching is known for its negative impact on animal numbers. As evidenced in PUWR, the uncontrolled human activities are the limiting factors in the survival of large ungulates (Kisame 2014). Poaching during the time of insurgency led to a decline in numbers of species (Lamprey and Michelmore 1996). The drivers to the poaching vice could be due to poverty and partly because the meat is a cheap source for animal proteins. Weber et al (2001) noted that the search for additional protein sources has exercabated poaching activities in the reserve. Andy (2013) observed that the removal of fauna from an area due to poaching flows from the immediate impact of killing an existing animal, the medium term effect of reducing breeding numbers and hence the rate of reproduction, and the long term effects of thinning the gene pool and the symbiotic- and often irreversible – impact this has on overall biodiversity. Andy (2013) further notes that in additional to the drop of animal populations, poaching affects ecosystems. Natural ecosystems often develop a rather delicate balance between different types of fauna and their local habitat. An estimated 348 heads of cattle were encountered in the park during the ground animal count (Figure 12). In 2014, 80 heads of cattle were encountered in the park during road count. Over the years the park management has been struggling to remove cattle from the park but this has continued to be a challenge. For example, the cattle containment areas were set up in LMNP and cattle found grazing in the park is driven to the containment areas for further management as exhibits (Figure 15) but cattle presence still manifests itself in the park. Domes- tication of cattle in wildlife protected areas is condemned in the UWA Act Cap 200 as a threat to conservation. Mishara et al (2003) also noted that the growing numbers in livestock populations create an overlap of diets and forage competition with wild herbivores, resulting in overgrazing and decline or local extinction in wild herbivore populations.

19 GROUND COUNTS FOR MEDIUM TO LARGE MAMMALS -LAKE MBURO NATIONAL PARK- 2018

Figure 15. Cattle in one of the containment areas (Kraal) in LMNP

20 UGANDA WILDLIFE AUTHORITY

5.0 CONCLUSION AND RECOMMENDATIONS Lake Mburo National Park is an important conservation area that boasts of large numbers of animals that survive in a challenging habitat currently dominated by the invasive acacia woodland. The re-designing of the survey provided good information for management compared to earlier designs that had a low sampling intensity and will provide the basis for future surveys in LMCA. On the other hand due to intensive law enforcement inside the park other than in the ranches, tend to make distribution and abundance of wild animals to be higher inside than outside the park. The following conservation measures are recommended: 1. Encourage conservation management enterprises that improve relations with the communities and mini- mize human-wildlife conflicts. 2. Construct more water holes (dams) that can store water long enough while ensuring adequate water supply at all times within the park. 3. Prioritize invasive species management in the park to improve wildlife habitat. 4. Translocate zebras and impala to other Protected Areas such as, PUWR and Katonga Wildlife Reserves whose populations on the ranches has grown to levels as those in the park to minimize Human-Wildlife Conflicts being caused by the increasing animal populations in the ranches. 5. Provide timely intelligence information to avert illegal activities before they happen in Lake mburo land- scape. 6. Strengthen community awareness protocols in LMCA.

21 GROUND COUNTS FOR MEDIUM TO LARGE MAMMALS -LAKE MBURO NATIONAL PARK- 2018

6.0 REFERENCES

Andy (2013). The problem of poaching. International anti-poaching foundation. Available at https// the problem of poaching.wordpress.com downloaded on 17 January, 2017 at 1700hrs. Averbeck, C (2001). Population ecology of Impala (Aepyceros melampus) and community- based wildlife . Dissertation wurde am 8.11.2001 bei der Technischen Universität München eingereicht. Buckland, S.T., Anderson, D.R., Burnham, K.P., & Laake, J.K (1993): Distance Sampling: Estimating Abundance of Biological Populations. Chapman & Hall, London & New York. Caro, T.M (1999). Conservation monitoring: estimating mammal densities in woodland habitats. Animal Conservation 2:305-315 East. R (1999). In: IUCN SSC Antelope Specialist Group. 2016. Tragelaphus scriptus (errata version published in 2017). The IUCN Red List of Threatened Species 2016: e.T22051A115165242. http://dx.doi. org/10.2305/IUCN.UK.20163.RLTS.T22051A50196111.en.Downloaded on 23 November 2017. Kisame F. E and Wanyama. F (2014) Sample Counts of Medium – Large Mammals in Lake Mburo Conservation Area, Uganda Ecological Monitoring and Research Unit. Unpublished Report to Uganda Wildlife Authority, P.O. Box 3530 KAMPALA Kisame, F.E (2014). The status and distribution of large ungulates in Pian Upe Wildlife Reserve, Uganda. Un published MSc. Dissertation, Makerere University, Kampala. Lamprey, R. H. & Michelmore, F. (1996). Surveys of Protected Areas Phase 1: Ministry of Tourism, Wildlife and Antiquities, Kampala. Mishra, C., Allen, P., McCarthy, T., Madhusadan, M., Bayarjargal, A., Prins, H. (2003). The Role of incentive Programmes in conserving the snow . Conservation Biology 17: 512-520. Norton-Griffiths, M (1978) Counting animals. Handbook No. 1: Serengeti ecological monitoring programme. Second edition. African Wildlife Foundation, Afropress. Nairobi, Kenya. Ogutu, J. O., Piepho, H. P., Reid, R. S., Rainy, M. E., Kruska, R. L., Worden, J. S., Nyabenge, M., Hobbs,N. T. (2010). “Large herbivore responses to water and Settlements in savannas” Ecological monographs 80 (2):241-266. Rwetsiba. A and Tumwesigye. C (2004). Aerial and Ground sample counts of medium–large mammals in lakr mburo conservation area, Uganda. Uganda Wildlife Authority, P.O Box 3530, KAMPALA. Wanyama. F and Kisame F. E (2012) Sample Counts of Medium – Large Mammals in Lake Mburo Conservation Area, Uganda Ecological Monitoring and Research Unit. Unpublished Report to Uganda Wildlife Authority, P.O. Box 3530 KAMPALA Weber, W., Lee, W. J. T., Vedder, A. & Lisa Naughton-Treves. (2001). African Rain Forest Ecology and Conservation: An Interdisciplinary Perspective. Yale University Press, New Haven & London.

22 UGANDA WILDLIFE AUTHORITY

7.0 APPENDICES

7.5.1. APPENDIX I. SOME OF THE SURVEY TEAM MEMBERS

7.5.2. APPENDIX II: ONE OF THE TRAINING SESSIONS BEFORE THE CENSUS

23 GROUND COUNTS FOR MEDIUM TO LARGE MAMMALS -LAKE MBURO NATIONAL PARK- 2018

7.5.3. APPENDIX III: SOME OF THE TEAMS PRACTICING USE OF GPS DURING THE TRAINING

7.5.4. APPENDIX IV: BRIEFING TEAMS AT THE END OF THE TRAINING

24 UGANDA WILDLIFE AUTHORITY

Uganda Wildlife Authority 7 Kira Road Kamwokya P.O.Box, Kampala Uganda Tel: +256 414 355 000 Fax: +256 414 546 291 Email: [email protected] facebook,.com/ugandawildlifeauthority twitter.com/ugandawildlife

25