干旱区生态环境知识资源中心

Major challenges for combined use of organic and mineral nutrient sources in smallholder agriculture include variable type and quality of the resources, their limited availability, timing of their relative application and the proportions at which the two should be combined. Short-term nutrient supply capacity of five different quality organic resources ranging from high to low quality, namely Crotalaria juncea, Calliandra calothyrsus, cattle manure, maize stover and Pinus patula sawdust were tested in the field using maize as a test crop. The study was conducted on two contrasting soil types at Makoholi and Domboshawa, which fall under different agro-ecological regions of Zimbabwe. Makoholi is a semi-arid area (< 650 mm yr(-1)) with predominantly coarse sandy soils containing approximately 90 g kg(-1) clay while Domboshawa (> 750 mm yr(-1)) soils are sandy-clay loams with 220 g kg(-1) clay. Each organic resource treatment was applied at low (2.5 t C ha(-1)) and high (7.5 t C ha(-1)) biomass rates at each site. Each plot was sub-divided into two with one half receiving 120 kg N ha(-1) against zero in the other. At Makoholi, there was a nine-fold increase in maize grain yield under high application rates of C. juncea over the unfertilized control, which yielded only 0.4 t ha(-1). Combinations of mineral N fertilizer with the leguminous resources and manure resulted in between 24% and 104% increase in grain yield against sole fertilizer, implying an increased nutrient recovery by maize under organic-mineral combinations. Maize biomass measured at 2 weeks after crop emergence already showed treatment differences, with biomass yields increasing linearly with soil mineral N availability (R-2 = 0.75). This 2-week maize biomass in turn gave a positive linear relationship (R-2 = 0.82) with grain yield suggesting that early season soil mineral N availability largely determined final yield. For low quality resources of maize stover and sawdust, application of mineral N fertilizer resulted in at least a seven-fold grain yield increase compared with sole application of the organic resources. Such nutrient combinations resulted in grain harvest indices of between 44% and 48%, up from a mean of 35% for sole application, suggesting the potential of increasing maize productivity from combinations of low quality resources with mineral fertilizer under depleted sandy soils. At Domboshawa, grain yields averaged 7 t ha(-1) and did not show any significant treatment differences. This was attributed to relatively high levels of fertility under the sandy-clay loams during this first year of the trial implementation. Differences in N supply by different resources were only revealed in grain and stover uptake. Grain N concentration from the high quality leguminous resources averaged 2% against 1.5% from sawdust treatments. We conclude that early season soil mineral N availability is the primary regulatory factor for maize productivity obtainable under poor sandy soils. Maize biomass at 2 weeks is a potential tool for early season assessment of potential yields under constrained environments. However, the likely impact on system productivity following repeated application of high N-containing organic materials on different soil types remains poorly understood.