DEVELOPMENT OF WOOD FUEL CONSUMPTION PREDICTIVE MODEL IN TANZANIA

Abstract

This study aimed to develop a wood fuel predictive model that could be used to give information which can be used to manage woodfuel supply with a view foster forest resources stewardship. The paper has briefly defined predictive modelling concepts, highlighted the significance of predictive modelling and described the salient steps involved in constructing predictive models. The paper has explicitly described how the predictive model was developed and validated. In light of the validation results, the paper also highlights the adjustment that has been made to the model to make it more plausible. It is concluded that in the current Tanzanian situation where there is no any model that can be used to predict and/or estimate wood fuel consumption, the developed wood fuel consumption predictive model can be useful in sustainable forest management strategies. Prior to its use, however, the constructed model needs to be further validated and adjusted accordingly using newly collected longitudinal data from the study area. Sufficient data should be collected from the strata (locations) commensurate with those used in the present study.