Search Results
You are looking at 1 - 5 of 5 items for :
- Author or Editor: Cameron R. Homeyer x
- Bulletin of the American Meteorological Society x
- Refine by Access: All Content x
Abstract
The diurnal cycle of coastal rainfall over western Puerto Rico was studied with high-frequency radiosondes launched by undergraduate students at the University of Puerto Rico at Mayagüez (UPRM). Thirty radiosondes were launched during a 3-week period as part of NASA’s Convective Processes Experiment—Aerosols and Winds (CPEX-AW) field project. The objective of the radiosonde launches over Puerto Rico was to understand the evolution of coastal convective systems that are often challenging to predict. Four different events were sampled: 1) a short-lived rainfall event during a Saharan air dust outbreak, 2) a 2-day period of limited rainfall activity under northeasterly wind conditions, 3) a 2-day period of heavy rainfall over land, and 4) a 2-day period of long-lived rainfall events that initiated over land and propagated offshore during the evening hours. The radiosondes captured the sea-breeze onset during the midmorning hours, an erosion of lower-tropospheric inversions, and substantial differences in column humidity between the four events. All radiosondes were launched by volunteer undergraduate students who were able to participate in person, while the coordination was done virtually with lead scientists located in Puerto Rico, Oklahoma, and Saint Croix. Overall, this initiative highlighted the importance of student–scientist collaboration in collecting critical observations to better understand complex atmospheric processes.
Abstract
The diurnal cycle of coastal rainfall over western Puerto Rico was studied with high-frequency radiosondes launched by undergraduate students at the University of Puerto Rico at Mayagüez (UPRM). Thirty radiosondes were launched during a 3-week period as part of NASA’s Convective Processes Experiment—Aerosols and Winds (CPEX-AW) field project. The objective of the radiosonde launches over Puerto Rico was to understand the evolution of coastal convective systems that are often challenging to predict. Four different events were sampled: 1) a short-lived rainfall event during a Saharan air dust outbreak, 2) a 2-day period of limited rainfall activity under northeasterly wind conditions, 3) a 2-day period of heavy rainfall over land, and 4) a 2-day period of long-lived rainfall events that initiated over land and propagated offshore during the evening hours. The radiosondes captured the sea-breeze onset during the midmorning hours, an erosion of lower-tropospheric inversions, and substantial differences in column humidity between the four events. All radiosondes were launched by volunteer undergraduate students who were able to participate in person, while the coordination was done virtually with lead scientists located in Puerto Rico, Oklahoma, and Saint Croix. Overall, this initiative highlighted the importance of student–scientist collaboration in collecting critical observations to better understand complex atmospheric processes.
Abstract
This paper synthesizes multiple methods for machine learning (ML) model interpretation and visualization (MIV) focusing on meteorological applications. ML has recently exploded in popularity in many fields, including meteorology. Although ML has been successful in meteorology, it has not been as widely accepted, primarily due to the perception that ML models are “black boxes,” meaning the ML methods are thought to take inputs and provide outputs but not to yield physically interpretable information to the user. This paper introduces and demonstrates multiple MIV techniques for both traditional ML and deep learning, to enable meteorologists to understand what ML models have learned. We discuss permutation-based predictor importance, forward and backward selection, saliency maps, class-activation maps, backward optimization, and novelty detection. We apply these methods at multiple spatiotemporal scales to tornado, hail, winter precipitation type, and convective-storm mode. By analyzing such a wide variety of applications, we intend for this work to demystify the black box of ML, offer insight in applying MIV techniques, and serve as a MIV toolbox for meteorologists and other physical scientists.
Abstract
This paper synthesizes multiple methods for machine learning (ML) model interpretation and visualization (MIV) focusing on meteorological applications. ML has recently exploded in popularity in many fields, including meteorology. Although ML has been successful in meteorology, it has not been as widely accepted, primarily due to the perception that ML models are “black boxes,” meaning the ML methods are thought to take inputs and provide outputs but not to yield physically interpretable information to the user. This paper introduces and demonstrates multiple MIV techniques for both traditional ML and deep learning, to enable meteorologists to understand what ML models have learned. We discuss permutation-based predictor importance, forward and backward selection, saliency maps, class-activation maps, backward optimization, and novelty detection. We apply these methods at multiple spatiotemporal scales to tornado, hail, winter precipitation type, and convective-storm mode. By analyzing such a wide variety of applications, we intend for this work to demystify the black box of ML, offer insight in applying MIV techniques, and serve as a MIV toolbox for meteorologists and other physical scientists.
The Stratosphere–Troposphere Analyses of Regional Transport 2008 (START08) experiment investigated a number of important processes in the extratropical upper troposphere and lower stratosphere (UTLS) using the National Science Foundation (NSF)–NCAR Gulfstream V (GV) research aircraft. The main objective was to examine the chemical structure of the extratropical UTLS in relation to dynamical processes spanning a range of scales. The campaign was conducted during April–June 2008 from Broomfield, Colorado. A total of 18 research flights sampled an extensive geographical region of North America (25°–65°N, 80°–120°W) and a wide range of meteorological conditions. The airborne in situ instruments measured a comprehensive suite of chemical constituents and microphysical variables from the boundary layer to the lower stratosphere, with flights specifically designed to target key transport processes in the extratropical UTLS. The flights successfully investigated stratosphere–troposphere exchange (STE) processes, including the intrusion of tropospheric air into the stratosphere in association with the secondary tropopause and the intrusion of stratospheric air deep into the troposphere. The flights also sampled the influence of convective transport and lightning on the upper troposphere as well as the distribution of gravity waves associated with multiple sources, including fronts and topography. The aircraft observations are complemented by satellite observations and modeling. The measurements will be used to improve the representation of UTLS chemical gradients and transport in Chemistry–Climate models (CCMs). This article provides an overview of the experiment design and selected observational highlights.
