skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


Title: A Machine‐Learning Approach to Classify Cloud‐to‐Ground and Intracloud Lightning
Abstract To know if a lightning discharge reaches the ground or remains within the thundercloud is critical for lightning safety as cloud‐to‐ground lightning poses the greatest threat to life and property. The current classification methods for most lightning detection networks, which are based on the classification of electromagnetic pulses produced by lightning, still have plenty of room to improve, including some known issues to be addressed. We present a machine‐learning approach to classify lightning discharges. The classification model used in this study is based on Support Vector Machines (SVMs). Compared with traditional multiparameter methods, our algorithm does not require extraction of individual pulse parameters and additionally provides a probability for each prediction. Using a representative lightning pulse data collected by the Cordoba Marx Meter Array in Argentina, we found the classification accuracy of our machine‐learning algorithm to be 97%, which is higher than that for the existing lightning detection networks.  more » « less
Award ID(s):
1654576
PAR ID:
10452946
Author(s) / Creator(s):
 ;  ;  ;  
Publisher / Repository:
DOI PREFIX: 10.1029
Date Published:
Journal Name:
Geophysical Research Letters
Volume:
48
Issue:
1
ISSN:
0094-8276
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; (, Earth and Space Science)
    Abstract The lightning data products generated by the low‐frequency (LF) radio lightning locating system (LLS) deployed during the Remote sensing of Electrification, Lightning, and Mesoscale/Microscale Processes with Adaptive Ground Observation (RELAMPAGO) field campaign in Argentina provide a valuable data set to research the lightning evolution and characteristics of convective storms that produce high‐impact weather. LF LLS data sets offer a practical range for mesoscale studies, allowing for the observation of lightning characteristics of storms such as mesoscale convective systems or large convective lines that travel longer distances which are not necessarily staying in range of regional VHF‐based lightning detection systems throughout their lifetime. LF LLSs also provide different information than optical space‐borne lightning detectors. Lightning measurements exclusive to LF systems include discharge peak current, lightning polarity, and lightning type classification based on the lightning‐emitted radio waveform. Furthermore, these measurements can provide additional information on flash rates (e.g., positive cloud‐to‐ground flash rate) or narrow bipolar events which may often be associated with dynamically intense convection. In this article, the geolocation and data processing of the LF data set collected during RELAMPAGO is fully described and its performance characterized, with location accuracy better than 10 km. The detection efficiency (DE) of the data set is compared to that of the Geostationary Lightning Mapper, and spatiotemporal DE losses in the LF data set are discussed. Storm case studies on November 10, 2018, highlight the strengths of the data set, which include robust flash clustering and insightful flash rate and peak current measures, while illustrating how its limitations, including DE losses, can be managed. 
    more » « less
  2. ; ; ; ; ; ; (, American Geophysical Union)
    Gigantic jets are a type of transient luminous event (TLE, Pasko 2010, doi: 10.1029/2009JA014860) that escape the cloud top of a thunderstorm and propagate up to the lower ionosphere (80-100 km altitude), transferring tens to hundreds of Coulombs of charge. Due to rarity of observations, it is still not understood how they affect the lower ionosphere and what storm systems produce them. In this presentation we will provide an overview and present preliminary results from a multi-institutional collaborative project, which seeks to detect gigantic jets over hemispheric scales using a combination of orbital and ground-based sensors and machine learning. Our pipeline has the potential to detect significantly more gigantic jets (thousands) than current methods, which relies on using ground-based cameras. We will build a large database of gigantic jet detections, and correlate the events with a Very Low Frequency (VLF) remote sensing network (Cohen et al. 2009, doi: 10.1109/TGRS.2009.2028334) to understand how they perturb the lower ionosphere – in addition to other meteorological datasets. Our detection methodology primarily uses the Geostationary Lightning Mapper (GLM), which is a staring optical imager in geostationary orbit that detects the 777.4 nm (OI) triplet commonly emitted by lightning (Goodman et al. 2013, doi: 10.1016/j.atmosres.2013.01.006). Gigantic jets have been shown to have unique signatures in the GLM data from past studies (Boggs et al. 2019, doi: 10.1029/2019GL082278; Boggs et al. 2022, doi: 10.1126/sciadv.abl8731). Thus far, we have built a preliminary, supervised machine learning model that detects potential gigantic jets using GLM, and begun development on a series of vetting techniques to confirm the detections as real gigantic jets. The vetting techniques use a combination of low frequency (LF) and extremely low frequency (ELF) sferic data, in combination with stereo GLM measurements that provide optical source altitude. In addition, we will soon be able to calculate optical stereo sources with GLM on GOES-16 and the newly launched Lightning Imager on EUMETSAT, significantly expanding the stereo region of detection. When our detection methodology grows in maturity, we will deploy it to all past GLM data (2018-present) and share the database publicly, allowing other researchers to use this data for their own research. 
