Food Selection and Habitat Association of Starck’s Hare (Lepus starcki Petter, 1963) in the Bale Mountains National Park, Ethiopia

Mekonnen, Tariku; Yaba, Mohammed; Bekele, Afework; Malcolm, J.

INTRODUCTION

The geographical location and physical features of Ethiopia have resulted in the diversification of wildlife (Tedla, 1995; Refera and Bekele, 2004). Thus, Ethiopia possesses a unique and characteristic fauna with a high level of endemism (Hillman, 1993).

Hares are important economically and aesthetically for humans (Nowak, 1999; Malcolm, 1997; Burgos and Burgos, 2007). They provide sport hunting, food and fur. Hares also form an important link in ecological food chains. Many predators rely on the abundance of hares for their diet. One example of this phenomenon is the linkage of Ethiopian wolf and Starck’s hares in the Bale Mountains (Sillero-Zubiri and Macdonald, 1997; Marino, 2003).

Starck’s hare (Lepus starcki) is one of the endemic mammals of Ethiopia (Yalden and Largen, 1992; Kingdon, 1997) and potentially a very important part of the ecosystem of the Bale Mountains National Park (BMNP). It is an important part of the diet of the endangered and endemic Ethiopian wolf. They are mainly found in afroalpine parts of the BMNP including Tullu Deemtu but also seen at low density in heather moorland and forest as well in the northern woodlands of the Park (Hillman, 1993). Yalden and Largen (1992) stated that Starck’s hares are quite abundant on the moorlands of the Bale Mountains. At the same time, they also occurred on both sides of the Rift Valley at altitude between 2140-4000 meter above sea level (m.a.s.l) in both grassland and afroalpine moorland. Sillero-Zubiri (1994) stated that Starck’s hares were particularly abundant on Sanetti Plateau and less in ericaceous heather and montane grassland.

Hares are herbivores and eat wide varieties of plants including grass, herbs and barks of plants. However, they often show preference to certain species of Vegetation. These preferences vary greatly between regions and are dependent on the local plant communities (Strevens and Rochford, 2004).

Currently, because of habitat loss and agriculture expansion wildlife population is decreasing in the world (Idris et al., 2001; Maan and Chaudhry, 2001; Gabol et al., 2005; Gundogdu, 2011). IUCN (2006) grouped Starck’s hares as least concern. And for effective research and management of wildlife species, the study of feeding ecology of mammals is in the center of population biology and ecology (Matrai et al., 1998). Therefore, the objective of this study was to identify the preferred diets and habitats of Starck’s hare.

MATERIALS AND METHOD

The Bale Mountains National Park (BMNP) is located Southeast of the Rift Valley about 400 km by road from Addis Ababa, between 6° 29^′ and 7° 10^′ N and 39° 28^′ and 39° 58^′E. It covers an area of 2200 km², with an altitudinal range from 1500-4377 masl. The typical afroalpine habitats in BMNP, Sanetti Plateau (3800-4050 m), Web Valley (3450-3550 m) and Tullu Deemtu (4000-4377 m.a.s.l.) formed the main study area.

Methods: Data for the availability of vegetation were obtained from 44 sample sites using line intercept method (Sutherland, 1996; Cummings and Smith, 2001).

To find the mean percentage coverage of each species (MPCSp), the following formula was used:

where, TCSp is total cover of the species, L is length of line.

To find the mean percentage vegetation coverage of the site (MPCS):

where, TCS is total cover of the site, L is length of line.

Diet analysis was carried out following the methods of Dingerkus and Montgomery (2001), Katona and Altbacker (2002) and Reichlin et al. (2005).

To identify the types and proportion of plant species used by L. starcki, the combination of faecal analysis and direct observation were used. Fresh pellets of Starck’s hare were collected from sites of different vegetation types during both seasons. In each season, two pellets were collected from 25 independent droppings at the minimum of 200 m interval. The samples were preserved in 70% ethanol and taken to the Department of Biology, Addis Ababa University, for further analysis. They were washed independently with distilled water to remove fine particles for proper identification and air dried. Two slides were prepared for each sample (mixing thoroughly) and observed under microscope to identify the diet. All fragments found on the slides were identified as monocotyledon, dicotyledon, or unidentified. Relative occurrence was determined by dividing the number of microscopic views in which a given species occurred by the total number of views multiplying by 100 (Katona and Altbacker, 2002).

