Ecological Factors Influencing Distribution Patterns Of Three Aristida Species And Their Associated Species In Kifuko Ranch, Laikipia County, Kenya

ABSTRACT

Species of Aristida are a common component of rangelands of Northern Kenya. They provide high value pasture to livestock, especially during their young stage of growth. The ecological data relating to the distribution of Aristida species at a local scale is scanty. The objectives of this research were; to assess the distribution patterns of the three Aristida species occurring at Kifuko ranch during wet and dry seasons, to determine plant species associated with the three species of Aristida and the possible local ecological factors that influence the distribution of the three species. Data collection was conducted during wet and dry seasons. The data collected included floral data and micro–habitats’ data of soil variables, slope directions, slope gradients, depressions, grazing intensity, soil depth and presence of cattle bomas, animal tracks, boulders, shade, termite hills and burning. This study used a total of 40 data collecting micro–sites and each micro–site measured 1 m2. Ten of the 40 used data collecting micro–sites were established on ten randomly selected points and the points were selected along a 3 km long transect. Four transects, each measuring 3 km long, were used in this study. Out of the four transects used in this study, two were established on deep soil habitats and the other two transects were established on shallow soil habitats. Analysis of paired t-tests was used to test abundance differences for each Aristida species, between wet and dry seasons. Detrended Correspondence Analysis (DCA) was used to cluster plant species associations. Variability of soil attributes among the four sampled transects was analyzed using Analysis of Variance (ANOVA) and their means were separated using Duncan’s multiple range (DMR) test. Descriptive analysis was used to analyze variability of other sampled micro–habitats’ variables among the sample transects. Multivariate analysis of Canonical Correspondence Analysis (CCA) was used to model Aristida species–environmental gradients relationship clusters. Analyses of paired t-tests and of ANOVA were performed using software of Microsoft Office Excel 2007, and the DCA and CCA analyses were performed using software of CONOCO version 4. Results of this study showed that each Aristida species distributed into a distinct cluster. The abundance trends of Aristida species, during wet and dry seasons, showed a significant abundance increases for Aristida. kenyensis and A. congesta (p = 0.01; and p = 0.04), respectively, during the wet season compared to the dry season, and no significant abundance change for A. adoensis (p = 0.26) during wet season compared to dry season. During wet and dry seasons, each Aristida species clustered with specific plant species associates. Distribution of A. kenyensis during the wet season was positively influenced by soil pH and was positively influenced by boulders during the dry season. Distribution of A. congesta individuals, during wet and dry seasons, was negatively influenced by soil depth and positively by clay content. Distribution of A. adoensis individuals, during wet and dry seasons, was positively influenced by slope direction and by soil depth, and was influenced negatively by clay content. This study recommends species of Aristida to be used by pastoralist communities living around Kifuko ranch to monitor quality of their pasture, as source of dryland pasture and as stabilizing species of degraded landscapes. In addition, a replica study should be carried out in another part of Northern Kenya to ascertain whether the variables sampled in this study were taken at optimum.