Articles | Volume 22, issue 2
https://doi.org/10.5194/nhess-22-577-2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/nhess-22-577-2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Nowcasting thunderstorm hazards using machine learning: the impact of data sources on performance
Jussi Leinonen
CORRESPONDING AUTHOR
Federal Office of Meteorology and Climatology MeteoSwiss, Locarno-Monti, Switzerland
Ulrich Hamann
Federal Office of Meteorology and Climatology MeteoSwiss, Locarno-Monti, Switzerland
Urs Germann
Federal Office of Meteorology and Climatology MeteoSwiss, Locarno-Monti, Switzerland
John R. Mecikalski
Atmospheric Science Department, University of Alabama in Huntsville, Huntsville, Alabama, USA
Related authors
Nathalie Rombeek, Jussi Leinonen, and Ulrich Hamann
Nat. Hazards Earth Syst. Sci., 24, 133–144, https://doi.org/10.5194/nhess-24-133-2024, https://doi.org/10.5194/nhess-24-133-2024, 2024
Short summary
Short summary
Severe weather such as hail, lightning, and heavy rainfall can be hazardous to humans and property. Dual-polarization weather radars provide crucial information to forecast these events by detecting precipitation types. This study analyses the importance of dual-polarization data for predicting severe weather for 60 min using an existing deep learning model. The results indicate that including these variables improves the accuracy of predicting heavy rainfall and lightning.
Jussi Leinonen, Jacopo Grazioli, and Alexis Berne
Atmos. Meas. Tech., 14, 6851–6866, https://doi.org/10.5194/amt-14-6851-2021, https://doi.org/10.5194/amt-14-6851-2021, 2021
Short summary
Short summary
Measuring the shape, size and mass of a large number of snowflakes is a challenging task; it is hard to achieve in an automatic and instrumented manner. We present a method to retrieve these properties of individual snowflakes using as input a triplet of images/pictures automatically collected by a multi-angle snowflake camera (MASC) instrument. Our method, based on machine learning, is trained on artificially generated snowflakes and evaluated on 3D-printed snowflake replicas.
Corey G. Amiot, Timothy J. Lang, Susan C. van den Heever, Richard A. Ferrare, Ousmane O. Sy, Lawrence D. Carey, Sundar A. Christopher, John R. Mecikalski, Sean W. Freeman, George Alexander Sokolowsky, Chris A. Hostetler, and Simone Tanelli
Atmos. Chem. Phys., 25, 12335–12355, https://doi.org/10.5194/acp-25-12335-2025, https://doi.org/10.5194/acp-25-12335-2025, 2025
Short summary
Short summary
Decoupling aerosol and environmental impacts on convection is challenging. Using airborne data, we correlated convective metrics with aerosol concentrations in several different environments. Results were mixed, but some comparisons suggest that medium-to-high aerosol concentrations were occasionally strongly correlated with convective intensity and prevalence, especially when the atmosphere was relatively unstable. It is important to consider storm environment when evaluating aerosol effects.
Frédéric G. Jordan, Clément Cosson, Marco Gabella, Ioannis V. Sideris, Adrien Liernur, Alexis Berne, and Urs Germann
Abstr. Int. Cartogr. Assoc., 9, 19, https://doi.org/10.5194/ica-abs-9-19-2025, https://doi.org/10.5194/ica-abs-9-19-2025, 2025
Alfonso Ferrone, Jérôme Kopp, Martin Lainer, Marco Gabella, Urs Germann, and Alexis Berne
Atmos. Meas. Tech., 17, 7143–7168, https://doi.org/10.5194/amt-17-7143-2024, https://doi.org/10.5194/amt-17-7143-2024, 2024
Short summary
Short summary
Estimates of hail size have been collected by a network of hail sensors, installed in three regions of Switzerland, since September 2018. In this study, we use a technique called “double-moment normalization” to model the distribution of diameter sizes. The parameters of the method have been defined over 70 % of the dataset and tested over the remaining 30 %. An independent distribution of hail sizes, collected by a drone, has also been used to evaluate the method.
