Hydrology and Earth System Sciences (Jun 2022)

Analysis of flash droughts in China using machine learning

  • L. Zhang,
  • L. Zhang,
  • L. Zhang,
  • Y. Liu,
  • Y. Liu,
  • L. Ren,
  • L. Ren,
  • A. J. Teuling,
  • Y. Zhu,
  • L. Wei,
  • L. Zhang,
  • S. Jiang,
  • X. Yang,
  • X. Fang,
  • H. Yin

DOI
https://doi.org/10.5194/hess-26-3241-2022
Journal volume & issue
Vol. 26
pp. 3241 – 3261

Abstract

Read online

The term “flash drought” describes a type of drought with rapid onset and strong intensity, which is co-affected by both water-limited and energy-limited conditions. It has aroused widespread attention in related research communities due to its devastating impacts on agricultural production and natural systems. Based on a global reanalysis dataset, we identify flash droughts across China during 1979–2016 by focusing on the depletion rate of weekly soil moisture percentile. The relationship between the rate of intensification (RI) and nine related climate variables is constructed using three machine learning (ML) technologies, namely, multiple linear regression (MLR), long short-term memory (LSTM), and random forest (RF) models. On this basis, the capabilities of these algorithms in estimating RI and detecting droughts (flash droughts and traditional slowly evolving droughts) were analyzed. Results showed that the RF model achieved the highest skill in terms of RI estimation and flash drought identification among the three approaches. Spatially, the RF-based RI performed best in southeastern China, with an average CC of 0.90 and average RMSE of the 2.6 percentile per week, while poor performances were found in the Xinjiang region. For drought detection, all three ML technologies presented a better performance in monitoring flash droughts than in conventional slowly evolving droughts. Particularly, the probability of detection (POD), false alarm ratio (FAR), and critical success index (CSI) of flash drought derived from RF were 0.93, 0.15, and 0.80, respectively, indicating that RF technology is preferable in estimating the RI and monitoring flash droughts by considering multiple meteorological variable anomalies in adjacent weeks to drought onset. In terms of the meteorological driving mechanism of flash drought, the negative precipitation (P) anomalies and positive potential evapotranspiration (PET) anomalies exhibited a stronger synergistic effect on flash droughts compared to slowly developing droughts, along with asymmetrical compound influences in different regions of China. For the Xinjiang region, P deficit played a dominant role in triggering the onset of flash droughts, while in southwestern China, the lack of precipitation and enhanced evaporative demand almost contributed equally to the occurrence of flash drought. This study is valuable to enhance the understanding of flash droughts and highlight the potential of ML technologies in flash drought monitoring.