No abstract

Rain-fed agriculture will remain the dominant source of staple food production and the livelihood foundation of the majority of the rural poor in sub-Saharan Africa (SSA). Greatly enhanced investment in agriculture by a broad range of stakeholders will be required if this sector is to meet the food security requirements of tomorrow's Africa. However, production uncertainty associated with between and within season rainfall variability remains a fundamental constraint to many investors who often over estimate the negative impacts of climate induced uncertainty. Climate change is likely to make matters worse with increases in rainfall variability being predicted. The ability of agricultural communities and agricultural stakeholders in SSA to cope better with the constraints and opportunities of current climate variability must first be enhanced for them to be able to adapt to climate change and the predicted future increase in climate variability. Tools and approaches are now available that allow for a better understanding, characterization and mapping of the agricultural implications of climate variability and the development of climate risk management strategies specifically tailored to stakeholders needs. Application of these tools allows the development and dissemination of targeted investment innovations that have a high probability of biophysical and economic success in the context of climate variability.

APSIM (Agricultural Production Systemsof both the total amounts in the whole projle and their distribution with depth. Since neither of these datasets included measurements of the runof component of the water balance, this aspect of model performance was evaluated, and shown to be generally good, using data from a third source where runoff had been measured from contour bay catchments. 0 1997

The mineralization/immobilization of nitrogen when organic sources are added to soil is represented in many simulation models as the outcome of decomposition of the added material and synthesis of soil organic matter. These models are able to capture the pattern of N release that is attributable to the N concentration of plant materials, or more generally the C:N ratio of the organic input. However the models are unable to simulate the more complex pattern of N release that has been reported for some animal manures, notably materials that exhibit initial immobilization of N even when the C:N of the material suggests it should mineralize N. The APSIM SoilN module was modified so that the three pools that constitute added organic matter could be specified in terms of both the fraction of carbon in each pool and also their C:N ratios (previously it has been assumed that all pools have the same C:N ratio). It is shown that the revised model is better able to simulate the general patterns on N mineralized that has been reported for various organic sources. By associating the model parameters with measured properties (the pool that decomposes most rapidly equates with water-soluble C and N; the pool that decomposes slowest equates with lignin-C) the model performed better than the unmodified model in simulating the N mineralization from a range of feeds and faecal materials measured in an incubation experiment.

No abstract

The productivity and residual benefits of four grain legumes to sorghum (Sorghum bicolor) grown in rotation were measured under semi-arid conditions over three cropping seasons. Two varieties of each of the grain legumes; cowpea (Vigna unguiculata); groundnut (Arachis hypogaea); pigeon pea (Cajanus cajan); Bambara groundnut (Vigna subterranea), and sorghum were grown during the first season. The same experiment was implemented three times in different, but adjacent fields that had similar soil types. At the end of the season the original plots were split in two and residues were either removed or incorporated into the subplots. The following season sorghum was planted in all subplots. In 2002/03 (314 mm rainfall) cowpeas produced the largest dry grain yield (0.98 and 1.36 t ha −1 ) among the legumes. During the wettest year (2003/04, 650 mm rainfall) groundnut had the highest yields (0.76 to 1.02 t ha −1 ). In 2004/05 (301 mm rainfall) most legume yields were less than 0.5 t ha −1 , except for pigeon pea. Estimates of % N from N 2 -fixation from the legumes were 15-50% (2002/03), 16-61% (2003/04) and 29-83% (2004/05). Soil water changes during the legume growth cycle were proportional to varietal differences in total legume biomass. Sorghum grain yield after legumes reached up to 1.62 t ha −1 in 2003/04 compared with 0.42 t ha −1 when following sorghum. In 2004/05, sorghum yields after legumes were also higher (up to 1.26 t ha −1 ) than sorghum after sorghum. Incorporation of crop residues had no significant effect on sorghum yield. Beneficial effect of legumes on yields of the subsequent sorghum crop were more readily explained by improvements in soil nitrogen supply than by the small observed changes in soil water relations. Our results demonstrate clear potential benefits for increasing grain legume cultivation in semi-arid environments through the use of improved germplasm, which also gave substantial increases in subsequent sorghum productivity (up 200% in a wet season and 30-100% in a dry season), compared with an unfertilized sorghum crop following sorghum.

Globally, a range of agronomic factors have been reported to have an impact on the performance of conservation agriculture (CA) and often determine its performance in relation to conventional agriculture (CONV). To assess this performance in Zimbabwe, 48 CA experiments were conducted by the International Crops Research Institute for the Semi-Arid Tropics in the semi-arid areas of southern Zimbabwe from 2004 to 2010, to calculate the weighted mean difference (WMD) through meta-analytical methods. The two CA practices, planting basins (Basins) and ripper tillage (Ripper), were compared with CONV. It was hypothesised that CA results improved yield compared with CONV and that the effect of CA practices on yield is affected by soil type, rainfall amount and distribution and selected management practices, which included rates of inorganic fertilisers and manures and mulching. Basins were superior to CONV in 59% of the experiments and the overall effect was significant (p < 0.001). The effect of Ripper was non-significant. The hypothesis that CA practices result in improved maize grain yield over CONV was accepted for Basins. The WMD for experiments conducted on sandy soils was 0.365 t ha −1 for Basins and 0.184 t ha −1 for Ripper, and in both cases was significant (p < 0.05). For clay soils, only the WMD for Basins was significant. A higher rainfall regime (500-830 mm) resulted in a lower WMD for Basins (0.095 t ha −1 ) and Ripper (0.105 t ha −1 ) compared with 0.151 t ha −1 for Basins and 0.110 t ha −1 for Ripper under lower rainfall (320-500 mm). The overall effect of Basins under the higher rainfall regime was not significant. There was better yield performance for Basins when the rainfall was well distributed; the reverse was noted for the Ripper. The application of 10-30 kg ha −1 of N (micro-dose range) resulted in a higher WMD for Basins than zero N application. Without N application, the WMD of Basins was not significant. For zero manure application in Basins, the WMD was 0.043 t ha −1 compared with 0.159 t ha −1 when manure was applied. The application of mulch depressed the WMD in Basins by 44% and Ripper by 89%. The hypothesis that yield performance under CA is influenced by soil type, rainfall amount and distribution, inorganic fertiliser and manure application was accepted.

Over the past 20 years, farming systems modelling has become an accessible tool for developing intervention strategies targeted at smallholder farmers in southern Africa. Applying the Agricultural Productions Systems sIMulator (APSIM) to credibly simulate key soil and crop processes in highly constrained, low yielding maize/legume systems has led to four distinct modes of use: (i) to add value to field experimentation and demonstration; (ii) in direct engagement with farmers; (iii) to explore key system constraints and opportunities with researchers and extension agencies; and (iv) in the generation of information for policy makers, bankers and insurance institutions. Examples of application in each of these modes are presented. Despite being demonstrated as an excellent tool for developing intervention strategies and extension material, the use of simulation is limited by a lack of competent local users. Better cooperation within the simulation community, sharing of climate, soil and crop parameterisation and validation datasets, and focussing of efforts on using models to benefit smallholder farmers are suggested as ways of increasing the use and relevance of simulation. Substantial investment in the training of agriculturalists and the further science development of systems simulation is required to tackle the enormous challenges facing agricultural development in the region.