Publications
"Downscaling Extremes: A Comparison of Extreme Value Distributions in Point-Source and Gridded Precipitation Data" - Annals of Applied Statistics, Mar 2010. Pre-Print Joint work with Richard L. Smith, Stephan Sain, Linda Mearns, Daniel Cooley. AOAS Supplemental Appendix Pre-Print"Spatio-Temporal Models for Large-scale Indicators of Extreme Weather." Heaton, M.J., Katzfuss, M., Ramachandar, S., Pedings, K., Gilleland, E., Mannshardt-Shamseldin, E., Smith, R.L. (To Appear, Environmetrics, 2010)
"Piecing together the past: Statistical insights into paleoclimatic reconstructions." Tingley, Martin P., Peter F. Craigmile, Murali Haran, Bo Li, Elizabeth Mannshardt-Shamseldin and Bala Rajaratnam. Manuscript submitted, and currently available as Technical Report No. 2010-09, Department of Statistics, Stanford University Abstract: Reconstructing the spatial pattern of a climate process through time from incomplete instrumental and climate proxy time series is a problem with clear societal relevance that poses both scientific and statistical challenges. The scope of these challenges, along with the interdisciplinary nature of the reconstruction problem, point to the need for greater cooperation between the earth science and statistics community – a sentiment echoed in recent parliamentary reports. (...) The key aims of this article are to 1) establish a general modeling and notational framework for the paleoclimate reconstruction problem that is transparent to both the earth science and statistics communities; 2) outline and distinguish between scientific and statistical challenges and indicate how modern statistical expertise can be brought to bear upon the problem; 3) offer, in broad strokes, some suggestions for model construction and how to perform the required statistical inference; and 4) identify issues that are important to both the earth science and applied statistics communities, and encourage greater collaboration between the two.
Papers in Preparation
“Extremes in Paleoclimate Proxy Data” Joint w/ Peter Craigmile and Martin Tingley. In preparation Abstract: There is much debate about the impact of a changing climate on extreme events. Proxy reconstructions for temperature series lead to several questions about long-term climate behavior, and how to interpret this behavior given the patterns seen in proxy series. For example, using proxy data to address questions such as "Were the 1990s the warmest decade of the last millennium", and "Is there evidence that the extreme events of recent decades are more extreme than previous decades?" The methodology of extreme value theory has not been widely applied to this problem. This paper looks at what the statistics of extremes has to o er paleoclimate reconstructions through modeling of the original proxy series. “Severe Weather Under a Changing Climate: Large Scale Indicators of Extreme Events” Joint work with Eric Gilleland – Paper in Preparation Abstract: One of the more critical issues with a changing climate is the behavior of extreme weather events. It is generally thought that such events would increase under a changing climate. However, climate models are currently at too coarse of a resolution to capture the very fine scale extreme events such as severe tornadic storms. One approach is to look at the behavior of large scale indicators of severe weather. Here convective available poten- tial energy and wind shear are considered as large scale indicators of severe weather, and their analysis presents some interesting statistical challenges. Over land, high values of both variables together are characteristic of an environment conducive to severe weather. Therefore, it is simplest to an- alyze their product as a univariate problem, but it is also important to investigate their behavior through a bivariate approach. Further, it is the extreme values of the processes that are of interest, which combined with the presence of spatial correlation, introduces new statistical challenges that are still being explored in the literature. Numerous approaches, in- cluding the use of the generalized extreme value distribution for annual maxima, the generalized Pareto distribution for threshold excesses and a bivariate analysis, are examined. Each approach is critiqued and com- pared for goodness of fit and predictive attributes on re-analysis data over the United States. Even for the univariate case there are numerous issues to be resolved. "Forecast Verification for Tornadoes with High-Resolution Models" Xiaoye Li, Vasilije Perovic, Ye Tian, Kazuki Uematsu, Yushu Yang, Harold Brooks, Elizabeth Mannshardt-Shamseldin Abstract: The behavior of extreme weather events, such as tornadoes, is of critical importance as these can cause loss of life, and have huge economic impacts. One approach to storm prediction is to look at the output from high resolution numerical weather prediction (NWP) models of severe weather. Several factors can be considered forecast variables associated with severe weather, including maximum updraft helicity and reectivity. We create models incorporating these forecast variables. Further, we evaluate these models based on well-developed forecast verification methods.Go to the ISDS home page.