skip to main content


Title: Airborne Radar Quality Control with Machine Learning
Abstract

Airborne Doppler radar provides detailed and targeted observations of winds and precipitation in weather systems over remote or difficult-to-access regions that can help to improve scientific understanding and weather forecasts. Quality control (QC) is necessary to remove nonweather echoes from raw radar data for subsequent analysis. The complex decision-making ability of the machine learning random-forest technique is employed to create a generalized QC method for airborne radar data in convective weather systems. A manually QCed dataset was used to train the model containing data from the Electra Doppler Radar (ELDORA) in mature and developing tropical cyclones, a tornadic supercell, and a bow echo. Successful classification of ∼96% and ∼93% of weather and nonweather radar gates, respectively, in withheld testing data indicate the generalizability of the method. Dual-Doppler analysis from the genesis phase of Hurricane Ophelia (2005) using data not previously seen by the model produced a comparable wind field to that from manual QC. The framework demonstrates a proof of concept that can be applied to newer airborne Doppler radars.

Significance Statement

Airborne Doppler radar is an invaluable tool for making detailed measurements of wind and precipitation in weather systems over remote or difficult to access regions, such as hurricanes over the ocean. Using the collected radar data depends strongly on quality control (QC) procedures to classify weather and nonweather radar echoes and to then remove the latter before subsequent analysis or assimilation into numerical weather prediction models. Prior QC techniques require interactive editing and subjective classification by trained researchers and can demand considerable time for even small amounts of data. We present a new machine learning algorithm that is trained on past QC efforts from radar experts, resulting in an accurate, fast technique with far less user input required that can greatly reduce the time required for QC. The new technique is based on the random forest, which is a machine learning model composed of decision trees, to classify weather and nonweather radar echoes. Continued efforts to build on this technique could benefit future weather forecasts by quickly and accurately quality-controlling data from other airborne radars for research or operational meteorology.

 
more » « less
NSF-PAR ID:
10488364
Author(s) / Creator(s):
 ;  
Publisher / Repository:
American Meteorological Society
Date Published:
Journal Name:
Artificial Intelligence for the Earth Systems
Volume:
3
Issue:
1
ISSN:
2769-7525
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Abstract

    High-resolution airborne cloud Doppler radars such as the W-band Wyoming Cloud Radar (WCR) have, since the 1990s, investigated cloud microphysical, kinematic, and precipitation structures down to 30-m resolution. These measurements revolutionized our understanding of fine-scale cloud structure and the scales at which cloud processes occur. Airborne cloud Doppler radars may also resolve cloud turbulent eddy structure directly at 10-m scales. To date, cloud turbulence has been examined as variances and dissipation rates at coarser resolution than individual pulse volumes. The present work advances the potential of near-vertical pulse-pair Doppler spectrum width as a metric for turbulent air motion. Doppler spectrum width has long been used to investigate turbulent motions from ground-based remote sensors. However, complexities of airborne Doppler radar and spectral broadening resulting from platform and hydrometeor motions have limited airborne radar spectrum width measurements to qualitative interpretation only. Here we present the first quantitative validation of spectrum width from an airborne cloud radar. Echoes with signal-to-noise ratio greater than 10 dB yield spectrum width values that strongly correlate with retrieved mean Doppler variance for a range of nonconvective cloud conditions. Further, Doppler spectrum width within turbulent regions of cloud also shows good agreement with in situ eddy dissipation rate (EDR) and gust probe variance. However, the use of pulse-pair estimated spectrum width as a metric for turbulent air motion intensity is only suitable for turbulent air motions more energetic than the magnitude of spectral broadening, estimated to be <0.4 m s−1for the WCR in these cases.

