Evaluation of flood prediction capability of the distributed Grid‐Xinanjiang model driven by weather research and forecasting precipitation

Abstract

The lead time of operational flood forecasting is critical for the effectiveness of flood alert and flood risk reduction. It is impossible to extend the lead time of flood forecasting by solely using rain gauge observations. However, the weather research and forecasting (WRF) model has the potential to produce quantitative precipitation forecasts that can facilitate the flood risk management by increasing the flood forecasting lead time. This study investigates the flood prediction capabilities of the well-tested Grid-Xinanjiang model (GXM) in a flood-prone area, located in the upper region of the Huaihe River Basin, when driven by gauge observations and WRF precipitation forecasts, respectively. The results indicate that GXM is capable of producing improved flood predictions by using the WRF precipitation forecasts. The incoming floods are difficult to be predicted in advance by using the gaugemeasured precipitation, especially when the lead time is larger than the flow concentration time. However, with the WRF forecasts, the occurrence of flood events can be predicted for longer lead times. This study also demonstrates that the temporal and spatial patterns of precipitation forecasts have an important impact on the prediction of both timing and magnitude of incoming floods. K E Y W O R D Sflood defence measures, forecasting and warning, hydrological modelling, precipitation

Evaluation of Flood Prediction Capability of the WRF-Hydro Model Based on Multiple Forcing Scenarios

“…The WRF model, a new generation of mesoscale NWP model, has been widely used in the past 20 years [36,53]. It can produce high-resolution (1-10 km) simulations of meteorological variables such as rainfall [33,34].…”

Physically-based landslide prediction over a large region: Scaling low-resolution hydrological model results for high-resolution slope stability assessment

“…Low-resolution but large scale hydrological models (typically applied at 1 km resolution or coarser) are widely available due to increasing availability of climate data and land surface datasets (Sun et al, 2017;Huang et al, 2019;Yao et al, 2019;van Beek et al, 2011;Wada et al, 2011;Luo et al, 2018). These models can be calibrated against observed discharge (Beck et al, 2016;Chao et al, 2018) so it captures the spatiotemporal variations across the basin.…”

Analysis of the specific water level of flood control and the threshold value of rainfall warning for small and medium rivers

“…The calculation of rainfall threshold by back extrapolation is based on the inverse calculation of the disaster flow to the catchment flow, under the condition of known confluence parameters, three undetermined elements are also involved. The first is the amount of precipitation, the second is the time distribution of precipitation, and the third is the soil moisture in the early stage (Yao et al, 2019). Under the combination of three uncertain factors, in order to form the same peak value, the solutions will be infinite, which is equal to "no solution".…”

“…Shitai county is non-data area, so the design flood uses the existing methods of territorial and peripheral areas (AS-DIWCH, 1984;Zhang et al, 2015). The area comparison method takes the gaotan hydrological station downstream of shitai of qiupu river as the reference station.…”