Files and links (1)
Published (Version of record)CC BY V4.0, Open Access
Journal article
Deep multi-task learning for early warnings of dust events implemented for the Middle East
NPJ climate and atmospheric science, Vol.6(1), 23
23/Mar/2023
Abstract
Events of high dust loading are extreme meteorological phenomena with important climate and health implications. Therefore, early forecasting is critical for mitigating their adverse effects. Dust modeling is a long-standing challenge due to the multiscale nature of the governing meteorological dynamics and the complex coupling between atmospheric particles and the underlying atmospheric flow patterns. While physics-based numerical modeling is commonly being used, we propose a meteorological-based deep multi-task learning approach for forecasting dust events. Our approach consists of forecasting the local PM10 (primary task) measured in situ, and simultaneously to predict the satellite-based regional PM10 (auxiliary task); thus, leveraging valuable information from a correlated task. We use 18 years of regional meteorological data to train a neural forecast model for dust events in Israel. Twenty-four hours before the dust event, the model can detect 76% of the events with even higher predictability of winter and spring events. Further analysis shows that local dynamics drive most misclassified events, meaning that the coherent driving meteorology in the region holds a predictive skill. Further, we use machine-learning interpretability methods to reveal the meteorological patterns the model has learned, thus highlighting the important features that govern dust events in the Middle East, being primarily lower-tropospheric winds, and Aerosol Optical Depth.
Details
- Title
- Deep multi-task learning for early warnings of dust events implemented for the Middle East
- Creators
- Ron Sarafian - 972WIS_INST___99Dori Nissenbaum - 972WIS_INST___99Shira Raveh-Rubin - 972WIS_INST___99Vikhyat Agrawal - Indian Institute of Technology BombayYinon Rudich - 972WIS_INST___99
- Resource Type
- Journal article
- Publication Details
- NPJ climate and atmospheric science, Vol.6(1), 23
- Number of pages
- 12
- Publisher
- NATURE PORTFOLIO
- Language
- English
- DOI
- https://doi.org/10.1038/s41612-023-00348-9
- Grant note
- The authors wish to thank Shai Bagon for his comments and ideas and Vered Silverman for extracting and preparing the reanalysis data. The European Centre for Medium-range Weather Forecasts and Israeli Meteorological Service are acknowledged for providing access to ERA5. The research was partially supported by a research grant from the Maggie Kaplan Research Fund, by the Israeli Council for Higher Education (CHE) via the Weizmann Data Science Research Center, and by the Helen Kimmel Center for Planetary Science at the Weizmann Institute of Science. Author contributions - R.S. and D.N. conceived of the presented idea; R.S. developed the theory with support from D.N. and performed the computations; D.N. built and processed the data; S.R.R. and Y.R. verified the analytical methods, the scientific insights, and supervised the findings of this study; R.S. wrote the manuscript with support from D.N., S.R.R., and Y.R.; V.A. programmed the visualization and explanation tools. All authors discussed the results and contributed to the final manuscript.
- Record Identifier
- 993442684503596
Metrics
3 Record Views