Jérôme Kopp, Alessandro Hering, Urs Germann, and Olivia Martius
Atmos. Meas. Tech., 17, 4529–4552, https://doi.org/10.5194/amt-17-4529-2024, https://doi.org/10.5194/amt-17-4529-2024, 2024
Short summary
Short summary
We present a verification of two products based on weather radars to detect the presence of hail and estimate its size. Radar products are remote detection of hail, so they must be verified against ground-based observations. We use reports from users of the Swiss Weather Services phone app to do the verification. We found that the product estimating the presence of hail provides fair results but that it should be recalibrated and that estimating the hail size with radar is more challenging.
Loris Foresti, Bernat Puigdomènech Treserras, Daniele Nerini, Aitor Atencia, Marco Gabella, Ioannis V. Sideris, Urs Germann, and Isztar Zawadzki
Nonlin. Processes Geophys., 31, 259–286, https://doi.org/10.5194/npg-31-259-2024, https://doi.org/10.5194/npg-31-259-2024, 2024
Short summary
Short summary
We compared two ways of defining the phase space of low-dimensional attractors describing the evolution of radar precipitation fields. The first defines the phase space by the domain-scale statistics of precipitation fields, such as their mean, spatial and temporal correlations. The second uses principal component analysis to account for the spatial distribution of precipitation. To represent different climates, radar archives over the United States and the Swiss Alpine region were used.
Martin Lainer, Killian P. Brennan, Alessandro Hering, Jérôme Kopp, Samuel Monhart, Daniel Wolfensberger, and Urs Germann
Atmos. Meas. Tech., 17, 2539–2557, https://doi.org/10.5194/amt-17-2539-2024, https://doi.org/10.5194/amt-17-2539-2024, 2024
Short summary
Short summary
This study uses deep learning (the Mask R-CNN model) on drone-based photogrammetric data of hail on the ground to estimate hail size distributions (HSDs). Traditional hail sensors' limited areas complicate the full HSD retrieval. The HSD of a supercell event on 20 June 2021 is retrieved and contains > 18 000 hailstones. The HSD is compared to automatic hail sensor measurements and those of weather-radar-based MESHS. Investigations into ground hail melting are performed by five drone flights.
Nathalie Rombeek, Jussi Leinonen, and Ulrich Hamann
Nat. Hazards Earth Syst. Sci., 24, 133–144, https://doi.org/10.5194/nhess-24-133-2024, https://doi.org/10.5194/nhess-24-133-2024, 2024
Short summary
Short summary
Severe weather such as hail, lightning, and heavy rainfall can be hazardous to humans and property. Dual-polarization weather radars provide crucial information to forecast these events by detecting precipitation types. This study analyses the importance of dual-polarization data for predicting severe weather for 60 min using an existing deep learning model. The results indicate that including these variables improves the accuracy of predicting heavy rainfall and lightning.
Jérôme Kopp, Agostino Manzato, Alessandro Hering, Urs Germann, and Olivia Martius
Atmos. Meas. Tech., 16, 3487–3503, https://doi.org/10.5194/amt-16-3487-2023, https://doi.org/10.5194/amt-16-3487-2023, 2023
Short summary
Short summary
We present the first study of extended field observations made by a network of 80 automatic hail sensors from Switzerland. The sensors record the exact timing of hailstone impacts, providing valuable information about the local duration of hailfall. We found that the majority of hailfalls lasts just a few minutes and that most hailstones, including the largest, fall during a first phase of high hailstone density, while a few remaining and smaller hailstones fall in a second low-density phase.
Monika Feldmann, Urs Germann, Marco Gabella, and Alexis Berne
Weather Clim. Dynam., 2, 1225–1244, https://doi.org/10.5194/wcd-2-1225-2021, https://doi.org/10.5194/wcd-2-1225-2021, 2021
Short summary
Short summary
Mesocyclones are the rotating updraught of supercell thunderstorms that present a particularly hazardous subset of thunderstorms. A first-time characterisation of the spatiotemporal occurrence of mesocyclones in the Alpine region is presented, using 5 years of Swiss operational radar data. We investigate parallels to hailstorms, particularly the influence of large-scale flow, daily cycles and terrain. Improving understanding of mesocyclones is valuable for risk assessment and warning purposes.
Hélène Barras, Olivia Martius, Luca Nisi, Katharina Schroeer, Alessandro Hering, and Urs Germann
Weather Clim. Dynam., 2, 1167–1185, https://doi.org/10.5194/wcd-2-1167-2021, https://doi.org/10.5194/wcd-2-1167-2021, 2021
Short summary
Short summary
In Switzerland hail may occur several days in a row. Such multi-day hail events may cause significant damage, and understanding and forecasting these events is important. Using reanalysis data we show that weather systems over Europe move slower before and during multi-day hail events compared to single hail days. Surface temperatures are typically warmer and the air more humid over Switzerland and winds are slower on multi-day hail clusters. These results may be used for hail forecasting.
