Real-Time Data Assimilation for Operational Ensemble Streamflow Forecasting

Abstract

Operational flood forecasting requires that accurate estimates of the uncertainty associated with modelgenerated streamflow forecasts be provided along with the probable flow levels. This paper demonstrates a stochastic ensemble implementation of the Sacramento model used routinely by the National Weather Service for deterministic streamflow forecasting. The approach, the simultaneous optimization and data assimilation method (SODA), uses an ensemble Kalman filter (EnKF) for recursive state estimation allowing for treatment of streamflow data error, model structural error, and parameter uncertainty, while enabling implementation of the Sacramento model without major modification to its current structural form. Model parameters are estimated in batch using the shuffled complex evolution metropolis stochasticensemble optimization approach (SCEM-UA). The SODA approach was implemented using parallel computing to handle the increased computational requirements. Studies using data from the Leaf River, Mississippi, indicate that forecast performance improvements on the order of 30% to 50% can be realized even with a suboptimal implementation of the filter. Further, the SODA parameter estimates appear to be less biased, which may increase the prospects for finding useful regionalization relationships.

Assimilation of stream discharge for flood forecasting: The benefits of accounting for routing time lags

“…The ensemble Kalman filter (EnKF) [Evensen, 1994], which originated from the discrete Kalman filter and the extended Kalman filter, has received significant attention in hydrological data assimilation because the hydrological model does not need to be linearized. The EnKF has been widely implemented for updating both lumped models [Moradkhani et al, 2005b;Pauwels and De Lannoy, 2009;Vrugt et al, 2006Vrugt et al, , 2005Weerts and El Serafy, 2006] and distributed models [Andreadis et al, 2007;Clark et al, 2008;Komma et al, 2008;Xie and Zhang, 2010] in streamflow forecasting applications. Another filtering method, the particle filter (PF), is receiving growing attention in hydrological data assimilation due to its ability to handle the propagation of non-Gaussian distributions through nonlinear models [Moradkhani et al, 2005a;Salamon and Feyen, 2009;Weerts and El Serafy, 2006].…”

Product‐error‐driven generator of probable rainfall conditioned on WSR‐88D precipitation estimates

“…While errors due to both factors have received much recent attention in the literature [e.g., Finnerty et al, 1997;Butts et al, 2004;Carpenter and Georgakakos, 2004;Wagener and Gupta, 2005;Borga et al, 2006;Huard and Mailhot, 2006;Kavetski et al, 2006aKavetski et al, , 2006bKuczera et al, 2006;Oudin et al, 2006], data assimilation and ensemble forecasting frameworks are considered effective means of improving forecasts in the face of uncertainty [e.g., Kavetski et al, 2002;Vrugt et al, 2005;Carpenter and Georgakakos, 2006a;Kavetski et al, 2006a;Moradkhani et al, 2005aMoradkhani et al, , 2005bMoradkhani et al, , 2006Oudin et al, 2006;Russo et al, 2006;Vrugt et al, 2006;Ajami et al, 2007;Gabellani et al, 2007;Smith et al, 2008]. Both approaches require specification of realistic uncertainty descriptions for the hydrologic models and their rainfall input.…”

Impact of climate change on water resources status: A case study for Crete Island, Greece

“…These processes are described by 16 parameters (Vrugt et al, 2006) that need to be 181 determined by the user or an optimization process using a suitable objective function. Because 182 of the number and nature of those parameters, objective function responses can contain 183 multiple optima.…”