    Significance Statement

    Doppler spectrum width is a widely available airborne radar measurement previously considered too uncertain to attribute to atmospheric turbulence. We validate, for the first time, the response of spectrum width to turbulence at and away from research aircraft flight level and demonstrate that under certain conditions, spectrum width can be used to diagnose atmospheric turbulence down to scales of tens of meters. These high-resolution turbulent air motion intensity measurements may better connect to cloud hydrometeor process and growth response seen in coincident radar reflectivity structures proximate to turbulent eddies.

     
    more » « less
  2. Missing or low-quality data regions usually happen to weather radars. One of the most common situations is beam blockage or partial beam blockage. Therefore, correction of weather radar observations that are partially or fully blocked is an indispensable step in radar data quality control and subsequent quantitative applications, especially in complex terrain environments such as the western United States. In this article, we propose a deep learning framework based on generative adversarial networks (GANs) for restoring partial beam blockage regions in polarimetric radar observations using local and global contextual information. Due to the diverse precipitation types, blockage conditions, and ground information in different areas, two radars deployed in two different regions characterized by different precipitation types are used to demonstrate the proposed methodology. Both are S-band operational Weather Surveillance Radar – 1988 Doppler (WSR-88D): KFWS located in Fort Worth, northern Texas, and KDAX located in Davis, northern California. For training the GAN model, this article simulates the partial beam blockage situations by manually cropping observation sectors of both KDAX and KFWS radar data. The trained models were tested using independent precipitation events in Texas and California to demonstrate the model effectiveness in inpainting “missing” data. In addition, this paper cross-tested the data with different precipitation features to examine the generalization capacity of the beam blockage correction models. The beam blockage correction performance is also compared with a traditional linear interpolation approach. The results show that for both domains the continuity of precipitation observations is greatly improved after applying the deep learning based inpainting approach. For the KFWS test data, some visible discrepancies exist between the results from models trained based on convective and stratiform precipitation events in Texas and California, respectively, yet both models outperform the traditional interpolation method. For the KDAX test data, both the model trained using the KFWS data from convective precipitation events in Texas and the model trained using KDAX data from stratiform precipitation events in California render a similar performance. Although ground truth is not available for the real blocked radar data, the repaired observations demonstrated a great potential for improved quantitative applications. 
    more » « less
  3. The Super Dual Auroral Radar Network (SuperDARN) is an international network of ground-based, space weather radars which have operated continuously in the Arctic and Antarctic regions for more than 30 years. These high-frequency (HF) radars use over-the-horizon (OTH) radio wave propagation to detect ionospheric plasma structures across ranges of several thousand kilometers (km). As a byproduct of this technique, the transmitted radar signals frequently reflect from the Earth's surface and can be observed as ground backscatter echoes. The monthly files in this dataset contain maps of daily ground backscatter observations from the Kapuskasing (KAP) SuperDARN HF radar binned onto an equal-area 24 km grid. The KAP radar is located in Ontario, Canada (49.39°N, 82.32°W) and is operated by Virginia Tech (Principal Investigator: J. Michael Ruohoniemi, mikeruo@vt.edu) with funding support from the National Science Foundation. 
    more » « less
  4. The Super Dual Auroral Radar Network (SuperDARN) is an international network of ground-based, space weather radars which have operated continuously in the Arctic and Antarctic regions for more than 30 years. These high-frequency (HF) radars use over-the-horizon (OTH) radio wave propagation to detect ionospheric plasma structures across ranges of several thousand kilometers (km). As a byproduct of this technique, the transmitted radar signals frequently reflect from the Earth's surface and can be observed as ground backscatter echoes. The monthly files in this dataset contain maps of daily ground backscatter observations from the Goose Bay (GBR) SuperDARN HF radar binned onto an equal-area 24 km grid. The GBR radar is located in Labrador, Canada (53.32°N, 60.46°W) and is operated by Virginia Tech (Principal Investigator: J. Michael Ruohoniemi, mikeruo@vt.edu) with funding support from the National Science Foundation. 
    more » « less
  5. The Super Dual Auroral Radar Network (SuperDARN) is an international network of ground-based, space weather radars which have operated continuously in the Arctic and Antarctic regions for more than 30 years. These high-frequency (HF) radars use over-the-horizon (OTH) radio wave propagation to detect ionospheric plasma structures across ranges of several thousand kilometers (km). As a byproduct of this technique, the transmitted radar signals frequently reflect from the Earth's surface and can be observed as ground backscatter echoes. The monthly files in this dataset contain maps of daily ground backscatter observations from the Iceland East (ICE) SuperDARN HF radar binned onto an equal-area 24 km grid. The ICE radar is located in Þykkvibær, Iceland (63.77°N (North), 20.54°W (West)) and is operated by Dartmouth College (Principal Investigator: Simon G. Shepherd, simon.g.shepherd@dartmouth.edu) with funding support from the National Science Foundation. 
    more » « less