Correction of CAMS PM<sub>10</sub> Reanalysis Improves AI-Based Dust Event Forecast
High dust loading significantly impacts air quality, climate, and public health. Early warning is crucial for mitigating short-term effects, and accurate dust field estimates are needed for forecasting. The Copernicus Atmosphere Monitoring Service (CAMS) offers global reanalysis datasets and forecas...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/17/2/222 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832587570955943936 |
|---|---|
| author | Ron Sarafian Sagi Nathan Dori Nissenbaum Salman Khan Yinon Rudich |
| author_facet | Ron Sarafian Sagi Nathan Dori Nissenbaum Salman Khan Yinon Rudich |
| author_sort | Ron Sarafian |
| collection | DOAJ |
| description | High dust loading significantly impacts air quality, climate, and public health. Early warning is crucial for mitigating short-term effects, and accurate dust field estimates are needed for forecasting. The Copernicus Atmosphere Monitoring Service (CAMS) offers global reanalysis datasets and forecasts of particulate matter with a diameter of under 10 μm (PM<sub>10</sub>), which approximate dust, but recent studies highlight discrepancies between CAMS data and ground <i>in-situ</i> measurements. Since CAMS is often used for forecasting, errors in PM<sub>10</sub> fields can hinder accurate dust event forecasts, which is particularly challenging for models that use <i>artificial intelligence</i> (AI) due to the scarcity of dust events and limited training data. This study proposes a machine-learning approach to correct CAMS PM<sub>10</sub> fields using <i>in-situ</i> data to enhance AI-based dust event forecasting. A correction model that links pixel-wise errors with atmospheric and meteorological variables was taught using gradient-boosting algorithms. This model is then utilized to predict CAMS error in previously unobserved pixels across the Eastern Mediterranean, generating CAMS error fields. Our bias-corrected PM<sub>10</sub> fields are, on average, 12 μg m<sup>−3</sup> more accurate, often reducing CAMS errors by significant percentages. To evaluate the contribution, we train a deep neural network to predict city-scale dust events (0–72 h) over the Balkans using PM<sub>10</sub> fields. Comparing the network’s performance when trained on both original and bias-corrected CAMS PM<sub>10</sub> fields, we show that the correction improves AI-based forecasting performance across all metrics. |
| format | Article |
| id | doaj-art-5d62438b76604226b8b26df6705f8fc7 |
| institution | Kabale University |
| issn | 2072-4292 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj-art-5d62438b76604226b8b26df6705f8fc72025-01-24T13:47:48ZengMDPI AGRemote Sensing2072-42922025-01-0117222210.3390/rs17020222Correction of CAMS PM<sub>10</sub> Reanalysis Improves AI-Based Dust Event ForecastRon Sarafian0Sagi Nathan1Dori Nissenbaum2Salman Khan3Yinon Rudich4Earth and Planetary Science Department, Weizmann Institute of Science, P.O. Box 26, Rehovot 7610001, IsraelDepartment of Statistics and Data Science, Hebrew University of Jerusalem, Jerusalem 91905, IsraelEarth and Planetary Science Department, Weizmann Institute of Science, P.O. Box 26, Rehovot 7610001, IsraelComputer Vision Department, Mohamed bin Zayed University of Artificial Intelligence, Building 1B, Masdar City, Abu Dhabi P.O. Box 7909, United Arab EmiratesEarth and Planetary Science Department, Weizmann Institute of Science, P.O. Box 26, Rehovot 7610001, IsraelHigh dust loading significantly impacts air quality, climate, and public health. Early warning is crucial for mitigating short-term effects, and accurate dust field estimates are needed for forecasting. The Copernicus Atmosphere Monitoring Service (CAMS) offers global reanalysis datasets and forecasts of particulate matter with a diameter of under 10 μm (PM<sub>10</sub>), which approximate dust, but recent studies highlight discrepancies between CAMS data and ground <i>in-situ</i> measurements. Since CAMS is often used for forecasting, errors in PM<sub>10</sub> fields can hinder accurate dust event forecasts, which is particularly challenging for models that use <i>artificial intelligence</i> (AI) due to the scarcity of dust events and limited training data. This study proposes a machine-learning approach to correct CAMS PM<sub>10</sub> fields using <i>in-situ</i> data to enhance AI-based dust event forecasting. A correction model that links pixel-wise errors with atmospheric and meteorological variables was taught using gradient-boosting algorithms. This model is then utilized to predict CAMS error in previously unobserved pixels across the Eastern Mediterranean, generating CAMS error fields. Our bias-corrected PM<sub>10</sub> fields are, on average, 12 μg m<sup>−3</sup> more accurate, often reducing CAMS errors by significant percentages. To evaluate the contribution, we train a deep neural network to predict city-scale dust events (0–72 h) over the Balkans using PM<sub>10</sub> fields. Comparing the network’s performance when trained on both original and bias-corrected CAMS PM<sub>10</sub> fields, we show that the correction improves AI-based forecasting performance across all metrics.https://www.mdpi.com/2072-4292/17/2/222PM<sub>10</sub>CAMSdust forecastingartificial intelligencemachine learning |
| spellingShingle | Ron Sarafian Sagi Nathan Dori Nissenbaum Salman Khan Yinon Rudich Correction of CAMS PM<sub>10</sub> Reanalysis Improves AI-Based Dust Event Forecast Remote Sensing PM<sub>10</sub> CAMS dust forecasting artificial intelligence machine learning |
| title | Correction of CAMS PM<sub>10</sub> Reanalysis Improves AI-Based Dust Event Forecast |
| title_full | Correction of CAMS PM<sub>10</sub> Reanalysis Improves AI-Based Dust Event Forecast |
| title_fullStr | Correction of CAMS PM<sub>10</sub> Reanalysis Improves AI-Based Dust Event Forecast |
| title_full_unstemmed | Correction of CAMS PM<sub>10</sub> Reanalysis Improves AI-Based Dust Event Forecast |
| title_short | Correction of CAMS PM<sub>10</sub> Reanalysis Improves AI-Based Dust Event Forecast |
| title_sort | correction of cams pm sub 10 sub reanalysis improves ai based dust event forecast |
| topic | PM<sub>10</sub> CAMS dust forecasting artificial intelligence machine learning |
| url | https://www.mdpi.com/2072-4292/17/2/222 |
| work_keys_str_mv | AT ronsarafian correctionofcamspmsub10subreanalysisimprovesaibaseddusteventforecast AT saginathan correctionofcamspmsub10subreanalysisimprovesaibaseddusteventforecast AT dorinissenbaum correctionofcamspmsub10subreanalysisimprovesaibaseddusteventforecast AT salmankhan correctionofcamspmsub10subreanalysisimprovesaibaseddusteventforecast AT yinonrudich correctionofcamspmsub10subreanalysisimprovesaibaseddusteventforecast |