The Stratosphere–Troposphere Analyses of Regional Transport 2008 (START08) experiment investigated a number of important processes in the extratropical upper troposphere and lower stratosphere (UTLS) using the National Science Foundation (NSF)–NCAR Gulfstream V (GV) research aircraft. The main objective was to examine the chemical structure of the extratropical UTLS in relation to dynamical processes spanning a range of scales. The campaign was conducted during April–June 2008 from Broomfield, Colorado. A total of 18 research flights sampled an extensive geographical region of North America (25°–65°N, 80°–120°W) and a wide range of meteorological conditions. The airborne in situ instruments measured a comprehensive suite of chemical constituents and microphysical variables from the boundary layer to the lower stratosphere, with flights specifically designed to target key transport processes in the extratropical UTLS. The flights successfully investigated stratosphere–troposphere exchange (STE) processes, including the intrusion of tropospheric air into the stratosphere in association with the secondary tropopause and the intrusion of stratospheric air deep into the troposphere. The flights also sampled the influence of convective transport and lightning on the upper troposphere as well as the distribution of gravity waves associated with multiple sources, including fronts and topography. The aircraft observations are complemented by satellite observations and modeling. The measurements will be used to improve the representation of UTLS chemical gradients and transport in Chemistry–Climate models (CCMs). This article provides an overview of the experiment design and selected observational highlights.
Abstract
Color vision deficiency (CVD) is a decreased ability to discern between particular colors. Eight percent of genetic males and half a percent of genetic females have some form of CVD, with many in the radar community falling into this group. When presenting data on a two-dimensional plane, it is common to use colors to represent values via a colormap. Colormap choice in the radar community is influenced by the ability to highlight scientifically interesting features in data, institutional choices, and domain dominance of legacy colormaps. The problem with these current colormaps is that many do not project well for those with CVD (i.e., green next to red). In working with the CVD community to address this problem, multiple colormaps for moments such as equivalent reflectivity factor and Doppler velocity were created for users with forms of CVD such as deuteranomaly, protanomaly, protanopia, and deuteranopia using Python tools such as colorspacious and viscm. We show how these colormaps can improve interpretability for four cases: a mesoscale convective system, a pyrocumulonimbus storm, a wintertime midlatitude cyclone, and widespread storms with a large bird migration. These new radar equivalent reflectivity factor, Doppler velocity, and polarization colormaps are designed to highlight rain, frozen precipitation, nonmeteorological targets, and velocity-based items, are perceptually uniform, and are visually friendly for those with CVD.
Abstract
Color vision deficiency (CVD) is a decreased ability to discern between particular colors. Eight percent of genetic males and half a percent of genetic females have some form of CVD, with many in the radar community falling into this group. When presenting data on a two-dimensional plane, it is common to use colors to represent values via a colormap. Colormap choice in the radar community is influenced by the ability to highlight scientifically interesting features in data, institutional choices, and domain dominance of legacy colormaps. The problem with these current colormaps is that many do not project well for those with CVD (i.e., green next to red). In working with the CVD community to address this problem, multiple colormaps for moments such as equivalent reflectivity factor and Doppler velocity were created for users with forms of CVD such as deuteranomaly, protanomaly, protanopia, and deuteranopia using Python tools such as colorspacious and viscm. We show how these colormaps can improve interpretability for four cases: a mesoscale convective system, a pyrocumulonimbus storm, a wintertime midlatitude cyclone, and widespread storms with a large bird migration. These new radar equivalent reflectivity factor, Doppler velocity, and polarization colormaps are designed to highlight rain, frozen precipitation, nonmeteorological targets, and velocity-based items, are perceptually uniform, and are visually friendly for those with CVD.
Abstract
The catastrophic derecho that occurred on 10 August 2020 across the midwestern United States caused billions of dollars of damage to both urban and rural infrastructure as well as agricultural crops, most notably across the state of Iowa. This paper documents the complex evolution of the derecho through the use of low-Earth-orbit passive-microwave imager and GOES-16 satellite-derived products complemented by products derived from NEXRAD weather radar observations. Additional satellite sensors including optical imagers and synthetic aperture radar (SAR) were used to observe impacts to the power grid and agriculture in Iowa. SAR improved the identification and quantification of damaged corn and soybeans, as compared to true-color composites and normalized difference vegetation index (NDVI). A statistical approach to identify damaged corn and soybean crops from SAR was created with estimates of 1.97 million acres of damaged corn and 1.40 million acres of damaged soybeans in the state of Iowa. The damage estimates generated by this study were comparable to estimates produced by others after the derecho, including two commercial agricultural companies.
Abstract
The catastrophic derecho that occurred on 10 August 2020 across the midwestern United States caused billions of dollars of damage to both urban and rural infrastructure as well as agricultural crops, most notably across the state of Iowa. This paper documents the complex evolution of the derecho through the use of low-Earth-orbit passive-microwave imager and GOES-16 satellite-derived products complemented by products derived from NEXRAD weather radar observations. Additional satellite sensors including optical imagers and synthetic aperture radar (SAR) were used to observe impacts to the power grid and agriculture in Iowa. SAR improved the identification and quantification of damaged corn and soybeans, as compared to true-color composites and normalized difference vegetation index (NDVI). A statistical approach to identify damaged corn and soybean crops from SAR was created with estimates of 1.97 million acres of damaged corn and 1.40 million acres of damaged soybeans in the state of Iowa. The damage estimates generated by this study were comparable to estimates produced by others after the derecho, including two commercial agricultural companies.
