Title: Modeling Errors in GPS Vertical Estimates
1Modeling Errors in GPS Vertical Estimates
- Signal propagation effects
- Signal scattering ( antenna phase
center/multipath ) - Atmospheric delay ( parameterization, mapping
functions ) - Unmodeled motions of the station
- Monument instability / local groundwater
- Loading of the crust by atmosphere, oceans, and
surface water - One-sided geometry increases vertical
uncertainties relative to horizontal and makes
the vertical more sensitive to session length
2(No Transcript)
3(No Transcript)
4Modeling Errors in GPS Vertical Estimates
- Signal propagation effects
- Signal scattering ( antenna phase
center/multipath ) - Atmospheric delay ( parameterization, mapping
functions ) - Unmodeled motions of the station
- Monument instability / local groundwater
- Loading of the crust by atmosphere, oceans, and
surface water
5Antenna Phase Patterns
6Modeling Antenna Phase-center Variations (PCVs)
- Ground antennas
- Relative calibrations by comparison with a
standard antenna (NGS, used by the IGS prior to
November 2006) - Absolute calibrations with mechanical arm (GEO)
or anechoic chamber - May be depend on elevation angle only or
elevation and azimuth - Current models are radome-dependent
- Errors for some antennas can be several cm in
height estimates - Satellite antennas (absolute)
- Estimated from global observations (T U Munich)
- Differences with evolution of SV constellation
mimic and scale change - Recommendation for GAMIT Use latest IGS
absolute ANTEX file (absolute) with AZ/EL for
ground antennas and ELEV (nadir angle) for SV
antennas
7Top PBO station near Lind, Washington. Bottom
BARD station CMBB at Columbia College, California
8Left Phase residuals versus elevation for
Westford pillar, without (top) and with (bottom)
microwave absorber. Right Change in height
estimate as a function of minimum elevation angle
of observations solid line is with the
unmodified pillar, dashed with microwave absorber
added
From Elosequi et al.,1995
9Antenna Ht
0.15 m
0.6 m
Simple geometry for incidence of a direct and
reflected signal
1 m
Multipath contributions to observed phase for
three different antenna heights From Elosegui
et al, 1995
10(No Transcript)
11Modeling Errors in GPS Vertical Estimates
- Signal propagation effects
- Signal scattering ( antenna phase
center/multipath ) - Atmospheric delay ( parameterization, mapping
functions ) - Unmodeled motions of the station
- Monument instability / local groundwater
- Loading of the crust by atmosphere, oceans, and
surface water
12(No Transcript)
13GPS adjustments to atmospheric zenith delay for
29 June, 2003 southern Vancouver Island (ALBH)
and northern coastal California (ALEN). Estimates
at 2-hr intervals.
14Effect of Neutral Atmosphere on GPS
Measurements Slant delay (Zenith Hydrostatic
Delay) (Dry Mapping Function)
(Zenith Wet Delay) (Wet Mapping
Function) ZHD well modeled by pressure (local
sensors or global model, GPT) Analytical
mapping functions (GMF) work well to 10 degrees
ZWD cannot be modeled with local temperature
and humidity - must estimate using the wet
mapping function as partial derivatives  Because
the wet and dry mapping functions are different,
errors in ZHD can cause errors in estimating
the wet delay (and hence total delay) .
15Percent difference (red) between hydrostatic and
wet mapping functions for a high latitude (dav1)
and mid-latitude site (nlib). Blue shows
percentage of observations at each elevation
angle. From Tregoning and Herring 2006.
16Difference between a) surface pressure derived
from standard sea level pressure and the mean
surface pressure derived from the GPT model. b)
station heights using the two sources of a priori
pressure. c) Relation between a priori pressure
differencesand height differences.
Elevation-dependent weighting was used in the GPS
analysis with a minimum elevation angle of 7 deg.
17Differences in GPS estimates of ZTD at Algonquin,
Ny Alessund, Wettzell and Westford computed using
static or observed surface pressure to derive the
a priori. Height differences will be about twice
as large. (Elevation-dependent weighting used).
18Modeling Errors in GPS Vertical Estimates
- Signal propagation effects
- Signal scattering ( antenna phase
center/multipath ) - Atmospheric delay ( parameterization, mapping
functions ) - Unmodeled motions of the station
- Monument instability / local groundwater
- Loading of the crust by atmosphere, oceans, and
surface water
19(No Transcript)
20Modeling Errors in GPS Vertical Estimates
- Signal propagation effects
- Signal scattering ( antenna phase
center/multipath ) - Atmospheric delay ( parameterization, mapping
functions ) - Unmodeled motions of the station
- Monument instability / local ground water
- Loading of the crust by atmosphere, oceans, and
surface water
21Atmosphere (purple) 2-5 mm Snow/water (blue)
2-10 mm Nontidal ocean (red) 2-3 mm
Annual vertical loading effects on site
coordinates From Dong et al. J. Geophys. Res.,
107, 2075, 2002
22Vertical (a) and north (b) displacements from
pressure loading at a low-latitude site (S.
Africa). Bottom is power spectrum. From Petrov
and Boy (2004)
23 Vertical (a) and north (b) displacements from
pressure loading at a mid-latitude site
(Germany). Bottom is power spectrum.
24Spatial and temporal autocorrelation of
atmospheric pressure loading
From Petrov and Boy, J. Geophys. Res., 109,
B03405, 2004
25Atmosphere (purple) 2-5 mm Snow/water (blue)
2-10 mm Nontidal ocean (red) 2-3 mm
Annual vertical loading effects on site
coordinates From Dong et al. J. Geophys. Res.,
107, 2075, 2002
26(No Transcript)
27Station height estimates for Rio Grande,
Argentina, using pressure from height-corrected
STP, GPT and actual observations (MET). Dashed
black line shows observed surface pressure pink
line shows atmospheric pressure loading
deformation (corrected for in the GPS analyses) ,
offset by 2.07 m.
28(No Transcript)
29Correlation between estimates of height and
zenith delay as function of minimum elevation
angle observed (VLBI, from Davis 1986)
30Uncertainty in estimated height as function of
minimum elevation angle observed (VLBI, from
Davis 1986 dotted line with no zenith delay
estimated)
31(No Transcript)
32Height (red simulated black estimated) and ZTD
(green simulated blue estimated) errors versus
latitude as a function of error in surface
pressure used to calculate the a priori ZHD.
Uniform 10 mm data weighting applied.
33Height (black/blue) and ZTD (red/green) errors at
Davis, Antarctica, for different elevation cutoff
angles as a function of error in surface pressure
used to calculate the a priori ZHD.. Results
shown for both elevation-dependent (blue and red
results) and constant data weighting (black and
green).