Deakin University

File(s) under permanent embargo

Meta-analysis evaluating water use, water saving, and water productivity in irrigated production of rice with SRI vs. standard management methods

journal contribution
posted on 2023-07-28, 04:14 authored by P Jagannath, Hemant PullabhotlaHemant Pullabhotla, N Uphoff
A meta-analysis was done of data from 29 published studies comparing SRI and non- SRI methods for irrigated rice production that had reported results from a total of 251 comparison trials. The purpose was to assess the differences in total and irrigation water use associated with SRI vs. non-SRI rice crop management practices, evaluating water reductions achieved with SRI management and calibrating differences in water productivity. To avoid purely nominal classifications, a SRI characterization matrix was used to assess the degree to which the respective trials had represented SRI management, based on the number and extent of specified agronomic practices used. Water use, yield and water productivity relationships were assessed based on the data extracted from the published literature. Data reported in different units were standardized to compare the use and productivity of water when producing rice in eight countries, assessing both irrigation water use and total water use (irrigation plus rainfall). Descriptive statistical analysis showed a clear advantage in water use and water productivity for SRI management compared to using more standard cultivation methods. Mean total water use in the trials with SRI management was 12.03 million liters ha-1, compared to 15.33 million liters ha-1 when more conventional methods were used, including continuous flooding of rice paddies. This average total water savings of 3.3 million liters ha-1 represented a 22% water saving. In the 17 studies that specifically reported and analyzed irrigation and not just total water use, the irrigation water applications averaged 7.2 million liters ha-1 with SRI management, compared to 11.1 million liters ha-1 under non-SRI practice, a reduction of 3.9 million liters ha-1 (35%). Average paddy in the trials using mostly SRI methods was 5.9 tons ha-1 compared to 5.3 tons ha-1 when using more standard rice-production practices. Higher yield was seen in all the studies with less input of water whenever mostly SRI crop management practices were introduced. Given the empirical methodology employed for classifying SRI, these trials did not necessarily make full use of the recommended methods. Only 30% of trials analyzed as 'SRI' employed 80% or more of the SRI recommendations. This evaluation of SRI water productivity was thus not based on ideal-type SRI crop and water management. Total water use efficiency (TWUE) was found to be 52% greater with use of most of the SRI methods, as mean productivity for SRI practices across all the trials was 0.6 grams of grain per liter of water, compared to 0.39 grams of grain per liter produced with non-SRI methods. The latter statistic is the same as the current average irrigation water productivity in rice production (Bouman et al. 2007). In terms of irrigation water use efficiency (IWUE), SRI trials showed an even greater advantage, producing on average 1.23 grams of grain per liter of irrigation water, compared to 0.69 gram of grain per liter produced with non-SRI crop management, an advantage of 78%. Further analysis showed that these advantages of water saving and water productivity with SRI management were manifested across widely varying contexts for rice production. The meta-analysis considered differences in water use efficiency across variations in cropping season (wet vs. dry); in climate, soil texture and soil pH; and in rice variety planted (length of crop cycle). Improvements in WUE associated with SRI management practices were tested and confirmed by multivariate regression analysis. Many interests will be served by being able to reduce water requirements for paddy cultivation. SRI has sometimes been characterized as too labor-intensive for widespread adoption, but a number of evaluations have shown SRI to be labor-neutral, or even laborsaving. The costs of SRI practice need to be further reduced and the means of SRI dissemination further refined. Improvements should be made in farmers' technical and organizational control over water so that they can apply smaller amounts more reliably to their paddy crops. Such investments become more economically justified as the value and cost of water increase in future. SRI management is not a solution to all problems and constraints in the rice sector, but it is an innovation in rice production freely available to benefit producers, consumers and the environment, increasing food production and the economic returns to farmers at the same time that it reduces the rice sector's requirements for irrigation water.



Taiwan Water Conservancy







Usage metrics

    Research Publications


    No categories selected


    Ref. manager