Improved estimation of carbon dioxide and methane using machine learning with satellite observations over the Arabian Peninsula
Abstract Estimating spatiotemporal maps of greenhouse gases (GHGs) is important for understanding climate change and developing mitigation strategies. However, current methods face challenges, including the coarse resolution of numerical models, and gaps in satellite data, making it essential to imp...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-024-84593-9 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841559749369266176 |
|---|---|
| author | Mariam Alcibahy Fahim Abdul Gafoor Farhan Mustafa Mutasem El Fadel Hamed Al Hashemi Ali Al Hammadi Maryam R. Al Shehhi |
| author_facet | Mariam Alcibahy Fahim Abdul Gafoor Farhan Mustafa Mutasem El Fadel Hamed Al Hashemi Ali Al Hammadi Maryam R. Al Shehhi |
| author_sort | Mariam Alcibahy |
| collection | DOAJ |
| description | Abstract Estimating spatiotemporal maps of greenhouse gases (GHGs) is important for understanding climate change and developing mitigation strategies. However, current methods face challenges, including the coarse resolution of numerical models, and gaps in satellite data, making it essential to improve the spatiotemporal estimation of GHGs. This study aims to develop an advanced technique to produce high-fidelity (1 km) maps of CO2 and CH4 over the Arabian Peninsula, a highly vulnerable region to climate change. Using XGBoost, columnar carbon dioxide (XCO2) and methane (XCH4) concentrations using satellite data from OCO-2 and Sentinel-5P (the target variables) were downscaled, with ancillary data from CarbonTracker, MODIS Terra, and ERA-5 (the input variables). The model is trained and validated against these datasets, achieving high performance for XCO2 (R2 = 0.98, RMSE = 0.58 ppm) and moderate accuracy for XCH4 (R2 = 0.63, RMSE = 13.26 ppb). Seasonal cycles and long-term trends were identified, with higher concentrations observed in summer, and emission hotspots in urban and industrial areas. Comparisons with the EDGAR inventory highlighted the significant contributions of the power, oil, and transportation sectors to GHG emissions. These results demonstrate the value of high-resolution data for local-scale monitoring, supporting targeted mitigation strategies and sustainable policymaking in the region. Future work could integrate ground-based observations to further enhance GHG monitoring accuracy. |
| format | Article |
| id | doaj-art-828700883a6f4ea9a71eb0c46abcf1e5 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-828700883a6f4ea9a71eb0c46abcf1e52025-01-05T12:15:48ZengNature PortfolioScientific Reports2045-23222025-01-0115112410.1038/s41598-024-84593-9Improved estimation of carbon dioxide and methane using machine learning with satellite observations over the Arabian PeninsulaMariam Alcibahy0Fahim Abdul Gafoor1Farhan Mustafa2Mutasem El Fadel3Hamed Al Hashemi4Ali Al Hammadi5Maryam R. Al Shehhi6Civil and Environmental Engineering Department, Khalifa UniversityCivil and Environmental Engineering Department, Khalifa UniversityHong Kong University of Science and TechnologyCivil and Environmental Engineering Department, Khalifa UniversitySpace Mission Department, UAE Space Agency Abu DhabiChemical and Petroleum Engineering Department, Khalifa UniversityCivil and Environmental Engineering Department, Khalifa UniversityAbstract Estimating spatiotemporal maps of greenhouse gases (GHGs) is important for understanding climate change and developing mitigation strategies. However, current methods face challenges, including the coarse resolution of numerical models, and gaps in satellite data, making it essential to improve the spatiotemporal estimation of GHGs. This study aims to develop an advanced technique to produce high-fidelity (1 km) maps of CO2 and CH4 over the Arabian Peninsula, a highly vulnerable region to climate change. Using XGBoost, columnar carbon dioxide (XCO2) and methane (XCH4) concentrations using satellite data from OCO-2 and Sentinel-5P (the target variables) were downscaled, with ancillary data from CarbonTracker, MODIS Terra, and ERA-5 (the input variables). The model is trained and validated against these datasets, achieving high performance for XCO2 (R2 = 0.98, RMSE = 0.58 ppm) and moderate accuracy for XCH4 (R2 = 0.63, RMSE = 13.26 ppb). Seasonal cycles and long-term trends were identified, with higher concentrations observed in summer, and emission hotspots in urban and industrial areas. Comparisons with the EDGAR inventory highlighted the significant contributions of the power, oil, and transportation sectors to GHG emissions. These results demonstrate the value of high-resolution data for local-scale monitoring, supporting targeted mitigation strategies and sustainable policymaking in the region. Future work could integrate ground-based observations to further enhance GHG monitoring accuracy.https://doi.org/10.1038/s41598-024-84593-9GCCGHGsClimate changeWarmingSenitnel-5PXGBoost |
| spellingShingle | Mariam Alcibahy Fahim Abdul Gafoor Farhan Mustafa Mutasem El Fadel Hamed Al Hashemi Ali Al Hammadi Maryam R. Al Shehhi Improved estimation of carbon dioxide and methane using machine learning with satellite observations over the Arabian Peninsula Scientific Reports GCC GHGs Climate change Warming Senitnel-5P XGBoost |
| title | Improved estimation of carbon dioxide and methane using machine learning with satellite observations over the Arabian Peninsula |
| title_full | Improved estimation of carbon dioxide and methane using machine learning with satellite observations over the Arabian Peninsula |
| title_fullStr | Improved estimation of carbon dioxide and methane using machine learning with satellite observations over the Arabian Peninsula |
| title_full_unstemmed | Improved estimation of carbon dioxide and methane using machine learning with satellite observations over the Arabian Peninsula |
| title_short | Improved estimation of carbon dioxide and methane using machine learning with satellite observations over the Arabian Peninsula |
| title_sort | improved estimation of carbon dioxide and methane using machine learning with satellite observations over the arabian peninsula |
| topic | GCC GHGs Climate change Warming Senitnel-5P XGBoost |
| url | https://doi.org/10.1038/s41598-024-84593-9 |
| work_keys_str_mv | AT mariamalcibahy improvedestimationofcarbondioxideandmethaneusingmachinelearningwithsatelliteobservationsoverthearabianpeninsula AT fahimabdulgafoor improvedestimationofcarbondioxideandmethaneusingmachinelearningwithsatelliteobservationsoverthearabianpeninsula AT farhanmustafa improvedestimationofcarbondioxideandmethaneusingmachinelearningwithsatelliteobservationsoverthearabianpeninsula AT mutasemelfadel improvedestimationofcarbondioxideandmethaneusingmachinelearningwithsatelliteobservationsoverthearabianpeninsula AT hamedalhashemi improvedestimationofcarbondioxideandmethaneusingmachinelearningwithsatelliteobservationsoverthearabianpeninsula AT alialhammadi improvedestimationofcarbondioxideandmethaneusingmachinelearningwithsatelliteobservationsoverthearabianpeninsula AT maryamralshehhi improvedestimationofcarbondioxideandmethaneusingmachinelearningwithsatelliteobservationsoverthearabianpeninsula |