A web based tool for operational real-time flood forecasting using data assimilation to update hydraulic states

This article describes an operational flood forecasting system set up for the city of Dijon, France. This system assimilates real-time flow data at an hourly time step with the stationary Kalman filter to update hydraulic states. It uses a semi-distributed hydrologic model to integrate rainfall measurements and forecasts and provide discharge forecasts at several points on the watershed. It also offers powerful data management tools and an elaborated graphical interface available from any computer connected to the Internet. The hydrologic model was calibrated using a semi-distributed approach. Its simulation and forecasting performances are analyzed. The performances of the system on a recent flood event are also investigated.     

Data assimilation: From photon counts to Earth System forecasts

Data assimilation - the set of techniques whereby information from observing systems and models is combined optimally - is rapidly becoming prominent in endeavours to exploit Earth Observation for Earth sciences, including climate prediction. This paper explains the broad principles of data assimilation, outlining different approaches (optimal interpolation, three-dimensional and four-dimensional variational methods, the Kalman Filter), together with the approximations that are often necessary to make them practicable. After pointing out a variety of benefits of data assimilation, the paper then outlines some practical applications of the exploitation of Earth Observation by data assimilation in the areas of operational oceanography, chemical weather forecasting and carbon cycle modelling. Finally, some challenges for the future are noted.     

A simple method for integrating a complex model into an ensemble data assimilation system using MPI

This paper details a strategy for modifying the source code of a complex model so that the model may be used in a data assimilation context, and gives the standards for implementing a data assimilation code to use such a model. The strategy relies on keeping the model separate from any data assimilation code, and coupling the two through the use of Message Passing Interface (MPI) functionality. This strategy limits the changes necessary to the model and as such is rapid to program, at the expense of ultimate performance. The implementation technique is applied in different models with state dimension up to .2.7 × 10⁸ The overheads added by using this implementation strategy in a coupled ocean-atmosphere climate model are shown to be an order of magnitude smaller than the addition of correlated stochastic random errors necessary for some nonlinear data assimilation techniques.     

A hybrid data assimilation scheme for model parameter estimation: Application to morphodynamic modelling

We present a novel algorithm for joint state-parameter estimation using sequential three dimensional variational data assimilation (3D Var) and demonstrate its application in the context of morphodynamic modelling using an idealised two parameter 1D sediment transport model. The new scheme combines a static representation of the state background error covariances with a flow dependent approximation of the state-parameter cross-covariances. For the case presented here, this involves calculating a local finite difference approximation of the gradient of the model with respect to the parameters. The new method is easy to implement and computationally inexpensive to run. Experimental results are positive with the scheme able to recover the model parameters to a high level of accuracy. We expect that there is potential for successful application of this new methodology to larger, more realistic models with more complex parameterisations.     

Setting up a hydrological model of Alberta: Data discrimination analyses prior to calibration

Failure to setup a large-scale hydrological model correctly may not allow proper calibration and uncertainty analyses, leading to inaccurate model prediction. To build a model with accurate accounting of hydrological processes, a data discrimination procedure was applied in this study. The framework uses a hydrological model of Alberta built with the Soil and Water Assessment Tool (SWAT) program. The model was used to quantify the causes and extents of biases in predictions due to different types of input data. Data types represented different sources of errors, including input data (e.g., climate), conceptual model (e.g., potholes, glaciers), and control structure (e.g., reservoirs, dams). The results showed that accounting for these measures leads to a better physical accounting of hydrological processes, significantly improving the overall model performance. The procedure used in this study helps to avoid unnecessary and arbitrary adjustment of parameters to compensate for the errors in the model structure.     

Hand tracking in a natural conversational environment by the interacting multiple model and probabilistic data association (IMM-PDA) algorithm

Traditional image based hand tracking algorithms use a single model Kalman filter to estimate and predict the hand state (position, velocity, and acceleration) and do not consider multiple measurements with noise and false alarms. However, these approaches may fail in the case of large maneuvers and/or a clutter measurement environment. In this paper, we apply the interacting multiple model (IMM) to catch hand maneuvers and the probabilistic data association (PDA) method to process noisy measurements and false alarms. A theoretical framework of image based hand tracking by the IMM-PDA algorithm is set up. Experiment results from several long video segments show that the IMM-PDA algorithm gives a superior performance compared to single model based Kalman filters.