Monitoring River Ice in the Northern Section of Yellow River Using Sentinel-1/2 Images

Autor: Xingwang Zhang, Haoming Xia, Yongwei Sheng
Jazyk: angličtina
Rok vydání: 2024
Předmět:
Zdroj: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol 17, Pp 13232-13243 (2024)
Druh dokumentu: article
ISSN: 1939-1404
2151-1535
DOI: 10.1109/JSTARS.2024.3408188
Popis: River ice monitoring plays a critical role in ice flood mitigation and control, hydrological research, and water resource regulation. The northern section of the Yellow River (NSYR) experiences the most frequent water level fluctuations due to river ice. However, accurate annual maps of river ice distribution are unavailable and their spatiotemporal changes in the NSYR are unknown. To address this issue, we developed an ice and water identification model. This model, based on Sentinel-1/2 images, leverages the river ice recognition index (RDRI) and the unique spectral characteristics of the Yellow River ice, and achieves accurate monitoring of ice and water with the NSYR, attaining an accuracy of 94.61% and a kappa coefficient of 0.8013. We applied this model to extract the time series of river ice area in NSYR from November 2018 to March 2022. Our results confirm that temperature significantly influences the formation of river ice, which, in turn, has a regulatory effect on river runoff. As the temperature decreases in winter, the river ice area expands rapidly. River ice area and water level usually reach their peak in January. Overall, for every 1°C decrease in average temperature, the river ice area expands by 8.47%, the water level rises by 0.14 m, and the discharge decreases by 18.38 m³/s. The findings of this study offer substantial methodological reinforcement for the identification and monitoring of river ice, along with crucial support in the realms of ice flood mitigation, hydrological research, and water resource management.
Databáze: Directory of Open Access Journals