Editorial Type:
Corrigendum
Article Type:
Correction

CORRIGENDUM

Jeffrey S. Whitaker Physical Sciences Division, NOAA/Earth System Research Laboratory, Boulder, Colorado

Search for other papers by Jeffrey S. Whitaker in
Current site
Google Scholar
PubMed Close
and
Thomas M. Hamill Physical Sciences Division, NOAA/Earth System Research Laboratory, Boulder, Colorado

Search for other papers by Thomas M. Hamill in
Current site
Google Scholar
PubMed Close
Print Publication:
01 Jun 2006
DOI:
https://doi.org/10.1175/MWR3156.1
Page(s):
1722
Full access

Corresponding author address: Thomas M. Hamill, Physical Sciences Division R/PSD 1, NOAA/Earth System Research Laboratory, 325 Broadway, Boulder, CO 80305-3328. Email: tom.hamill@noaa.gov

Corresponding author address: Thomas M. Hamill, Physical Sciences Division R/PSD 1, NOAA/Earth System Research Laboratory, 325 Broadway, Boulder, CO 80305-3328. Email: tom.hamill@noaa.gov

We have discovered a coding error in the underlying computations that were used to produce Fig. 3 from Whitaker and Hamill (2002). This figure showed the ensemble-mean analysis error from the ensemble Kalman filter (EnKF) and the ensemble square root filter (EnSRF) for a 40-dimensional version of the model of Lorenz and Emanuel (1998) with a forcing of 8.0 and a time step of 0.05. In the production of this figure, our ensemble data assimilation code generated synthetic observations at each model grid point at each time that should have been independent. While observation errors changed with each data assimilation cycle, we discovered a coding bug that inappropriately caused all observation errors at a given time to be the same. A corrected Fig. 3 is reproduced below (Fig. 4 results were essentially unchanged). There was a slight degradation in the relative improvement of our proposed EnSRF relative to the EnKF when compared with the original published version of the figure.

REFERENCES

  • Lorenz, E. N., and K. A. Emanuel, 1998: Optimal sites for supplemental weather observations: Simulation with a small model. J. Atmos. Sci, 55 , 399–414.

    • Search Google Scholar
    • Export Citation

  • Whitaker, J. S., and T. M. Hamill, 2002: Ensemble data assimilation without perturbed observations. Mon. Wea. Rev, 130 , 1913–1924.

    • Search Google Scholar
    • Export Citation

Fig. 3.
Fig. 3.

Ensemble-mean error as a function of the distance at which the covariance filter goes to zero, and the covariance inflation factor, for (a) the EnKF and (b) the EnSRF. Results are for a 10-member ensemble averaged over 5000 assimilation cycles using the model of Lorenz and Emanuel (1998), with observations of every state variable. Observations have unit error variance. Shaded areas indicate regions in parameter space where the filter diverges.

Citation: Monthly Weather Review 134, 6; 10.1175/MWR3156.1

Save
Cover Monthly Weather Review
Monthly Weather Review

  • Lorenz, E. N., and K. A. Emanuel, 1998: Optimal sites for supplemental weather observations: Simulation with a small model. J. Atmos. Sci, 55 , 399–414.

    • Search Google Scholar
    • Export Citation

  • Whitaker, J. S., and T. M. Hamill, 2002: Ensemble data assimilation without perturbed observations. Mon. Wea. Rev, 130 , 1913–1924.

    • Search Google Scholar
    • Export Citation
  • <sc>Fig</sc>. 3.
    View in gallery
    Fig. 3.

    Ensemble-mean error as a function of the distance at which the covariance filter goes to zero, and the covariance inflation factor, for (a) the EnKF and (b) the EnSRF. Results are for a 10-member ensemble averaged over 5000 assimilation cycles using the model of Lorenz and Emanuel (1998), with observations of every state variable. Observations have unit error variance. Shaded areas indicate regions in parameter space where the filter diverges.

Fig. 3.

Ensemble-mean error as a function of the distance at which the covariance filter goes to zero, and the covariance inflation factor, for (a) the EnKF and (b) the EnSRF. Results are for a 10-member ensemble averaged over 5000 assimilation cycles using the model of Lorenz and Emanuel (1998), with observations of every state variable. Observations have unit error variance. Shaded areas indicate regions in parameter space where the filter diverges.
All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 2628 2465 1015
PDF Downloads 84 33 0