V' Zavorotny

1
Potential Signals from Ocean Reflections

• V. Zavorotny
• NOAA/Earth System Research Laboratory

2
Previous works on using GPS reflections for
coastal ocean observations

• K. Anderson, (2000), Determination of water level
  and tides using interferometric observations of
  GPS signals, Journal of Atmospheric and Oceanic
  Technology, vol.17(8), pp. 11181127.
• K. Anderson, (1997), Method for remotely
  detecting tides and the height of other surfaces,
  US Patent 5703594.
• Anderson suggested to use correlator output (SNR)
  of the GPS receiver as a signal for ocean level,
  tide observations.
• Anderson used 20-deg tilted antennas to increase
  a reflected signal strength.
• He observed interferometric fringes and retrieved
  from them antenna height above the ocean surface
• The actual shape of antenna gain was not
  accounted for in his simulations.
• In his calculations he neglected the effect of
  surface roughness.
• At the same time, he estimated tropospheric
  effects on visibility of interferometric fringes
  and found it negligible for the typical
  conditions of observation.

3
Andersons results
Measurements (crosses) made at SIO (19251953 UTC
23 Sep 1998). The best-fit correlation to
theoretical model is indicated by the dark solid
line and corresponds to an antenna height of
10.30 m above the water surface.
One GPS receiver (Allen Osborne Associates
SNR-8000 TurboRouge) was installed at the pier of
Scripps Institution of Oceanography, La Jolla,
CA.
4
Andersons results (contd)
Time series comparison for SIO measurements. The
standard error in the comparison of GPS
measurements to the in situ SIO tide gauge is
less than 12 cm, which is less than the typical
standard deviation of the tide gauge
measurements.
Wallops Island, VA, test site and radials to GPS
satellites rising or setting on the horizon.
Comparison of GPS measurements made at Wallops
Island, VA, from 4 to 18 Mar 1998, to the scaled
Wachapreague, VA, tide gauge reference. Excluding
three outliers, the standard deviation in
comparing to the scaled Wachapreague data is 10.2
cm.
5
Geometry of the problem and two phenomena of
interest
Phase delay between direct and reflected signals
causes interference
Surface roughness causes decorrelation between
these two signals it destroys interference
Direct GPS signal
RHCP
?
LHCP
Reflected GPS signal
N
?
h
?
?
?
?
l
h
?L
Rayleigh parameter
Phase delay
6
Simplified equations based on geometric optics
Complex amplitude of the received signal at a
given polarization
G , antenna gains at positive and negative
elevation angles
, decorrelation factor due to surface roughness
R, Fresnel reflection coefficient at given
polarization
, total received power.
A more accurate approach would make use of a
surface diffraction integral in the Kirchhoff
approximation but GO gives a qualitatively
correct result.
7
GPS reflections for the case of Harvest Oil
Platform
14 GPS satellites were available at Harvest Oil
Platform location on Oct 26, 2008
Elevation angles of GPS satellites between 0000
UTC and 0100 UTC
The antenna height is assumed to be at 50 m
height above the WGS 84 ellipsoid.
8
Geometry of GPS specular reflection zones for
the case of Harvest Oil Platform
California
HOP
500 m
50 km
Specular reflection tracks for high-elevation
satellites
Specular reflection tracks for low-elevation
satellites
9
Typical gain patterns of GPS antennas
L1 1.5754 GHz
L2 1.2276 GHz
Red right-hand circular polarization Blue
left-hand circular polarization
10
Sensitivity of fringes phase to a sea level (no
roughness) for L1and L2 choke ring antennas
L1
L2
The case of the the oil platform antenna height
above the sea level is about 50 m. The fringes
for h 50 m and h 50.1 m start to separate.
11
Sensitivity of interference fringes to sea
roughness for L1and L2 choke ring antennas (h
50 m)
L1
L2
Higher winds lead to stronger surface roughness
which leads to a decorrelation of reflected
signal and, as a result, to the suppression of
fringes at higher elevation angles.
12
Sensitivity to a sea level (without sea
roughness) for L1and L2 choke ring antennas at H
5 m
L1
L2
The case of the low-platform GPS antenna height
above the sea level is about 5 m. The fringes for
h 5 m and h 5.1 m are well separated.
13
Sensitivity of interference fringes to sea
roughness for L1and L2 choke ring antennas (h
5 m)
L2
L1
Higher winds lead to the suppression of fringes
at higher elevation angles.
14
Conclusion

• Previous works on using GPS reflections for
  coastal ocean observations show 12-cm altimetric
  accuracy, no significant effect of tropospheric
  or ionospheric refractive effects. At the same
  time, no attempts were made to measure surface
  roughness/sea state/wind using this GPS
  reflection technique.
• We propose to revitalize this technique for
  purposes of coastal ocean observing systems using
  standard geodetic GPS receivers and the SNR as a
  probing signal.
• Currently we are developing a reliable modeling
  of both effects of sea level changes and surface
  roughness on GPS signal interference .
• Simple algorithms of sea level and surface
  roughness/wind retrievals can be expected based
  on the developed theoretical model.
• Analysis of retrieval errors due to the finite
  observation time and stochastic nature of the
  wind-generated surface gravity waves, comparisons
  with field measuremnts would be the next step in
  proving feasibility of this technique.