Short Course on Statistical Analysis of Weather and Climate Extremes

Sunday, 2 February 2014 Room C202

Program

Extreme events are of primary concern when studying climate variability and change because of their societal impacts. Well-established statistical methods are available for analyzing univariate data, yet these methods are seldom taught in university classes. Moreover, new developments in analyzing persistent extreme phenomena, such as heat waves and droughts, and spatial patterns in extremes have been proposed, but have not fully made their way into atmospheric studies. This course will aid the participants in shortening the learning curve toward the proper analysis of extremes in both univariate and more complex settings, and bring them up to speed on current research.

The AMS short course will provide an introduction to statistical methods for analyzing extreme events with applications to weather and climate studies. The course will include instruction in using the R programming language for implementing extreme value analyses.  The course is aimed at students and researchers in atmospheric and climate sciences, among other fields, and at stakeholders such as water managers.   A minimal pre-requisite for participants is completion of an introductory statistics course, but no training in extreme value analysis per se is required.

The format of the course will consist of a background lecture on atmospheric science challenges concerning extreme events to be addressed, an introductory lecture on extreme value analysis, and several hands-on exercise sessions using weather and climate data. Experience using R is desirable, but not required for this course.

Computers, laptops or internet access will not necessarily be available.  It is required that participants bring their own laptop computer with the R programming language and the extRemes package already installed.  Installation of R, and the required package, is straightforward (see instructions at the R-project web site, http://www.r-project.org, at http://cran.r-project.org/doc/manuals/r-release/R-admin.html).

 Instructors for the course include: Barbara Casati, Ouranos, Montreal Canada, Eric Gilleland and Rick Katz both from NCAR, Boulder, Colorado, U.S.A. For more information, please contact Barbara Casati ([email protected]) or Eric Gilleland (http://www.ral.ucar.edu/staff/ericg).