Between 16 and 24 September 2001, we assessed the extent, distribution and condition of the remaining grassland habitat, and estimated the distribution and density of two globally threatened, restricted-range species at the Mau Narok/Molo Grasslands, Kenya. Cultivation covered the largest proportion (48%), whereas grasslands constituted only 33% of the entire Important Bird Area (IBA). More continuous grasslands were found in the southernmost part of the IBA, mostly occupied by the pastoral Masai community. Large-scale conversion to barley and wheat plantations poses a major threat to the grasslands. Sharpe's Longclaw Macronyx sharpei occurred at densities of 1.2 individuals sq ha on grasslands and was the second-most commonly observed species. Aberdares Cisticola Cisticola aberdare was recorded in only three plots, where seven individuals were counted. The IBA has great potential for conserving highland grassland endemics as it holds significant areas of large, high-quality grassland patches. Future conservation action should concentrate on areas where such grasslands persist.
The Mau Narok/Molo Grasslands Important Bird Area (IBA) holds significant areas of Kenya's unique highland grasslands. This and Kinangop Grasslands IBA (Fig 1), either side of the central Rift Valley, are the only sites that hold significant areas of Kenya's unique highland grasslands, which lack any formal protection and are rapidly vanishing. They are home to several migratory bird species and various specialised grassland birds, and are of key importance as they harbour the restricted-range Sharpe's Longclaw Macronyx sharpei and Aberdare Cisticola Cisticola aberdare, both classified as Endangered by BirdLife International2. No adequate information concerning the status of either species or their grassland habitat is available from Mau Narok/Molo Grasslands. Between 16 and 24 September 2001, we surveyed the extent, distribution and condition of the remaining grassland habitat, and estimated the distribution and density of these species. We also used the survey to (1) increase awareness among the local community of the conservation value of the IBA, (2) make opportunistic contacts with interested members of this community with the aim of using them to initiate a local conservation group or Site Support Group (SSG), and (3) involve and train three members of the Kinangop Grasslands SSG in grassland survey methods. Here, we present the basic results of our work.
Mau Narok/Molo Grasslands
The Mau Narok/Molo Grasslands IBA is an extensive montane grassland situated on the crest of the Mau escarpment, which forms the western wall of the central Rift Valley in Kenya. This high, open plateau occupies c80 km south-east to north-west, and is bounded (and partially interrupted) by the Mau Forest complex. Rainfall is c1,000 mm per year, and the original vegetation is short grassland, with some heather and scrub on ridges where soils are deeper. The area has high potential for agriculture, and has gradually been settled since the 1950s; it is now heavily populated, with a landscape severely modified by cultivation. Cereals are the main crops.
The IBA is sited in Nakuru and (a small part of) Narok districts, within Rift Valley province. Population density in Nakuru district is high (164 per km2) . The main occupants of the study area are smallholders of the Kikuyu and Kalenjin communities (Molo and northern Mau Narok) and the Masai community (who are traditionally livestock herders) in the southern Mau Narok.
We continuously mapped broad land-use categories over a predetermined route through the grasslands by estimating the percentage cover of each habitat around points en route, at intervals of 1.5 km. Transects were conducted every 3 km to survey grassland birds in selected 1-ha plots within larger grasslands. Thirty-seven plots were surveyed. The area of each larger patch was recorded as <5, 5 to 20 or >20 ha, and the plots were surveyed by a team of six people spread across the width of the area, walking slowly across it 10-20 m apart. All species and individuals flushed were recorded. Measures of grassland quality were also noted within the plot using a system developed for monitoring grassland habitat in Kinangop. Grass height was recorded as G1 (short), G2 (medium length) and G3 (tall). Percentage tussock cover was recorded as T1 (no or just a few scattered tussocks), T2 (moderate tussock covering up to 30% of area), T3 (considerable tussock covering 30-60% of area), and T4 (dense tussock covering >60% of area). The number of livestock and other bird species within the entire patch were also recorded.