    more » « less
  3. ; ; ; ; (, Journal on Multimodal User Interfaces)
    Automatic detection of an individual’s mind-wandering state has implications for designing and evaluating engaging and effective learning interfaces. While it is difficult to differentiate whether an individual is mind-wandering or focusing on the task only based on externally observable behavior, brain-based sensing offers unique insights to internal states. To explore the feasibility, we conducted a study using functional near-infrared spectroscopy (fNIRS) and investigated machine learning classifiers to detect mind-wandering episodes based on fNIRS data, both on an individual level and a group level, specifically focusing on automated window selection to improve classification results. For individual-level classification, by using a moving window method combined with a linear discriminant classifier, we found the best windows for classification and achieved a mean F1-score of 74.8%. For group-level classification, we proposed an individual-based time window selection (ITWS) algorithm to incorporate individual differences in window selection. The algorithm first finds the best window for each individual by using embedded individual-level classifiers and then uses these windows from all participants to build the final classifier. The performance of the ITWS algorithm is evaluated when used with eXtreme gradient boosting, convolutional neural networks, and deep neural networks. Our results show that the proposed algorithm achieved significant improvement compared to the previous state of the art in terms of brain-based classification of mind-wandering, with an average F1-score of 73.2%. This builds a foundation for mind-wandering detection for both the evaluation of multimodal learning interfaces and for future attention-aware systems. 
    more » « less
  4. ; ; ; (, Weather and Forecasting)
    Abstract Over the past decade the use of machine learning in meteorology has grown rapidly. Specifically neural networks and deep learning have been used at an unprecedented rate. To fill the dearth of resources covering neural networks with a meteorological lens, this paper discusses machine learning methods in a plain language format that is targeted to the operational meteorological community. This is the second paper in a pair that aim to serve as a machine learning resource for meteorologists. While the first paper focused on traditional machine learning methods (e.g., random forest), here a broad spectrum of neural networks and deep learning methods is discussed. Specifically, this paper covers perceptrons, artificial neural networks, convolutional neural networks, and U-networks. Like the Part I paper, this manuscript discusses the terms associated with neural networks and their training. Then the manuscript provides some intuition behind every method and concludes by showing each method used in a meteorological example of diagnosing thunderstorms from satellite images (e.g., lightning flashes). This paper is accompanied with an open-source code repository to allow readers to explore neural networks using either the dataset provided (which is used in the paper) or as a template for alternate datasets. 
    more » « less
  5. ; ; ; ; ; (, Journal of Geophysical Research: Atmospheres)
    Abstract This work examines a severe weather event that took place over central Argentina on 11 December 2018. The evolution of the storm from its initiation, rapid organization into a supercell, and eventual decay was analyzed with high‐temporal resolution observations. This work provides insight into the spatio‐temporal co‐evolution of storm kinematics (updraft area and lifespan), cloud‐top cooling rates, and lightning production that led to severe weather. The analyzed storm presented two convective periods with associated severe weather. An overall decrease in cloud‐top local minima IR brightness temperature (MinIR) and lightning jump (LJ) preceded both periods. LJs provided the highest lead time to the occurrence of severe weather, with the ground‐based lightning networks providing the maximum warning time of around 30 min. Lightning flash counts from the Geostationary Lightning Mapper (GLM) were underestimated when compared to detections from ground‐based lightning networks. Among the possible reasons for GLM's lower detection efficiency were an optically dense medium located above lightning sources and the occurrence of flashes smaller than GLM's footprint. The minimum MinIR provided the shorter warning time to severe weather occurrence. However, the secondary minima in MinIR that preceded the absolute minima improved this warning time by more than 10 min. Trends in MinIR for time scales shorter than 6 min revealed shorter cycles of fast cooling and warming, which provided information about the lifecycle of updrafts within the storm. The advantages of using observations with high‐temporal resolution to analyze the evolution and intensity of convective storms are discussed. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email