A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Generating Synthetic Streamflow Forecasts with Specified Precision
Synthetic hydrologic forecasts are often needed to evaluate water resources planning and management strategies when appropriate historical forecasts are not available. Synthetic forecasts can be generated to evaluate system performance with existing forecast products over historical periods when they were not available, or with different forecast products that do not yet exist. Synthetic forecast generation procedures should produce forecasts that are realistic and have the desired statistical properties. Two synthetic forecast generation techniques are proposed that create a time series of forecasts with (1) the desired mean, (2) the correct variance, and (3) the desired forecast precision. One uses a classical stochastic hydrology approach, and the other the generalized maintenance of variance extension (GMOVE) concept using historical hydrologic series as input. A critique is provided of several published synthetic forecast generation algorithms that produced unrealistic results. The GMOVE methodology is used in a stochastic optimization model of a single reservoir hydropower system. Using forecasts of varying precision, the example illustrates the ability of more precise forecasts to improve system operations.
Generating Synthetic Streamflow Forecasts with Specified Precision
Synthetic hydrologic forecasts are often needed to evaluate water resources planning and management strategies when appropriate historical forecasts are not available. Synthetic forecasts can be generated to evaluate system performance with existing forecast products over historical periods when they were not available, or with different forecast products that do not yet exist. Synthetic forecast generation procedures should produce forecasts that are realistic and have the desired statistical properties. Two synthetic forecast generation techniques are proposed that create a time series of forecasts with (1) the desired mean, (2) the correct variance, and (3) the desired forecast precision. One uses a classical stochastic hydrology approach, and the other the generalized maintenance of variance extension (GMOVE) concept using historical hydrologic series as input. A critique is provided of several published synthetic forecast generation algorithms that produced unrealistic results. The GMOVE methodology is used in a stochastic optimization model of a single reservoir hydropower system. Using forecasts of varying precision, the example illustrates the ability of more precise forecasts to improve system operations.
Generating Synthetic Streamflow Forecasts with Specified Precision
Lamontagne, J. R. (author) / Stedinger, J. R. (author)
2018-01-19
Article (Journal)
Electronic Resource
Unknown
Generating Synthetic Streamflow Forecasts with Specified Precision
| British Library Online Contents | 2018
Comparison of Precipitation and Streamflow Correcting for Ensemble Streamflow Forecasts
| DOAJ | 2018
Ensemble Combination of Seasonal Streamflow Forecasts
| Online Contents | 2016
Ensemble Combination of Seasonal Streamflow Forecasts
| ASCE | 2015
A Bayesian approach to probabilistic streamflow forecasts
| British Library Online Contents | 2013