Search Results

You are looking at 1 - 10 of 22 items for

  • Author or Editor: Amy McGovern x
  • All content x
Clear All Modify Search
Ryan Lagerquist, Amy McGovern, and Travis Smith

Abstract

Thunderstorms in the United States cause over 100 deaths and $10 billion (U.S. dollars) in damage per year, much of which is attributable to straight-line (nontornadic) wind. This paper describes a machine-learning system that forecasts the probability of damaging straight-line wind (≥50 kt or 25.7 m s−1) for each storm cell in the continental United States, at distances up to 10 km outside the storm cell and lead times up to 90 min. Predictors are based on radar scans of the storm cell, storm motion, storm shape, and soundings of the near-storm environment. Verification data come from weather stations and quality-controlled storm reports. The system performs very well on independent testing data. The area under the receiver operating characteristic (ROC) curve ranges from 0.88 to 0.95, the critical success index (CSI) ranges from 0.27 to 0.91, and the Brier skill score (BSS) ranges from 0.19 to 0.65 (>0 is better than climatology). For all three scores, the best value occurs for the smallest distance (inside storm cell) and/or lead time (0–15 min), while the worst value occurs for the greatest distance (5–10 km outside storm cell) and/or lead time (60–90 min). The system was deployed during the 2017 Hazardous Weather Testbed.

Full access
Ryan Lagerquist, John T. Allen, and Amy McGovern

Abstract

This paper describes the development and analysis of an objective climatology of warm and cold fronts over North America from 1979 to 2018. Fronts are detected by a convolutional neural network (CNN), trained to emulate fronts drawn by human meteorologists. Predictors for the CNN are surface and 850-hPa fields of temperature, specific humidity, and vector wind from the ERA5 reanalysis. Gridded probabilities from the CNN are converted to 2D frontal regions, which are used to create the climatology. Overall, warm and cold fronts are most common in the Pacific and Atlantic cyclone tracks and the lee of the Rockies. In contrast with prior research, we find that the activity of warm and cold fronts is significantly modulated by the phase and intensity of El Niño–Southern Oscillation. The influence of El Niño is significant for winter warm fronts, winter cold fronts, and spring cold fronts, with activity decreasing over the continental United States and shifting northward with the Pacific and Atlantic cyclone tracks. Long-term trends are generally not significant, although we find a poleward shift in frontal activity during the winter and spring, consistent with prior research. We also identify a number of regional patterns, such as a significant long-term increase in warm fronts in the eastern tropical Pacific Ocean, which are characterized almost entirely by moisture gradients rather than temperature gradients.

Restricted access
G. Eli Jergensen, Amy McGovern, Ryan Lagerquist, and Travis Smith

Abstract

We demonstrate that machine learning (ML) can skillfully classify thunderstorms into three categories: supercell, part of a quasi-linear convective system, or disorganized. These classifications are based on radar data and environmental information obtained through a proximity sounding. We compare the performance of five ML algorithms: logistic regression with the elastic-net penalty, random forests, gradient-boosted forests, and support-vector machines with both a linear and nonlinear kernel. The gradient-boosted forest performs best, with an accuracy of 0.77 ± 0.02 and a Peirce score of 0.58 ± 0.04. The linear support-vector machine performs second best, with values of 0.70 ± 0.02 and 0.55 ± 0.05, respectively. We use two interpretation methods, permutation importance and sequential forward selection, to determine the most important predictors for the ML models. We also use partial-dependence plots to determine how these predictors influence the outcome. A main conclusion is that shape predictors, based on the outline of the storm, appear to be highly important across ML models. The training data, a storm-centered radar scan and modeled proximity sounding, are similar to real-time data. Thus, the models could be used operationally to aid human decision-making by reducing the cognitive load involved in manual storm-mode identification. Also, they could be run on historical data to perform climatological analyses, which could be valuable to both the research and operational communities.

