Improving dryland maize (Zea mays L.) water productivity in the Chokwe District of Mozambique through better nutrient management

Abstract

The southern region of Mozambique is characterized by arid to semi-arid climatic conditions with soils of poor fertility and low water retention capacity. The rainfall season is from September to April. In some areas, the rain season accommodates two production cycles, which is augmented by extended or unexpected rains in May and June. Maize is the main crop in this region. The major limiting factors for maize production in the Chókwè District under rainfed agriculture are rainfall amount and its distribution and soil fertility. Water productivity in this region is very low. The Challenge Program on Water and Food (CPWF), for which the slogan was “more crop per drop”, has attempted to identify and address water productivity constraints throughout the Limpopo River Basin (LRB). This study considers the water productivity in dryland areas, assuming that yields may not be only limited by water, but also by soil fertility. The study was aimed at investigating the improvement of water productivity by correcting nutrient deficiencies and recommending strategies to mitigate these deficiencies. A field experiment was conducted at Chókwè Agrarian Research Centre with maize cultivar (cv. Matuba). Matuba was selected because of its high tolerance to drought. Treatments were based on the most limiting soil nutrients at the experimental site. Crop parameters measured included total dry matter, fractional interception of photosynthetically active radiation (FIPAR), leaf area and grain yield. In addition, the Soil Water Balance (SWB) model was used to simulate potential yields with no nutrient limitations. Results of this study illustrated that the application of N resulted in improvements in total dry matter yield, leaf area index (LAI), FIPAR and water use efficiency (WUE). Application of both N and P improved the grain yield, leaf area duration (LAD) and WUE. SWB model simulations indicate that in only 1 out of 5 years in Chókwè District, the simulated yields were not higher compared to actual yields (0.2 - 1 ton ha-1). In conclusion, grain yield improvements are expected if nutrition is kept at optimum levels. This implies that in most years dryland yields are in fact nutrient limited and better nutrition can be used as a strategy to improve water productivity (WP) and grain yield