Time: Tuesday and Thursday 1:00-2:15
Room: NWC 5720
When and where: Tuesday and Thursday 1:00-‐2:15pm, NWC 5720
Office hours: Tuesday 2:15-‐3pm, NWC 5341 or upon email request
Prerequisites: MATH 3113 (ODE), 4163 (PDE), and 3333 (linear algebra); ENGR 3723 (numerical methods) or equivalent or permission of instructor. Programming experience is useful. Not limited to Meteorology students.
Selected journal articles.
- Kalnay, 2002 (or later edition): Atmospheric Modeling, Data Assimilation and Predictability.
- Lewis, S. Lakshmivarahan, and S. K. Dhall, 2006 (or later edition): Dynamic data assimilation: A least square approach.
- Daley, 1995 (or later edition): Atmospheric Data Analysis.
In-‐class exams (40%): Mid-‐term exam (15%), Final exam (25%)
4-‐5 Homework assignments (40%), due in class two weeks after assigned In-‐class presentation (20%)
The course is designed to introduce students to the world of ensemble Kalman filter data assimilation technique (EnKF), an advanced data assimilation method that has become popular in Meteorology, hydrology, Ecology, etc. The students will learn the most popular EnKF techniques through lectures and hands-‐on project assignments, and learn the EnKF applications in different fields through in-‐class presentations and discussions. The students will not only learn various EnKF techniques and their applications, but also develop their skills in scientific thinking and synthesis, written and oral communication throughout the course
Basic concepts of data assimilation
Mathematical preparation: matrix algebra
Least square and Bayesian contexts
Brief review of statistical interpolation, 3DVAR and 4DVAR Classic Kalman filter and Extended Kalman filter
Basic concepts of Ensemble Kalman filter
Ensemble Kalman filter with perturbed observations Ensemble square root filter
(Local) Ensemble Transform Kalman filter
Common problems and treatments in ensemble Kalman filters
Hybrid ensemble-‐variational method, ensemble smoother and other special topics Applications of ensemble Kalman filters