Jussi Leinonen, Jacopo Grazioli, and Alexis Berne
Atmos. Meas. Tech., 14, 6851–6866, https://doi.org/10.5194/amt-14-6851-2021, https://doi.org/10.5194/amt-14-6851-2021, 2021
Short summary
Short summary
Measuring the shape, size and mass of a large number of snowflakes is a challenging task; it is hard to achieve in an automatic and instrumented manner. We present a method to retrieve these properties of individual snowflakes using as input a triplet of images/pictures automatically collected by a multi-angle snowflake camera (MASC) instrument. Our method, based on machine learning, is trained on artificially generated snowflakes and evaluated on 3D-printed snowflake replicas.
Daniel Wolfensberger, Marco Gabella, Marco Boscacci, Urs Germann, and Alexis Berne
Atmos. Meas. Tech., 14, 3169–3193, https://doi.org/10.5194/amt-14-3169-2021, https://doi.org/10.5194/amt-14-3169-2021, 2021
Short summary
Short summary
In this work, we present a novel quantitative precipitation estimation method for Switzerland that uses random forests, an ensemble-based machine learning technique. The estimator has been trained with a database of 4 years of ground and radar observations. The results of an in-depth evaluation indicate that, compared with the more classical method in use at MeteoSwiss, this novel estimator is able to reduce both the average error and bias of the predictions.
Maxi Boettcher, Andreas Schäfler, Michael Sprenger, Harald Sodemann, Stefan Kaufmann, Christiane Voigt, Hans Schlager, Donato Summa, Paolo Di Girolamo, Daniele Nerini, Urs Germann, and Heini Wernli
Atmos. Chem. Phys., 21, 5477–5498, https://doi.org/10.5194/acp-21-5477-2021, https://doi.org/10.5194/acp-21-5477-2021, 2021
Short summary
Short summary
Warm conveyor belts (WCBs) are important airstreams in extratropical cyclones, often leading to the formation of intense precipitation. We present a case study that involves aircraft, lidar and radar observations of water and clouds in a WCB ascending from western Europe across the Alps towards the Baltic Sea during the field campaigns HyMeX and T-NAWDEX-Falcon in October 2012. A probabilistic trajectory measure and an airborne tracer experiment were used to confirm the long pathway of the WCB.
Cited articles
Abrams, M., Crippen, R., and Fujisada, H.: ASTER Global Digital Elevation
Model (GDEM) and ASTER Global Water Body Dataset (ASTWBD), Remote Sens., 12,
1156, https://doi.org/10.3390/rs12071156, 2020. a
Autonès, F. and Claudon, M.: Algorithm Theoretical Basis Document for the
Convection Product Processors of the NWC/GEO, Tech. Rep.
SAF/NWC/CDOP/MFT/SCI/ATBD/11, Meteo-France, Toulouse,
https://www.nwcsaf.org/Downloads/GEO/2018.1/Documents/Scientific_Docs/NWC-CDOP2-GEO-MFT-SCI-ATBD-Convection_v2.2.pdf
(last access: 21 February 2022), 2012. a
Ayzel, G., Scheffer, T., and Heistermann, M.: RainNet v1.0: a convolutional
neural network for radar-based precipitation nowcasting, Geosci. Model Dev.,
13, 2631–2644, https://doi.org/10.5194/gmd-13-2631-2020, 2020. a
Barras, H., Hering, A., Martynov, A., Noti, P.-A., Germann, U., and Martius,
O.: Experiences with Crowdsourced Hail Reports in Switzerland, B.