Open access
Amy McGovern, Christopher D. Karstens, Travis Smith, and Ryan Lagerquist

Abstract

Real-time prediction of storm longevity is a critical challenge for National Weather Service (NWS) forecasters. These predictions can guide forecasters when they issue warnings and implicitly inform them about the potential severity of a storm. This paper presents a machine-learning (ML) system that was used for real-time prediction of storm longevity in the Probabilistic Hazard Information (PHI) tool, making it a Research-to-Operations (R2O) project. Currently, PHI provides forecasters with real-time storm variables and severity predictions from the ProbSevere system, but these predictions do not include storm longevity. We specifically designed our system to be tested in PHI during the 2016 and 2017 Hazardous Weather Testbed (HWT) experiments, which are a quasi-operational naturalistic environment. We considered three ML methods that have proven in prior work to be strong predictors for many weather prediction tasks: elastic nets, random forests, and gradient-boosted regression trees. We present experiments comparing the three ML methods with different types of input data, discuss trade-offs between forecast quality and requirements for real-time deployment, and present both subjective (human-based) and objective evaluation of real-time deployment in the HWT. Results demonstrate that the ML system has lower error than human forecasters, which suggests that it could be used to guide future storm-based warnings, enabling forecasters to focus on other aspects of the warning system.

Open access
Amy McGovern, Andrea Balfour, Marissa Beene, and David Harrison

Abstract

We have developed and released an iPad application, Storm Evader, to demonstrate to youth how technology can be used as a tool and to teach youth about weather and its impact on real-world activities, including flying. As technology becomes more widespread in modern society, many young people overlook the usefulness of technology and have instead come to see it primarily as a provider of entertainment. Storm Evader exposes children to meteorology and aviation in an engaging way. The game requires players to route planes across the United States while avoiding dangerous storms and conserving fuel. The artificial intelligence inside the game suggests routes around a storm. To maximize their scores, players should take the computer’s suggestions into account. Players are exposed to actual radar data and are aided by computer-generated forecast models and storm outlooks from the Storm Prediction Center. In the future, we will include a turbulence model. Exposure to the variety of weather forecasts and their use in the game introduces children to new weather concepts. We also report results of surveys of learning from groups of children playing the game.

Full access
David John Gagne II, Amy McGovern, and Ming Xue

Abstract

Probabilistic quantitative precipitation forecasts challenge meteorologists due to the wide variability of precipitation amounts over small areas and their dependence on conditions at multiple spatial and temporal scales. Ensembles of convection-allowing numerical weather prediction models offer a way to produce improved precipitation forecasts and estimates of the forecast uncertainty. These models allow for the prediction of individual convective storms on the model grid, but they often displace the storms in space, time, and intensity, which results in added uncertainty. Machine learning methods can produce calibrated probabilistic forecasts from the raw ensemble data that correct for systemic biases in the ensemble precipitation forecast and incorporate additional uncertainty information from aggregations of the ensemble members and additional model variables. This study utilizes the 2010 Center for Analysis and Prediction of Storms Storm-Scale Ensemble Forecast system and the National Severe Storms Laboratory National Mosaic & Multi-Sensor Quantitative Precipitation Estimate as input data for training logistic regressions and random forests to produce a calibrated probabilistic quantitative precipitation forecast. The reliability and discrimination of the forecasts are compared through verification statistics and a case study.

Full access
Ryan Lagerquist, Amy McGovern, and David John Gagne II

Abstract

This paper describes the use of convolutional neural nets (CNN), a type of deep learning, to identify fronts in gridded data, followed by a novel postprocessing method that converts probability grids to objects. Synoptic-scale fronts are often associated with extreme weather in the midlatitudes. Predictors are 1000-mb (1 mb = 1 hPa) grids of wind velocity, temperature, specific humidity, wet-bulb potential temperature, and/or geopotential height from the North American Regional Reanalysis. Labels are human-drawn fronts from Weather Prediction Center bulletins. We present two experiments to optimize parameters of the CNN and object conversion. To evaluate our system, we compare the objects (predicted warm and cold fronts) with human-analyzed warm and cold fronts, matching fronts of the same type within a 100- or 250-km neighborhood distance. At 250 km our system obtains a probability of detection of 0.73, success ratio of 0.65 (or false-alarm rate of 0.35), and critical success index of 0.52. These values drastically outperform the baseline, which is a traditional method from numerical frontal analysis. Our system is not intended to replace human meteorologists, but to provide an objective method that can be applied consistently and easily to a large number of cases. Our system could be used, for example, to create climatologies and quantify the spread in forecast frontal properties across members of a numerical weather prediction ensemble.

