DEVELOPING IMPROVED DRYLAND CROPPING SYSTEMS FOR MAIZE IN SEMI-ARID TANZANIA. PART II. USE OF A MODEL TO EXTRAPOLATE AND ADD VALUE TO EXPERIMENTAL RESULTS

Abstract

Dryland agriculture is critically important to food security and rural livelihoods in Tanzania, but crop production is seriously constrained throughout the semi-arid lowlands by the rainfall regime. A major challenge is to develop improved cropping systems to alleviate the moisture constraint. Experimental evidence indicates that adoption of rainwater harvesting systems can bring benefits, but the restricted spatial and temporal extent of the experimental work leads to difficulties in extrapolation. This paper shows how the PARCHED-THIRST model can add value to the experimental results and provide important insights into their transferability. The model is seen as an aid to researchers, planners and extensionists in interpreting experimental results and designing locally appropriate interventions. Simulation based on 30 years of daily meteorological data provides an opportunity for temporal extrapolation. The long-term simulation allows an objective assessment of the risks and benefits associated with alternative rainwater harvesting systems. Simulation for different soils and modified rainfall regimes permits objective analysis of spatial transferability of experimental results to any other site for which rainwater harvesting interventions might be considered. It is shown that macrocatchment rainwater harvesting reduces drought risk within the target area, but may bring a serious risk of erosion due to excessively high flow rates. The overall assessment of the twin-track approach (experimentation+simulation) is that rainwater harvesting has potential for increasing productivity and sustainability of maize cropping systems in semi-arid Tanzania provided that the innovations are properly matched to the site-specific environmental conditions.

Abstract from the paper’s OpenAlex record, shown for indexing and research. Copyright remains with the authors or the publisher. Read at the source ↗

The text the model read, with the source record’s inline markup (italics, sub- and superscripts) rendered and its character entities decoded; quote offsets refer to the raw text.