How does a GPS Tide Gauge work?

Many GPS tide gauges consist of two antennas and receivers. One antenna (RHCP) is installed like normal geodetic antennas - i.e. pointing to zenith. This antenna/receiver is used to track the direct signal. A second antenna is pointed down - towards the sea surface. This antenna (usually designed for LHCP signals) is meant to primarily pick up the reflected signal. 


Here we examine the potential to measure sea surface reflections using only a regular geodetic antenna/receiver. A geodetic antenna is designed to preferentially receive energy from the direct signal, but inevitably, reflected energy is also received. The SNR data track the interference between the direct and reflected signals. 

 

The site we use in this example, SC02, is near the water, but provides unobstructed views only at certain azimuths. Only data from the four satellite tracks shown below will be analyzed.

The SNR data for four tracks are superimposed on tide gauge data from a nearby NOAA sensor.  Because of the ground surface near the GPS site, only the “low elevation” data from 5-15 degrees above the horizon are used. These elevation angles reflect from the water. The frequency of the oscillations seen in these data correspond to different water levels.  When the water level is high, the oscillations are lower frequency and vice versa.  We currently extract the primary frequency using a Lomb Scargle Periodogram (see below), but plan to modify it so that we can extract sea level and sea level rate.

What does that interference pattern look like?  The simulation below shows you hypothetical reflections from a sea surface that are 4 meters and 7 meters below the antenna.

Peak frequencies are converted to an equivalent water level (blue circles) and then plotted below. A NOAA tide gauge ~300 meters away is shown in black.

See Larson et al., 2012 for more details.