Research papersEstimating hydrological parameters based on rainfall patterns in river basins with no long-term historical observations

- Estimation of hydrological parameters with limited observed data.
- Comprehensive evaluation of simulation results as a group of events.
- Establishment of statistical equations based on rainfall patterns.
- Mapping the distribution of hydrological parameters with rainfall patterns.

Small and medium river basins may frequently suffer from the destructive hydrological extremes (e.g., floods). However, the common problem in such regions is a lack of long-term historical observations. Meteorological and hydrological station networks in some river basins in China were newly-built only a few years ago, and it is infeasible to estimate hydrological parameters from calibration and validation with a long time period directly. This paper aims to develop a method to estimate the feasible hydrological parameters based on rainfall patterns in such regions. Digital Yellow River Integrated Model (DYRIM) is adopted as the hydrological model, and the feasible hydrological parameters can be estimated based on limited rainfall-runoff events. First, for each rainfall-runoff event, the parameters are independently calibrated with the observed rainfall and hydrological data using a double-layer parallel system. Then, the performances of the simulation results are comprehensively evaluated, and the value ranges of the parameters can be obtained. Finally, the statistical relationships between hydrological parameters and rainfall patterns (i.e., amount and intensity) are established, which are expressed by the statistical equations and the distribution of hydrological parameters with the rainfall patterns. From a sample demonstration, it is concluded that this parameter estimation method will be useful to estimate the feasible hydrological parameters for future rainfall-runoff events in river basins with no long-term historical observations.