Water level measurement with a low-cost smartphone using GNSS-IR: an over 2-year study case in Buenos Aires, Argentina

This work evaluates the capability of a low-cost smartphone to measure variations in water level caused by astronomical tide and storm surges along the coast of Buenos Aires, Argentina, over a period of 27 months, employing the GNSS-IR technique. To achieve this, the smartphone Signal to Noise Ratio-derived heights are contrasted with the heights acquired from a co-located tide gauge of the Argentine Naval Hydrographic Service. A novel methodology is applied to obtain the water levels from the smartphone observations by identifying density maximums in reflections calculated using the Lomb-Scargle periodogram. The density maximums are obtain using LOWESS regression, generating a water level series sampled every 5 min referenced to the Chart Datum. Water levels were calculated for 97.7% of the time, exhibiting a standard deviation of 0.042 m against tide gauge observations. Differences between the daily (monthly) mean levels of the analyzed series showed a standard deviation of 0.016 m (0.007 m), and the amplitude differences for the tidal harmonic constituents were smaller than 0.011 m. While the results found in this work show that these devices cannot compete with traditional tide gauges, they can complement them or be used independently to monitor water level changes in regions where installing and maintaining conventional tide gauges is challenging.