In addition to faecal analysis, direct observation was used to identify different species of grasses consumed by Starck’s hare. Different types of habitats, where hares usually fed were selected carefully for hide observation. Using 8x42 binocular, species of plant eaten were examined for 120 h. The plant species on the habitat was directly observed, before and after, to see the plant eaten by looking at the bite.

To identify habitat selection, a total of 10 transects were systematically located on 40 km² on each of the study area at the interval of 1 km. The length of each transect was 4 km. Transects were conducted on foot stopping and carefully observing at approximately 50 m interval. Each time, a hare was seen, habitat type, activity at a time and weather condition were recorded. The survey time was at dusk and dawn.

Data analysis: Data analysis was done using SPSS Vession 16. Significance difference between wet and dry season diet composition and plant coverage were analyzed using chi-square.

RESULTS

Forage availability: During the wet season, Starck’s hare had most access to different species of herbs and grasses (Table 1). However, the proportion of plant species changed dramatically during the dry season. The major components of plant species in the study area were Festuca spp., Alchemilla abyssinica, Helichrysum spp. and Trifolium species. Usually, Starck’s hares use Helichrysum spp. as a shelter. During the dry season, wetland plants were available for feeding. These include Carex monostachya, Ranuculus oreophytus, Haplocarpha rueppelli and Trifolium acaule. Although, Festuca spp. and Alchemilla spp. decreased in coverage during the dry season, their relative abundance was high. There is significant difference (p<0.05) between plant cover during the wet season (65.21%) and during the dry season (<30%).

In this study, the most abundant plant species during both seasons were Festuca spp., Alchemilla byssinica, Helichrysum gofanse, H. citrispinum, H. splendidum and Trifolium acaule. Other studies have also noted all these species, as a dominant group in the area (Gashaw and Fetene, 1996; Puccetti et al., 1996).

Food use: Of the 27 plant species identified from the study area, Starck’s hares entirely fed on monocotyledons (grasses) during the wet season. These were Festuca spp., Koeleria capensis, Agrostis gracilifolia and Carex monostachya. During the dry season, dicotyledon plants were observed in the diet of Starck’s hares, however, the proportion of monocotyledons was still high (Table 2).

Starck’s hares were highly selective for monocot plants with limited use of dicots during the dry season (Table 3).

Habitat: Starck’s hares were mostly restricted to rocky grassland with low wind areas. However, during the dry season when grasslands dried out, most were observed feeding in wetland habitats (Table 4).

Table 1:	Coverage of plant species identified from the study area using line intercept method from sanetti plateau

DISCUSSION

In this study, the most abundant plant species during both seasons were Festuca spp., Alchemilla abyssinica, Helichrysum gofanse, H. citrispinum, H. splendidum and Trifolium acaule. Other studies have also noted all these species, as a dominant group in the area (Gashaw and Fetene, 1996; Puccetti et al., 1996).

Most literature implies that hares are generalized herbivores, with a diet consisting primarily of grasses and shrubs but also barks, fruits, seeds, leaves and buds based on the habitat types (Rao et al., 2002). However, in the present study, Starck’s hares have a preferences to grasses as these occurred at high frequencies in all independent droppings. Supporting this, Hewson and Hinge (1990) concluded that hares spent a large part of their time grazing on wild grassland, indicating a preference for grasses. During the dry season, they fed on more diverse plant species than during the wet season, even though there was relatively high plant species diversity during the wet season. This might be due to the low availability of preferred food (grasses). Plants consumed during the dry season were based on the availability in the area (wetland plants). However, certain species such as Alchemilla spp. were never eaten although abundant.

Table 2:	Relative frequency of occurrence of plants identified from independent droppings

MN: Monocotyledon, DC: Dicotyledon, UN: Unidentified

Table 3:	Plant coverage and percentage dietary composition of starck’s hare from faecal analysis

Table 4:	Number and percentage of starck’s hares recorded from different types of habitats at different level of wind

The pattern of Starck’s hares habitat use observed was associated with seasons. Though rocky grassland is the most preferred areas for starck’s hare to protect themselves from predator and get grasses, during dry season mostly they are observed in wetland areas to get food.

From all the grass species identified from the study area, Starck’s hares fed frequently on Festuca species. Soft and green parts of the plants were the most preferred compared to other parts during the study period in both seasons.

ACKNOWLEDGMENTS

We would like to acknowledge the Ethiopian Wolf Conservation Program (EWCP) and Addis Ababa University for providing financial and material support.