Search Results
You are looking at 21 - 30 of 40 items for
- Author or Editor: Cameron R. Homeyer x
- Refine by Access: All Content x
Abstract
Geostationary satellite imagers provide historical and near-real-time observations of cloud-top patterns that are commonly associated with severe convection. Environmental conditions favorable for severe weather are thought to be represented well by reanalyses. Predicting exactly where convection and costly storm hazards like hail will occur using models or satellite imagery alone, however, is extremely challenging. The multivariate combination of satellite-observed cloud patterns with reanalysis environmental parameters, linked to Next Generation Weather Radar (NEXRAD) estimated maximum expected size of hail (MESH) using a deep neural network (DNN), enables estimation of potentially severe hail likelihood for any observed storm cell. These estimates are made where satellites observe cold clouds, indicative of convection, located in favorable storm environments. We seek an approach that can be used to estimate climatological hailstorm frequency and risk throughout the historical satellite data record. Statistical distributions of convective parameters from satellite and reanalysis show separation between nonsevere and severe hailstorm classes for predictors that include overshooting cloud-top temperature and area characteristics, vertical wind shear, and convective inhibition. These complex, multivariate predictor relationships are exploited within a DNN to produce a likelihood estimate with a critical success index of 0.511 and Heidke skill score of 0.407, which is exceptional among analogous hail studies. Furthermore, applications of the DNN to case studies demonstrate good qualitative agreement between hail likelihood and MESH. These hail classifications are aggregated across an 11-yr Geostationary Operational Environmental Satellite (GOES) image database from GOES-12/13 to derive a hail frequency and severity climatology, which denotes the central Great Plains, the Midwest, and northwestern Mexico as being the most hail-prone regions within the domain studied.
Abstract
Geostationary satellite imagers provide historical and near-real-time observations of cloud-top patterns that are commonly associated with severe convection. Environmental conditions favorable for severe weather are thought to be represented well by reanalyses. Predicting exactly where convection and costly storm hazards like hail will occur using models or satellite imagery alone, however, is extremely challenging. The multivariate combination of satellite-observed cloud patterns with reanalysis environmental parameters, linked to Next Generation Weather Radar (NEXRAD) estimated maximum expected size of hail (MESH) using a deep neural network (DNN), enables estimation of potentially severe hail likelihood for any observed storm cell. These estimates are made where satellites observe cold clouds, indicative of convection, located in favorable storm environments. We seek an approach that can be used to estimate climatological hailstorm frequency and risk throughout the historical satellite data record. Statistical distributions of convective parameters from satellite and reanalysis show separation between nonsevere and severe hailstorm classes for predictors that include overshooting cloud-top temperature and area characteristics, vertical wind shear, and convective inhibition. These complex, multivariate predictor relationships are exploited within a DNN to produce a likelihood estimate with a critical success index of 0.511 and Heidke skill score of 0.407, which is exceptional among analogous hail studies. Furthermore, applications of the DNN to case studies demonstrate good qualitative agreement between hail likelihood and MESH. These hail classifications are aggregated across an 11-yr Geostationary Operational Environmental Satellite (GOES) image database from GOES-12/13 to derive a hail frequency and severity climatology, which denotes the central Great Plains, the Midwest, and northwestern Mexico as being the most hail-prone regions within the domain studied.
Abstract
The quantification of storm updrafts remains unavailable for operational forecasting despite their inherent importance to convection and its associated severe weather hazards. Updraft proxies, like overshooting top area from satellite images, have been linked to severe weather hazards but only relate to a limited portion of the total storm updraft. This study investigates if a machine learning model, namely, U-Nets, can skillfully retrieve maximum vertical velocity and its areal extent from three-dimensional gridded radar reflectivity alone. The machine learning model is trained using simulated radar reflectivity and vertical velocity from the National Severe Storm Laboratory’s convection permitting Warn-on-Forecast System (WoFS). A parametric regression technique using the sinh–arcsinh–normal distribution is adapted to run with U-Nets, allowing for both deterministic and probabilistic predictions of maximum vertical velocity. The best models after hyperparameter search provided less than 50% root mean squared error, a coefficient of determination greater than 0.65, and an intersection over union (IoU) of more than 0.45 on the independent test set composed of WoFS data. Beyond the WoFS analysis, a case study was conducted using real radar data and corresponding dual-Doppler analyses of vertical velocity within a supercell. The U-Net consistently underestimates the dual-Doppler updraft speed estimates by 50%. Meanwhile, the area of the 5 and 10 m s−1 updraft cores shows an IoU of 0.25. While the above statistics are not exceptional, the machine learning model enables quick distillation of 3D radar data that is related to the maximum vertical velocity, which could be useful in assessing a storm’s severe potential.
