Comparative evaluation of maximum likelihood ensemble filter and ensemble Kalman filter for real-time assimilation of streamflow data into operational hydrologic models

Various data assimilation (DA) methods have been used and are being explored for use in operational streamflow forecasting. For ensemble forecasting, ensemble Kalman filter (EnKF) is an appealing candidate for familiarity and relative simplicity. EnKF, however, is optimal in the second-order sense, only if the observation equation is linear. As such, without an iterative approach, EnKF may not be appropriate for assimilating streamflow data for updating soil moisture states due to the strong nonlinear relationships between the two. Maximum likelihood ensemble filter (MLEF), on the other hand, is not subject to the above limitation. Being an ensemble extension of variational assimilation (VAR), MLEF also offers a strong connection with the traditional single-valued forecast process through the control, or the maximum likelihood, solution. In this work, we apply MLEF and EnKF as a fixed lag smoother to the Sacramento (SAC) soil moisture accounting model and unit hydrograph (UH) for assimilation of streamflow, mean areal precipitation (MAP) and potential evaporation (MAPE) data for updating soil moisture states. For comparative evaluation, three experiments were carried out. Comparison between homoscedastic vs. heteroscedastic modeling of selected statistical parameters for DA indicates that heteroscedastic modeling does not improve over homoscedastic modeling, and that homoscedastic error modeling with sensitivity analysis may suffice for application of MLEF for soil moisture updating using streamflow data. Comparative evaluation with respect to the model errors associated with soil moisture dynamics, the ensemble size and the number of streamflow observations assimilated per cycle showed that, in general, MLEF outperformed EnKF under varying conditions of observation and model errors, and ensemble size, and that MLEF performed well with an ensemble size as small as 5 while EnKF required a much larger ensemble size to perform closely to MLEF. Also, MLEF was not very sensitive to the uncertainty parameters and performed reasonably well over relatively wide ranges of parameter settings, an attribute desirable for operational applications where accurate estimation of such parameters is often difficult.

[1]  Victor Koren,et al.  Assimilation of streamflow and in situ soil moisture data into operational distributed hydrologic models: Effects of uncertainties in the data and initial model soil moisture states , 2011 .

[2]  G. Evensen,et al.  Data assimilation and inverse methods in terms of a probabilistic formulation , 1996 .

[3]  Kuolin Hsu,et al.  Uncertainty assessment of hydrologic model states and parameters: Sequential data assimilation using the particle filter , 2005 .

[4]  Oldrich Rakovec,et al.  State updating of a distributed hydrological model with Ensemble Kalman Filtering: Effects of updating frequency and observation network density on forecast accuracy , 2012 .

[5]  Ionel M. Navon,et al.  Optimality of variational data assimilation and its relationship with the Kalman filter and smoother , 2001 .

[6]  N. Verhoest,et al.  State and bias estimation for soil moisture profiles by an ensemble Kalman filter: Effect of assimilation depth and frequency , 2007 .

[7]  William R. Cotton,et al.  Development of an Aerosol Retrieval Method: Description and Preliminary Tests , 2008 .

[8]  Miguel A. Rico-Ramirez,et al.  EGU General Assembly , 2011 .

[9]  M. Zupanski.  Maximum Likelihood Ensemble Filter: Theoretical Aspects , 2005 .

[10]  Dong Jun Seo,et al.  Automatic state updating for operational streamflow forecasting via variational data assimilation , 2009 .

[11]  Geir Nævdal,et al.  An Iterative Ensemble Kalman Filter , 2011, IEEE Transactions on Automatic Control.

[12]  Mekonnen Gebremichael,et al.  Rainfall: State of the Science , 2010 .

[13]  Martyn P. Clark,et al.  HEPEX: The Hydrological Ensemble Prediction Experiment , 2007 .

[14]  Jutta Thielen,et al.  Aims, challenges and progress of the Hydrological Ensemble Prediction Experiment (HEPEX) following the third HEPEX workshop held in Stresa 27 to 29 June 2007 , 2008 .

[15]  Wade T. Crow,et al.  Improving hydrologic predictions of a catchment model via assimilation of surface soil moisture , 2011 .

[16]  Haksu Lee,et al.  Variational assimilation of streamflow into operational distributed hydrologic models: effect of spatiotemporal scale of adjustment , 2012 .

[17]  Konstantine P. Georgakakos,et al.  Impacts of parametric and radar rainfall uncertainty on the ensemble streamflow simulations of a distributed hydrologic model , 2004 .

[18]  Fred C. Schweppe,et al.  Uncertain dynamic systems , 1973 .

[19]  Martyn P. Clark,et al.  Incorporating Medium-Range Numerical Weather Model Output into the Ensemble Streamflow Prediction System of the National Weather Service , 2005 .

[20]  James D. Brown,et al.  The Science of NOAA's Operational Hydrologic Ensemble Forecast Service , 2014 .

