Statistical analysis of the relationship between summer monsoon precipitation extremes and foodgrain yield over India

The Indian economy largely depends on agriculture which is highly influenced by the spatio-temporal variability of precipitation. Kharif and rabi are the two main crop-growing seasons which require major proportion of rainfall. Increase in heavy precipitation events, however, can have adverse effects on the crops. This study, therefore, is mainly focused on understanding the variation of extremes in precipitation during summer monsoon season and its impact on kharif foodgrain yield over India. For this, several objectively defined indices of observed precipitation extremes, in terms of frequencies and intensities, have been computed for the period 1951-2003 using daily gridded data of 1° latitude � 1° longitude resolution. Efforts have also been made to prepare the climatology of extremes in precipitation (1961-1990) to determine their basic characteristics over the Indian region. Correlation analysis reveals that a large part of the country exhibits positive relationship between kharif foodgrain yield and the indices of extreme precipitation, significant at 5 level. Correlations with the indices of frequencies defined as seasonal count of days when rainfall exceeds 85th, 75th, and 65th percentiles (and 30, 20, and 10 mm) show that spatial extent and strength of the positive relationship decreases with increase in threshold values. This indicates that very heavy daily rainfall is less useful than the moderate daily rainfall. Strong negative relationship between the yield and indices of extreme precipitation observed over heavy precipitation areas of northeastern parts of the country indicate that very heavy daily rainfall have an adverse effect on crop. The study also points out that increase in heavy rainfall activities during July are favourable for the yield, but correlations become weak and statistically insignificant during the month of August. Individual crops show similar results as that obtained from aggregate kharif foodgrain, however, magnitude of the correlations varies from crop to crop