A review of altimetry waveform retracking for inland water levels

Reliable surface height observations over inland water bodies are useful for understanding the hydrological cycle. Satellite radar altimetry particularly contributed with its long-term archive and minimal cloud interference. Specialized inland water altimetry, developed from oceanography and geodesy, is still being extensively investigated. By synthesizing pioneering studies on “retracking algorithms”, this review demonstrates, from a user perspective, why optimizing conventional retracking is still important and how it can extend reliable historical water level retrieval over more ungauged sites. Numerous unrevealed inland water bodies have small sizes or complex surroundings, posing challenges to maintaining accuracy. Applications have shown that a critical key lies in the retracking correction during range retrieval (uncertainty likely on the order of meters), compared with other corrections (on the order of centimeters or decimeters). From multiple uncertainty factors in range retrieval, signal entanglements from land contamination and off-nadir effects are core issues. We evaluate and compared key strategies from prototype retrackers to improved retrackers, especially the empirical ones optimized for inland waters. Sub-waveform extraction and adjustment for Delay-Doppler modes has advanced range retrieval to a new stage. Four innovative inland-water-compatible retrackers are introduced in detail, with a highlight on their distinct approaches to robustly improve performance. Considering the selection of different data and retrackers in varying scenarios, a synthesis analysis is conducted based on results reported in previous literature. In conclusion, the empirical retracking has been enhanced to offer stable decimeter-level accuracy in intricate landscapes (e.g., small lakes and rivers with varied surroundings). In comparison, the physical retracking has been upgraded to provide greater precision for homogeneous surface in large lakes. For future inland water altimetry, we articulate how additionally retracked results can benefit hydrological applications, and what difficulties would arise when extending study scales.