Data assimilation method based on the constraints of confidence region

Yong Li, Siming Li, Yao Sheng, Luheng Wang

Research output: Contribution to journalArticle

Abstract

The ensemble Kalman filter (EnKF) is a distinguished data assimilation method that is widely used and studied in various fields including methodology and oceanography. However, due to the limited sample size or imprecise dynamics model, it is usually easy for the forecast error variance to be underestimated, which further leads to the phenomenon of filter divergence. Additionally, the assimilation results of the initial stage are poor if the initial condition settings differ greatly from the true initial state. To address these problems, the variance inflation procedure is usually adopted. In this paper, we propose a new method based on the constraints of a confidence region constructed by the observations, called EnCR, to estimate the inflation parameter of the forecast error variance of the EnKF method. In the new method, the state estimate is more robust to both the inaccurate forecast models and initial condition settings. The new method is compared with other adaptive data assimilation methods in the Lorenz-63 and Lorenz-96 models under various model parameter settings. The simulation results show that the new method performs better than the competing methods.

Original languageEnglish (US)
Pages (from-to)334-345
Number of pages12
JournalAdvances in Atmospheric Sciences
Volume35
Issue number3
DOIs
StatePublished - Mar 1 2018

Fingerprint

data assimilation
Kalman filter
inflation
method
oceanography
divergence
filter
methodology
simulation
forecast

All Science Journal Classification (ASJC) codes

  • Atmospheric Science

Cite this

Li, Yong ; Li, Siming ; Sheng, Yao ; Wang, Luheng. / Data assimilation method based on the constraints of confidence region. In: Advances in Atmospheric Sciences. 2018 ; Vol. 35, No. 3. pp. 334-345.
@article{5c5090a3bc904bfe811e8b8374869a01,
title = "Data assimilation method based on the constraints of confidence region",
abstract = "The ensemble Kalman filter (EnKF) is a distinguished data assimilation method that is widely used and studied in various fields including methodology and oceanography. However, due to the limited sample size or imprecise dynamics model, it is usually easy for the forecast error variance to be underestimated, which further leads to the phenomenon of filter divergence. Additionally, the assimilation results of the initial stage are poor if the initial condition settings differ greatly from the true initial state. To address these problems, the variance inflation procedure is usually adopted. In this paper, we propose a new method based on the constraints of a confidence region constructed by the observations, called EnCR, to estimate the inflation parameter of the forecast error variance of the EnKF method. In the new method, the state estimate is more robust to both the inaccurate forecast models and initial condition settings. The new method is compared with other adaptive data assimilation methods in the Lorenz-63 and Lorenz-96 models under various model parameter settings. The simulation results show that the new method performs better than the competing methods.",
author = "Yong Li and Siming Li and Yao Sheng and Luheng Wang",
year = "2018",
month = "3",
day = "1",
doi = "10.1007/s00376-017-7045-y",
language = "English (US)",
volume = "35",
pages = "334--345",
journal = "Advances in Atmospheric Sciences",
issn = "0256-1530",
publisher = "Science Press",
number = "3",

}

Data assimilation method based on the constraints of confidence region. / Li, Yong; Li, Siming; Sheng, Yao; Wang, Luheng.

In: Advances in Atmospheric Sciences, Vol. 35, No. 3, 01.03.2018, p. 334-345.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Data assimilation method based on the constraints of confidence region

AU - Li, Yong

AU - Li, Siming

AU - Sheng, Yao

AU - Wang, Luheng

PY - 2018/3/1

Y1 - 2018/3/1

N2 - The ensemble Kalman filter (EnKF) is a distinguished data assimilation method that is widely used and studied in various fields including methodology and oceanography. However, due to the limited sample size or imprecise dynamics model, it is usually easy for the forecast error variance to be underestimated, which further leads to the phenomenon of filter divergence. Additionally, the assimilation results of the initial stage are poor if the initial condition settings differ greatly from the true initial state. To address these problems, the variance inflation procedure is usually adopted. In this paper, we propose a new method based on the constraints of a confidence region constructed by the observations, called EnCR, to estimate the inflation parameter of the forecast error variance of the EnKF method. In the new method, the state estimate is more robust to both the inaccurate forecast models and initial condition settings. The new method is compared with other adaptive data assimilation methods in the Lorenz-63 and Lorenz-96 models under various model parameter settings. The simulation results show that the new method performs better than the competing methods.

AB - The ensemble Kalman filter (EnKF) is a distinguished data assimilation method that is widely used and studied in various fields including methodology and oceanography. However, due to the limited sample size or imprecise dynamics model, it is usually easy for the forecast error variance to be underestimated, which further leads to the phenomenon of filter divergence. Additionally, the assimilation results of the initial stage are poor if the initial condition settings differ greatly from the true initial state. To address these problems, the variance inflation procedure is usually adopted. In this paper, we propose a new method based on the constraints of a confidence region constructed by the observations, called EnCR, to estimate the inflation parameter of the forecast error variance of the EnKF method. In the new method, the state estimate is more robust to both the inaccurate forecast models and initial condition settings. The new method is compared with other adaptive data assimilation methods in the Lorenz-63 and Lorenz-96 models under various model parameter settings. The simulation results show that the new method performs better than the competing methods.

UR - http://www.scopus.com/inward/record.url?scp=85041003537&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85041003537&partnerID=8YFLogxK

U2 - 10.1007/s00376-017-7045-y

DO - 10.1007/s00376-017-7045-y

M3 - Article

VL - 35

SP - 334

EP - 345

JO - Advances in Atmospheric Sciences

JF - Advances in Atmospheric Sciences

SN - 0256-1530

IS - 3

ER -