Nitrogen and soil organic matter benefits to maize by fast growing pure and mixed species legume fallows in Western Kenya

Abstract

Declining yields as a result of soil fertility depletion associated with continuous cropping without external inputs of nutrients is a major problem facing smallholder farmers in sub-Saharan Africa. SignificCiflt increases in crop yields have been reported following short duration leguminous fallows (9-18 months) with substantial residual benefits for following maize crops over at least two cropping seasons. However, despite large yield increases it is hypothesized that incorporation of fallow species residues may decompose too rapidly due to their high quality (high N, low lignin, low polyphenols) and N release will be asynchronous with crop demand. This may result in substantial N losses during the crop growth phase. Single species fallows may contribute less to long-term build-up of soil organic matter than associated with mixed species resulting in shorter residual yield benefits. It is also hypothesized that widespread use of single species fallows may lead to problems of pests and diseases and risks of fallow establishment failure. Mixed species fallows also have the potential to provide multiple products and increased biodiversity. Field and laboratory experiments were undertaken to investigate the effects of pure and mixed fallows on tree biomass productivity, maize yield, synchrony between N release and crop demand and soil organic matter build-up after biomass incorporation. Short and long-term soil fertility and crop yield sustainability of improved fallow systems were predicted under current and alternative management options. Field experiments were conducted on farmers fields in western Kenya using Sesbania sesban, Grota/aria grahamiana, Gajanus cajan and Macropti/ium atropurpureum as single species fallows and mixtures of these with sesbania. Maize yields were increased by 175-309 % in the first crop following 15 months of pure and mixed legume fallows and increased by 37-197 % in the second crop compared with continuous maize cropping. These increases were comparable to an application of at least 100 kg N ha-1 of inorganic fertilizer in each of the two cropping seasons after the fallows indicating that fallows were able to supply large amounts of N. Leaching tube experiments showed that incorporation of legume fallow residues resulted in N mineralization patterns related to their qualities. Incorporation of calliandra, a species high in polyphenols, resulted in temporary immobilization of N but did not affect N release from sesbania when applied in a mixture. In field experiments, net mineralization of residue N was greatest two weeks after biomass incorporation and maize planting. Continuous maize cropping, natural weed fallow and calliandra residues resulted in the lowest mineral N releases and sesbania, macroptilium, cajanus and crotalaria residues and the mixtures with sesbania resulted in the highest mineral N releases. Total N uptake by maize ranged between 40 to 120 kg N ha-1 and was lowest in the continuous maize cropping and natural weed fallow and highest in the improved fallow treatments. Application of 15N-labelled residues showed that maize recovered more N from the higher quality (sesbania and macroptilium) than from lower quality calliandra residues. A large fraction (50-60%) of the applied residue 15N remained in the soil after incorporation, particularly in the single and mixed calliandra treatments. About 14-15% of N applied as sesbania and macroptilium residues was recovered by first maize crop and 3-6% was recovered in the second crop. About 30-40 % of applied N could not be accounted for. 15N distribution in particle size fractions showed that most calliandra N was found in the >20 IJ.mfraction but sesbania and macroptilium N was mostly in the < 20 IJ.mfraction. The 15Nstem injection technique showed that the contribution of fallow roots to maize N nutrition was less than 5%. Simulation of maize growth and yield using the WaNuLCAS model suggested maize growth to be N restricted after the first cropping season after the fallow, with significant leaching of N below 1 m. The simulated SOM-C and SOM-N build-up was greater in the high biomass crotalaria treatment. Fallowing with sesbania after every two cropping seasons or with crotalaria every 4 seasons, or mixing of the two species are recommended as possible management options to remediate soil nutrient depletion in smallholder farming systems in sub-Saharan Africa.