Gaseitsiwe Smollie Masunga, Banyana Kegoeng, Theophilus Kgosithaba, and Lucas Pius Rutina
C/N, herbivory, litter decomposition, lignin/N
The decomposition rates and litter chemistry carbon (C), nitrogen (N), phosphorus (P), potassium (K) and lignin of senesced leaf litter of seven dominant tree species in three elephant-transformed woodlands were examined in semi-arid Botswana. The woodlands were selected on the basis of a gradient in woodland cover mainly caused by elephant herbivory. The species studied were Combretum elaeagnoides, Croton megalobotrys, Philenoptera nelsii from the heavily browsed and open canopy Shrubland, C. elaeagnoides, Friesodielsia obovata, Baikiaea plurijuga from the ecotone and moderately browsed Mixed woodland, and B. plurijuga, Combretum apiculatum and Croton gratissimus from the least browsed and closed canopy Teak woodland. Overall, there were significant differences in decomposition rates of litter species between the three vegetation communities (Kruskall Wallis 48.585, p<0.001). Significant differences in decomposition rates were observed between litter species, with P. nelsii (k = 0.22 y-1) in Shrubland and B. plurijuga in Teak woodland (k = 0.231 y-1) showing lower rates than other litter species. Despite having lower decomposition rates P. nelsii suprisingly had a higher N concentration and lower C:N and lignin-to-N ratios suggesting that these ratios may not have been the only ones influencing the decomposition rates of the litter species. For similar litter species placed across different communities, B. plurijuga decomposed faster in moderately open Mixed woodland than in closed canopy Teak woodland. The decomposition rates of C. eleagnoides in Shrubland and in Mixed woodland were marginally different, with litter in Shrubland decomposing slightly faster than the one in Mixed woodland. C. elaeagnoides in the Shrubland also had lower C:N and lignin-to-N ratios than in Mixed woodland. The lignin-to-N ratio was generally a better predictor of the decomposition rates than C:N
ratio, suggesting that future studies on decomposition of tree leaf litter can largely use lignin-to-N ratios rather C:N for predicting plant litter decomposition rates. Generally, litter species in Shrubland had a higher N concentration pointing to the positive feedback mechanisms of herbivory by megafauna where dung and urine deposited increase soil N availability and subsequently increase forage quality and high decomposition rates.