U.S. CLIVAR Extremes Workshop

extremes workshop banner

Workshop Information

Download Final Agenda

Summary:

The US CLIVAR Working Group on Extremes held a workshop on Analyses, Dynamics, and Modeling of Large Scale Meteorological Patterns Associated with Extreme Temperature and Precipitation Events. The 2.5 day workshop was August 20-22, 2013 at the Lawrence Berkeley National Laboratory in Berkeley, California. The workshop had oral and poster sessions along with invited presentations.

Group of people standing in the woods
U.S. CLIVAR Extremes Workshop Participants

Objectives:

The motivation of this workshop were to explore short-term, extreme, temperature and precipitation events that occur in North America with an emphasis upon the large scale meteorological patterns (LSMPs) associated with such events. To improve our understanding of such events this workshop brought together experts in synoptic and dynamic meteorology, atmospheric modeling, and statistics. Extreme events with a time scale of 5 days or less will be emphasized. Observational studies often emphasize statistical analyses of surface data. Modeling studies often emphasize the skill in simulating the surface data. However, improvements in understanding and simulation follow from applying synoptic and dynamic analyses, motivating the common interface of understanding and simulating the LSMPs associated with extreme events. Conversely, model assessment and statistical tools can be applied to LSMPs. Hence, LSMPs provide a unifying workshop locus for bringing together modelers, statisticians, synopticians, and dynamicists.

The specific objectives are:

  • Establish methodology and research protocols for incorporating LSMPs in statistical, dynamical, and synoptic analyses;
  • Provide preliminary assessments of where current climate models stand in their simulation of LSMPs and downscale connection to T/P extreme events.

Topics:

Data

What issues arise in data quality or quantity? How well do station observations compare with reanalyses? What observation-based or model data are needed for extreme event identification? What ETCCDI indices are relevant?

Statistics

What data-handling techniques are relevant -- self-organizing maps, composites, etc. -- to identify large scale meteorological patterns (LSMPs)? What statistical methods apply to these extreme events? What statistical connections are there between extreme event and large-scale phenomena ­ such as low frequency phenomena like ENSO, NAO, etc.?

Synoptics and Dynamics

What are the physical mechanisms responsible for LSMPs?  What roles do local dynamical processes and remote forcing play? What dynamical diagnostic tools -- such as wave activity flux, E-vectors, energy budgets etc. -- are useful to understand the formation and maintenance of LSMPs?

Modeling

How well can models simulate LSMPs and the associated T/P extremes? What are the uncertainties in simulating/predicting the LSMP and extreme T/P? What insights can models provide to better understand the relationships between LSMP and extreme T/P? Assessment of model simulation of extremes in the LSMP context (such as: LSMPs in models versus analyses, developing LSMP-based metrics, assessment of model dynamics in LSMP formation.)

Participants:

The workshop welcomed abstracts related to temperature and/or precipitation extremes on a 5 days or less time scale. Application of one or more analysis techniques to the following provided datasets are especially welcome: California Central Valley max/min temperatures, Iowa precipitation, and climate model output. The invited speakers were encouraged to highlight their approaches and findings using one or more of the provided datasets as part of their talks. Poster sessions and breakout sessions were held.

Scientific Organizing Committee:

Richard Grotjahn, Chair, University of California, Davis
Matt Barlow, University of Massachusetts
Rob Black, Georgia Tech University
Ruby Leung, Pacific Northwest National Laboratory
Russ Schumacher, Colorado State University
Michael Wehner, University of Berkeley/LBL