Am. Meteorol. Soc., 100, 1429–1440, https://doi.org/10.1175/BAMS-D-18-0090.1, 2019. a
Bedka, K., Murillo, E. M., Homeyer, C. R., Scarino, B., and Mersiovsky, H.: The Above-Anvil Cirrus Plume: An Important Severe Weather Indicator in Visible and Infrared Satellite Imagery, Weather Forecast., 33, 1159–1181,
https://doi.org/10.1175/WAF-D-18-0040.1, 2018. a
Bedka, K. M. and Khlopenkov, K.: A Probabilistic Multispectral Pattern
Recognition Method for Detection of Overshooting Cloud Tops Using Passive
Satellite Imager Observations, J. Appl. Meteorol. Clim., 55, 1983–2005,
https://doi.org/10.1175/JAMC-D-15-0249.1, 2016. a
Chai, T. and Draxler, R. R.: Root mean square error (RMSE) or mean absolute
error (MAE)? – Arguments against avoiding RMSE in the literature, Geosci. Model Dev., 7, 1247–1250, https://doi.org/10.5194/gmd-7-1247-2014, 2014. a
Changnon, S. A.: Relationships between Thunderstorms and Cloud-to-Ground
Lightning in the United States, J. Appl. Meteorol., 32, 88–105,
https://doi.org/10.1175/1520-0450(1993)032<0088:RBTACT>2.0.CO;2, 1993. a
Czernecki, B., Taszarek, M., Marosz, M., Półrolniczak, M., Kolendowicz, L.,
Wyszogrodzki, A., and Szturc, J.: Application of machine learning to large
hail prediction – The importance of radar reflectivity, lightning occurrence
and convective parameters derived from ERA5, Atmos. Res., 227, 249–262,
https://doi.org/10.1016/j.atmosres.2019.05.010, 2019. a
Dixon, M. and Wiener, G.: TITAN: Thunderstorm Identification, Tracking,
Analysis, and Nowcasting – A Radar-based Methodology, J. Atmos. Ocean. Tech., 10, 785–797, https://doi.org/10.1175/1520-0426(1993)010<0785:TTITAA>2.0.CO;2, 1993. a
Foresti, L., Sideris, I. V., Nerini, D., Beusch, L., and Germann, U.: Using a
10-Year Radar Archive for Nowcasting Precipitation Growth and Decay: A
Probabilistic Machine Learning Approach, Weather Forecast., 34, 1547–1569, https://doi.org/10.1175/WAF-D-18-0206.1, 2019. a
Franch, G., Nerini, D., Pendesini, M., Coviello, L., Jurman, G., and
Furlanello, C.: Precipitation Nowcasting with Orographic Enhanced Stacked
Generalization: Improving Deep Learning Predictions on Extreme Events,
Atmosphere, 11, 267, https://doi.org/10.3390/atmos11030267, 2020. a
GOES-R Algorithm Working Group and GOES-R Series Program Office: NOAA GOES-R
Series Advanced Baseline Imager (ABI) Level 2 Cloud Top Height (ACHA) [data set], https://doi.org/10.7289/V5HX19ZQ, 2018a. a
GOES-R Algorithm Working Group and GOES-R Series Program Office: NOAA GOES-R
Series Advanced Baseline Imager (ABI) Level 2 Cloud Optical Depth (COD) [data set], https://doi.org/10.7289/V58G8J02, 2018b. a
GOES-R Algorithm Working Group and GOES-R Series Program Office: NOAA GOES-R
Series Advanced Baseline Imager (ABI) Level 2 Cloud Top Pressure (CTP) [data set], https://doi.org/10.7289/V5D50K85, 2018c. a
GOES-R Algorithm Working Group and GOES-R Series Program Office: NOAA GOES-R
Series Advanced Baseline Imager (ABI) Level 2 Derived Stability Indices [data set], https://doi.org/10.7289/V50Z71KF, 2018d. a
GOES-R Calibration Working Group and GOES-R Series Program: NOAA GOES-R
Series Advanced Baseline Imager (ABI) Level 1b Radiances [data set], https://doi.org/10.7289/V5BV7DSR, 2017. a
Greene, D. R. and Clark, R. A.: Vertically Integrated Liquid Water – A New
Analysis Tool, Mon. Weather Rev., 100, 548–552,
https://doi.org/10.1175/1520-0493(1972)100<0548:VILWNA>2.3.CO;2, 1972. a
Handwerker, J.: Cell tracking with TRACE3D – a new algorithm, Atmos. Res.,
61, 15–34, https://doi.org/10.1016/S0169-8095(01)00100-4, 2002. a
Heidinger, A. K., Pavolonis, M. J., Calvert, C., Hoffman, J., Nebuda, S.,
