{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/125828"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/125828","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"The influence of thunderstorm type on extreme near-surface wind characteristics","abstract":"Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2026-08-01","abstract_html":"Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2026-08-01","abstract_has_math":false,"creators":["Roegner, David Timothy"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"M.S.","degree_level":"Thesis","degree_discipline":"Atmospheric Sciences","degree_department":null,"school":null,"contributors":["Lombardo, Franklin T","Trapp, Robert J","Haberlie, Alex M"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2024,"date_issued":"2024-07-16","date_published":"2024-07-16","updated_at":"2026-07-22T22:25:02Z","subjects":["Thunderstorm","Near-surface Wind","Mesoscale Convective System","Supercell","Single-cell","Machine Learning","Atmospheric Sciences","Extreme Winds","Extreme Value Analysis","Wind","Wind Engineering","Civil Engineering","Damaging Winds"],"languages":["en","eng"],"rights":["Copyright 2024 David Roegner"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/2142/125828","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Lombardo, Franklin T","Trapp, Robert J","Haberlie, Alex M"]},{"key":"dc:creator","label":"Author","values":["Roegner, David Timothy"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2024-07-16","2024-08"]},{"key":"dc:type","label":"Dc Type","values":["text","Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Atmospheric Sciences"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["M.S."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois at Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Thunderstorm","Near-surface Wind","Mesoscale Convective System","Supercell","Single-cell","Machine Learning","Atmospheric Sciences","Extreme Winds","Extreme Value Analysis","Wind","Wind Engineering","Civil Engineering","Damaging Winds"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en","eng"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2024 David Roegner"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://hdl.handle.net/2142/125828"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2026-08-01","The student, David Roegner, accepted the attached license on 2024-07-12 at 19:40.","The student, David Roegner, submitted this Thesis for approval on 2024-07-12 at 22:29.","This Thesis was approved for publication on 2024-07-16 at 16:01.","DSpace SAF Submission Ingestion Package generated from Vireo submission #21108 on 2025-02-04 at 21:25:56","The wind speed used for wind load design in the United States is based on a binary classification approach where the probability distributions of non-thunderstorm and thunderstorm wind speeds are analyzed independently in a mixed distribution. These wind speeds utilize data from the Automated Surface Observing System (ASOS). Beyond wind speed, thunderstorm wind fields are primarily modeled utilizing a simple downburst model. Neither the wind field model nor wind speed prediction accounts for the different dynamical processes forming these extreme winds in individual thunderstorm types. Previous work has shown different thunderstorm types have different physical properties (e.g., wind profile and gust factor) at the surface. In an effort to bridge the gap between wind engineering and atmospheric sciences, this study seeks to determine 1) should individual thunderstorm types be considered separately for wind speed predictions in wind load design? and 2) do different thunderstorm types have different characteristics such that ASCE should treat thunderstorm winds differently in wind load design? The first objective is to determine if individual thunderstorm type can be considered separately when predicting extreme winds for wind load design. Using data from the ASOS network and radar data from the National Weather Service (NWS) in Iowa between 1996 and 2022, independent thunderstorm events that produced measured peak wind speeds > 58 mph were classified as being generated from single-cell, multicellular, or supercell thunderstorms. Classification was done using a subjective scheme that utilizes radar reflectivity and Doppler velocity GIFs. The categorized events were combined into a single superstation, checked for independence, and then an extreme value analysis was done for each thunderstorm type. Multicellular storms dominate the extreme wind climatology in Iowa, suggesting that thunderstorm type can be considered separately for wind load design. The addition of the 118 mph estimated peak wind speed that was lost due to power loss at the ASOS station from the 10 August 2020 derecho suggests that the ASOS network may not be accurately capturing the extreme wind climatology. The second objective is to determine if wind characteristics important for wind loading differ by thunderstorm type. Using a mostly objective classification scheme developed for the study, reflectivity mosaics of flagged thunderstorm events captured by over 700 ASOS stations with wind speeds > 46 mph were classified as multicellular organized, multicellular unorganized, cellular, synoptic, not an event, or West of -105 (west of -105 degree longitude). Utilizing the decision tree guess, and statistics from the categorized reflectivity mosaics, an XGBoost and CatBoost machine learning model was trained. Both models had high accuracies with the CatBoost model obtaining a 95.1% accuracy rate. Some misclassification of the multicellular unorganized category occurred, resulting in less cellular events than truth. Machine learning was then utilized to classify the independent thunderstorm events associated with winds of > 46 mph in Iowa, Illinois, and Indiana. The 1-minute ASOS records were then used, and engineering relevant wind characteristics such as the duration of extreme winds and gust factor were calculated. A Mann-Whitney U test found statistically significant differences between thunderstorm categories for multiple wind characteristics, including the gust factor and duration of extreme winds, supporting the notion of considering thunderstorms and the respective types separately in wind load design."