Water use of Guinea grass as affected by rainfed local conditions and practices in Sub-Saharan Africa

Abstract

Ongoing efforts are currently being made to rehabilitate drought-affected pastures in Sub-Saharan Africa. One approach being explored is the introduction of non-native grass species, such as Megathyrsus maximus (Guinea grass). This study aims to investigate the water use of Guinea grass in semi-arid environments under rainfed conditions. Additionally, it aims to a better understanding of the variability of water use in Guinea grass through the utilization of the Bagging machine learning algorithm. Split-plot field experiments were carried out over two consecutive rainy seasons (2020-2021). The treatments included two in-situ rainwater harvesting practices, RWH (ridging plus terracing and terracing alone), three seeding rates, SR (1.5, 2.5, and 3.5 kg ha-1), and two soil nitrogen fertilization rates, SF (95 kg N ha-1 and 0 kg N ha-1). These treatments were compared to a control plot that involved zero-tillage, no fertilization, and no rainwater harvesting. The collected datasets were analyzed using R, SPSS 15, and spreadsheets. The results showed significant differences in plant indices and soil moisture content among the treatments. However, the treatments had insignificant effects on seasonal actual crop evapotranspiration (ETa), which ranged from 1.93 to 3.29 mm day-1. The interactions between SR and RWH were found to have significant impacts on water use. The Bagging algorithm revealed that the variability in ETa could be attributed to SR (42%), RWH (31%), and SF (26%), respectively. The implementation of rainwater harvesting practices resulted in a significant reduction in water usage, saving 86% of the green water used with a water footprint of 0.25 m3 kg-1, compared to 1.7 m3 kg-1 for no adoption of RWH conditions. The water use of rainfed Guinea grass was also found highly sensitive to dry spells. Further detailed studies using multiple-layer models are recommended to gain a better understanding of the non-linear interactions in semi-arid environments.