In much of southern Africa, lack of access to chemical fertilizers makes crop production dependent on biological nitrogen sources. One promising source is leguminous trees such as Tephrosia candida and Gliricidia sepium, which can be intercropped with maize and incorporated as green manure. However, these species’ resilience to climate variability – especially as seedlings – has not been evaluated. Furthermore, labor demands may constrain farmers’ ability to optimally manage these systems, resulting in late planting. We conducted a field experiment to test the effects of drought and planting date on productivity of Tephrosia and Gliricidiaseedlings in Malawi.
The experiment, established in November 2008, included four cropping treatments (sole maize; maize + Gliricidia; maize + Tephrosia; and Tephrosia fallow), two rainfall treatments (ambient; early cessation imposed by rainout shelters), and two treatments for seedling planting date (normal; late) in a full factorial design. Thirty-six plots (each 6 × 6 m) were arranged in randomized complete blocks with four replications. Survival and growth of individual seedlings was monitored monthly. Total seedling biomass was measured at incorporation (each year for Tephrosia intercropping, and after two years for Gliricidia intercropping and Tephrosia fallows). Maize yield was recorded annually (in May 2009, 2010, and 2011).
Results/Conclusions
By 2011, the agroforestry systems had boosted maize yields from 508 kg/ha to 947 kg/ha (Tephrosia intercropping), 2019 kg/ha (Gliricidia intercropping), and 2224 kg/ha (Tephrosia fallows). However, both drought and late planting decreased (p <0.01) maize yield by ~25%, although the drought was discontinued in 2010 and thus did not affect maize directly. Environmental stresses affected seedling performance: both drought and late planting significantly worsened Tephrosia mortality and reduced seedling height and biomass of both species. Late planting had a more drastic effect than drought. Gliricidia seedlings, however, showed no mortality (except from mechanical damage). Accordingly, residual effects of drought were less pronounced in Gliricidia plots.
Each agroforestry system struck a different balance between labor demands, maize yields, and resilience to stress. Gliricidia performed well under drought and suboptimal management, but it depressed maize yield during establishment and had high labor demands. Tephrosia relay intercropping yielded prompt benefits, but it was vulnerable to drought and late planting. Tephrosia improved fallows required little labor and gave the highest maize yields, but they precluded two years of cropping. In conclusion, although resilience during the seedling stage is an important characteristic for selecting agroforestry species, it may involve tradeoffs with other beneficial characteristics.