AOSC630: Statistical Methods in Meteorology and Oceanography
This webpage will be updated as needed.
- Time of class: W 11:30am-2:00pm (3 credits)
- Instructor: Takemasa Miyoshi
- Office Hours: By appointment
STAT400 or equivalent introductory statistics course is a prerequisite.
Catalog description: "Parametric and non-parametric tests; time series analysis and filtering; wavelets. Multiple regression and screening; neural networks. Empirical orthogonal functions and teleconnections. Statistical weather and climate prediction, including MOS, constructed analogs. Ensemble forecasting and verification."
- Wilks, D. S., 2011: Statistical methods in the atmospheric sciences, 3rd ed., Academic Press, 676pp. ISBN: 0123850223
- Van den Dool, H., 2007: Empirical Methods in Short-Term Climate Prediction, Oxford University Press, 240pp. ISBN: 0199202788
- There are very useful class notes in Professor Eugenia Kalnay's webpage.
- 3/7: Dr. Krasnopolsky on Neural Networks (PPT, Paper)
- 3/14: Dr. Pena on Ensemble Forecasts (PPT)
- 3/28: Dr. Tangborn on Wavelets (Notes)
- 4/4: Dr. Antolik on MOS (PPT)
The textbook chapters shown in the schedule do not necessarily correspond to the class contents very well. However, since the class time is limited, reading the textbook is an essential part of the course work. Students have a chance to ask questions about the textbook in each class except for guest lectures.
Homework must be submitted in the class.
Exam is on April 18. The exam covers the course materials except for the guest lectures.
4. Course Project
At the graduate level, it is important to develop an ability to apply the statistical methods learned through the course to the problems of student's own interest. Choose and obtain any data of your interest (must be related to atmospheric/oceanic science), apply appropriate statistical methods to investigate the data, and draw a conclusion from the investigation. Summarize what you did and what you found to be a 20-min oral presentation. Additional 10 min is given for questions and discussions. The project presentation is graded based on the overall quality. Active participation to other students' presentations such as asking interesting questions adds a small bonus.
Students must abide by the University of Maryland Code of Academic Integrity. Here is the link to the Student Honor Council.
Grading is based on two homework (20% x 2), an exam (30%), and a project presentation (30%). Late submission of homework results zero grading point for the corresponding part.
Grading criteria: A+: 95.00% -- 100.00% A : 85.00% -- 94.99% A-: 80:00% -- 84.99% B+: 77.50% -- 79.99% B : 72.50% -- 77.49% B-: 70.00% -- 72.49% C+: 67.50% -- 69.99% C : 62.50% -- 67.49% C-: 60.00% -- 62.49% D+: 57.50% -- 59.99% D : 52.50% -- 57.49% D-: 50.00% -- 52.49% F : 0.00% -- 49.99%