Vegetation data on South Nandi Forest collected in 2016
Description
Five sample plots (30m X 20m) along each transect at intervals of 200m were established to facilitate forest species counts to enable generation of information on species representation in the specific forest areas. For the whole forest, seven 1-km transects were done adjacent to the villages. For each plot, there was a main plot A(30m X20m), where all trees with over 10cm DBH were measured using a diameter tape and height also measured; two Sub-plots of B (10m X 5m) – where all saplings, lianas and shrubs, DBH = 2 cm - 9.99 cm, at least 1.5 m height were assessed and two Sub-plots of C (1m X 2m) - seedlings counted. The population structure of the species was analysed across ten DBH classes whereas the status of regeneration was determined based on the population size of seedlings, saplings and adults (Gebrehiwot & Hundera, 2014). The identification of the plants was done by at least two local scouts with wide knowledge of the species in vernacular names and by referring to flora books (Dale & Greenway, 1961; Beentje, 1994; KFS, 2015). Finally, three Focused group discussions (FGD) based on the procedure by Krueger & Casey (2009) was done to document uses of plants in South Nandi forest. For each plant species, all its known uses were mentioned and consensus reached among members of the group before recording them. Focused group discussions per site were done per site were done separately in the three different locations in September 2016. Quantitative analysis was done using either Excel 2010, SPSS version 21 or Genstat version 19 depending of the type of analysis, for example, Genstat was used in doing Principal Component Analysis (PCA). PCA is a multivariate analysis which is widely used in biology where high-dimensional data are very common. In this study, PCA was used to analyze tree/shrub species diversity trends in the forest based on the method described in Lever et al. (2017). Forest data was first subjected to Kolmogorov-Smirnov and Shapiro-Wilk normality tests in SPSS Version 21 and if data was not normally distributed, it was transformed Log (x + 4) then a two-way ANOVA done. Descriptive statistics and homogeneity test (Levene’s test) were generated; where the results indicated a significant difference (p<0.05), Post Hoc test (Tukey’s test) were done to find out which means were significantly different, and for qualitative analysis, descriptive statistics was used. Forest plot data were summarized according to standard protocols; the stems for seedlings, saplings, and trees were expressed as the number of stems per hectare (density). Basal area (BA), which is the cross-sectional area of a tree, was calculated from dbh and presented for various size classes as m2ha-1.