Soil fertility depletion ranks as the most important drawback to crop productivity in Sub-Saharan Africa. On-farm experiments were conducted during the 2014/2015 and 2015/2016 seasons to explore the effect of nitrogen (N), phosphorus (P) and rhizobia inoculation on biological nitrogen fixation, yields, nutritional and mineral components of common bean in smallholder farming areas of Eastern Zimbabwe. Two improved common bean cultivars (Gloria and NUA45), widely grown in Zimbabwe were tested in a split-plot arranged in randomized complete block design. The main plot factor was the combination of N (0 and 60 kg ha-1) and P (0 and 20 kg ha-1), and the sub-plot factors were cultivar (Gloria and NUA 45) and inoculation (+/- inoculum). Both N and P were applied at 20 kg ha-1 at planting, with an additional 40 kg N ha-1 top dressing applied at flowering for the treatments receiving N. At peak flowering, nitrogen fixation was estimated using the 15N natural abundance method for the P, NP, I and NPI treatments. At physiological maturity, grain samples were analyzed for protein, trypsin and phytate in response to N and P. The other micronutrients tested for the grain were iron, zinc, manganese, copper and boron. Number of nodules per plant, and active nodules were all both significantly increased by P application. Nitrogen application was particularly effective in increasing dry biomass, pod loading and number of seeds per pod. Using Bidens pilosa as the reference plant, the proportion of nitrogen fixed was not significant in the all treatments. Analysis of variance showed that variety, N and P fertilizers have no influence on nutritional components of grain grown in degraded sites. In non-degraded fields showed significant varietal differences –NUA 45 had higher Cu and Mn while Gloria was richer in Zn. Fertilization significantly increased grain Zn content but there was no benefit of co-application of N and P. Both variety and fertilization had no influence on grain protein, trypsin, phytate or iron content. On degraded sites that had approximately 0.32% SOC, none of the factors significantly increased grain yields (P > 0.05). The combined analysis of variance of grain yield obtained on the two sites that had SOC > 0.6% showed significant simple effects of N and P, as well as a significant NP interaction (P = 0.03), but with neither inoculation nor cultivar effect. These results suggest that farmers can invest in both N and P for common bean production, but only targeting soils that are not acutely degraded. Improved common bean cultivars currently on the market barely respond to the local rhizobia inoculum. Farmers can invest in P only for increased common bean zinc, in undegraded soils. Further investigations are recommended to ascertain the exact conditions and management under which the most nutrition gains for common bean are made.
MSc and Bcs thesis, internship reports