Spatio-temporal Assessment and Monitoring of Agricultural Drought in Karnataka, India

Abstract

Agricultural drought monitoring is crucial as it affects food production and fodder, especially in countries like India; consequently, it affects the country’s economy, where nearly 70% of the population depends on agriculture for livelihood. Conventional drought indices consider the mean monthly rainfall values to assess the drought conditions by ignoring the intra-monthly rainfall variations. Due to climate change and erratic rainfall patterns, monthly mean values are not a suitable representation of the rainfall that occurred in the corresponding month. To address this challenge, the current study employed a methodology for calculating agricultural drought using a standardized net-precipitation evapotranspiration index (SNEPI) from 2000 to 2022 by accounting for rainfall variations at an intra-monthly scale. This study employed daily gridded rainfall data and monthly evapotranspiration obtained from the India Meteorological Department (IMD) and NASA’s global land data assimilation system (GLDAS), respectively, at 0.25° × 0.25° spatial resolution for the calculation of SNEPI. Intra-monthly variation of rainfall pattern is addressed by deriving the uniformity coefficient and multiplying it with mean monthly rainfall values. The results were compared with the widely used drought index, the standardized precipitation evapotranspiration index (SPEI). The spatial (agroclimatic zones and whole Karnataka level) and temporal (annual and monthly scale) analyses of SNEPI and SPEI were performed. According to the yearly reports of the Karnataka State Natural Disaster Management Centre (KSNDMC), the highest negative rainfall departure occurred in 2003 and 2016, both termed as deficiency periods. The results showed that in 2006, the drought was observed; however, the annual rainfall was near normal magnitude. Therefore, this study presented the detailed results of 2003, 2006, and 2016. A higher magnitude (0.98) of the correlation coefficient was observed for October, the monsoon season’s termination month. Also, the decreased correlations of 0.88, 0.88, and 0.84 were observed for the months of July, August, and September, respectively. This can be interpreted as increased intra-monthly variations, which SNEPI successfully captures, whereas SPEI ignores the variation. SNEPI succeeds in the early detection of drought events due to its ability to detect short-term dry and wet spells, which correlates well with SPEI at all considered months, inferred that it can be used for drought identification. The results suggest that this index is better for understanding the agricultural drought patterns spatially and temporally across Karnataka’s agro-climatic zones, which incorporates the intra-monthly rainfall variations and helps the agricultural community and policymakers. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.