Straka, W., Walther, A., and Wanzong, S.: ABI Cloud Products from the GOES-R Series, in: The GOES-R Series: A New Generation of Geostationary Environmental Satellites, chap. 6, edited by: Goodman, S. J., Schmit, T. J., Daniels, J., and Redmon, R. J., Elsevier, 43–62,
https://doi.org/10.1016/B978-0-12-814327-8.00006-8, 2020. a
Heiss, W. H., McGrew, D. L., and Sirmans, D.: Nexrad: next generation weather
radar (WSR-88D), Microwave J., 33, 79+, 1990. a
Helmus, J. J. and Collis, S. M.: The Python ARM Radar Toolkit (Py-ART), a
library for working with weather radar data in the Python programming language, J. Open Res. Software, 4, e25, https://doi.org/10.5334/jors.119, 2016. a
Hering, A., Morel, C., Galli, G., Sénési, S., Ambrosetti, P., and Boscacci, M.: Nowcasting thunderstorms in the Alpine region using a radar based adaptive thresholding scheme, in: Proceedings of ERAD 2004, https://www.copernicus.org/erad/2004/online/ERAD04_P_206.pdf (last access: 21 February 2022), 2004. a, b, c
Hering, A., Sénési, S., Ambrosetti, P., and Bernard-Bouissières, I.: Nowcasting thunderstorms in complex cases using radar data, in: WMO Symposium on Nowcasting and Very Short Range Forecasting, https://www.researchgate.net/publication/228609271_Nowcasting_thunderstorms_in_complex_cases_using_radar_data
(last access: 21 February 2022), 2005. a, b
Hering, A., Germann, U., Boscacci, M., and Sénési, S.: Operational
thunderstorm nowcasting in the Alpine region using 3D-radar severe weather
parameters and lightning data, in: Proceedings of ERAD 2006, http://www.crahi.upc.edu/ERAD2006/proceedingsMask/00122.pdf (last
access: 21 February 2022), 2006. a, b
Hoffmann, J.: Entwicklung und Anwendung von statistischen Vorhersage – Interpretationsverfahren für Gewitternowcasting und Unwetterwarnungen unter Einbeziehung von Fernerkundungsdaten, PhD thesis, Freie Universität Berlin, Berlin, https://doi.org/10.17169/refubium-15903, 2008. a
Huang, W., Jiang, Y., Liu, X., Pan, Y., Li, X., Guo, R., Huang, Y., and Duan,
B.: Classified Early-warning and Nowcasting of Hail Weather Based on Radar
Products and Random Forest Algorithm, in: 2019 International Conference on
Meteorology Observations (ICMO), https://doi.org/10.1109/ICMO49322.2019.9026039, 2019. a
James, P. M., Reichert, B. K., and Heizenreder, D.: NowCastMIX: Automatic
Integrated Warnings for Severe Convection on Nowcasting Time Scales at the
German Weather Service, Weather Forecasti., 33, 1413–1433,
https://doi.org/10.1175/WAF-D-18-0038.1, 2018. a
Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., and Liu,
T.-Y.: LightGBM: a highly efficient gradient boosting decision tree, in:
Proceedings of the 31st International Conference on Neural Information
Processing Systems, 3149–3157, https://dl.acm.org/doi/abs/10.5555/3294996.3295074 (last access: 21 February 2022), 2017. a
Kelly, D. L., Schaefer, J. T., and Doswell, C. A.: Climatology of Nontornadic
Severe Thunderstorm Events in the United States, Mon. Weather Rev., 113,
1997–2014, https://doi.org/10.1175/1520-0493(1985)113<1997:CONSTE>2.0.CO;2, 1985. a
Kober, K. and Tafferner, A.: Tracking and Nowcasting of Convective Cells Using Remote Sensing Data from Radar and Satellite, Meteorol. Z., 1, 75–84, https://doi.org/10.1127/0941-2948/2009/359, 2009. a, b
Kober, K., Craig, G. C., Keil, C., and Dörnbrack, A.: Blending a probabilistic nowcasting method with a high-resolution numerical weather prediction ensemble for convective precipitation forecasts, Q. J. Roy. Meteorol. Soc., 138, 755–768, https://doi.org/10.1002/qj.939, 2012. a
Kumar, A., Islam, T., Sekimoto, Y., Mattmann, C., and Wilson, B.: Convcast: An embedded convolutional LSTM based architecture for precipitation nowcasting using satellite data, PLOS One, 15, 1–18,
