ZHANG Jianyun, XIE Kang, JIN Junliang, LIU Yanli, HE Ruimin, WANG Guoqing. Transformation and innovation of flood forecasting methods under a changing environmentJ. Advances in Water Science.
Citation: ZHANG Jianyun, XIE Kang, JIN Junliang, LIU Yanli, HE Ruimin, WANG Guoqing. Transformation and innovation of flood forecasting methods under a changing environmentJ. Advances in Water Science.

Transformation and innovation of flood forecasting methods under a changing environment

  • Frequent flood disasters have long been a critical hidden danger for the Chinese nation. Driven by the combined effects of global climate change and human activities, China’s temperature rise rate exceeds the global average, and extreme short-duration heavy rainfall is becoming more frequent and intense. The underlying surface of river basins has undergone drastic changes due to urbanization, water conservancy project construction and other human activities, which has broken the hypotheses of "consistency" and "stationarity" of hydrological processes. As a result, the applicability of traditional flood forecasting methods has been impaired, posing severe challenges to flood forecasting work. Aiming at the new problems of flood forecasting under a changing environment, this paper proposes that flood forecasting methods need to realize eight core transformations: shifting from lumped models to distributed models in model structure; from single observation sources to multi-source data assimilation and fusion in data input; from long-series calibration in well-gauged basins to appropriate methods for data-scarce basins in parameter determination; from one-way forecasting to precipitation-flood-scheduling-rolling forecasting in forecasting mode; from single flood forecasting to comprehensive forecasting for multiple disaster types and scenarios in forecasting business; from control section nodes to full hydrological elements of digital twin basins in forecasting elements; from deterministic forecasting to probabilistic and ensemble forecasting in forecasting results; and from physical mechanism models to the combination of physical mechanism models and data-driven models in forecasting models. Artificial intelligence technology shows remarkable application potential in flood forecasting and engineering practice, yet it is necessary to consolidate the foundation of hydrological physical process models, construct flood forecasting methods adapted to the changing environment, and improve the comprehensive defense capacity against flood disasters in river basins..
  • loading

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return