Significance Statement
All convective storm hazards (tornadoes, hail, heavy rain, straight line winds) can be related to a storm’s updraft. Yet, there is no direct measurement of updraft speed or area available for forecasters to make their warning decisions from. This paper addresses the lack of observational data by providing a machine learning solution that skillfully estimates the maximum updraft speed within storms from only the radar reflectivity 3D structure. After further vetting the machine learning solutions on additional real-world examples, the estimated storm updrafts will hopefully provide forecasters with an added tool to help diagnose a storm’s hazard potential more accurately.
Abstract
The quantification of storm updrafts remains unavailable for operational forecasting despite their inherent importance to convection and its associated severe weather hazards. Updraft proxies, like overshooting top area from satellite images, have been linked to severe weather hazards but only relate to a limited portion of the total storm updraft. This study investigates if a machine learning model, namely, U-Nets, can skillfully retrieve maximum vertical velocity and its areal extent from three-dimensional gridded radar reflectivity alone. The machine learning model is trained using simulated radar reflectivity and vertical velocity from the National Severe Storm Laboratory’s convection permitting Warn-on-Forecast System (WoFS). A parametric regression technique using the sinh–arcsinh–normal distribution is adapted to run with U-Nets, allowing for both deterministic and probabilistic predictions of maximum vertical velocity. The best models after hyperparameter search provided less than 50% root mean squared error, a coefficient of determination greater than 0.65, and an intersection over union (IoU) of more than 0.45 on the independent test set composed of WoFS data. Beyond the WoFS analysis, a case study was conducted using real radar data and corresponding dual-Doppler analyses of vertical velocity within a supercell. The U-Net consistently underestimates the dual-Doppler updraft speed estimates by 50%. Meanwhile, the area of the 5 and 10 m s−1 updraft cores shows an IoU of 0.25. While the above statistics are not exceptional, the machine learning model enables quick distillation of 3D radar data that is related to the maximum vertical velocity, which could be useful in assessing a storm’s severe potential.
Significance Statement
All convective storm hazards (tornadoes, hail, heavy rain, straight line winds) can be related to a storm’s updraft. Yet, there is no direct measurement of updraft speed or area available for forecasters to make their warning decisions from. This paper addresses the lack of observational data by providing a machine learning solution that skillfully estimates the maximum updraft speed within storms from only the radar reflectivity 3D structure. After further vetting the machine learning solutions on additional real-world examples, the estimated storm updrafts will hopefully provide forecasters with an added tool to help diagnose a storm’s hazard potential more accurately.
Abstract
Hurricane Harvey produced unprecedented widespread rainfall amounts over 1000 mm in portions of southeast Texas, including Houston, from 26 to 31 August 2017. The highly efficient and prolonged warm rain processes associated with Harvey played a key role in the catastrophic flooding that occurred throughout the region. Precipitation efficiency (PE) is widely referred to in the scientific literature when discussing excessive precipitation events that lead to catastrophic flash flooding, but has yet to be explored or quantified in tropical cyclones coincident with polarimetric radar observations. With the introduction of dual-polarization radar to the NEXRAD WSR-88D network, polarimetric radar variables such as Z H , Z DR, and K DP can be used to gain insight into the precipitation processes that contribute to enhanced PE. It was found that 6-h mean values of Z H between 35 and 45 dBZ, Z DR between 1 and 1.5 dB, and K DP greater than 1° km−1 were collocated with the regions of PE greater than 100% between 27 and 29 August. Additionally, supercell thunderstorms embedded in the outer bands of Harvey were identified via 3–6 km Multi-Radar Multi-Senor (MRMS) rotation tracks and were collocated with swaths of enhanced positive Z H , Z DR, and K DP. A polarimetric rainfall relationship estimates that 1-h mean rainfall rates in these supercells were as high as 85 mm h−1 and made a significant contribution to the excessive precipitation event that occurred over the region.