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["The influence of thunderstorm type on extreme near-surface wind characteristics"]}]}],"canonical_facts":{"dc:contributor":["Lombardo, Franklin T","Trapp, Robert J","Haberlie, Alex M"],"dc:creator":["Roegner, David Timothy"],"dc:date":["2024-07-16","2024-08"],"dc:description":["Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2026-08-01","The student, David Roegner, accepted the attached license on 2024-07-12 at 19:40.","The student, David Roegner, submitted this Thesis for approval on 2024-07-12 at 22:29.","This Thesis was approved for publication on 2024-07-16 at 16:01.","DSpace SAF Submission Ingestion Package generated from Vireo submission #21108 on 2025-02-04 at 21:25:56","The wind speed used for wind load design in the United States is based on a binary classification approach where the probability distributions of non-thunderstorm and thunderstorm wind speeds are analyzed independently in a mixed distribution. These wind speeds utilize data from the Automated Surface Observing System (ASOS). Beyond wind speed, thunderstorm wind fields are primarily modeled utilizing a simple downburst model. Neither the wind field model nor wind speed prediction accounts for the different dynamical processes forming these extreme winds in individual thunderstorm types. Previous work has shown different thunderstorm types have different physical properties (e.g., wind profile and gust factor) at the surface. In an effort to bridge the gap between wind engineering and atmospheric sciences, this study seeks to determine 1) should individual thunderstorm types be considered separately for wind speed predictions in wind load design? and 2) do different thunderstorm types have different characteristics such that ASCE should treat thunderstorm winds differently in wind load design? The first objective is to determine if individual thunderstorm type can be considered separately when predicting extreme winds for wind load design. Using data from the ASOS network and radar data from the National Weather Service (NWS) in Iowa between 1996 and 2022, independent thunderstorm events that produced measured peak wind speeds > 58 mph were classified as being generated from single-cell, multicellular, or supercell thunderstorms. Classification was done using a subjective scheme that utilizes radar reflectivity and Doppler velocity GIFs. The categorized events were combined into a single superstation, checked for independence, and then an extreme value analysis was done for each thunderstorm type. Multicellular storms dominate the extreme wind climatology in Iowa, suggesting that thunderstorm type can be considered separately for wind load design. The addition of the 118 mph estimated peak wind speed that was lost due to power loss at the ASOS station from the 10 August 2020 derecho suggests that the ASOS network may not be accurately capturing the extreme wind climatology. The second objective is to determine if wind characteristics important for wind loading differ by thunderstorm type. Using a mostly objective classification scheme developed for the study, reflectivity mosaics of flagged thunderstorm events captured by over 700 ASOS stations with wind speeds > 46 mph were classified as multicellular organized, multicellular unorganized, cellular, synoptic, not an event, or West of -105 (west of -105 degree longitude). Utilizing the decision tree guess, and statistics from the categorized reflectivity mosaics, an XGBoost and CatBoost machine learning model was trained. Both models had high accuracies with the CatBoost model obtaining a 95.1% accuracy rate. Some misclassification of the multicellular unorganized category occurred, resulting in less cellular events than truth. Machine learning was then utilized to classify the independent thunderstorm events associated with winds of > 46 mph in Iowa, Illinois, and Indiana. The 1-minute ASOS records were then used, and engineering relevant wind characteristics such as the duration of extreme winds and gust factor were calculated. A Mann-Whitney U test found statistically significant differences between thunderstorm categories for multiple wind characteristics, including the gust factor and duration of extreme winds, supporting the notion of considering thunderstorms and the respective types separately in wind load design."],"dc:format":["application/pdf"],"dc:identifier":["https://hdl.handle.net/2142/125828"],"dc:language":["en","eng"],"dc:rights":["Copyright 2024 David Roegner"],"dc:subject":["Thunderstorm","Near-surface Wind","Mesoscale Convective System","Supercell","Single-cell","Machine Learning","Atmospheric Sciences","Extreme Winds","Extreme Value Analysis","Wind","Wind Engineering","Civil Engineering","Damaging Winds"],"dc:title":["The influence of thunderstorm type on extreme near-surface wind characteristics"],"dc:type":["text","Thesis"],"thesis:degree_discipline":["Atmospheric Sciences"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["M.S."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:25:02Z"}