Retrieving all-weather precipitable water vapor using near-infrared and thermal infrared observations
Near-infrared (NIR) and thermal infrared (TIR) remote sensing are primary methods for monitoring large-scale, high-resolution precipitable water vapor (PWV); however, their application is limited to clear-sky conditions. To overcome this limitation and enable all-weather PWV retrieval, this study de...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-08-01
|
| Series: | International Journal of Applied Earth Observations and Geoinformation |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1569843225003462 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Near-infrared (NIR) and thermal infrared (TIR) remote sensing are primary methods for monitoring large-scale, high-resolution precipitable water vapor (PWV); however, their application is limited to clear-sky conditions. To overcome this limitation and enable all-weather PWV retrieval, this study develops a new satellite-based PWV retrieval model to derive all-weather PWV at a spatial resolution of 300 m based on the synergistic use of NIR and TIR data from the Ocean and Land Colour Instrument (OLCI) and Sea and Land Surface Temperature Radiometer (SLSTR) instruments onboard the Sentinel-3B satellite. The proposed model consists of two sub-models for PWV derivation: one each for clear- and cloudy-sky conditions. The clear-sky PWV retrieval model integrates water vapor information from both NIR and TIR observations, whereas the cloudy-sky PWV retrieval model uses cloud-top properties derived from TIR data to correct the NIR water vapor within or above the clouds to the ground. Tested against the GNSS PWV, the new model has a root-mean-square error (RMSE) of 1.01 mm and bias of 0.09 mm for clear-sky conditions, representing a 51 % improvement in accuracy over the operational OLCI NIR products. For cloudy-sky conditions, the RMSE and bias are 2.66 mm and 0.02 mm, respectively, thereby filling the gap in high-quality PWV retrieval products under cloudy-sky conditions. The combined all-weather PWV product demonstrates robust spatiotemporal performance, with an RMSE and Bias of 1.92 mm and 0.16 mm, respectively. Given that the GNSS PWV has an accuracy of 1.37 mm, the accuracy achieved in this study is satisfactory. |
|---|---|
| ISSN: | 1569-8432 |