https://doi.org/10.1371/journal.pone.0230114, 2020. a
Lagerquist, R., McGovern, A., and Smith, T.: Machine Learning for Real-Time
Prediction of Damaging Straight-Line Convective Wind, Weather Forecast., 32, 2175–2193, https://doi.org/10.1175/WAF-D-17-0038.1, 2017. a
Lagerquist, R., McGovern, A., Homeyer, C. R., Gagne II, D. J., and Smith, T.:
Deep Learning on Three-Dimensional Multiscale Data for Next-Hour Tornado
Prediction, Mon. Weather Rev., 148, 2837–2861, https://doi.org/10.1175/MWR-D-19-0372.1, 2020. a
Leinonen, J., Hamann, U., Germann, U., and Mecikalski, J. R.: Machine learning code and dataset for “Nowcasting thunderstorm hazards using machine learning: the impact of data sources on performance”, Zenodo [code and data set], https://doi.org/10.5281/zenodo.6206919, 2021. a
Li, J., Li, Z., and Schmit, T. J.: ABI Legacy Atmospheric Profiles and
Derived Products from the GOES-R Series, in: The GOES-R Series: A New
Generation of Geostationary Environmental Satellites, chap. 7, edited by: Goodman, S. J., Schmit, T. J., Daniels, J., and Redmon, R. J., Elsevier, 63–77, https://doi.org/10.1016/B978-0-12-814327-8.00007-X, 2020. a
Marshall, J. S. and Radhakant, S.: Radar Precipitation Maps as Lightning
Indicators, J. Appl. Meteorol. Clim., 17, 206–212,
https://doi.org/10.1175/1520-0450(1978)017<0206:RPMALI>2.0.CO;2, 1978. a
Martner, B. E., Yuter, S. E., White, A. B., Matrosov, S. Y., Kingsmill, D. E., and Ralph, F. M.: Raindrop Size Distributions and Rain Characteristics in
California Coastal Rainfall for Periods with and without a Radar Bright Band,
J. Hydrometeorol., 9, 408–425, https://doi.org/10.1175/2007JHM924.1, 2008. a
Mecikalski, J. R. and Bedka, K. M.: Forecasting Convective Initiation by
Monitoring the Evolution of Moving Cumulus in Daytime GOES Imager, Mon. Weather Rev., 134, 49–78, https://doi.org/10.1175/MWR3062.1, 2006. a
Mecikalski, J. R., MacKenzie, W. M., Koenig, M., and Muller, S.: Cloud-Top
Properties of Growing Cumulus prior to Convective Initiation as Measured by
Meteosat Second Generation. Part I: Infrared Fields, J. Appl. Meteorol. Clim., 4, 521–534, https://doi.org/10.1175/2009JAMC2344.1, 2010. a
Mecikalski, J. R., Williams, J. K., Jewett, C. P., Ahijevych, D., LeRoy, A.,
and Walker, J. R.: Probabilistic 0–1-h Convective Initiation Nowcasts that
Combine Geostationary Satellite Observations and Numerical Weather Prediction
Model Data, J. Appl. Meteorol. Clim., 54, 1039–1059,
https://doi.org/10.1175/JAMC-D-14-0129.1, 2015. a
Mecikalski, J. R., Sandmæl, T. N., Murillo, E. M., Homeyer, C. R., Bedka, K. M., Apke, J. M., and Jewett, C. P.: Random Forest Model to Assess Predictor Importance and Nowcast Severe Storms using High-Resolution Radar–GOES Satellite–Lightning Observations, Mon. Weather Rev., 149, 1725–1746,
https://doi.org/10.1175/MWR-D-19-0274.1, 2021. a
Mostajabi, A., Finney, D. L., Rubinstein, M., and Rachidi, F.: Nowcasting
lightning occurrence from commonly available meteorological parameters using
machine learning techniques, Clim. Atmos. Sci., 2, 41,
https://doi.org/10.1038/s41612-019-0098-0, 2019. a
Mueller, C., Saxen, T., Roberts, R., Wilson, J., Betancourt, T., Dettling, S., Oien, N., and J., Y.: NCAR Auto-Nowcast System, Weather Forecast., 18, 545–561, https://doi.org/10.1175/1520-0434(2003)018<0545:NAS>2.0.CO;2, 2003. a
NASA/METI/AIST/Japan Spacesystems and US/Japan ASTER Science Team: ASTER
Global Digital Elevation Model V003 [data set], https://doi.org/10.5067/ASTER/ASTGTM.003, 2019. a
Natekin, A. and Knoll, A.: Gradient boosting machines, a tutorial, Front.
