Computer Methods in Meteorological Analysis and Forecasting

METEO 474 Syllabus: Computer Methods in Meteorological Analysis and Forecasting 


Prof. Steven J. Greybush
618 Walker Building 

Teaching Assistant (TA):

Dandan Wei
410 Walker Building 

Course Information: 

  • Course Hours: Monday and Friday, 1:25 PM – 3:20 PM
  • Course Location: 126 Walker Building
  • Professor Office Hours: Wednesday, 11:00 AM – 12:00 PM; Monday 3:20-4:00 PM
  • TA Office Hours: Tuesday, 2:00 PM – 3:00 PM

Course Description: Distribution of scalars and vectors; sampling; regression and correlation in two and three dimensions; time series, statistical forecasting; forecast verification. 

Required Materials: None

Required textbooks:

Data Mining, 4th Edition: Practical Machine Learning Tools and Techniques

by Witten, Frank, Hall, and Pal

Internet materials and links: CANVAS 

Course Objectives: 

  1. To learn those computer methods needed for statistical analysis and forecasting of the weather.
  2. To learn to apply those methods to develop accurate and robust weather forecasting systems. 

Course Outcomes: 

  1. To demonstrate knowledge of the components of automated analysis and forecasting systems, and several techniques used to construct them.
  2. To demonstrate knowledge of several metrics for statistical forecast verification.
  3. To demonstrate the ability to apply statistical models and machine learning techniques to meteorological datasets.
  4. To demonstrate the ability to statistically verify the performance of forecast models. 

Statistics Prerequisite: Stat 301 or Stat 401 or E B F 472 

Note: Meteo 474 is an elective course for undergraduate Meteorology majors. 

Students who do not meet the prerequisites may be dis-enrolled during the first 10-day free add-drop period after being informed in writing by the instructor (see:  If you have not completed the listed prerequisites, then consult with the instructor.  

Computer Methods of Meteorological Analysis and Forecasting explores the computationally intensive statistical methods used in the development of automated weather analysis and forecasting systems. The focus of the course is on learning to develop and use artificially intelligent automated systems to perform data quality control, quantitative analysis of large meteorological data sets, and weather forecasting. Coverage will include the relevant statistical, mathematical, and computational methods including matrix operations, data quality control, regression analysis, neural network construction, decision tree growth, and forecast system verification. Students will leave the course with an understanding of how to efficiently develop accurate and robust statistical weather analysis and prediction systems. Thus, the course serves as a professional elective for those students wishing to pursue careers in statistical weather forecasting, meteorological data analysis, and associated fields. Meteorology 474 uses a project oriented lecture/lab format to provide students with hands-on experience in developing and testing weather analysis and forecast systems. Students will both code their own forecast system development programs and use off-the-shelf software designed for rapid development and testing of forecast systems. To tackle these assignments, students will use the computer laboratory facilities of the Meteorology Department and meteorological data sets of current interest. A key element of the resulting project reports will be an investigation into the origin of the observed forecast system errors. The class size is tailored to in-class training with the software tools and open discussion with the instructor and classmates. Grading will be based on the assignments and on a mid-term and final examination. 

Assessment Tools: 

Required written/oral assignments 

There will be classroom assignments, which will count toward the course grade. 

There will be an immediate 25% penalty for any assignment handed in late, a 50% penalty after six hours, and no credit will be given for an assignment handed in after the start of the next class.  Professor maintains the right to decline acceptance of a late assignment beyond a certain time.  Neatness, organization, technical soundness, spelling and grammar are important.  While students may consult with their classmates on these assignments, the final product should represent the student’s own work. 

Examination Policy 

One midterm and one final exam will be given.  These will be closed-book, individual written assessments. 

Grading Policy 

  • Exam 1                                                30%
  • Final Exam                                         30%
  • Assignments                                       35%
  • Attendance / Participation                   5%

Attendance and Participation: Students are required to attend class and participate in all exercises.  Active, thoughtful contributions to class discussions are encouraged.


Regular Add / Drop Deadline is Jan 13, Late Drop Deadline is Apr 6.
Final exam will be scheduled by the university during exams week (April 30 – May 4). 

The course content, topics, and timeline listed here is intended as a guideline, and is subject to modification by the instructors. 

