Skip to content

Background

Hydrological forecasting is used for prediction of the future state of a hydrological system (e.g. river runoff and groundwater levels). It is relevant at different temporal scales, ranging from short- and medium-range (few hours to some days ahead) to long-range or seasonal forecasting (several months ahead). Accurate real-time hydrological forecasts are essential for protection against water-related hazards (e.g. floods), operation of infrastructure and water resources management. Recent advances in radar rainfall estimation and forecasting, numerical weather predictions, satellite remote sensing, and faster computing facilities are opening up new opportunities in real-time hydrological forecasting. More effective use of the different information sources via data assimilation will provide the basis for producing more accurate and more reliable forecasts.

For short-range hydrological forecasting quantitative precipitation estimates and forecasts from weather radars have proven to be valuable. However, as the forecast is extrapolated in time, the quality deteriorates. Extrapolation type forecast models, i.e. COTREC (Sokul,2011) , are typically based on the lowest part of the atmosphere in order to achieve good near-ground precipitation estimates, but they do not account for upper atmosphere dynamics.

Quantitative precipitation forecasts (QPF) from numerical weather prediction (NWP) models are, on the other hand, more skilful for larger lead times. Thus, there is a potential to bridge the temporal gap between NWP and radar based QPF by assimilation of radar data into NWP models (Bowler et al., 2006). A widely applied approach for data assimilation is based on nudging (e.g. Davioli and Buzzi, 2004). A key scientific challenge is related to the assimilation of the measured radar reflectivity into the NWP model, which requires efficient data quality algorithms to distinguish reflections of the actual precipitation from reflection from ground, sea-surface, external sources, etc., and to correct for the absorption depending on the strength and type of precipitation. In this regard, it is a challenge to account for the high spatial and temporal variation of the precipitation field and integrate this into the forecast model (Fabry and Seed, 2009). This requires a more refined 3D extrapolation model for the measured radar reflectivity and better methods for handling uncertainties in the measured data.

Within short- and medium-range flood forecasting recent research has focussed on the use of ensembles of QPF from NWP models to produce ensemble hydrological forecasts (Cloke and Pappenberger, 2009). Medium-range NWP ensemble forecast products, such as the ECMWF (European Centre for Medium-Range Weather Forecasts) ensemble prediction system, have been operational for more than two decades, but its application in hydrological forecasting is still relatively new. In recent years, development of high-resolution, short-range ensemble prediction systems is rapidly increasing as a result of increased computer power and research in ensemble prediction methods (e.g. Iversen et al., 2011). Use of NWP ensembles has a large potential in improving hydrological forecast skills, and, importantly, provide estimation of forecast uncertainty. Key scientific challenges are understanding and quantification of the different uncertainty sources in the forecast system and updating of the hydrological forecast model using data assimilation (Cloke and Pappenberger, 2009).

Seasonal forecasting of hydrological variables has been based on different probabilistic climate forecast approaches, including statistical models and ensemble prediction systems based on coupled atmosphere-ocean models (e.g. Barnston et al, 2010). Recent developments in climate forecasts have provided potential in improving seasonal forecasting for more efficient water resources management, which has not yet been fully explored (Mishra and Singh, 2011).

Within hydrological data assimilation research has primarily focused on assimilation of single data sources, including river runoff in flood forecasting models (e.g. Madsen and Skotner, 2005;Weerts and El Sarafy, 2006), groundwater levels in groundwater models (Hendricks Franssen et al., 2011), and soil moisture in land-surface models (Reichle et al., 2008). To fully utilise the complementary nature of different types of in-situ and remote sensing measurements of hydrological variables, multi-variate data assimilation within integrated hydrological modelling systems is required. In this regard, the increasing availability of remote sensing data of hydrological variables provides new opportunities for data assimilation that are largely unexplored (Reichle, 2008).

