feature
Grassroots
August 2017
Vol. 17, No. 3
in every three years (tri-annual) and also depending whether there was enough grass fuel after each burn to support a hot fire for the subsequent burn (Balch et al. 2015).
Sampling procedure
Two fixed belt transects of 30 m long were randomly placed in the trial where changes in the woody and herbaceous species were recorded at each field plot during data collection.
Determination of herbaceous species composition
Step-point method was used to determine the herbaceous species composition (Colon et al. 2016). Here, a researcher moves two steps forward and inserted a rod on the ground just in front of the tip of the living herbaceous species (grasses and forbs/sedges) on which the rod had strike upon was recorded in the data sheet. In the event that the rod has strike on a bare area the nearest living herbaceous plant to the rod was recorded because of the small size of the plots (3600 m2). Botanical composition was determined along the fixed transects used for bush surveys where only 50 points were made on each transect to make 100 points of each sample site (plot) from the two transects in a plot. Botanical composition of the sample was expressed as the percentage frequency of the plants recorded during survey while, basal cover refer to the percentage of the number of strikes in relation to the total number of points recorded. Standing biomass was estimated using a disc pasture meter. After all the data was collected from both herbaceous and woody components and fire applied, all the treatments except controls were grazed with cattle as means of simulating practical farming practices.
Determination of woody species composition
Changes in the woody species were monitored along the two fixed 30 m long chains laid out straight in the two areas of each field plot representing the entire plot (Mudzengi et al. 2013). The centre of 3 m long rod (1.5 m left & 1.5 m right) was moved down the chain at an angle of 90°, where every woody plant within the 3 m width along the chain was recorded in terms of species, lowest browse material, canopy height and canopy diameter and also classified as palatable or non-palatable. The recorded woody species data provided an estimate of tree equivalents per hectare (TE/ha), number of woody plants per hectare (P/ha) and browse units per hectare (BU/ha). There was no data collected before the experiment site was rolled with a Bosvark machine.
Analysis of collected data
Changes and responses in herbaceous and woody species to treatments applied were analysed by two-way analysis of variance (ANOVA) or repeated measures of analysis at a significance level of 5% (0.05) using GLM procedure of SAS (Van Coller et al. 2013). Mean separation was tested using Least-Square Means (LSM) procedure of SAS. The response of woody species to the treatments applied are explained through the analysis of variance (ANOVA) on the variables measured (tree equivalents, browse units and density) while, response of the herbaceous species is explained through composition and biomass production.
Results
Composition of herbaceous species
Eleven grass species were recorded; various forbs and sedges were grouped together and recorded to determine their relative abundances. The following indicators were also noted; form of growth, forage factor and ecological status and are presented in Table 2. The identified herbaceous species were in the form of perennial tufted species with the exception of Cynodon dactylon and Digitaria eriantha (stoloniferous sub-species) which were perennial creeping species. In terms of their ecological grades, only four species which were in the category of Decreaser, with six species in the category of Increaser-IIs and one species classified as Increaser-I. However, in terms of their relative abundances, Panicum maximum and Digitaria eriantha which are both Decreaser species with a high forage factor of 10, were remarkably abundant in the trial with means of 34.5 and 24.4% respectively. Again,
f 34.5 and 24.4% respectively. Again, a relative abundance of Increaser-IIs was observed from Cynodondactylon and forbs/sedges with means of 11.6 and 8.8% respectively. However the forage factor of these species was remarkably low to the values of 2 and 0.
36