Abstract
Hurricane Harvey produced unprecedented widespread rainfall amounts over 1000 mm in portions of southeast Texas, including Houston, from 26 to 31 August 2017. The highly efficient and prolonged warm rain processes associated with Harvey played a key role in the catastrophic flooding that occurred throughout the region. Precipitation efficiency (PE) is widely referred to in the scientific literature when discussing excessive precipitation events that lead to catastrophic flash flooding, but has yet to be explored or quantified in tropical cyclones coincident with polarimetric radar observations. With the introduction of dual-polarization radar to the NEXRAD WSR-88D network, polarimetric radar variables such as Z H , Z DR, and K DP can be used to gain insight into the precipitation processes that contribute to enhanced PE. It was found that 6-h mean values of Z H between 35 and 45 dBZ, Z DR between 1 and 1.5 dB, and K DP greater than 1° km−1 were collocated with the regions of PE greater than 100% between 27 and 29 August. Additionally, supercell thunderstorms embedded in the outer bands of Harvey were identified via 3–6 km Multi-Radar Multi-Senor (MRMS) rotation tracks and were collocated with swaths of enhanced positive Z H , Z DR, and K DP. A polarimetric rainfall relationship estimates that 1-h mean rainfall rates in these supercells were as high as 85 mm h−1 and made a significant contribution to the excessive precipitation event that occurred over the region.
Abstract
Changes in land surface and aerosol characteristics from urbanization can affect dynamic and microphysical properties of severe storms, thus affecting hazardous weather events resulting from these storms such as hail and tornadoes. We examine the joint and individual effects of urban land and anthropogenic aerosols of Kansas City on a severe convective storm observed during the 2015 Plains Elevated Convection At Night (PECAN) field campaign, focusing on storm evolution, convective intensity, and hail characteristics. The simulations are carried out at the cloud-resolving scale (1 km) using a version of WRF-Chem in which the spectral-bin microphysics (SBM) is coupled with the Model for Simulating Aerosol Interactions and Chemistry (MOSAIC). It is found that the urban land effect of Kansas City initiated a much stronger convective cell and the storm got further intensified when interacting with stronger turbulence induced by the urban land. The urban land effect also changed the storm path by diverting the storm toward the city, mainly resulting from enhanced urban land-induced convergence in the urban area and around the urban–rural boundaries. The joint effect of urban land and anthropogenic aerosols enhances occurrences of both severe hail and significant severe hail by ~20% by enhancing hail formation and growth from riming. Overall the urban land effect on convective intensity and hail is relatively larger than the anthropogenic aerosol effect, but the joint effect is more notable than either of the individual effects, emphasizing the importance of considering both effects in evaluating urbanization effects.
Abstract
Changes in land surface and aerosol characteristics from urbanization can affect dynamic and microphysical properties of severe storms, thus affecting hazardous weather events resulting from these storms such as hail and tornadoes. We examine the joint and individual effects of urban land and anthropogenic aerosols of Kansas City on a severe convective storm observed during the 2015 Plains Elevated Convection At Night (PECAN) field campaign, focusing on storm evolution, convective intensity, and hail characteristics. The simulations are carried out at the cloud-resolving scale (1 km) using a version of WRF-Chem in which the spectral-bin microphysics (SBM) is coupled with the Model for Simulating Aerosol Interactions and Chemistry (MOSAIC). It is found that the urban land effect of Kansas City initiated a much stronger convective cell and the storm got further intensified when interacting with stronger turbulence induced by the urban land. The urban land effect also changed the storm path by diverting the storm toward the city, mainly resulting from enhanced urban land-induced convergence in the urban area and around the urban–rural boundaries. The joint effect of urban land and anthropogenic aerosols enhances occurrences of both severe hail and significant severe hail by ~20% by enhancing hail formation and growth from riming. Overall the urban land effect on convective intensity and hail is relatively larger than the anthropogenic aerosol effect, but the joint effect is more notable than either of the individual effects, emphasizing the importance of considering both effects in evaluating urbanization effects.
