{"id":{"repo_id":"vt","oai_identifier":"oai:vtechworks.lib.vt.edu:10919/134297"},"canonical_url":"https://search.dev.ndltd.org/etd/vt/oai:vtechworks.lib.vt.edu:10919/134297","repository":{"repo_id":"vt","name":"Virginia Tech","base_url":"https://vtechworks.lib.vt.edu/oai/request"},"display":{"title":"Development of Data-driven Models to Predict Vs30 from mHVSR","abstract":"This work investigates the potential of using microtremor horizontal-to-vertical spectral ratio (mHVSR) measurements to predict the time-averaged shear-wave velocity in the upper 30 meters (VS30) using data-driven models. We develop a dataset comprising 536 sites with 2,861 three-component ambient noise recordings from global regions, including New Zealand, Taiwan, Italy, Ecuador, Mexico and the United States. The identically processed three-component ambient noise recordings are used to make mHVSR measurements. To predict VS30 from mHVSR, we consider two types of models: low-dimensional models, which use features of the mHVSR curve such as the fundamental site frequency (f0,HVSR) and peak amplitude (A0,HVSR), and high-dimensional models, which use the entire mHVSR mean curve to predict VS30. In addition, we integrate topographic features from a 1 arc-second digital elevation model (DEM) for both model types using binned elevation as a proxy for geologic composition and relative elevation as a proxy for topography. The low-dimensional models are shown to reasonably predict VS30, coefficient of determination (R²) up to 0.69 on the testing set, when considering both mHVSR and topographic features. The high-dimensional models are shown to achieve improved accuracy to the low-dimensional models (R² of 0.82 on the testing set) when using the mean mHVSR curve regardless of whether the mean mHVSR curve is supplemented with the additional topographic features. These findings demonstrate that while low-dimensional features of the mHVSR curve are informative, leveraging the full shape of the mHVSR curve leads to improved prediction of VS30. Furthermore, the use of the mHVSR mean curve has the additional advantage that it does not require the extraction of features and can be used at all sites including those with and without resonant peaks. We compare the results with a model developed to predict VS30 from remote sensing data and demonstrate the utility of the models by predicting VS30 at 1,855 broadband recording stations across North America.","abstract_html":"This work investigates the potential of using microtremor horizontal-to-vertical spectral ratio (mHVSR) measurements to predict the time-averaged shear-wave velocity in the upper 30 meters (VS30) using data-driven models. We develop a dataset comprising 536 sites with 2,861 three-component ambient noise recordings from global regions, including New Zealand, Taiwan, Italy, Ecuador, Mexico and the United States. The identically processed three-component ambient noise recordings are used to make mHVSR measurements. To predict VS30 from mHVSR, we consider two types of models: low-dimensional models, which use features of the mHVSR curve such as the fundamental site frequency (f0,HVSR) and peak amplitude (A0,HVSR), and high-dimensional models, which use the entire mHVSR mean curve to predict VS30. In addition, we integrate topographic features from a 1 arc-second digital elevation model (DEM) for both model types using binned elevation as a proxy for geologic composition and relative elevation as a proxy for topography. The low-dimensional models are shown to reasonably predict VS30, coefficient of determination (R²) up to 0.69 on the testing set, when considering both mHVSR and topographic features. The high-dimensional models are shown to achieve improved accuracy to the low-dimensional models (R² of 0.82 on the testing set) when using the mean mHVSR curve regardless of whether the mean mHVSR curve is supplemented with the additional topographic features. These findings demonstrate that while low-dimensional features of the mHVSR curve are informative, leveraging the full shape of the mHVSR curve leads to improved prediction of VS30. Furthermore, the use of the mHVSR mean curve has the additional advantage that it does not require the extraction of features and can be used at all sites including those with and without resonant peaks. We compare the results with a model developed to predict VS30 from remote sensing data and demonstrate the utility of the models by predicting VS30 at 1,855 broadband recording stations across North America.","abstract_has_math":false,"creators":["Sharma Wagle, Kushal"],"institution":"Virginia Tech","degree_name":"Master of Science","degree_level":"masters","degree_discipline":"Civil Engineering","degree_department":"Civil and Environmental Engineering","school":null,"contributors":[],"advisors":[],"committee_chairs":["Vantassel, Joseph Philip"],"committee_members":["Rodriguez-Marek, Adrian","Green, Russell A."],"year":2025,"date_issued":"2025-05-29","date_published":"2025-05-29","updated_at":"2026-07-22T22:18:42Z","subjects":["mHVSR","HVSR","VS30","Machine Learning","Site Characterization"],"languages":["en"],"rights":["In Copyright"],"rights_urls":["http://rightsstatements.org/vocab/InC/1.0/"],"identifier_entries":[{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["vt_gsexam:44056"],"render_values":[{"text":"vt_gsexam:44056","href":null,"code":true}]}]},"links":{"outbound_url":"https://hdl.handle.net/10919/134297","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.committeechair","label":"Committee Chair","values":["Vantassel, Joseph Philip"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Rodriguez-Marek, Adrian","Green, Russell A."]