Quantification of Yield Gaps in Rain-fed Rice, Wheat, Cotton and Mustard in India: Global Theme on Agroecosystems Report no. 43

In order to meet increasing demands of food due to rising population and income, food production in India and other south Asian countries need to be increased. Rain-fed agriculture in India, practiced on 94 million hectares (M ha), is considered a major source of production increase in future. This report analyses the magnitude of rain-fed potential yield gaps of rice, wheat, mustard and cotton crops, considering the spatial and temporal variation in climatic features. These yields can be interpreted as the upper limit that can be achieved by the current varieties in a rain-fed scenario with soil and weather as the only yield reducing factors. InfoCrop, a generic dynamic crop simulation model with sensitivity to variety, agronomic management, soil, weather, flooding, frost and pests and calibrated and validated in typical rain-fed and irrigated cultivating areas of these crops was used for quantification of rain-fed potential yields in different regions. Yields in technology maximization experiments (Plant Breeder’s fields) and on-farm technology demonstration plots (Front-line demonstrations) have also been used as additional measures of potential yield. Yield gaps were calculated as the difference between these yield levels and the region’s average measured yields. The results showed that irrespective of the definition of potential yield, there was considerable yield gap across all states in all crops indicating large scope for increasing rain-fed yields in future. On an average, the gap relative to simulated rain-fed potential yields was 2560 kg ha-1 for rice, 1120 kg ha-1 for cotton and 860 kg ha-1 for mustard. Such national average rain-fed yield gaps could not be estimated in wheat because of large percent of irrigated area in all states. The mean yield gap based on the average of simulated, experimental and on-farm rain-fed potential yields was 1670 kg ha-1 for rice, 770 kg ha-1 for cotton, 460 kg ha-1 for mustard and 70 kg ha-1 for wheat. It remains to be quantified if these biophysical estimates of yield gaps can be bridged economically