Models for estimating aboveground biomass of four dryland woody species in Burkina Faso, West Africa. (September 2020)

• Record Type: Journal Article
• Authors: Bayen, Philippe; Noulèkoun, Florent; Bognounou, Fidèle; Lykke, Anne Mette; Djomo, Adrien; Lamers, John P.A.; Thiombiano, Adjima
• Is Part Of: Journal of arid environments. Volume 180(2020)
• Publication Date: 2020-09
• Subjects: Senegalia -- Vachellia, Seemingly Unrelated Regressions (SUR) -- crown area -- carbon stock -- savanna; Arid regions ecology -- Periodicals; Arid regions -- Periodicals; Écologie des régions arides -- Périodiques; Régions arides -- Périodiques; 577.54
• DOI: 10.1016/j.jaridenv.2020.104205
• Languages: English
• ISSNs: 0140-1963

Abstract

Abstract: Reliable estimates of carbon stocks for drylands are constrained by a limited number of biomass models for predicting accurately the biomass of dominant tree and shrub species. In this study, 119 trees with a diameter at breast height (dbh) between 3.82 and 45.48 cm were destructively sampled to evaluate aboveground biomass (AGB) allocation and prediction for four dryland species ( Senegalia dudgeonii, Senegalia gourmaensis, Vachellia nilotica and Vachellia tortilis ) commonly found in the Sudanian and Sahelian zones of Burkina Faso. Biomass allocation to leaf, branch and stem was examined using a classmometric analysis. Allometric equations relating biomass to dbh alone, and to dbh in combination with height and/or crown area were developed with and without consideration of the additivity property. Branches stored the largest (up to 90%) proportion of AGB. For three of the species, the branch mass fraction increased with increasing tree size at the expense of stem and leaf mass fractions. The statistical evaluations of the selected additive and non-additive models were satisfactory with no severe deficiencies. Diagnostics of model performance revealed that non-additive models performed better than additive models. The biomass models developed can be used for improved quantification of carbon stock and dynamics in dry savanna ecosystems. Highlights: Additive and non-additive biomass models were developed for four dryland woody species Biomass allocation influenced the predictive ability of allometric equations Branches were the tree component with the largest share (up to 90%) of aboveground biomass Consideration of tree architectural variables improved models' predictions Mixed-species equations provided reliable species-specific biomass estimations