Leveraging a Novel Straightforward Integration Approach for Independent CYGNSS Soil Moisture Retrieval across Vegetated Regions
The state-of-the-art Cyclone Global Navigation Satellite System (CYGNSS) constellation offers an effective approach for characterizing diurnal soil moisture (SM) dynamics at sub-daily intervals. However, SM retrieval from this constellation still faces 2 underlying challenges, including the vegetati...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
American Association for the Advancement of Science (AAAS)
2025-01-01
|
| Series: | Journal of Remote Sensing |
| Online Access: | https://spj.science.org/doi/10.34133/remotesensing.0726 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The state-of-the-art Cyclone Global Navigation Satellite System (CYGNSS) constellation offers an effective approach for characterizing diurnal soil moisture (SM) dynamics at sub-daily intervals. However, SM retrieval from this constellation still faces 2 underlying challenges, including the vegetation attenuation effects and the excessive dependence on supplementary SM data from other systems. To address these limitations, this study introduces a novel CYGNSS SM retrieval method that combines a 2-step calibration approach for eliminating vegetation effects, and the relative signal-to-noise ratio (rSNR) method adapted to surface reflectivity (SR) for direct SM retrieval from this observable without the needs for supplementary datasets. The performance of this integrated CYGNSS SM retrieval was assessed via intercomparison to its single-component methods against global in situ SM network measurements under land use and land cover (LULC) types, vegetation water content (VWC), sand fraction (SF), land surface temperature (LST), and overall conditions based on 3 common metrics of root mean square error (RMSE), unbiased RMSE (ubRMSE), and bias. The intercomparison across LULC indicates that the proposed integration method outperformed its single-component methods in most LULC types, particularly in croplands and savannahs. The integration approach was found to be more superior under sparse canopies, together with high-SF and LST areas. The overall evaluation proved the robustness of applying our integration approach for improving CYGNSS SM retrieval rather than single-step methods, where the best performance with the lowest average RMSE, ubRMSE, and bias of 0.117, 0.088, and 0.092 m3 m−3, respectively, was observed with the integrated one. |
|---|---|
| ISSN: | 2694-1589 |