Determining Optimal Soil Moisture for Irrigated Rice in Indonesia with System of Rice Intensification

Abstract

In this study, an optimal combination of soil moisture for irrigated rice in Indonesia that maximizes both yield and water productivity of system of rice intensification (SRI) paddy field was determined by genetic algorithm (GA) model-based optimization. Before performing optimization, a formula to describe yield by soil moisture and meteorological parameters was identified using multiple non-linear regression analysis. The GA model was performed based on the identification process according to the empirical data during three cropping seasons. Here, we classified soil moisture level into three levels i.e. wet (W), medium (M) or dry (D) based on the soil water retention curve. As the results, the optimal soil moisture was a combination of wet, wet, medium, and dry levels for initial, crop development, mid-season and late season growth stages, respectively. We called this regime as W-W-M-D regime. The wet level in the initial and crop development growth stages should be achieved providing enough water for the plant to develop root, stem and tiller, and then the field can be drained into the medium level with the irrigation threshold of field capacity to avoid spike let sterility in mid-season stage and finally, let the field in the dry level to save more water in the late season stage when plant water requirement is minimum. By this scenario, it was simulated that the yield can be increased up to 8.35% and water productivity up to 13.49% with saving water up to 12.28% compared to the empirical data.