Abstract
An improved understanding of common differences between tornadic and nontornadic supercells is sought using a large set of observations from the operational NEXRAD WSR-88D polarimetric radar network in the contiguous United States. In particular, data from 478 nontornadic and 294 tornadic supercells during a 7-yr period (2011–17) are used to produce probability-matched composite means of microphysical and kinematic variables. Means, which are centered on echo-top maxima and in a horizontal coordinate system rotated such that storm motion points in the positive x dimension, are created in altitude relative to ground level at times of peak echo-top altitude and peak midlevel rotation for nontornadic supercells and times at and prior to the first tornado in tornadic supercells. Robust differences between supercell types are found, with consistent characteristics at and preceding tornadogenesis in tornadic storms. In particular, the mesocyclone is found to be vertically aligned in tornadic supercells and misaligned in nontornadic supercells. Microphysical differences found include a low-level radar reflectivity hook echo aligned with and ~10 km right of storm center in tornadic supercells and displaced 5–10 km down-motion in nontornadic supercells, a low-to-midlevel differential radar reflectivity dipole that is oriented more parallel to storm motion in tornadic supercells and more perpendicular in nontornadic supercells, and a separation between enhanced differential radar reflectivity and specific differential phase (with unique displacement-relative correlation coefficient reductions) at low levels that is more perpendicular to storm motion in tornadic supercells and more parallel in nontornadic supercells.
Abstract
An improved understanding of common differences between tornadic and nontornadic supercells is sought using a large set of observations from the operational NEXRAD WSR-88D polarimetric radar network in the contiguous United States. In particular, data from 478 nontornadic and 294 tornadic supercells during a 7-yr period (2011–17) are used to produce probability-matched composite means of microphysical and kinematic variables. Means, which are centered on echo-top maxima and in a horizontal coordinate system rotated such that storm motion points in the positive x dimension, are created in altitude relative to ground level at times of peak echo-top altitude and peak midlevel rotation for nontornadic supercells and times at and prior to the first tornado in tornadic supercells. Robust differences between supercell types are found, with consistent characteristics at and preceding tornadogenesis in tornadic storms. In particular, the mesocyclone is found to be vertically aligned in tornadic supercells and misaligned in nontornadic supercells. Microphysical differences found include a low-level radar reflectivity hook echo aligned with and ~10 km right of storm center in tornadic supercells and displaced 5–10 km down-motion in nontornadic supercells, a low-to-midlevel differential radar reflectivity dipole that is oriented more parallel to storm motion in tornadic supercells and more perpendicular in nontornadic supercells, and a separation between enhanced differential radar reflectivity and specific differential phase (with unique displacement-relative correlation coefficient reductions) at low levels that is more perpendicular to storm motion in tornadic supercells and more parallel in nontornadic supercells.
Abstract
This paper describes the development of convolutional neural networks (CNN), a type of deep-learning method, to predict next-hour tornado occurrence. Predictors are a storm-centered radar image and a proximity sounding from the Rapid Refresh model. Radar images come from the Multiyear Reanalysis of Remotely Sensed Storms (MYRORSS) and Gridded NEXRAD WSR-88D Radar dataset (GridRad), both of which are multiradar composites. We train separate CNNs on MYRORSS and GridRad data, present an experiment to optimize the CNN settings, and evaluate the chosen CNNs on independent testing data. Both models achieve an area under the receiver-operating-characteristic curve (AUC) well above 0.9, which is considered to be excellent performance. The GridRad model achieves a critical success index (CSI) of 0.31, and the MYRORSS model achieves a CSI of 0.17. The difference is due primarily to event frequency (percentage of storms that are tornadic in the next hour), which is 3.52% for GridRad but only 0.24% for MYRORSS. The best CNN predictions (true positives and negatives) occur for strongly rotating tornadic supercells and weak nontornadic cells in mesoscale convective systems, respectively. The worst predictions (false positives and negatives) occur for strongly rotating nontornadic supercells and tornadic cells in quasi-linear convective systems, respectively. The performance of our CNNs is comparable to an operational machine-learning system for severe weather prediction, which suggests that they would be useful for real-time forecasting.
