Validation of the system of rice intensification (SRI) in Bangladesh

Abstract

A series of experiments was conducted in eastern Bangladesh to investigate the system of rice intensification (SRI). In the first, various components of SRI were studied: seedling raising methods, varieties, seedling age, plant spacing, organic manure application, and herbicides in rice grown with SRI management practices. In the second, the performance of SRI at the experimental station was compared to existing best management recommendations (BRRI recommendations). The third was a comparative study of the effects of the SRI, the BRRI and farmer's practices on rice yield in on-farm trials conducted in two sub-districts of Comilla district. Among the seedling raising methods, the normal seedbed method gave significantly higher yield than a modified dapog method, but was similar to polyethylene-covered seedbeds. In comparison of short- and long-duration varieties, the long-duration variety BRRI dhan29 yielded highest with SRI practices. In an experiment comparing plant spacing subjected to SRI, the grain yield of BRRI dhan29 was highest (7.53 t ha −1) at a spacing of 25 cm × 15 cm, and lowest at a spacing of 40 cm × 40 cm (5.10 t ha −1). Partial substitution of urea with organic manure applications at different rates had no effect on grain yield. Among weed control methods tested, the highest yields were obtained from a combination of herbicide application and a single hand weeding. In an on-station comparison, grain yield with conventional management following BRRI recommendations was significantly higher than under SRI management. In on-farm trials, BRRI recommended management performed significantly better than SRI and resulted in higher grain yield, lower cost and higher profit. SRI practice only had higher yield than farmers’ practices. SRI practice yielded an average of 5.95 t ha −1 while BRRI recommended and farmers’ practices resulted in 6.88 and 5.05 t ha −1, respectively. The total cost of rice production was highest for SRI, followed by the BRRI recommended management and farmers’ practices in two sites of Comilla district. However, the highest net return was obtained from BRRI management, whereas SRI and farmers’ practice had similar net returns. SRI required 12.9% more labor than the BRRI practice and 19.2% more than farmers’ practice. In SRI, 25 and 35% more labor was needed for weeding compared to BRRI and farmers’ practices, respectively. Results of these studies suggest that several of the key management principles stated in SRI had in fact little effect on rice yields. Moreover, SRI did not increase rice yields when it was compared to existing best management practices. This and the increased labor demand and poor economic performance may make it an unattractive choice for rice farmers in Bangladesh.