CLIM Fall 2017: Statistics for Climate Research

NOTE: The September 5th class will be held at the Archie K. Davis Conference Center (RTP Headquarters, 12 Davis Dr. RTP, NC) 


Tuesdays 4:30 pm-7:00pm at SAMSI, Research Triangle Park, NC, beginning Tuesday, August 29, 2017
No class during week of Thanksgiving (November 21, 2017)
Last class: Tuesday, November 28, 2017
Final Presentation: December 5, 2017

NOTE: Below are listed tentative dates of classes and instructors  for this course.

Date Class Speaker Presentation
Aug 29 Introduction: Statistics and Computing Background Brian Reich
Sep 05 Detection and Attribution Richard Smith
Sep 12 Guest lecture: Analysis of Climate Model data Dr. Murali Haran, Penn State University
Sep 19 Climate Informatics Brian Reich
Sep 26 Guest lecture: Data Fusion Dr. Veronica Berrocal, University of Michigan
Oct 03 Spatial and Spatio-temporal Statistics Doug Nychka, Brian Reich and Bo Li, University of Illinois
Estimating Curves and Surfaces Doug Nychka, University of Illinois
Oct 10 Spatial and Spatio-temporal Statistics Doug Nychka, Brian Reich and Bo Li, University of Illinois
Spatial Data: Models and Analysis Doug Nychka, University of Illinois
Oct 17 Spatial and Spatio-temporal Statistics Doug Nychka, Brian Reich and Bo Li, University of Illinois
Nonstationary Covariance Modeling Bo Li, University of Illinois
Oct 24 Spatial and Spatio-temporal Statistics Doug Nychka, Brian Reich and Bo Li, University of Illinois
Geostats for Large Data Sets Brian Reich, University of Illinois
Oct 31 Quantile Regression Surya Tokdar, Duke University
Nov 07 Extremes Richard Smith
Nov 14 Climate Trends Richard Smith
Nov 21 NO CLASS – Thanksgiving
Nov 28 Trend Detection Using Time Series Methods Richard Smith
Dec 05 Final Presentations Richard Smith


Course Outline: This course is being offered in conjunction with the SAMSI year-long research program on Mathematical and Statistical Methods for Climate and the Earth System. The course will cover statistical and computational methods for the analysis of data arising in climate research. Specific topics will include:

  • Time series methods and assessments of trends in climatological data
  • Analysis of large spatial datasets in climate research
  • Methods based on Empirical Orthogonal Functions
  • Climate Informatics: the application of machine learning methods and high-performance computing in climate research
  • Statistics for climate extremes
  • Climate and Health

Prerequisites: The course is intended for graduate students (MS and PhD) in Statistics, Mathematics or other mathematically related disciplines. Students in fields such as Environmental Sciences or Geography are welcome to attend provided they have exposure to Mathematics and Statistics courses at an appropriate level, such as a standard undergraduate calculus sequence, a course in linear algebra, and a graduate-level course in statistical topics such as Linear Models or Regression Analysis. Additional courses in topics such as time series or spatial statistics will be very helpful but are not required. Apart from the statistical prerequisites, participants are expected to have sufficient computing expertise to allow them to run statistical analyses in R. In cases of doubt, feel free to contact one of the instructors.

Assessment: Each student will be expected to conduct a small project applying the techniques of the course to some relevant dataset developed in consultation with the instructors. Depending on the number of students in the course, students may be grouped into teams of two or three students. Each student or team will present its findings at the final class session on Tuesday, December 5, and should also prepare a short written report of its findings.

Grading: Final grades will be based primarily on course projects. There will be no regular homeworks or exams.

Registration: (processed through the respective university)

  • UNC-CH: STOR 893 and MATH 793
  • Duke: STA 790.01
  • NCSU: MA 810.001 and ST 810.001

Questions: email [email protected]