Evaluation of numerical weather prediction model rainfall forecasts for streamflow forecasting
Durga Lal Shrestha, David Ewen Robertson, Quan Jun Wang, Thomas C. Pagano, Prasantha Hapuarachchi
Abstract
Durga Lal Shrestha, David Ewen Robertson, Quan Jun Wang, Thomas C. Pagano, Prasantha Hapuarachchi
Abstract
This study evaluates rainfall forecasts from the first generation Australian Numerical Weather Prediction (NWP) models over the Ovens catchment in southeast Australia. It also evaluates streamflow forecasts produced using the NWP rainfall forecasts as input to a hydrological model. The rainfall forecasts of four NWP models (ACCESS-G/80 km, ACCESS-R/37.5 km, ACCESS-A/12 km and ACCESS-VT/5 km) are compared to observed rainfall at gauging stations and to derived catchment rainfall for one year period of 2010-2011. The 5 and 12 km resolution NWP models appear to be relatively free from bias (i.e. As an experiment, the raw NWP model rainfall forecasts are used to force a semi-distributed conceptual rainfall runoff model to produce hourly streamflow forecasts. The rainfall forecasts have little influence on streamflow forecasts for lead times shorter than the time of concentration of the forecast point. For longer lead times, however, the rainfall forecasts result in better streamflow forecasts than assuming zero future rain. The streamflow forecasts may be further improved if better rainfall forecasts are obtained through post-processing of the NWP outputs.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
This study evaluates rainfall forecasts from the first generation Australian Numerical Weather Prediction (NWP) models over the Ovens catchment in southeast Australia. It also evaluates streamflow forecasts produced using the NWP rainfall forecasts as input to a hydrological model. The rainfall forecasts of four NWP models (ACCESS-G/80 km, ACCESS-R/37.5 km, ACCESS-A/12 km and ACCESS-VT/5 km) are compared to observed rainfall at gauging stations and to derived catchment rainfall for one year period of 2010-2011. The 5 and 12 km resolution NWP models appear to be relatively free from bias (i.e. As an experiment, the raw NWP model rainfall forecasts are used to force a semi-distributed conceptual rainfall runoff model to produce hourly streamflow forecasts. The rainfall forecasts have little influence on streamflow forecasts for lead times shorter than the time of concentration of the forecast point. For longer lead times, however, the rainfall forecasts result in better streamflow forecasts than assuming zero future rain. The streamflow forecasts may be further improved if better rainfall forecasts are obtained through post-processing of the NWP outputs.
Key concepts: Streamflow, Numerical weather prediction, Environmental science, Climatology, Meteorology, Flood forecasting, Drainage basin, Geography