Rice yield and its relation to root growth and nutrient-use efficiency under SRI and conventional cultivation: an evaluation in Madagascar

Abstract

While plant growth and productivity are known to derive from the interaction between genetic potential (G) and environmental factors (E), efforts to improve rice production have usually proceeded assuming a standard E that is created by conventional rice-growing practices. Genotypes have been assessed for their performance in continuously flooded paddy soils, with optimally dense plant populations, with reliance on inorganic fertilization to raise yields. The System of Rice Intensification (SRI) developed in Madagascar and now becoming accepted in much of Asia proposes that GxE interactions can be made more productive with different management practices: optimally sparse populations, established with very young seedlings carefully transplanted, intermittent flooding of paddies, with active soil aeration and with soil organic matter enhanced as much as possible. This article evaluates the effects of alternative SRI cultural practices on grain yield with particular attention to their impact on the growth and functioning of rice plant roots and on associated nutrient-use efficiencies that could be contributing to the observed higher grain yields. On-station experiments and on-farm surveys were conducted in Madagascar to evaluate SRI practices in comparison with standard cultural methods, considering how rice plants’ expression of their genetic potential was affected by different crop management practices. Controlling for both soil and farmer effects, rice plants cultivated with SRI methods produced average yields more than double those from standard practice (6.26 vs. 2.63 t ha⁻¹). The most evident phenotypic difference was in plant root growth, assessed by root-pulling resistance (RPR), a summary measure of root system development. On average, uprooting single SRI plants required 55.2 kg of force plant⁻¹, while pulling up clumps of three conventionally grown plants required 20.7 kg hill⁻¹, or 6.9 kg plant⁻¹. SRI plants thus offered 8 times more resistance per plant to uprooting. Direct measurements confirmed that SRI methods induced both greater and deeper root growth, which could be contributing to increased nutrient uptake throughout the crop cycle, compared with the shallower rooting and shorter duration of root functioning under continuous flooding. Rice plants grown with SRI methods took up more macronutrients than did the roots of conventionally managed plants, which was reflected in the higher SRI yields. When grain yield was regressed on nutrient uptake to assess nutrient-use efficiency, SRI plants achieved higher grain yield per unit of N taken up, compared to plants grown with conventional methods. The internal efficiency (IE) of SRI plants in utilizing macronutrients was 69.2 for N, 347.2 for P, and 69.7 for K, while the IE in plants conventionally grown was 74.9, 291.1, and 70.4 for these three macronutrients, respectively. Although no significant differences in IE were observed for N and K, the uptake of P was significantly greater, indicating more efficient use of P by SRI plants for grain production. More research needs to be done on such relationships, but this study indicates that productive changes in the structure and functioning of rice plants, particularly their roots, can be induced by alternative management methods.