Welcome to Chinese Journal of Applied Ecology! Today is Share:

Chinese Journal of Applied Ecology ›› 2018, Vol. 29 ›› Issue (4): 1051-1060.doi: 10.13287/j.1001-9332.201804.009

• Special Features of Hydrological Variability and Inconsistency • Previous Articles     Next Articles

Hydrologic variability and sensitivity based on Hurst coefficient and Bartels statistic

LEI Xu1, XIE Ping1,2*, WU Zi-yi1, SANG Yan-fang3, ZHAO Jiang-yan1, LI Bin-bin4   

  1. 1State Key Laboratory of Water Resources and Hydropower Engineering Science, Wuhan University, Wuhan 430072, China;
    2Collaborative Innovation Center for Territorial Sovereignty and Maritime Rights, Wuhan 430072, China;
    3Key Laboratory of Water Cycle and Related Land Surface Process, Chinese Academy of Sciences, Beijing 100101, China;
    4Middle Yangtze River Bureau of Hydrology and Water Resources Survey, Wuhan 430012, China;
  • Received:2018-01-18 Online:2018-04-18 Published:2018-04-18
  • Contact: * E-mail: pxie@whu.edu.cn
  • Supported by:

    This work was supported by the National Natural Science Foundation of China (91547205, 91647110, 51579181), the Water Engineering and Science Project of Hunan Province (Xiangshuikeji [2015]13-21), and the ‘Bingwei’ Youth Innovation Promotion Association of Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences.

Abstract: Due to the global climate change and frequent human activities in recent years, the pure stochastic components of hydrological sequence is mixed with one or several of the variation ingredients, including jump, trend, period and dependency. It is urgently needed to clarify which indices should be used to quantify the degree of their variability. In this study, we defined the hydrological variability based on Hurst coefficient and Bartels statistic, and used Monte Carlo statistical tests to test and analyze their sensitivity to different variants. When the hydrological sequence had jump or trend variation, both Hurst coefficient and Bartels statistic could reflect the variation, with the Hurst coefficient being more sensitive to weak jump or trend variation. When the sequence had period, only the Bartels statistic could detect the mutation of the sequence. When the sequence had a dependency, both the Hurst coefficient and the Bartels statistics could reflect the variation, with the latter could detect weaker dependent variations. For the four variations, both the Hurst variability and Bartels variability increased with the increases of variation range. Thus, they could be used to measure the variation intensity of the hydrological sequence. We analyzed the temperature series of different weather stations in the Lancang River basin. Results showed that the temperature of all stations showed the upward trend or jump, indicating that the entire basin had experienced warming in recent years and the temperature variability in the upper and lower reaches was much higher. This case study showed the practicability of the proposed method.