This lesson introduces the processes of model data assimilation. It also discusses the impacts of errors in the data assimilation on model forecasts and how a human forecaster can compensate for them.
After completing this lesson, you will be able to:
Describe the goal of data assimilation (DA), and why a model might produce a better forecast if its analysis does not fit a perfect observation too closely
State the major assumptions of DA and why NWP is used as the background field
Describe the major shortcomings of DA
Explain how errors in observations and model increments are applied to produce the analysis
Explain how 3D-VAR and 4D-VAR handle background error covariances
Identify examples of bad analysis
data assimilation, observations, 3D-VAR analysis, 4D-VAR analysis
Support Form
Please inform us of any problem that you are having by visiting our Support Form.
In order to better serve our users and sponsors, access to our online materials requires registration. To learn more about registering or to get help with common questions about our site, please visit our FAQs.
Most of the material made available on MetEd is free for non-commercial, educational use. We encourage you to use our material to enhance your education and training efforts. To review the legal-speak with all the nitty gritty details, see our Legal Notices page.