},{"key":"dc:contributor.department","label":"Department","values":["Civil and Environmental Engineering"]},{"key":"dc:creator","label":"Author","values":["Sharma Wagle, Kushal"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2025-05-30T08:02:38Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2025-05-30T08:02:38Z"]},{"key":"dc:date.issued","label":"Date","values":["2025-05-29"]},{"key":"dc:publisher","label":"Institution","values":["Virginia Tech"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Civil Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["masters"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Virginia Polytechnic Institute and State University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["mHVSR","HVSR","VS30","Machine Learning","Site Characterization"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["In Copyright"]},{"key":"dc:rights.uri","label":"Rights URI","values":["http://rightsstatements.org/vocab/InC/1.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["vt_gsexam:44056"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/10919/134297"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["This work investigates the potential of using microtremor horizontal-to-vertical spectral ratio (mHVSR) measurements to predict the time-averaged shear-wave velocity in the upper 30 meters (VS30) using data-driven models. We develop a dataset comprising 536 sites with 2,861 three-component ambient noise recordings from global regions, including New Zealand, Taiwan, Italy, Ecuador, Mexico and the United States. The identically processed three-component ambient noise recordings are used to make mHVSR measurements. To predict VS30 from mHVSR, we consider two types of models: low-dimensional models, which use features of the mHVSR curve such as the fundamental site frequency (f0,HVSR) and peak amplitude (A0,HVSR), and high-dimensional models, which use the entire mHVSR mean curve to predict VS30. In addition, we integrate topographic features from a 1 arc-second digital elevation model (DEM) for both model types using binned elevation as a proxy for geologic composition and relative elevation as a proxy for topography. The low-dimensional models are shown to reasonably predict VS30, coefficient of determination (R²) up to 0.69 on the testing set, when considering both mHVSR and topographic features. The high-dimensional models are shown to achieve improved accuracy to the low-dimensional models (R² of 0.82 on the testing set) when using the mean mHVSR curve regardless of whether the mean mHVSR curve is supplemented with the additional topographic features. These findings demonstrate that while low-dimensional features of the mHVSR curve are informative, leveraging the full shape of the mHVSR curve leads to improved prediction of VS30. Furthermore, the use of the mHVSR mean curve has the additional advantage that it does not require the extraction of features and can be used at all sites including those with and without resonant peaks. We compare the results with a model developed to predict VS30 from remote sensing data and demonstrate the utility of the models by predicting VS30 at 1,855 broadband recording stations across North America."]},{"key":"dc:description.abstractgeneral","label":"General Abstract","values":["The amount of damage caused by earthquakes is strongly controlled by local ground conditions. Soft ground can dramatically amplify earthquake shaking compared to firm bedrock, making those areas more vulnerable to damage. To understand which areas may be most at risk, engineers use a parameter called VS30, which measures how soft the ground at a site is in the top 30 meters. VS30 plays a central role in earthquake hazard maps, building codes, and infrastructure design. However, traditional methods for measuring VS30 in the field are expensive, labor-intensive, and spatially limited, leaving large portions of the world unmapped and unprepared. My research proposes a practical solution: using a method called the Horizontal-to-Vertical Spectral Ratio (HVSR), to estimate VS30 in a low-cost, non-invasive way. HVSR involves recording weak, natural ground vibrations, called microtremors, that are present in the ground, even when there is no earthquake. By analyzing the way these vibrations behave, we can learn about the layers of soil and rock beneath the surface without drilling or using heavy equipment. While HVSR-based models have been developed previously using smaller regional datasets, this thesis develops a global dataset of HVSR and VS30 measurements. These data cover diverse geological environments that span multiple continents. Using this rich dataset, I developed machine learning models that combine HVSR with topographic features to produce better predictions of VS30. The result is a generalized, data-driven set of models capable of predicting VS30 from HVSR measurements. The impact of this work extends beyond earthquake engineering. It supports global efforts in urban planning, infrastructure development, and disaster risk reduction, especially in rapidly growing cities where formal geotechnical studies are lacking. It can enhance the resolution of global seismic hazard maps, inform safer land-use policies, and serve as a foundational layer for climate-resilient infrastructure planning, particularly in areas prone to multiple hazards such as landslides or flooding. By lowering the barriers to site characterization, this research contributes to a more informed and prepared global community, where science and data can guide decisions that ultimately save lives and reduce economic losses."