Abstract
This paper describes the development of convolutional neural networks (CNN), a type of deep-learning method, to predict next-hour tornado occurrence. Predictors are a storm-centered radar image and a proximity sounding from the Rapid Refresh model. Radar images come from the Multiyear Reanalysis of Remotely Sensed Storms (MYRORSS) and Gridded NEXRAD WSR-88D Radar dataset (GridRad), both of which are multiradar composites. We train separate CNNs on MYRORSS and GridRad data, present an experiment to optimize the CNN settings, and evaluate the chosen CNNs on independent testing data. Both models achieve an area under the receiver-operating-characteristic curve (AUC) well above 0.9, which is considered to be excellent performance. The GridRad model achieves a critical success index (CSI) of 0.31, and the MYRORSS model achieves a CSI of 0.17. The difference is due primarily to event frequency (percentage of storms that are tornadic in the next hour), which is 3.52% for GridRad but only 0.24% for MYRORSS. The best CNN predictions (true positives and negatives) occur for strongly rotating tornadic supercells and weak nontornadic cells in mesoscale convective systems, respectively. The worst predictions (false positives and negatives) occur for strongly rotating nontornadic supercells and tornadic cells in quasi-linear convective systems, respectively. The performance of our CNNs is comparable to an operational machine-learning system for severe weather prediction, which suggests that they would be useful for real-time forecasting.
Abstract
Intense tropopause-penetrating updrafts and gravity wave breaking generate cirrus plumes that reside above the primary anvil. These “above anvil cirrus plumes” (AACPs) exhibit unique temperature and reflectance patterns in satellite imagery, best recognized within 1-min “super rapid scan” observations. AACPs are often evident during severe weather outbreaks and, due to their importance, have been studied for 35+ years. Despite this research, there is uncertainty regarding why some storms produce AACPs but other nearby storms do not, exactly how severe are storms with AACPs, and how AACP identification can assist with severe weather warning. These uncertainties are addressed through analysis of severe weather reports, NOAA/National Weather Service (NWS) severe weather warnings, metrics of updraft cloud height, intensity, and rotation derived from Doppler radars, as well as ground-based total lightning observations for 4583 storms observed by GOES super rapid scanning, 405 of which produced an AACP. Datasets are accumulated throughout storm lifetimes through radar object tracking. It is found that 1) AACP storms generated 14 times the number of reports per storm compared to non-AACP storms; 2) AACPs appeared, on average, 31 min in advance of severe weather; 3) 73% of significant severe weather reports were produced by AACP storms; 4) AACP recognition can provide comparable warning lead time to that provided by a forecaster; and 5) the presence of an AACP can increase forecaster confidence that large hail will occur. Given that AACPs occur throughout the world, and most of the world is not observed by Doppler radar, AACP-based severe storm identification and warning would be extremely helpful for protecting lives and property.
Abstract
Intense tropopause-penetrating updrafts and gravity wave breaking generate cirrus plumes that reside above the primary anvil. These “above anvil cirrus plumes” (AACPs) exhibit unique temperature and reflectance patterns in satellite imagery, best recognized within 1-min “super rapid scan” observations. AACPs are often evident during severe weather outbreaks and, due to their importance, have been studied for 35+ years. Despite this research, there is uncertainty regarding why some storms produce AACPs but other nearby storms do not, exactly how severe are storms with AACPs, and how AACP identification can assist with severe weather warning. These uncertainties are addressed through analysis of severe weather reports, NOAA/National Weather Service (NWS) severe weather warnings, metrics of updraft cloud height, intensity, and rotation derived from Doppler radars, as well as ground-based total lightning observations for 4583 storms observed by GOES super rapid scanning, 405 of which produced an AACP. Datasets are accumulated throughout storm lifetimes through radar object tracking. It is found that 1) AACP storms generated 14 times the number of reports per storm compared to non-AACP storms; 2) AACPs appeared, on average, 31 min in advance of severe weather; 3) 73% of significant severe weather reports were produced by AACP storms; 4) AACP recognition can provide comparable warning lead time to that provided by a forecaster; and 5) the presence of an AACP can increase forecaster confidence that large hail will occur. Given that AACPs occur throughout the world, and most of the world is not observed by Doppler radar, AACP-based severe storm identification and warning would be extremely helpful for protecting lives and property.