Key scientific challenges in hydrological data assimilation are related to the quantification of model and measurement uncertainties and implementation of computationally efficient data assimilation algorithms that are feasible for real-time applications. Proper treatment of model and measurement bias is critical for the performance of the data assimilation system (Drecourt et al., 2006). Recent developments in ensemble-based filtering algorithms with statistical regularisation (Sørensen et al., 2004; Evensen, 2007) have shown the potential for effective assimilation of large amounts of monitoring data in real-time forecasting systems. The combination of filtering and forecasting of model errors has shown an important potential in increasing forecast lead times (Madsen and Skotner, 2005).
The HydroCast project will address a number of the scientific challenges outlined above for further advancing hydrological forecasting and data assimilation. Specific innovative contributions of the project include:

  1. Development of new types of weather radar forecast models and data quality algorithms for assimilation of weather radar in NWP models and integration with hydrological modelling and data assimilation for short-range hydrological forecasting.
  2. Further advancement of the use of ensemble precipitation forecasts and hydrological data assimilation for providing seamless probabilistic hydrological forecasts, considering short-range, medium-range and seasonal forecasting.
  3. Development of a multi-variate data assimilation system for assimilation of in-situ and remote sensing measurements of hydrological variables in an integrated hydrological modelling system.

References:

Barnston, A.G., Li, S., Mason, S.J., DeWitt, D.G., Goddard, L., Gong, X., 2010, Verification of the first 11 years of IRI's seasonal climate forecasts, J. Appl. Meteor. Climatol., 49, 493–520, doi: 10.1175/2009JAMC2325.1.
Bowler, N.E., Pierce, C.E., Seed, A.W., 2006, STEPS: A probabilistic precipitation forecasting scheme which merges an extrapolation nowcast with downscaled NWP, Quar. J. Royal Met. Soc., 132 (620), 2127-2155.
Cloke, H.L., Pappenberger, F., 2009, Ensemble flood forecasting: A review, J. Hydrol., 375, 613-626.
Davolio, S., Buzzi, A., 2004, A Nudging scheme for the assimilation of precipitation data into a mesoscale model, Wea. Forecast., 19, 855-871.
Drécourt, J.P., Madsen, H., Rosbjerg, D., 2006, Bias aware Kalman filters: Comparison and improvements, Adv. Water Resour., 29, 707-718.
Evensen, G., 2007, Data Assimilation: The Ensemble Kalman Filter, Springer.
Fabry, F., Seed, A.W., 2009, Quantifying and predicting the accuracy of radar-based quantitative precipitation forecasts, Adv. Water Resour., 32(7), 1043-1049.
Hendricks-Franssen, H.J., Kaiser, H.P., Kuhlmann, U., Bauser, G., Stauffer, F., Müller, R., Kinzelbach, W., 2011, Operational real-time modeling with ensemble Kalman filter of variably saturated subsurface flow including stream-aquifer interaction and parameter updating, Water Resour. Res., 47, W02532, doi:10.1029/2010WR009480.
Iversen, T., Deckmyn, A., Santos, C., Sattler, K., Bremnes, J.B., Feddersen, H., Frogner, I., 2011, Evaluation of 'GLAMEPS'–a proposed multimodel EPS for short range forecasting, Tellus 63A, 513-530.
Madsen, H., Skotner, C., 2005, Adaptive state updating in real-time river flow forecasting - A combined filtering and error forecasting procedure, J. Hydrol., 308, 302-312.
Mishra, A.K., Singh, V.P., 2011, Drought modeling – A review, J. Hydrol., 403, 157-175.
Reichle, R., 2008, Data assimilation methods in the Earth sciences, Adv. Water Resour., 31, 1411–1418.
Reichle, R. H., Crow, W.T., Keppenne, C.L., 2008, An adaptive ensemble Kalman filter for soil moisture data assimilation, Water Resour. Res., 44, W03423, doi:10.1029/2007WR006357.
Sokul, Z., 2011, Assimilation of extrapolated radar reflectivity into a NWP model and its impact on a precipitation forecast at high resolution, Atm. Res., 100, 201–212.
Sørensen, J.V.T., Madsen H., Madsen H., 2004, Efficient sequential techniques for the assimilation of tide gauge data in three dimensional modeling of the North Sea and Baltic Sea system, J. Geophys. Res., 109, 10.1029/2003JC002144.
Weerts, A.H., El Serafy, G.Y.H., 2006, Particle filtering and ensemble Kalman filtering for state updating with hydrological conceptual rainfall-runoff models, Water Resour. Res., 42, doi:10.1029/2005WR004093.