Sixty species were recorded in the grassland patches surveyed. Sharpe's Longclaw occurred at a relatively high overall density, of 1.2 individuals per sq ha, throughout the 37 plots, i.e. slightly higher than the 0.85 per sq ha from sample plots in Kinangop recorded by Muchai et al . However, the Mau Narok/Molo density compares well with that noted only in those plots with short grass and tussocks at Kinangop . The species occurred at 2.2 per sq ha in plots within grasslands of >20 ha, compared to those of <20 ha where the species occurred at densities of 0.2 per sq ha. Based on overall density, it was the fourth most abundant species after Black-winged Lapwing Vanellus melanopterus, Red-capped Lark Calandrella cinerea and Grassland Pipit Anthus cinnamomeus. Jackson's Widowbird Euplectes jacksoni, a Near-Threatened species, was the fifth most abundant species. The longclaw was encountered in 17 out of 37 survey plots, and was the second most frequently encountered species after Grassland Pipit, which was recorded in 22 plots.
Aberdare Cisticola was recorded in only three plots, with a total of seven individuals, suggesting a low density. However, it is probable that the species was under-recorded during the counts as the transect method used was designed for surveying Sharpe's Longclaw and may not be as appropriate for the cisticola, which may also utilise different habitat. The few individuals to be located were all on or close to slightly bushy grassland. Other commonly encountered species (and the number of plots in which they were recorded) were Common Fiscal Lanius collaris (17), Red-capped Lark (14), Stonechat Saxicola torquata (ten), Baglafecht Weaver Ploceus baglafecht (nine) and Wing-snapping Cisticola Cisticola ayresii (eight).
As grasslands were estimated to cover one-third of the IBA (c13,000 ha), the area surveyed was c0.3% of the IBA's grassland area. The plots were, however, within patches totalling approximately 2,150 ha (16.5 %).
Grasslands covered only one-third of the estimated area (c40,000 ha)  of the entire IBA (Table 1), while cultivation covered approximately 50% of the IBA. The spatial distribution of grasslands varied between the Molo and Mau Narok areas of the IBA. Most (75%) of the 20 grassland patches surveyed in Narok district (southern Mau Narok) were >20 ha. Only 35% of the 17 patches in Nakuru district (Molo block and northern Mau Narok) were >20 ha in size. Grasslands in Molo occurred as patches of varying sizes and were mostly isolated by cultivation. The main crops in Molo were maize and pyrethrum. Wheat, barley and potatoes were also common.
Grassland extent and distribution in Mau Narok varied between that occupied by smallholders and that by the mainly pastoral Masai community. The north-west Mau Narok was heavily cultivated and only 14.4% held grasslands. Smallholders of the Kalenjin and Kikuyu communities, who mainly cultivate maize, potatoes and vegetables, occupied most of this area. Elsewhere in Mau Narok, where the Masai are the principal land-users, grasslands covered 43.2%. In particular, more extensive grasslands were found in two regions of Mau Narok: west of Mau Narok town around Olokurto, and south of Mau Narok between Olorropil and south through Ol Chorro and Ol Joro.
Principal crops in cultivated areas of Masai land were barley and wheat, which mainly occurred on extensive farms accessible to agricultural machinery (Fig 2). The Masai community traditionally practised pastoralism and have tended to abandon extensive pasture to their livestock (Fig 3). This practice is, however, becoming rarer due to the advent of large-scale crop cultivation. Barley and wheat, which are potentially more lucrative than livestock farming, are currently being increasingly cultivated and pose a major threat especially to large grasslands in southern Mau Narok. The remaining grasslands are now heavily grazed as the pastoralists still herd large numbers of cattle.
In comparison to Kinangop Grasslands IBA, the remnant grasslands at Mau Narok/Molo IBA were larger, although isolated by large areas under cereals, probably because most land holdings are larger in Mau Narok and Molo than in Kinangop6.