Abstract
Remote sensing observations, especially those from ground-based radars, have been used extensively to discriminate between severe and nonsevere storms. Recent upgrades to operational remote sensing networks in the United States have provided unprecedented spatial and temporal sampling to study such storms. These networks help forecasters subjectively identify storms capable of producing severe weather at the ground; however, uncertainties remain in how to objectively identify severe thunderstorms using the same data. Here, three large-area datasets (geostationary satellite, ground-based radar, and ground-based lightning detection) are used over 28 recent events in an attempt to objectively discriminate between severe and nonsevere storms, with an additional focus on severe storms that produce tornadoes. Among these datasets, radar observations, specifically those at mid- and upper levels (altitudes at and above 4 km), are shown to provide the greatest objective discrimination. Physical and kinematic storm characteristics from all analyzed datasets imply that significantly severe [≥2-in. (5.08 cm) hail and/or ≥65-kt (33.4 m s−1) straight-line winds] and tornadic storms have stronger upward motion and rotation than nonsevere and less severe storms. In addition, these metrics are greatest in tornadic storms during the time in which tornadoes occur.
Abstract
Remote sensing observations, especially those from ground-based radars, have been used extensively to discriminate between severe and nonsevere storms. Recent upgrades to operational remote sensing networks in the United States have provided unprecedented spatial and temporal sampling to study such storms. These networks help forecasters subjectively identify storms capable of producing severe weather at the ground; however, uncertainties remain in how to objectively identify severe thunderstorms using the same data. Here, three large-area datasets (geostationary satellite, ground-based radar, and ground-based lightning detection) are used over 28 recent events in an attempt to objectively discriminate between severe and nonsevere storms, with an additional focus on severe storms that produce tornadoes. Among these datasets, radar observations, specifically those at mid- and upper levels (altitudes at and above 4 km), are shown to provide the greatest objective discrimination. Physical and kinematic storm characteristics from all analyzed datasets imply that significantly severe [≥2-in. (5.08 cm) hail and/or ≥65-kt (33.4 m s−1) straight-line winds] and tornadic storms have stronger upward motion and rotation than nonsevere and less severe storms. In addition, these metrics are greatest in tornadic storms during the time in which tornadoes occur.
Abstract
Few studies have assessed combined satellite, lightning, and radar databases to diagnose severe storm potential. The research goal here is to evaluate next-generation, 60-s update frequency geostationary satellite and lightning information with ground-based radar to isolate which variables, when used in concert, provide skillful discriminatory information for identifying severe (hail ≥ 2.5 cm in diameter, winds ≥ 25 m s−1, and tornadoes) versus nonsevere storms. The focus of this study is predicting severe thunderstorm and tornado warnings. A total of 2004 storms in 2014–15 were objectively tracked with 49 potential predictor fields related to May, daytime Great Plains convective storms. All storms occurred when 1-min Geostationary Operational Environmental Satellite (GOES)-14 “super rapid scan” data were available. The study used three importance methods to assess predictor importance related to severe warnings and used random forests to provide a model and skill evaluation measuring the ability to predict severe storms. Three predictor importance methods show that GOES mesoscale atmospheric-motion-vector-derived cloud-top divergence and above-anvil cirrus plume presence provide the most satellite-based discriminatory power for diagnosing severe warnings. Other important fields include Earth Networks Total Lightning flash density, GOES estimated cloud-top vorticity, and overshooting-top presence. Severe warning predictions are significantly improved at the 95% confidence level when a few important satellite and lightning fields are combined with radar fields, versus when only radar data are used in the random-forest model. This study provides a basis for including satellite and lightning fields within machine-learning models to help forecast severe weather.
