A Deep Learning-based Velocity Dealiasing Algorithm Derived from the WSR-88D Open Radar Product Generator

Radial velocity estimates provided by Doppler weather radar are critical measurements used by operational forecasters for the detection and monitoring of life-impacting storms. The sampling methods used to produce these measurements are inherently susceptible to aliasing, which produces ambiguous velocity values in regions with high winds, and needs to be corrected using a velocity dealiasing algorithm (VDA). In the US, the Weather Surveillance Radar – 1988 Doppler (WSR-88D) Open Radar Product Generator (ORPG) is a processing environment that provides a world-class VDA; however, this algorithm is complex and can be difficult to port to other radar systems outside of the WSR-88D network. In this work, a Deep Neural Network (DNN) is used to emulate the 2-dimensionalWSR-88D ORPG dealiasing algorithm. It is shown that a DNN, specifically a customized U-Net, is highly effective for building VDAs that are accurate, fast, and portable to multiple radar types. To train the DNN model, a large dataset is generated containing aligned samples of folded and dealiased velocity pairs. This dataset contains samples collected from WSR-88D Level-II and Level-III archives, and uses the ORPG dealiasing algorithm output as a source of truth. Using this dataset, a U-Net is trained to produce the number of folds at each point of a velocity image. Several performance metrics are presented using WSR-88D data. The algorithm is also applied to other non-WSR-88D radar systems to demonstrate portability to other hardware/software interfaces. A discussion of the broad applicability of this method is presented, including how other Level-III algorithms may benefit from this approach.

[1]  P. Kollias,et al.  A Primer on Phased Array Radar Technology for the Atmospheric Sciences , 2022, Bulletin of the American Meteorological Society.

[2]  Chunlin Huang,et al.  Spatiotemporal Reconstruction of MODIS Normalized Difference Snow Index Products Using U-Net with Partial Convolutions , 2022, Remote. Sens..

[3]  M. Biggerstaff,et al.  Hurricane Florence (2018): Long duration single‐ and dual‐Doppler observations and wind retrievals during landfall , 2021, Geoscience Data Journal.

[4]  Precious Jatau,et al.  A machine learning approach for classifying bird and insect radar echoes with S-band Polarimetric Weather Radar , 2021, Journal of Atmospheric and Oceanic Technology.

[5]  David Warde,et al.  A Neural-Network Quality Control scheme for improved Quantitative Precipitation Estimation accuracy on the UK weather radar network , 2021 .

[6]  John Y. N. Cho,et al.  Towards the Next Generation Operational Meteorological Radar , 2021, Bulletin of the American Meteorological Society.

[7]  Raia Hadsell,et al.  Skilful precipitation nowcasting using deep generative models of radar , 2021, Nature.

[8]  Pierre Gentine,et al.  PrecipGAN: Merging Microwave and Infrared Data for Satellite Precipitation Estimation Using Generative Adversarial Network , 2021, Geophysical Research Letters.

[9]  Ting‐Shuo Yo,et al.  A Deep Learning Approach to Radar‐Based QPE , 2021, Earth and Space Science.

[10]  Jonathan A. Weyn,et al.  Sub‐Seasonal Forecasting With a Large Ensemble of Deep‐Learning Weather Prediction Models , 2021, Journal of Advances in Modeling Earth Systems.

[11]  Daniel Leuenberger,et al.  R2D2: A Region-Based Recursive Doppler Dealiasing Algorithm for Operational Weather Radar , 2020 .

[12]  Michael Jones,et al.  Compute, Time and Energy Characterization of Encoder-Decoder Networks with Automatic Mixed Precision Training , 2020, 2020 IEEE High Performance Extreme Computing Conference (HPEC).

[13]  Jonathan J. Helmus,et al.  UNRAVEL: A Robust Modular Velocity Dealiasing Technique for Doppler Radar , 2020 .

[14]  Steven D. Miller,et al.  Development and Interpretation of a Neural Network-Based Synthetic Radar Reflectivity Estimator Using GOES-R Satellite Observations , 2020, Journal of Applied Meteorology and Climatology.

[15]  Haonan Chen,et al.  A Machine Learning System for Precipitation Estimation Using Satellite and Ground Radar Network Observations , 2020, IEEE Transactions on Geoscience and Remote Sensing.

[16]  M. Biggerstaff,et al.  Mobile ground‐based SMART radar observations and wind retrievals during the landfall of Hurricane Harvey (2017) , 2019, Geoscience Data Journal.

[17]  Ji Yang,et al.  A Bayesian Hydrometeor Classification Algorithm for C-Band Polarimetric Radar , 2019, Remote. Sens..

[18]  Mark S. Veillette,et al.  Creating Synthetic Radar Imagery Using Convolutional Neural Networks , 2018, Journal of Atmospheric and Oceanic Technology.

[19]  S. Gulev,et al.  Deep Convolutional Neural Networks Capabilities for Binary Classification of Polar Mesocyclones in Satellite Mosaics , 2018, Atmosphere.

[20]  Timothy J. Lang,et al.  It's Time for Color Vision Deficiency Friendly Color Maps in the Radar Community , 2018 .

[21]  Alexey Shvets,et al.  TernausNet: U-Net with VGG11 Encoder Pre-Trained on ImageNet for Image Segmentation , 2018, Computer-Aided Analysis of Gastrointestinal Videos.

[22]  James M. Kurdzo,et al.  WSR-88D Chaff Detection and Characterization using an Optimized Hydrometeor Classification Algorithm , 2017 .

