The ensemble Kalman filter and its relations to other nonlinear filters

Roth, Michael; Fritsche, Carsten; Hendeby, Gustaf; Gustafison, Fredrik

doi:10.1109/eusipco.2015.7362581

Cited by 17 publications

(12 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…According to the Monte Carlo method, an optimal solution is found between the numerical model solution and the snow depth data, and it is used to update the initial snow depth field. A continuous loop such as this can update the field and obtain the optimal solution (Roth et al 2015). Then, the error is corrected, the snow depth prediction value of the numerical dynamic model close to the observed snow depth value is maximized, and the process of snow depth assimilation is completed (Sun 2012).…”

Section: Results Of Snow Depth Assimilationmentioning

confidence: 99%

Assimilation of D-InSAR snow depth data by an ensemble Kalman filter

Yang

2021

Arab J Geosci

View full text Add to dashboard Cite

Snow depth mirrors regional climate change and is a vital parameter for medium- and long-term numerical climate prediction, numerical simulation of land-surface hydrological process, and water resource assessment. However, the quality of the available snow depth products retrieved from remote sensing is inevitably affected by cloud and mountain shadow, and the spatiotemporal resolution of the snow depth data cannot meet the need of hydrological research and decision-making assistance. Therefore, a method to enhance the accuracy of snow depth data is urgently required. In the present study, three kinds of snow depth data which included the D-InSAR data retrieved from the remote sensing images of Sentinel-1 synthetic aperture radar, the automatically measured data using ultrasonic snow depth detectors, and the manually measured data were assimilated based on ensemble Kalman filter. The assimilated snow depth data were spatiotemporally consecutive and integrated. Under the constraint of the measured data, the accuracy of the assimilated snow depth data was higher and met the need of subsequent research. The development of ultrasonic snow depth detector and the application of D-InSAR technology in snow depth inversion had greatly alleviated the insufficiency of snow depth data in types and quantity. At the same time, the assimilation of multi-source snow depth data by ensemble Kalman filter also provides high-precision data to support remote sensing hydrological research, water resource assessment, and snow disaster prevention and control program.

show abstract

Section: Results Of Snow Depth Assimilationmentioning

confidence: 99%

Assimilation of D-InSAR snow depth data by an ensemble Kalman filter

Yang

2021

Arab J Geosci

View full text Add to dashboard Cite

show abstract

“…Both approaches belong to the class of sampling‐based filter algorithms in that the samples are generated and propagated through the state‐space model to approximate the error statistics, and the procedures of UKF and EnKF are quite similar (Roth et al, 2015), but we highlight the following essential differences.…”

Section: Methodsmentioning

confidence: 99%

Improving Forecast Skill of Lowland Hydrological Models Using Ensemble Kalman Filter and Unscented Kalman Filter

Sun

Bao

Valk

et al. 2020

Water Resources Research

View full text Add to dashboard Cite

For operational water management in lowlands and polders (for instance, in the Netherlands), lowland hydrological models are used for flow prediction, often as an input for a real‐time control system to steer water with pumps and weirs to keep water levels within acceptable bounds. Therefore, proper initialization of these models is essential. The ensemble Kalman filter (EnKF) has been widely used due to its relative simplicity and robustness, while the unscented Kalman filter (UKF) has received little attention in the operational context. Here, we test both UKF and EnKF using a lowland lumped hydrological model. The results of a reforecast experiment in an operational context using an hourly time step show that when using nine ensemble members, both filters can improve the accuracy of the forecast by updating the state of a lumped hydrological model (Wageningen Lowland Runoff Simulator, WALRUS) based on the observed discharge, while UKF has achieved better performance than EnKF. Additionally, we show that an increase in the ensemble members does not necessarily mean a significant increase in performance. WALRUS model with either UKF or EnKF could be considered for hydrological forecasting for supporting water management of polders and lowlands, with UKF being the computationally leaner option.

show abstract

“…Another variant of the Kalman filter, called the ensemble Kalman filter (EnKF), has been introduced to tackle high-dimensional dynamical problems in geophysics, see, e.g. [15][16][17][18][19].…”

Section: Stochastic Filteringmentioning

confidence: 99%