Drought propagation in Northern China Plain: A comparative analysis of GLDAS and MERRA-2 datasets

Drought impact is closely related to water deficit phases within the hydrological cycle, and to drought propagation (the complex evolutions from meteorological drought to hydrological drought and to soil moisture drought). Besides the qualitative description, the quantitative analysis of drought propagation is still limited so far. Therefore, in this study, the propagation from meteorological to hydrological and to soil moisture droughts have been quantitatively analyzed. In particular, run theory has been utilized for quantifying key drought features including its duration and magnitude. The two available land assimilation datasets from 1981-2018 were selected for comparison purpose in Northern China Plain. Our results showed that although drought events identified by two datasets were not the same, both datasets revealed that meteorological drought occurred more frequently than hydrological and soil moisture droughts. And both demonstrated meteorological drought, hydrological drought and soil moisture drought were not synchronous. In particular, more than 91.89% of meteorological droughts led to hydrological droughts, and more than 87.10% of hydrological droughts caused soil moisture droughts. Furthermore, linear models showed the best for drought duration and magnitude. We found when meteorological drought was prolonged or shortened by one month, more than 91.89% probability it would lead to the extension or shortening of hydrological drought by 0.992 months with 1.687 unit in magnitude. And if the duration of hydrological drought changed by one month, more than 80.56% probability of the soil moisture drought duration would change by 1.006 months with 0.992 unit in magnitude. The results of duration and magnitude fitted well across the whole study area. Building on the above drought evolution information, practical drought mitigation measures and early warning system could be established.