Detecting the persistence of drying trends under changing climate conditions using four meteorological drought indices

Abstract

Numerous drought indices based on single or multiple variables have been developed for the assessment of drought. This study aimed to detect the variability of droughts by applying four drought indices, the standardized precipitation index, the reconnaissance drought index, the standardized precipitation evaporative drought index and the effective drought index, at five meteorological stations in the Songhua River basin (SRB) in China during 1962–2013 and at two meteorological stations in the Indus River basin (IRB) during 1980–2013 using different timescales (3, 6, 9 and 12 months). Moreover, a Mann−Kendall trend test at the 5% significance level was used for the assessment of drought trends in the basins. The results indicated that the 12 month data series for the SRB and the 6 month data series for the IRB were appropriate for assessing dry conditions and for comparing the drought indices. Pearson's correlation analyses between the drought indices results and the meteorological variables revealed that the standardized precipitation index is a better drought index for the assessment of drought in the SRB (r = 0.99). The results also showed that the standardized precipitation evaporative drought index versus the reconnaissance drought index is better at identifying drought in the IRB with a Pearson's correlation value r of 0.98. Severe and extremely dry conditions prevailed between 1990 and 2001 in the SRB, and there were extreme drought conditions with an intensity of −2.00 in the IRB during 2001–2002. Overall, this study provides insights into choosing the appropriate time interval and drought index for a specific region to identify the severity of droughts.