Files in this item



application/pdfSCHLIE-DISSERTATION-2017.pdf (12MB)
(no description provided)PDF


Title:Weather extremes in a changing climate: analyses for extreme precipitation, severe hail, and tornadoes
Author(s):Schlie, Emily Elizabeth-Janssen
Director of Research:Wuebbles, Donald J.
Doctoral Committee Chair(s):Wuebbles, Donald J.
Doctoral Committee Member(s):Trapp, Robert J.; Sriver, Ryan; Jewett, Brian F.; Kunkel, Kenneth; Brooks, Harold
Department / Program:Atmospheric Sciences
Discipline:Atmospheric Sciences
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):Climate change
Atmospheric science
Severe hail
Abstract:This thesis is broken into three parts, examining three distinct hazards that often accompany severe thunderstorms: extreme precipitation, severe hail, and tornados. First, observational data and an ensemble of climate change model experiments (from the Coupled Model Intercomparison Project Phase 5 (CMIP5)) are used to examine past and potential future seasonal changes in extreme precipitation event frequency over the United States. Using the extreme precipitation index as a metric for extreme precipitation event frequency change, we find key differences between models and observations. In particular, the CMIP5 models tend to overestimate the number of spring events and underestimate the number of summer events. This seasonal shift in the models is amplified in projections. These results provide a basis for evaluating climate model skill in simulating observed seasonality and changes in regional extreme precipitation. Additionally, we highlight key sources of variability and uncertainty that can potentially inform regional impact analyses and adaptation planning. Severe hail, another impactful hazard associated with severe storms, is also investigated. A radar-based hail climatology, with superior coverage and resolution, is possible using the Next-Generation Weather Radar (NEXRAD) reanalysis through the application of multi-radar multisensory (MRMS) algorithms, such as Maximum Expected Size of Hail (MESH). Using 12-years of MESH data we define a “severe hail outbreak day” and analyze characteristics of the severe hail and severe hail outbreak dataset, including an analysis of hail swaths. When comparing severe hail days in MESH to reports, we find a linear relationship between MESH and reports. Several case studies are also included to highlight the utility of MESH when studying outbreaks of severe hail, specifically regarding outbreak events that occur in lowpopulation areas. We find that severe hail days decrease while severe hail outbreak days increase over the 12-years examined. The increase in outbreaks is happening primarily in the month of June, where the number of severe hail days stays fairly constant over the 12-years. This suggests that the increase in outbreaks is mainly taking place on days when severe hail already occurs. When examining hail swath characteristics we found that there are a greater number of hail swaths, with a Major-Axis-Length (MAL) of at least 15km, on outbreak versus non-outbreak days. Additionally, hail swaths with the largest MALs occur on outbreak days. Lastly, the frequency and spatial extent of environments supportive of tornado and severe hail outbreak days, utilizing reanalysis data, are investigated over a 38-year historical period. A better understanding of the meteorological reason for observed trends in severe weather events is provided. The MESH-based severe hail dataset and tornado reports from the Storm Prediction Center storm reports database are considered ground truth. Composite parameters and thresholds, from the North American Regional Reanalysis (NARR), signifying outbreak environments are then determined. Severe hail and tornado outbreak days and areal extent of supportive environments are assessed. Specifically, the changing nature of tornado outbreaks is addressed utilizing these environmental parameters. The existence of long term trends in severe hail and tornado outbreaks, independent of reporting biases are identified. A long-term (1980-2015), statistically significant, positive trend in environments favoring severe hail and tornado outbreak days is found. This suggests that the recent short-term, positive trend in MESH based severe hail outbreak days extends further back in time. Additionally the yearly total outbreak days are positively correlated with the yearly median outbreak area. This correlation is statistically significant and supports our hypothesis that the increase in days supportive of outbreaks coincides with increases in the areal extent of outbreak environments. The variability of tornado outbreak days is also found to be increasing with time, supporting prior studies utilizing reports.
Issue Date:2017-12-07
Rights Information:Copyright 2017 Emily Elizabeth-Janssen Schlie
Date Available in IDEALS:2018-03-13
Date Deposited:2017-12

This item appears in the following Collection(s)

Item Statistics