A joint research team from the National Time Service Center(NTSC) of the Chinese Academy of Sciences(CAS), Xinjiang Institute of Ecology and  Geography CAS, Shandong University of Science and Technology, and Hohai  University successfully achieved centimeter-level water level monitoring  of the Kalabuli Reservoir in Xinjiang using Global Navigation Satellite  System Interferometric Reflectometry (GNSS-IR) technology.

The related results were published in the journal Advances in Space Research under the title "GNSS-IR Water Level Monitoring for Complex Environments: Application to Kalabeily Reservoir in Xinjiang, China."

GNSS-IR  is a technique that extracts characteristics of reflecting surfaces by  analyzing the interference phenomenon generated when direct and  reflected satellite signals superimpose at the antenna. It offers the  advantages of low cost and high precision. Currently, this technology  has been widely applied in water level monitoring for various types of  water bodies such as oceans, reservoirs, rivers, and lakes. It is also  utilized in the retrieval of environmental parameters including soil  moisture, snow depth, and sea wind, and has become a key technological  field actively promoted at the national level.

The  Kalabelly Reservoir is located in the Kezi River Basin of the Xinjiang  Uygur Autonomous Region. Influenced by the operational mode of "sediment  discharge during flood seasons and water storage during dry seasons"  the annual water level of the reservoir fluctuates significantly. f As a  result, the effective reflection zone of GNSS signals varies notably  with the rise and fall of the water level. In such complex environments,  the traditional uniformly selected elevation angle range (USEAR) method  cannot adequately adapt to the dynamically changing reflection zone.  This often leads to situations where the reflection points deviate from  the water surface, which leads to abnormal inversion outcomes and maks  it difficult to achieve long-term, continuous, high-precision  monitoring.

To  address this challenge, researchers proposed a method for dynamically  selecting the elevation angle range (DSEAR). This method is grounded in  the geometric properties of the Fresnel zone. It dynamically determines  the suitable elevation angle range for each day by integrating the  median reflector height (RH) retrieved from the previous day's data with  pre-defined boundaries of the effective water surface reflection area.  Compared to conventional methods, the DSEAR approach effectively  excludes non-aqueous reflected signals, broadens the usable elevation  angle range, and increases the effective reflection area. Consequently,  it significantly enhances both the success rate and the reliability of  water level retrieval, laying a solid foundation for real-time reservoir  water level monitoring.

Figure  1 Comparison of static water level retrieval sequences using two  different elevation angle selection methods (Imaged by NTSC)

To  further enhance retrieval accuracy, the researchers also introduced a  multi-station multi-signal water level retrieval algorithm based on a  robust estimation strategy. The final water level retrieval achieved an  accuracy of 4.07 cm. Compared to the single-station multi-signal  retrieval method, the accuracy was improved by up to 44.70%. These  results confirm the significant potential of GNSS-IR technology in  complex environments.

Figure  2 Water level retrieval sequence and error analysis from the  multi-station multi-signal combination at Kalabeily Reservoir. (Imaged  by NTSC)