Throughout the study area continuous grasslands appear to mostly survive along shallow sloping valleys. Seventeen (46%) of the surveyed grasslands were along watercourses and none was <5 ha. Most grasslands were heavily grazed and had short grass, as the Masai keep large herds of cattle on these areas. We counted 2,351 cows, 7,307 sheep and goats, and 42 donkeys within an area of c2,000 ha, at a grazing intensity of nine large animal units (LAU) per sq ha. We assume that a single LAU is equivalent to one cow or five sheep, and represents the metabolic equivalent of a 454-kg cow .
A higher proportion (15) of the plots had moderate tussock cover. Only two plots had dense tussock, whereas nine plots completely lacked or had just a few scattered tussocks. Sharpe's Longclaw strongly prefers short grass with tussocks [3, 4]. Most of the survey plots (62%) had short grass with tussocks.
Most (20) of the surveyed grasslands were >20 ha, whereas 13 were 5-20 ha. Only four patches were <5 ha in size. The 82 points where land-use was surveyed covered a total area of 14,496 ha (c36% of the IBA).
Training SSG members, awareness and contacts
The three SSG members from Kinangop gained skills in grassland bird survey techniques. Posters have been developed by the Friends of Kinangop Plateau to raise awareness concerning the importance of conserving Kinangop Grassland IBA, which faces similar threats to Mau Narok/Molo. Thirteen of the latter posters, 22 national posters with information concerning all Kenyan IBAs and 25 Kenyan IBA fact sheets were distributed. Talks outlining the conservation importance of the IBA were made to children and teachers at two schools (By-Gum and Segututou primary schools), and informative material presented to a student and headmaster at two additional schools (Fig 4). Seven contacts keen to form a Site Support Group were made.
- Initiate Site Support Group(s) for the IBA, based on existing and other, strategically sought contacts.
- Identify landowners and/or community leaders who control the management of large areas of land, and use them as focal points to influence management of such areas to the benefit of conservation.
- Determine those grassland management practices (eg stocking levels) that are compatible with conservation of key bird species, and the economic costs of their adoption, through well-focused research.
We are grateful to the Friends of Kinangop Plateau (FoKP), in particular Andrew Mwangi, Jacob Njoroge and Pius Njoroge for their immense contribution in the field, and those farmers who permitted access to their land. Funding for the field work was provided by African Bird Club to Nature Kenya and the Ornithology Department of the National Museums of Kenya. The Ornithology Department provided equipment used during the survey.
- Bennun, L. and Njoroge, P. 1999. Important Bird Areas in Kenya. Nairobi: Nature Kenya.
- BirdLife International. 2000. Threatened Birds of the World. Cambridge, UK: BirdLife International & Barcelona: Lynx Edicions.
- Lens, L., Muchai, M., Bennun, L.A. and Duchateau, L. 2000. Conservation planning in an agricultural landscape: the case of Sharpe's Longclaw. Ostrich 71: 300-303.
- Muchai, S.M. 1998. Some aspects of the conservation biology of Sharpe's Longclaw (Macronyx sharpei Jackson 1904): a Kenyan grassland endemic bird. MSc. thesis. Eldoret: Moi University.
- Muchai, M., Lens, L. and Bennun, L. 2002. Habitat selection and conservation of Sharpe's Longclaw (Macronyx sharpei), a threatened Kenyan grassland endemic. Biol. Conserv. 105: 271-277.
- Ndang'ang'a, P.K., du Plessis, M.A., Ryan, P.G. and Bennun, L.A. 2002. Grassland decline in Kinangop Plateau, Kenya. Implications for conservation of Sharpe's Longclaw Macronyx sharpei. Biol. Conserv. 107: 341-350.
- Owen-Smith, N. and Danckwerts, J.E. 1997. Herbivory. In: Cowling R.M., Richardson, D.M. and Pierce, S.M. (eds) Vegetation of Southern Africa. Cambridge, UK: Cambridge University Press.
- Republic of Kenya. 2001. The 1999 Population and Housing Census 1999. Vol 1. Nairobi: Central Bureau of Statistics, Ministry of Finance and Planning.