[21]  S. Sorooshian,et al.  Stochastic parameter estimation procedures for hydrologie rainfall‐runoff models: Correlated and heteroscedastic error cases , 1980 .

[22]  G. Evensen,et al.  An ensemble Kalman smoother for nonlinear dynamics , 2000 .

[23]  R. Ibbitt,et al.  Hydrological data assimilation with the ensemble Kalman filter: Use of streamflow observations to update states in a distributed hydrological model , 2007 .

[24]  Xianhong Xie,et al.  Data assimilation for distributed hydrological catchment modeling via ensemble Kalman filter , 2010 .

[25]  Dusanka Zupanski,et al.  Estimation of global CO2 fluxes at regional scale using the maximum likelihood ensemble filter , 2008 .

[26]  J. Whitaker,et al.  Ensemble Data Assimilation without Perturbed Observations , 2002 .

[27]  Yuqiong Liu,et al.  Uncertainty in hydrologic modeling: Toward an integrated data assimilation framework , 2007 .

[28]  Jaap Schellekens,et al.  The use of MOGREPS ensemble rainfall forecasts in operational flood forecasting systems across England and Wales. , 2011 .

[29]  G. Evensen.  Sequential data assimilation with a nonlinear quasi‐geostrophic model using Monte Carlo methods to forecast error statistics , 1994 .

[30]  Julie Demargne,et al.  Generation of ensemble precipitation forecast from single-valued quantitative precipitation forecast for hydrologic ensemble prediction , 2011 .

[31]  R. E. Kalman,et al.  A New Approach to Linear Filtering and Prediction Problems , 2002 .

[32]  Stephen E. Cohn,et al.  An Introduction to Estimation Theory (gtSpecial IssueltData Assimilation in Meteology and Oceanography: Theory and Practice) , 1997 .

[33]  J. Schaake,et al.  Precipitation and temperature ensemble forecasts from single-value forecasts , 2007 .

[34]  Peter R. Oke,et al.  A deterministic formulation of the ensemble Kalman filter: an alternative to ensemble square root filters , 2008 .

[35]  A. Weerts,et al.  Particle filtering and ensemble Kalman filtering for state updating with hydrological conceptual rainfall‐runoff models , 2006 .

[36]  Murugesu Sivapalan,et al.  Soil moisture controls on patterns of grass green-up in Inner Mongolia: an index based approach , 2012 .

[37]  Haksu Lee,et al.  Utility of SCaMPR Satellite versus Ground-Based Quantitative Precipitation Estimates in Operational Flood Forecasting: The Effects of TRMM Data Ingest , 2014 .

[38]  Geir Evensen,et al.  The Ensemble Kalman Filter: theoretical formulation and practical implementation , 2003 .

[39]  Roman Krzysztofowicz,et al.  Bayesian theory of probabilistic forecasting via deterministic hydrologic model , 1999 .

[40]  D. McLaughlin,et al.  Hydrologic Data Assimilation with the Ensemble Kalman Filter , 2002 .

[41]  Ricardo Todling,et al.  A fixed-lag Kalman smoother for retrospective data assimilation , 1994 .

[42]  A. Denning,et al.  Carbon flux bias estimation employing Maximum Likelihood Ensemble Filter (MLEF) , 2007 .

[43]  Jeffrey L. Anderson.  An Ensemble Adjustment Kalman Filter for Data Assimilation , 2001 .

[44]  Marc Bocquet,et al.  Ensemble Kalman filtering without the intrinsic need for inflation , 2011 .

[45]  Dong-Jun Seo,et al.  Real-Time Variational Assimilation of Hydrologic and Hydrometeorological Data into Operational Hydrologic Forecasting , 2003 .

[46]  A. Jazwinski.  Stochastic Processes and Filtering Theory , 1970 .

[47]  Dong-Jun Seo,et al.  Short-term ensemble streamflow forecasting using operationally-produced single-valued streamflow forecasts , 2010 .

[48]  W. Wagner,et al.  Improving runoff prediction through the assimilation of the ASCAT soil moisture product , 2010 .

[49]  Florian Pappenberger,et al.  Ensemble flood forecasting: a review. , 2009 .

[50]  D. Zupanski,et al.  Uncertainty analysis using the WRF maximum likelihood ensemble filter system and comparison with dropwindsonde observations in Typhoon Sinlaku (2008) , 2010 .

[51]  Yuqiong Liu,et al.  A wavelet-based approach to assessing timing errors in hydrologic predictions , 2011 .

[52]  Jaap Schellekens,et al.  Recent developments in operational flood forecasting in England, Wales and Scotland , 2009 .

[53]  Dong-Jun Seo,et al.  Variational assimilation of streamflow into operational distributed hydrologic models: effect of spatiotemporal adjustment scale , 2012 .