No abstract

In the Ethiopian Central Highlands, a serious soil degradation occurs while fuelwood demand is high. This study consists of an evaluation of seven tree species for fuelwood and soil restoration under three soil management options: control, manure and manure + mulch, in degraded highlands of Ethiopia. The experimental design was a split‐plot, species as the main plot and treatment as subplot, with three replicates. Survival count, height and root collar diameter growth measurements were measured annually until 48 months. Biomass production for fuelwood was inferred at the end of the experiment. Before and after the experiment, soil parameters (pH, organic carbon, N, P, K and cation exchange capacity) were measured to test changes in soil condition because of species plantation. A mixed model and repeated analysis of variance was performed. Grevillea robusta A. Cunn. ex R. Br. showed maximum survival (100%), followed by H. abyssinica (Bruce) J.F.Gmel. (93·52%); while the lowest survival rate was recorded for A. decurrens Willd. (57·41%). Hagenia abyssinica (Bruce) J.F.Gmel. and Chamaecytisus palmensis (Christ.) Hutch showed the lowest growth rates but both species showed the highest soil condition improvement. E. globulus Labill. and Acacia species presented the highest growth rates and biomass although Eucalyptus depleted soil nitrogen. Hagenia abyssinica (Bruce) J.F.Gmel. is recommended for soil rehabilitation, whereas Grevillea robusta A. Cunn. ex R. Br. can be used for simultaneous fuelwood production and soil rehabilitation. An ecological based study on E. globulus’ Labill. effects in Central Highlands is recommended before recommendation for large scale fuelwood plantations. Copyright © 2014 John Wiley & Sons, Ltd.

Abstract& Key message Biomass equations are presented for five tree species growing in a natural forest in Ethiopia. Fitted models showed more accurate estimations than published generalized models for this dry tropical forest. & Context Biomass equations are needed to correctly quantify harvestable stock and biomass for sustainability efforts in forest management, but this kind of information is scarce in Ethiopia. & Aims This study sought to develop biomass models for five of the most common native tree species in the Chilimo dry afro-montane mixed forest in the central highlands of Ethiopia: Allophyllus abyssinicus, Olea europaea ssp. cuspidata, Olinia rochetiana, Rhus glutinosa, and Scolopia theifolia. Comparison with generalized models was intended to show the greater accuracy of the specific models. & Methods A total of 90 trees from different diameter classes were selected, felled, and divided into different biomass compartments. Biomass equation models were fitted using jointgeneralized least squares regression to ensure the additivity property between the biomass compartments and total biomass. & Results These were the first models developed for these species in African tropical forests. Models were including diameter at breast height and total height as independent variables, obtaining more accurate biomass estimations using these models than from generalized models. & Conclusion Fitted models are reliable for estimating aboveground biomass in the Chilimo forest and for more general application in similar forest types. Model applicability for biomass or carbon estimation is high within forest inventory data contexts.