Application of GNSS Reflectometry for the Monitoring of Lake Ice Cover

Abstract

Lakes cover vast expanses of land in many regions of the Northern Hemisphere. Their presence has been shown to have a significant impact on local weather and climate. The seasonal presence of ice cover affects the transfer of energy and heat between lakes and the overlaying atmosphere, as well as various socio-economic activities, including transportation, recreation, and cultural practices. However, climate change is rapidly altering lake ice cover and its phenology and ice thickness with meaningful implications for both human activities and the ecosystems they support. Monitoring the seasonal variability and changes in lake ice cover and thickness is crucial for understanding the impacts of climate change; however, traditional in-situ observations have declined over the last few decades, creating a need for innovative remote sensing approaches. Global Navigation Satellite System Reflectometry (GNSS-R) offers a novel, cost-effective method for monitoring lake ice dynamics. Unlike traditional remote sensing techniques, GNSS-R utilizes existing satellite signals (known as signals of opportunity), providing high temporal and spatial resolution data that can be used to detect and analyze lake ice conditions. This thesis investigates the application of GNSS-R for lake ice remote sensing, focusing on the signals' sensitivity to different phases of lake ice and its potential ice detection and the monitoring of lake ice phenology. The thesis begins with an exploration of the fundamentals of GNSS-R and its relevance to lake ice remote sensing. Through understanding the scattering mechanisms involved when GNSS signals interact with lake ice, this research establishes a theoretical framework that supports the subsequent experimental analyses. The research then evaluates the sensitivity of GNSS-R signals, particularly the Signal-to-Noise Ratio (SNR), to various lake ice conditions. Using data from the Cyclone Global Navigation Satellite System (CYGNSS) mission over Qinghai Lake, the thesis examines how SNR values change in response to freeze-up, ice cover, spring melt onset, and breakup. The results demonstrate that GNSS-R can effectively monitor lake ice dynamics with high temporal resolution, making it a valuable tool for tracking the seasonal evolution, inter-annual variability and changes in ice cover. Further investigation is conducted on the potential of hybrid compact polarimetry in GNSS-R for analyzing lake ice cover properties. Data from the Soil Moisture Active Passive Reflectometry (SMAP-R) mission acquired over large Canadian lakes are analyzed to assess the sensitivity of polarimetric GNSS-R signals to ice cover conditions. The study finds that hybrid compact polarimetry enhances the interpretation of GNSS-R data, particularly in distinguishing between different ice and water conditions. In addition, the use of machine learning, specifically random forest, combining several polarimetric parameters, improves the accuracy of lake ice phenology detection compared to using each parameter alone. To deepen our understanding of how lake ice modifies the reflectivity of GNSS signals, a multi-layer scattering regime model is developed and validated against CYGNSS data. The model simulates the interaction of GNSS signals with various lake ice layers and the underlying water interface, providing insights into the complex scattering processes that influence GNSS-R measurements. The successful validation of the model demonstrates its utility for improving the accuracy of lake ice phenology analysis and offers a robust tool for broader cryospheric applications. This thesis contributes to the field of remote sensing by demonstrating the effectiveness of GNSS-R in monitoring lake ice and advancing the understanding of GNSS signal interactions with ice-covered surfaces. The findings highlight the potential of GNSS-R as a low-cost, high-resolution tool for tracking lake ice dynamics, with implications for climate monitoring, weather forecasting and environmental management.