Fire temperature and residence time during dry season burning in a Sudanian savanna-woodland of West Africa with implication for seed germination

Prescribed fire is used in the Sudanian savanna-woodland of West Africa as a forest management tool. An experiment was carried out to assess the effects of season of burning, and different vertical probe positions on maximum fire temperature and temperature residence time above 60℃ which is considered lethal for plant tissues. Seasons of burning considered were： an early season fire set at the beginning of the dry season （beginning of December）, mid-season fire set at the peak of the dry season （mid-January）, and a late season fire at the end of the dry season （end of March）. The effects of these fires on the germination of buried seeds of three socio-economically valuable tree species were also examined. Results indicated significant differences in maximum fire temperature and residence time with respect to season of burning and vertical probe position （p 〈 0.001）. The highest and longest lasting temperatures were observed at 20 cm above ground during early fire and at the soil surface during mid-season and late fires. This, in turn, affected germination responses of seeds buried at different soil depths. Implications of these findings in the current management practices are discussed.

Allometric prediction of above-ground biomass of eleven woody tree species in the Sudanian savanna-woodland of West Africa

Allometric models are necessary for estimating biomass in terrestrial ecosystems. Generalized allometric relationship exists for many tropical trees, but species- and region-specific models are often lacking. We developed species-specific allometric models to predict aboveground biomass for 11 native tree species of the Sudanian savanna- woodlands. Diameters at the base and at breast height, with species means ranging respectively from 11 to 28 cm and 9 to 19 cm, and the height of the trees were used as predictor variables. Sampled trees spanned a wide range of sizes including the largest sizes these species can reach. As a response variable, the biomass of the trees was obtained through destructive sampling of 4 754 trees during wood harvesting. We used a stepwise multiple regression analysis with backward elimination procedure to develop models separately predicting, total biomass of the trees, stem biomass, and biomass of branches and twigs. All species- specific regression models relating biomass with measured tree dimen- sions were highly significant （p 〈 0.001）. The biomass of branches and twigs was less predictable compared to stem biomass and total biomass, although their models required fewer predictors and predictor interac- tions. The best-fit equations for total above-ground biomass and stem biomass bad R2 〉 0.70, except for the Acacia species; for branches including twig biomass, R2-values varied from 0.749 for Anogeissus leiocarpa to 0.183 for Acacia macrostachya. The use of these equations in estimating available biomass will avoid destructive sampling, and aid in planning for sustainable use of these species.