Neurorobot., 7, 21, https://doi.org/10.3389/fnbot.2013.00021, 2013. a
NOAA National Weather Service (NWS) Radar Operations Center: Next Generation Radar (NEXRAD) Level 2 Base Data [data set], https://doi.org/10.7289/V5W9574V, 1991. a
Pulkkinen, S., Nerini, D., Pérez Hortal, A. A., Velasco-Forero, C., Seed, A.,
Germann, U., and Foresti, L.: Pysteps: an open-source Python library for
probabilistic precipitation nowcasting (v1.0), Geosci. Model Dev., 12,
4185–4219, https://doi.org/10.5194/gmd-12-4185-2019, 2019. a
Raspaud, M., Hoese, D., Dybbroe, A., Lahtinen, P., Devasthale, A., Itkin, M.,
Hamann, U., Rasmussen, L. O., Nielsen, E. S., Leppelt, T., Maul, A., Kliche,
C., and Thorsteinsson, H.: PyTroll: An Open-Source, Community-Driven Python Framework to Process Earth Observation Satellite Data, B. Am. Meteorol. Soc., 99, 1329–1336, https://doi.org/10.1175/BAMS-D-17-0277.1, 2018. a
Roberts, R. D. and Rutledge, S.: Nowcasting Storm Initiation and Growth Using
GOES-8 and WSR-88D Data, Weather Forecast., 18, 562–584,
https://doi.org/10.1175/1520-0434(2003)018<0562:NSIAGU>2.0.CO;2, 2003. a
Rudlosky, S. D., Goodman, S. J., and Virts, K. S.: Lightning Detection: GOES-R Series Geostationary Lightning Mapper, in: The GOES-R Series: A New
Generation of Geostationary Environmental Satellites, chap. 16, edited by: Goodman, S. J., Schmit, T. J., Daniels, J., and Redmon, R. J., Elsevier, 193–202, https://doi.org/10.1016/B978-0-12-814327-8.00016-0, 2020. a
Schmit, T. J. and Gunshor, M. M.: ABI Imagery from the GOES-R Series, in:
The GOES-R Series: A New Generation of Geostationary Environmental Satellites, chap. 4, edited by: Goodman, S. J., Schmit, T. J., Daniels, J., and Redmon, R. J., Elsevier, 23–34, https://doi.org/10.1016/B978-0-12-814327-8.00004-4, 2020. a
Shi, X., Chen, Z., Wang, H., Yeung, D.-Y., Wong, W.-K., and Woo, W.-C.:
Convolutional LSTM Network: A Machine Learning Approach for Precipitation
Nowcasting, in: Advances in Neural Information Processing Systems 28, edited
by Cortes, C., Lawrence, N. D., Lee, D. D., Sugiyama, M., and Garnett, R.,
Curran Associates, Inc., 802–810,
http://papers.nips.cc/paper/5955-convolutional-lstm-network-a-machine-learning-approach
-for-precipitation-nowcasting.pdf (last access: 21 February 2022), 2015. a
Shi, X., Gao, Z., Lausen, L., Wang, H., Yeung, D.-Y., Wong, W.-K., and Woo,
W.-C.: Deep learning for precipitation nowcasting: a benchmark and a new
model, in: Proceedings of the 31st International Conference on Neural Information Processing Systems, 5622–5632, https://dl.acm.org/doi/abs/10.5555/3295222.3295313 (last access: 21 February 2022), 2017. a
Smith, T. M., Lakshmanan, V., Stumpf, G. J., Ortega, K. L., Hondl, K., Cooper, K., Calhoun, K. M., Kingfield, D. M., Manross, K. L., Toomey, R., and
Brodgen, J.: Multi-Radar Multi-Sensor (MRMS) Severe Weather and Aviation
Products: Initial Operating Capabilities, B. Am. Meteorol. Soc., 97, 1617–1630, https://doi.org/10.1175/BAMS-D-14-00173.1, 2016. a
Snyder, J. P.: Map Projections – A Working Manual, United States Government
Printing Office, Washington, DC, USA, https://doi.org/10.3133/pp1395, 1987. a
Sprenger, M., Schemm, S., Oechslin, R., and Jenkner, J.: Nowcasting Foehn Wind Events Using the AdaBoost Machine Learning Algorithm, Weather Forecast., 32, 1079–1099, https://doi.org/10.1175/WAF-D-16-0208.1, 2017. a