[23]  Prabhat,et al.  ExtremeWeather: A large-scale climate dataset for semi-supervised detection, localization, and understanding of extreme weather events , 2016, NIPS.

[24]  Alexei A. Efros,et al.  Image-to-Image Translation with Conditional Adversarial Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[25]  Jonathan J. Helmus,et al.  The Python ARM Radar Toolkit (Py-ART), a Library for Working with Weather Radar Data in the Python Programming Language , 2016 .

[26]  A. Ryzhkov,et al.  Polarimetric Radar Characteristics of Melting Hail. Part III: Validation of the Algorithm for Hail Size Discrimination , 2016 .

[27]  Andrew D. Byrd,et al.  Observations of Severe Local Storms and Tornadoes with the Atmospheric Imaging Radar , 2015 .

[28]  Dit-Yan Yeung,et al.  Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting , 2015, NIPS.

[29]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[30]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[31]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[32]  Roger M. Wakimoto,et al.  An Integrated Damage, Visual, and Radar Analysis of the 2013 Moore, Oklahoma EF5 Tornado , 2014 .

[33]  J. B. Mead,et al.  A Mobile Rapid-Scanning X-band Polarimetric (RaXPol) Doppler Radar System , 2013 .

[34]  Mark B. Yeary,et al.  PX-1000: A Solid-State Polarimetric X-Band Weather Radar and Time–Frequency Multiplexed Waveform for Blind Range Mitigation , 2013, IEEE Transactions on Instrumentation and Measurement.

[35]  X. Zou,et al.  A Velocity Dealiasing Scheme for Synthetic C-Band Data from China's New Generation Weather Radar System (CINRAD) , 2012 .

[36]  John Y. N. Cho,et al.  Terminal Doppler Weather Radar enhancements , 2010, 2010 IEEE Radar Conference.

[37]  J. Y. N. Cho,et al.  Signal Processing Algorithms for the Terminal Doppler Weather Radar: Build 2 , 2010 .

[38]  D. Burgess.  Observed failure modes of the WSR-88D velocity dealiasing algorithm during severe weather outbreaks , 2009 .

[39]  Arthur Witt,et al.  Performance of a new velocity dealiasing algorithm for the WSR-88D , 2009 .

[40]  Stanley G. Benjamin,et al.  CONVECTIVE-SCALE WARN-ON-FORECAST SYSTEM: A vision for 2020 , 2009 .

[41]  Alexander V. Ryzhkov,et al.  The Hydrometeor Classification Algorithm for the Polarimetric WSR-88D: Description and Application to an MCS , 2009 .

[42]  Yadong Wang,et al.  Characterization of Tornado Spectral Signatures Using Higher-Order Spectra , 2007 .

[43]  Francesc Junyent,et al.  Close-Range Observations of Tornadoes in Supercells Made with a Dual-Polarization, X-Band, Mobile Doppler Radar , 2007 .

[44]  Jian Zhang,et al.  An Automated 2D Multipass Doppler Radar Velocity Dealiasing Scheme , 2006 .

[45]  David A. R. Kristovich,et al.  Convective evolution across Lake Michigan during a widespread lake-effect snow event , 2003 .

[46]  Donald W. Burgess,et al.  Radar Observations of the 3 May 1999 Oklahoma City Tornado , 2002 .

[47]  Robert A. Houze,et al.  A Real-Time Four-Dimensional Doppler Dealiasing Scheme , 2001 .

[48]  Arthur Witt,et al.  The National Severe Storms Laboratory Tornado Detection Algorithm , 1998 .

[49]  Donald W. Burgess,et al.  The National Severe Storms Laboratory Mesocyclone Detection Algorithm for the WSR-88D* , 1998 .

[50]  Erik N. Rasmussen,et al.  Fine-Scale Doppler Radar Observations of Tornadoes , 1996, Science.

[51]  Gerry Wiener,et al.  Two-Dimensional Dealiasing of Doppler Velocities , 1993 .

[52]  Michael D. Eilts,et al.  Efficient Dealiasing of Doppler Velocities Using Local Environment Constraints , 1990 .

[53]  J. V. Evans,et al.  Development of an automated windshear detection system using Doppler weather radar , 1989, Proc. IEEE.

[54]  D. Zrnic,et al.  Doppler Radar and Weather Observations , 1984 .

[55]  M. Skolnik,et al.  Introduction to Radar Systems , 2021, Advances in Adaptive Radar Detection and Range Estimation.

[56]  Conrad L. Ziegler,et al.  De-Aliasing First-Moment Doppler Estimates , 1977 .

[57]  Xuan Peng,et al.  CNGAT: A Graph Neural Network Model for Radar Quantitative Precipitation Estimation , 2022, IEEE Transactions on Geoscience and Remote Sensing.

[58]  Jeremiah Neubert,et al.  Deep learning approaches to biomedical image segmentation , 2020 .

[59]  Ekaba Bisong,et al.  TensorFlow 2.0 and Keras , 2019, Building Machine Learning and Deep Learning Models on Google Cloud Platform.

[60]  Jeremy Kepner,et al.  Llgrid: Supercomputer for sensor processing , 2013 .

[61]  XU QIN,et al.  A VAD-Based Dealiasing Method for Radar Velocity Data Quality Control , 2011 .

[62]  D. Burgess,et al.  Severe thunderstorm detection by radar , 1990 .

[63]  D. Burgess,et al.  Tornado Detection by Pulsed Doppler Radar , 1978 .