Estimating fine resolution shortwave broadband albedo of croplands from Harmonized Landsat and Sentinel-2 data.
Saved in:
| Title: | Estimating fine resolution shortwave broadband albedo of croplands from Harmonized Landsat and Sentinel-2 data. |
|---|---|
| Authors: | Dai, Jie1 (AUTHOR) daijie2@msu.edu, Chen, Jiquan1,2,3 (AUTHOR), Lei, Cheyenne4 (AUTHOR), Falvo, Grant5 (AUTHOR), Robertson, G Philip1,3,6 (AUTHOR) |
| Source: | Journal of Applied Meteorology & Climatology. May2026, Vol. 65 Issue 5, p1-13. 13p. |
| Subjects: | Albedo, Satellite-based remote sensing, Machine learning, Farms, Remote sensing, Random forest algorithms, Boosting algorithms |
| Abstract: | Altered surface albedo due to land cover conversions and management is a significant driver of global climate change. Albedo can be directly measured at ground stations, and remote sensing data can be used to scale-up albedo values to regional and global levels. Some previous studies have retrieved fine resolution (10–30 m) instantaneous albedo and coarse resolution (500–1000 m) daily mean albedo from remote sensing data, but they all required the input of MODIS albedo information at 500 m resolution, and none have assembled both instantaneous and daily albedo based exclusively on fine resolution satellite data. To address this issue, we compiled 387 instantaneous and 346 daily albedo records using field net radiometer measurements from the bioenergy croplands at the W. K. Kellogg Biological Station in southwest Michigan. We then connected these albedo records with a suite of variables derived from Harmonized Landsat and Sentinel-2 data through two machine learning algorithms (random forest regression and extreme gradient boosting) to retrieve clear-sky instantaneous and daily shortwave broadband albedo. The performance statistics indicate reasonable accuracy of model results (RMSE around or below 0.03 except for snow-covered surfaces), suggesting that the retrieval of both instantaneous and daily albedo based exclusively on fine resolution satellite data is promising. To facilitate the use of fine resolution albedo products at the global level, future efforts need to include more albedo records of diverse surface cover types, as well as to accurately model daily albedo for cloudy days to address the "clear-sky bias". [ABSTRACT FROM AUTHOR] |
| Copyright of Journal of Applied Meteorology & Climatology is the property of American Meteorological Society and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Altered surface albedo due to land cover conversions and management is a significant driver of global climate change. Albedo can be directly measured at ground stations, and remote sensing data can be used to scale-up albedo values to regional and global levels. Some previous studies have retrieved fine resolution (10–30 m) instantaneous albedo and coarse resolution (500–1000 m) daily mean albedo from remote sensing data, but they all required the input of MODIS albedo information at 500 m resolution, and none have assembled both instantaneous and daily albedo based exclusively on fine resolution satellite data. To address this issue, we compiled 387 instantaneous and 346 daily albedo records using field net radiometer measurements from the bioenergy croplands at the W. K. Kellogg Biological Station in southwest Michigan. We then connected these albedo records with a suite of variables derived from Harmonized Landsat and Sentinel-2 data through two machine learning algorithms (random forest regression and extreme gradient boosting) to retrieve clear-sky instantaneous and daily shortwave broadband albedo. The performance statistics indicate reasonable accuracy of model results (RMSE around or below 0.03 except for snow-covered surfaces), suggesting that the retrieval of both instantaneous and daily albedo based exclusively on fine resolution satellite data is promising. To facilitate the use of fine resolution albedo products at the global level, future efforts need to include more albedo records of diverse surface cover types, as well as to accurately model daily albedo for cloudy days to address the "clear-sky bias". [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 15588424 |
| DOI: | 10.1175/JAMC-D-25-0081.1 |