Course content: (details subject to adjustment as semester progresses) 

Weeks / Topics  

  • 1-2            
    • Introduction to analysis / forecasting systems
    • Basic Statistics: Mean, Standard Deviation
    • Linear Regression, Logistic Regression
  • 3-5                  
    • Statistical Basis of Forecast Verification
    • Verification Techniques
    • Verifying Distribution Forecasts
    • Forecast System Robustness
  • 6-9   
    • Decision Trees; Growing and Pruning
    • Boosted Trees
    • Data Quality Control
    • Probabilistic Forecasting
    • Data Assimilation
  • 10-12
    • Neural Networks
    • Weighting, Objective Functions, and Iterative Improvement
    • Training via Back Propagation
    • Learning Rate vs Convergence
  • 13-14  
    • Clustering
    • Valuation of Forecasts
    • Value of Probability Forecasts
  • 15
    • Forecast System Intercomparisons 

Lecture notes will often be placed on CANVAS (, although students are ultimately responsible for their own note-taking.  It is reasonable that material covered during lectures, and through assignments, may appear on tests. Reading the corresponding sections in the textbook may aid understanding of the course material. 

Academic Integrity Statement:
Academic integrity is fundamental not only to one’s experience at the university, but remains essential throughout one’s career.  Students are not to receive unauthorized assistance on any course quizzes or individual assessments.  Students are not to misrepresent the work of others as their own.  Serious offenses may warrant a zero on the assignment or assessment.

Students in this class are expected to write up their problem sets individually, to work the exams on their own, and to write their papers in their own words using proper citations.  Class members may work on the problem sets in groups, but then each student must write up the answers separately.  Students are not to copy problem or exam answers from another person's paper and present them as their own; students may not plagiarize text from any sources (e.g. papers or solutions or websites) written by others.  Students who present other people's work as their own will receive at least a 0 on the assignment and may well receive an F or XF in the course.  Please see: Earth and Mineral Sciences Academic Integrity Policy:, which this course adopts.  To learn more, see Penn State's "Plagiarism Tutorial for Students." 

If in doubt about how the academic integrity policy applies to a specific situation, students are encouraged to consult with the professor or TA. 

Attendance Policy
Students who miss class for legitimate reasons will be given a reasonable opportunity to make up missed work, including exams and quizzes.  Students are not required to secure the signature of medical personnel in the case of illness or injury and should use their best judgment on whether they are well enough to attend class or not; the University Health Center will not provide medical verification for minor illnesses or injuries. Other legitimate reasons for missing class include religious observance, military service, family emergencies, regularly scheduled university-approved curricular or extracurricular activities, and post-graduate, career-related interviews when there is no opportunity for students to re-schedule these opportunities (such as employment and graduate school final interviews).  Students who encounter serious family, health, or personal situations that result in extended absences should contact the Office of Student and Family Services for help:   Whenever possible, students participating in University-approved activities should submit to the instructor a Class Absence Form available from the Registrar's Office:, at least one week prior to the activity. This course abides by the Penn State Attendance Policy E-11:, and Conflict Exam Policy 44-35: Please also see Illness Verification Policy:, and Religious Observance Policy:

Course Copyright 

All course materials students receive or to which students have online access are protected by copyright laws. Students may use course materials and make copies for their own use as needed, but unauthorized distribution and/or uploading of materials without the instructor’s express permission is strictly prohibited. 

Weather Delays and Campus Emergencies: 

Campus emergencies, including weather delays, are announced on Penn State News: http:/ and communicated to cellphones, email, the Penn State Facebook page, and Twitter via PSUAlert (Sign up at: Students will not be required to attend class if campus is closed during any part of the scheduled class time. 

Accommodations for students with disabilities: 

Penn State welcomes students with disabilities into the University's educational programs. Every Penn State campus has an office for students with disabilities. The Office for Disability Services (ODS) website provides contact information for every Penn State campus: ( For further information, please visit the Office for Disability Services website ( In order to receive consideration for reasonable accommodations, you must contact the appropriate disability services office at the campus where you are officially enrolled, participate in an intake interview, and provide documentation based on the documentation guidelines ( If the documentation supports your request for reasonable accommodations, your campus’s disability services office will provide you with an accommodation letter. Please share this letter with your instructors and discuss the accommodations with them as early in your courses as possible. You must follow this process for every semester that you request accommodations. 

Disclaimer Statement 

Please note that the specifics of this Course Syllabus can be changed at any time, and you will be responsible for abiding by any such changes. Changes will be posted to the course discussion forum. 

Acknowledgements: We would like to thank previous instructors of Meteo 474, including George Young, for their contributions to the development and structure of this course.