Steinacker, R., Dorninger, M., Wölfelmaier, F., and Krennert, T.: Automatic Tracking of Convective Cells and Cell Complexes from Lightning and Radar Data, Meteorol. Atmos. Phys., 72, 101–110, https://doi.org/10.1007/s007030050009, 2000. a
Süli, E. and Mayers, D. F.: An Introduction to Numerical Analysis, Cambridge University Press, Cambridge, UK, https://doi.org/10.1017/CBO9780511801181, 2003. a
Sullivan, P. C.: GOES-R Series Spacecraft and Instruments, in: The GOES-R
Series: A New Generation of Geostationary Environmental Satellites, chap. 3, edited by: Goodman, S. J., Schmit, T. J., Daniels, J., and Redmon, R. J.,
Elsevier, 13–21, https://doi.org/10.1016/B978-0-12-814327-8.00003-2, 2020. a
Waldvogel, A., Federer, B., and Grimm, P.: Criteria for the Detection of Hail
Cells, J. Appl. Meteorol., 18, 1521–1525,
https://doi.org/10.1175/1520-0450(1979)018<1521:CFTDOH>2.0.CO;2, 1979. a
Willmott, C. J. and Matsuura, K.: Advantages of the mean absolute error (MAE)
over the root mean square error (RMSE) in assessing average model performance, Clim. Res., 30, 79–82, https://doi.org/10.3354/cr030079, 2005. a
Wilson, J. W. and Mueller, C. K.: Nowcasts of Thunderstorm Initiation and
Evolution, Weather Forecast., 8, 113–131, https://doi.org/10.1175/1520-0434(1993)008<0113:NOTIAE>2.0.CO;2, 1993. a
Wilson, J. W., Crook, N. A., Mueller, C. K., Sun, J., and Dixon, M.: Nowcasting Thunderstorms: A Status Report, B. Amer. Meteorol. Soc., 79, 2079–2100, https://doi.org/10.1175/1520-0477(1998)079<2079:NTASR>2.0.CO;2, 1998. a
Yeung, J. K., Smith, J. A., Baeck, M. L., and Villarini, G.: Lagrangian
Analyses of Rainfall Structure and Evolution for Organized Thunderstorm
Systems in the Urban Corridor of the Northeastern United States, J. Hydrometeorol., 16, 1575–1595, https://doi.org/10.1175/JHM-D-14-0095.1, 2015. a
Zhang, J., Howard, K., Langston, C., Kaney, B., Qi, Y., Tang, L., Grams, H.,
Wang, Y., Cocks, S., Martinaitis, S., Arthur, A., Cooper, K., Brogden, J.,
and Kitzmiller, D.: Multi-Radar Multi-Sensor (MRMS) Quantitative
Precipitation Estimation: Initial Operating Capabilities, B. Am. Meteorol. Soc., 97, 621–638, https://doi.org/10.1175/BAMS-D-14-00174.1, 2016. a
Zhou, K., Zheng, Y., Dong, W., and Wang, T.: A Deep Learning Network for
Cloud-to-Ground Lightning Nowcasting with Multisource Data, J. Atmos. Ocean.
Tech., 37, 927–942, https://doi.org/10.1175/JTECH-D-19-0146.1, 2020.
a
Zinner, T., Mannstein, H., and Tafferner, A.: Cb-TRAM: Tracking and monitoring severe convection from onset over rapid development to mature phase using multi-channel Meteosat-8 SEVIRI data, Meteorol. Atmos. Phys., 101, 191–210, https://doi.org/10.1007/s00703-008-0290-y, 2008. a
Short summary
We evaluate the usefulness of different data sources and variables to the short-term prediction (
nowcasting) of severe thunderstorms using machine learning. Machine-learning models are trained with data from weather radars, satellite images, lightning detection and weather forecasts and with terrain elevation data. We analyze the benefits provided by each of the data sources to predicting hazards (heavy precipitation, lightning and hail) caused by the thunderstorms.
We evaluate the usefulness of different data sources and variables to the short-term prediction...
Altmetrics
Final-revised paper
Preprint