Full access
David John Gagne II, Amy McGovern, and Jerry Brotzge

Abstract

This paper presents an automated approach for classifying storm type from weather radar reflectivity using decision trees. Recent research indicates a strong relationship between storm type (morphology) and severe weather, and such information can aid in the warning process. Furthermore, new adaptive sensing tools, such as the Center for Collaborative Adaptive Sensing of the Atmosphere’s (CASA’s) weather radar, can make use of storm-type information in real time. Given the volume of weather radar data from those tools, manual classification of storms is not possible when dealing with real-time data streams. An automated system can more quickly and efficiently sort through real-time data streams and return value-added output in a form that can be more easily manipulated and understood. The method of storm classification in this paper combines two machine learning techniques: K-means clustering and decision trees. K-means segments the reflectivity data into clusters, and decision trees classify each cluster. The K means was used to separate isolated cells from linear systems. Each cell received labels such as “isolated pulse,” “isolated strong,” or “multicellular.” Linear systems were labeled as “trailing stratiform,” “leading stratiform,” and “parallel stratiform.” The classification scheme was tested using both simulated and observed storms. The simulated training and test datasets came from the Advanced Regional Prediction System (ARPS) simulated reflectivity data, and observed data were collected from composite reflectivity mosaics from the CASA Integrative Project One (IP1) network. The observations from the CASA network showed that the classification scheme is now ready for operational use.

Full access
Shawn L. Handler, Heather D. Reeves, and Amy McGovern

ABSTRACT

In this study, a machine learning algorithm for generating a gridded CONUS-wide probabilistic road temperature forecast is presented. A random forest is used to tie a combination of HRRR model surface variables and information about the geographic location and time of day per year to observed road temperatures. This approach differs from its predecessors in that road temperature is not deterministic (i.e., provides a forecast of a specific road temperature), but rather it is probabilistic, providing a 0%–100% probability that the road temperature is subfreezing. This approach can account for the varying controls on road temperature that are not easily known or able to be accounted for in physical models, such as amount of traffic, road composition, and differential shading by surrounding buildings and terrain. The algorithm is trained using road temperature observations from one winter season (October 2016–March 2017) and calibrated/evaluated using observations from the following winter season (October 2017–March 2018). Case-study analyses show the algorithm performs well for various scenarios and captures the temporal and spatial evolution of the probability of subfreezing roads reliably. Statistical evaluation for the predicted probabilities shows good skill as the mean area under the receiver operating characteristics curve is 0.96 and the Brier skill score is 0.66 for a 2-h forecast and only degrades slightly as lead time is increased. Additionally, the algorithm produces well-calibrated probabilities, and consistent discrimination between clearly above-freezing and subfreezing environments.

Restricted access
Amanda Burke, Nathan Snook, David John Gagne II, Sarah McCorkle, and Amy McGovern

Abstract

In this study, we use machine learning (ML) to improve hail prediction by postprocessing numerical weather prediction (NWP) data from the new High-Resolution Ensemble Forecast system, version 2 (HREFv2). Multiple operational models and ensembles currently predict hail, however ML models are more computationally efficient and do not require the physical assumptions associated with explicit predictions. Calibrating the ML-based predictions toward familiar forecaster output allows for a combination of higher skill associated with ML models and increased forecaster trust in the output. The observational dataset used to train and verify the random forest model is the Maximum Estimated Size of Hail (MESH), a Multi-Radar Multi-Sensor (MRMS) product. To build trust in the predictions, the ML-based hail predictions are calibrated using isotonic regression. The target datasets for isotonic regression include the local storm reports and Storm Prediction Center (SPC) practically perfect data. Verification of the ML predictions indicates that the probability magnitudes output from the calibrated models closely resemble the day-1 SPC outlook and practically perfect data. The ML model calibrated toward the local storm reports exhibited better or similar skill to the uncalibrated predictions, while decreasing model bias. Increases in reliability and skill after calibration may increase forecaster trust in the automated hail predictions.

Free access