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["Master of Science"]},{"key":"dc:format.medium","label":"Dc Format Medium","values":["ETD"]},{"key":"dc:title","label":"Title","values":["Development of Data-driven Models to Predict Vs30 from mHVSR"]}]}],"canonical_facts":{"dc:contributor.committeechair":["Vantassel, Joseph Philip"],"dc:contributor.committeemember":["Rodriguez-Marek, Adrian","Green, Russell A."],"dc:contributor.department":["Civil and Environmental Engineering"],"dc:creator":["Sharma Wagle, Kushal"],"dc:date.accessioned":["2025-05-30T08:02:38Z"],"dc:date.available":["2025-05-30T08:02:38Z"],"dc:date.issued":["2025-05-29"],"dc:description.abstract":["This work investigates the potential of using microtremor horizontal-to-vertical spectral ratio (mHVSR) measurements to predict the time-averaged shear-wave velocity in the upper 30 meters (VS30) using data-driven models. We develop a dataset comprising 536 sites with 2,861 three-component ambient noise recordings from global regions, including New Zealand, Taiwan, Italy, Ecuador, Mexico and the United States. The identically processed three-component ambient noise recordings are used to make mHVSR measurements. To predict VS30 from mHVSR, we consider two types of models: low-dimensional models, which use features of the mHVSR curve such as the fundamental site frequency (f0,HVSR) and peak amplitude (A0,HVSR), and high-dimensional models, which use the entire mHVSR mean curve to predict VS30. In addition, we integrate topographic features from a 1 arc-second digital elevation model (DEM) for both model types using binned elevation as a proxy for geologic composition and relative elevation as a proxy for topography. The low-dimensional models are shown to reasonably predict VS30, coefficient of determination (R²) up to 0.69 on the testing set, when considering both mHVSR and topographic features. The high-dimensional models are shown to achieve improved accuracy to the low-dimensional models (R² of 0.82 on the testing set) when using the mean mHVSR curve regardless of whether the mean mHVSR curve is supplemented with the additional topographic features. These findings demonstrate that while low-dimensional features of the mHVSR curve are informative, leveraging the full shape of the mHVSR curve leads to improved prediction of VS30. Furthermore, the use of the mHVSR mean curve has the additional advantage that it does not require the extraction of features and can be used at all sites including those with and without resonant peaks. We compare the results with a model developed to predict VS30 from remote sensing data and demonstrate the utility of the models by predicting VS30 at 1,855 broadband recording stations across North America."],"dc:description.abstractgeneral":["The amount of damage caused by earthquakes is strongly controlled by local ground conditions. Soft ground can dramatically amplify earthquake shaking compared to firm bedrock, making those areas more vulnerable to damage. To understand which areas may be most at risk, engineers use a parameter called VS30, which measures how soft the ground at a site is in the top 30 meters. VS30 plays a central role in earthquake hazard maps, building codes, and infrastructure design. However, traditional methods for measuring VS30 in the field are expensive, labor-intensive, and spatially limited, leaving large portions of the world unmapped and unprepared. My research proposes a practical solution: using a method called the Horizontal-to-Vertical Spectral Ratio (HVSR), to estimate VS30 in a low-cost, non-invasive way. HVSR involves recording weak, natural ground vibrations, called microtremors, that are present in the ground, even when there is no earthquake. By analyzing the way these vibrations behave, we can learn about the layers of soil and rock beneath the surface without drilling or using heavy equipment. While HVSR-based models have been developed previously using smaller regional datasets, this thesis develops a global dataset of HVSR and VS30 measurements. These data cover diverse geological environments that span multiple continents. Using this rich dataset, I developed machine learning models that combine HVSR with topographic features to produce better predictions of VS30. The result is a generalized, data-driven set of models capable of predicting VS30 from HVSR measurements. The impact of this work extends beyond earthquake engineering. It supports global efforts in urban planning, infrastructure development, and disaster risk reduction, especially in rapidly growing cities where formal geotechnical studies are lacking. It can enhance the resolution of global seismic hazard maps, inform safer land-use policies, and serve as a foundational layer for climate-resilient infrastructure planning, particularly in areas prone to multiple hazards such as landslides or flooding. By lowering the barriers to site characterization, this research contributes to a more informed and prepared global community, where science and data can guide decisions that ultimately save lives and reduce economic losses."],"dc:description.degree":["Master of Science"],"dc:format.medium":["ETD"],"dc:identifier.other":["vt_gsexam:44056"],"dc:identifier.uri":["https://hdl.handle.net/10919/134297"],"dc:language.iso":["en"],"dc:publisher":["Virginia Tech"],"dc:rights":["In Copyright"],"dc:rights.uri":["http://rightsstatements.org/vocab/InC/1.0/"],"dc:subject":["mHVSR","HVSR","VS30","Machine Learning","Site Characterization"],"dc:title":["Development of Data-driven Models to Predict Vs30 from mHVSR"],"dc:type":["Thesis"],"thesis:degree_discipline":["Civil Engineering"],"thesis:degree_level":["masters"],"thesis:degree_name":["Master of Science"],"thesis:institution_name":["Virginia Polytechnic Institute and State University"]},"updated_at":"2026-07-22T22:18:42Z"}