VOL.Ghana 59 J. Sci. 59 (2018), 23 - 30 GHANA JOURNAL OF SCIENCE https://dx.doi.org/10.4314/gjs.v59i1.223 ABIOTIC AND ANTHROPOGENIC FACTORS AFFECTING THE DISTRIBUTION OF FOUR SYMPATRIC LARGE HERBIVORES IN THE MOLE NATIONAL PARK, GHANA K. B. DAKWA Department of Conservation Biology and Entomology, School of Biological Sciences, University of Cape Coast, Ghana Email: [email protected] /[email protected] Abstract The impact of abiotic and anthropogenic factors on the distributions of buffalo (Syncerus caffer), harte- beest (Alcelaphus buselaphus), roan antelope (Hippotragus equinus) and waterbuck (Kobus defassa) at Mole National Park was assessed by transect survey. Generalized linear mixed effects logistic re- gression was used to model mammal presence/absence as a function of ecological factors. Hartebeest inhabited highlands and avoided floodplains but buffalo and roan avoided floodplains by selecting both lowlands and highlands while waterbuck inhabited lowlands but not necessarily the floodplains. Fire, water availability and anthropogenic activities were limiting factors, which constrained habitat use to make some areas unexplored for foraging. Buffalo, roan and hartebeest did not inhabit areas close to the park’s boundaries. Herbivores need optimal environment almost free of constraints to construct their distribution patterns. Therefore, management should address the problems identified in this study to ensure the herbivores’ redistribution to maximise their use of resources for their effective conservation. Introduction data on distribution of wildlife in a reserve, Mole National Park (MNP) was established conservation planning will remain inefficient in 1958 for the protection and conservation of and ineffective (Bauer et al., 2010). If we are representative savanna fauna in the northern to manage landscapes to conserve wildlife, part of Ghana. Though the herbivorous mam- it is important that we understand the role of mals form the majority of the mammalian spe- landscape factors in determining their distribu- cies richness of the reserve, no previous survey tions. Thus, predicting the distribution of wild- has specifically assessed the ecology of these life populations is an important component of herbivores. Wildlife research in and around the development of management strategies for MNP is scanty in general including evalua- their conservation. tion of population trends and status of lions The aim of this study was to provide re- (Burton et al., 2010), censuses (Wilson et al., search data for the effective conservation of 1993), cost of raids caused by wildlife (Dakwa, some large herbivores at MNP, specifically to: 2016a), allometry in sympatric grazers (Dak- 1) determine the effect of abiotic (water and wa, 2016b), and evaluation of disturbances to altitude) and anthropogenic factors on the dis- water birds (Aikins et al., 2018) and few stud- tribution of four sympatric large herbivores of ies on community and wildlife conflict issues, MNP, namely, buffalo (Syncerus caffer), roan which remain unpublished. Without research antelope (Hippotragus equinus), waterbuck GJS is an Open Access Journal and distributed under the terms of the Creative Commons (CC) License [CC BY 4.0] 24 GHANA JOURNAL OF SCIENCE VOL. 59 (Kobus defassa) and hartebeest (Alcelaphus tive humidity in the rainy season reaches 90% buselaphus) and 2) investigate possible con- at night and 70% day time but the figures drop straints resulting from the influence of those to 50% and 20% respectively in the dry sea- factors in the herbivores’ distribution. The son (Wildlife Division, 2011). The main habitat results of this study are expected to inform types in MNP are the Burkea-Terminalia open management decisions on effective strategies savanna woodland with Vitellaria paradoxa aimed at the redistribution of these mammals (BTV), the Burkea-Terminalia savanna wood- in order to maximise the use of resources at land with Detarium microcarpum (BTD), the MNP. Knowledge of the distribution of these Anogeissus with Vitellaria paradoxa (A), the charismatic mammals at MNP will also boost boval vegetation (Loudetiopsis kerstingii-Poly- tourism as this will be integrated into manage- carpaea tenuifolia community on rocky sub- ment planning and implementation strategies strates, riverine forest (R) along most of the for effective tour guiding. rivers in the park, and swamp (S) (Schmitt & Experimental Adu-Nsiah, 1993). The Park is home to over 93 mammal species including the savanna el- Study Area ephant (Loxodonta africana), hippopotamus (Hippopotamus amphibious), buffalo (Synce- rus caffer), Defassa waterbuck (Kobus defas- sa), roan (Hippotragus equinus), and harte- beest (Alcelaphus bucelaphus). Data Collection Sixteen strip transects 200 m wide, 100 