Understanding cassava yield response to soil and fertilizer nutrient supply in West Africa

Enhanced understanding of plant and nutrient interactions is key to improving yields. We adapted the model for QUantitative Evaluation of the Fertility of Tropical Soils (QUEFTS) to assess cassava yield response to soil and fertilizer nutrients in West Africa. Data from on-station and farmer's field experiments across Togo and Ghana were used. Soil nutrient supply ranged from 86 to 177, 18-24 and 70-104 kg ha(-1) of N, P and K, respectively, and induced variable cassava responses to fertilizer application. Considering harvest (HI) in deriving the physiological nutrient use efficiencies for maximum dilution (PhEmax) and for maximum accumulation (PhEmin), which are key QUEFTS parameters, improved model predictions with reduction in normalized root mean square error from 32 to 13% at Davi, (Southern Togo) and from 18 to 13% at Kumasi (Southern Ghana). Model overestimated yields in Nyankpala (Northern Ghana) where drought stress reduced yields. Estimated PhEmin and PhEmax at HI of 0.50 were 41 and 96 kg kg(-1) N, 232 and 589 kg kg(-1) P, and 34 and 160 kg kg(-1) K. QUEFTS can be used for site-specific estimates of cassava yield responses to fertilizers under rain-fed conditions in West Africa, provided that yield is primarily constrained by N, P and K supplies, and not by drought or other nutrients.