An evaluation of the Standardized Precipitation Index for assessing inter‐annual rice yield variability in the Ganges–Brahmaputra–Meghna region. (14th September 2015)

• Record Type: Journal Article
• Authors: Kattelus, Mirja; Salmivaara, Aura; Mellin, Ilkka; Varis, Olli; Kummu, Matti
• Is Part Of: International journal of climatology. Volume 36:Number 5(2016)
• Page Start: 2210
• Page End: 2222
• Publication Date: 2015-09-14
• Subjects: SPI -- drought -- precipitation -- climate variation -- rice yield -- Ganges -- Brahmaputra -- Meghna; Climatology -- Periodicals; Climat -- Périodiques; Climatologie -- Périodiques; 551.605
• DOI: 10.1002/joc.4489
• Languages: English
• ISSNs: 0899-8418

Abstract

ABSTRACT: Climate variability has major impacts on crop yields and food production in South Asia. The spatial differences of the impact are not, however, well understood. In this study, we thus aim to analyse the spatio‐temporal relationship between precipitation and rice yields in the Ganges–Brahmaputra–Meghna region. The effects of rainfall variation on yields were analysed with regression models using the Standardized Precipitation Index (SPI) as an explanatory variable. Our results indicate that in large part of the study region, a strong relationship between precipitation and rice yields exists and the SPI at various lags chosen as the predictor variable performed well in describing the inter‐annual yield variability. However, the study demonstrated large spatial variations in the strength of this relationship or optionally in the suitability of the chosen methodology for investigating it. In the mid‐plains of the Ganges, which represent very important agricultural areas, precipitation variability has a strong impact on rice yields, while in downstream Ganges as well as in Brahmaputra, where precipitation is more abundant, the relationship was less pronounced. Where the performance of the regression models was weaker, it is likely that yield variation depended on other factors such as management practices or on other climate factors such as temperature. The results further showed that the SPI at 1, 3, 6 and 12 month lags calculated for the monsoon time (June–October) are most commonly the best at explaining the rice yield variability. The SPI can thus be considered a very useful predictor of rice yield variability in some parts of the study region, demonstrating that they could be used for agricultural applications and policy decisions to improve the region's food security.