m either side of the central line and of varying lengths, along existing roads and paths traversing the main habitat types and 20 point transects, mea- suring 500 x 500 m each were used to survey the large herbivores. The transects were at least 1 km apart and were distributed to en- sure as much spatial representation as possible in the study area and their locations were then mapped by a GPS (all Garmin Ltd., Olathe, Fig. 1: Study area showing locations of transects iden- tified by numbers. KA, USA). Presence/absence data on buffalo, roan antelope, waterbuck and hartebeest were The Mole National Park (Fig. 1) covers an collected by walking point transects or driving area of about 4,840 km2 as Ghana’s largest strip transects at a speed of 20 km/h between wildlife refuge. The Park is located in north- 07.30-10.30 GMT and 16.00-18.00 GMT, west Ghana on grassland savanna and ripari- when herbivores are active, at least twice each an ecosystems. This area of Ghana receives an season (Table 1) between September 2013 and average of 1,100 mm per year of rainfall; the August 2014. Mammals were viewed within mean annual temperature of 28°C varies from the transects with the help of binoculars when 26°C in December to 31°C in March; and rela- it was necessary. VOL. 59 GHANA JOURNAL OF SCIENCE 25 TABLE 1 Different Seasons at Mole National Park. Season Description Period Dry Harmattan with low humidity October - December Fire Dry harmattan with low humidity; windy January - April Rainy Heavy rains with lightning and thunder May - July Flood Slight rains; Lowlands are flooded. August - September Measurement of abiotic factors Data analysis The most important abiotic factors are thought All data were entered into Microsoft Excel ta- to be water availability and altitude (Bailey et bles (version 2013), before being exported into al., 1996; Groom & Harris, 2010). Distanc- R statistical software package version 3.1.2 (R es of transect habitat and mammal locations core team, 2014). from the nearest sources of water were calcu- lated by using the nearest-features extension Generalised linear model (GLM) in ArcGIS (v9.3), after waypoint data stored The aim of the implementation of the GLM in GPSs were exported into an ArcGIS (v9.3) was to model mammal presence/absence (re- environment. Altitudes of transect habitats and sponse variable) as a function of abiotic and mammal locations were recorded by GPSs and anthropogenic parameters recorded (explan- the data transferred to a computer. For small atory variables), namely: (i) nearest distance habitats along transects, information on the from a source of water (ii) altitude (iii) near- transect habitat location such as distance from est distance from the park’s boundary, and (iv) the nearest source of water and altitude were nearest distance from a community or farm. assumed to be the same for their mammal loca- The idea behind this analysis was to investi- tions. However, because broad habitats occu- gate how these factors affected the distribution pied wide distance variations, such information of the mammals using presence/absence data were taken at the location of each mammal. as a function of distribution. For the analysis of these kinds of nested data with a sufficiently Measurement of anthropogenic factors large sample size, mixed modelling, which will Human activities in and around PAs may affect ensure convergence of maximum likelihood species distribution. In this study, the park’s estimates is the optimal statistical technique boundary, communities and farms represented recommended; and as this is a binary data, a possible human presence or activities. There- generalized linear mixed effects logistic regres- fore, the distributions of mammals in relation sion model was used. R packages multcomp to the locations of the boundaries, communi- (Hothorn et al., 2008), stats (R Core Team, ties or farms were investigated by taking GPS 2014) and lme4 (Bates et al., 2014) were used. readings of the mammals or their habitat lo- The large grazers’ resource locating distances cations and calculating their distances from were expected to be substantially smaller than the boundaries, farms or communities by us- the spatial extent of the study area, therefore ing the nearest-features extension in ArcGIS spatial auto-correlation between transects was (v9.3) after waypoint data stored in GPSs were possible. However, the mixed effects logistic exported into an ArcGIS (v9.3) environment. regression used dealt with spatial auto-correla- tion within and between transects (McCulloch 26 GHANA JOURNAL OF SCIENCE VOL. 59 et al., 2001). In all analyses for any particular ρ < 0.001); and the fire season very highly sig- mammal, the standard deviation (a measure of nificantly predicted the distributions