Abstract
Few studies have assessed combined satellite, lightning, and radar databases to diagnose severe storm potential. The research goal here is to evaluate next-generation, 60-s update frequency geostationary satellite and lightning information with ground-based radar to isolate which variables, when used in concert, provide skillful discriminatory information for identifying severe (hail ≥ 2.5 cm in diameter, winds ≥ 25 m s−1, and tornadoes) versus nonsevere storms. The focus of this study is predicting severe thunderstorm and tornado warnings. A total of 2004 storms in 2014–15 were objectively tracked with 49 potential predictor fields related to May, daytime Great Plains convective storms. All storms occurred when 1-min Geostationary Operational Environmental Satellite (GOES)-14 “super rapid scan” data were available. The study used three importance methods to assess predictor importance related to severe warnings and used random forests to provide a model and skill evaluation measuring the ability to predict severe storms. Three predictor importance methods show that GOES mesoscale atmospheric-motion-vector-derived cloud-top divergence and above-anvil cirrus plume presence provide the most satellite-based discriminatory power for diagnosing severe warnings. Other important fields include Earth Networks Total Lightning flash density, GOES estimated cloud-top vorticity, and overshooting-top presence. Severe warning predictions are significantly improved at the 95% confidence level when a few important satellite and lightning fields are combined with radar fields, versus when only radar data are used in the random-forest model. This study provides a basis for including satellite and lightning fields within machine-learning models to help forecast severe weather.
Abstract
Rapid acceleration of cloud-top outflow near vigorous storm updrafts can be readily observed in Geostationary Operational Environmental Satellite-14 (GOES-14) super rapid scan (SRS; 60 s) mode data. Conventional wisdom implies that this outflow is related to the intensity of updrafts and the formation of severe weather. However, from an SRS satellite perspective, the pairing of observed expansion and updraft intensity has not been objectively derived and documented. The goal of this study is to relate GOES-14 SRS-derived cloud-top horizontal divergence (CTD) over deep convection to internal updraft characteristics, and document evolution for severe and nonsevere thunderstorms. A new SRS flow derivation system is presented here to estimate storm-scale (<20 km) CTD. This CTD field is coupled with other proxies for storm updraft location and intensity such as overshooting tops (OTs), total lightning flash rates, and three-dimensional flow fields derived from dual-Doppler radar data. Objectively identified OTs with (without) matching CTD maxima were more (less) likely to be associated with radar-observed deep convection and severe weather reports at the ground, suggesting that some OTs were incorrectly identified. The correlation between CTD magnitude, maximum updraft speed, and total lightning was strongly positive for a nonsupercell pulse storm, and weakly positive for a supercell with multiple updraft pulses present. The relationship for the supercell was nonlinear, though larger flash rates are found during periods of larger CTD. Analysis here suggests that combining CTD with OTs and total lightning could have severe weather nowcasting value.
Abstract
Rapid acceleration of cloud-top outflow near vigorous storm updrafts can be readily observed in Geostationary Operational Environmental Satellite-14 (GOES-14) super rapid scan (SRS; 60 s) mode data. Conventional wisdom implies that this outflow is related to the intensity of updrafts and the formation of severe weather. However, from an SRS satellite perspective, the pairing of observed expansion and updraft intensity has not been objectively derived and documented. The goal of this study is to relate GOES-14 SRS-derived cloud-top horizontal divergence (CTD) over deep convection to internal updraft characteristics, and document evolution for severe and nonsevere thunderstorms. A new SRS flow derivation system is presented here to estimate storm-scale (<20 km) CTD. This CTD field is coupled with other proxies for storm updraft location and intensity such as overshooting tops (OTs), total lightning flash rates, and three-dimensional flow fields derived from dual-Doppler radar data. Objectively identified OTs with (without) matching CTD maxima were more (less) likely to be associated with radar-observed deep convection and severe weather reports at the ground, suggesting that some OTs were incorrectly identified. The correlation between CTD magnitude, maximum updraft speed, and total lightning was strongly positive for a nonsupercell pulse storm, and weakly positive for a supercell with multiple updraft pulses present. The relationship for the supercell was nonlinear, though larger flash rates are found during periods of larger CTD. Analysis here suggests that combining CTD with OTs and total lightning could have severe weather nowcasting value.
