The tropospheric delay estimation for EPN/IGS permanent stations

Witold Rohm Jarosław Bosy
Institute of Geodesy and Geoinformatics
Wroclaw University of Environmental and Life Sciences
[email protected]; [email protected]
The tropospheric delay estimation for
EPN/IGS permanent stations located in
the Sudety Mountains and in the
adjacent areas.
Aim:
• Use of the GPS observation to build the spatio –temporal
(4D) troposphere tomography model on the Sudety
Mountains.
Data sources:
• meteorological parameters,
• GPS observations.
Products:
• Spatial and temporal distribution of water vapour in the
troposphere over GPS network.
GPS meteorology: MHES Grant
• Evaluation of EPN ZTD product as a proper
reference for the Sudety subnetwork.
• Meteorological GPT model accuracy
assessment
• Meteorological data realiability assesment
GPS Meteoreology: Stage I
Period: year 2006
Meteo stations: Wroclaw II, Śnieżka, Cervena, Praha-Kbely
GPS stations: WROC SNEC BISK GOPE
• T temperature [0C,0K]
• p pressure [hPa]
• h humidity [%,hPa]
Meteorological parameters
• pressure: 5 hPa
• temperature: 3 0C
– GPT model
• pressure: 0.2 hPa
• temperature: 0.2 0C
• humidity: 2%
– Polish Meteorological Service (IMGW) and Czech
Hydrometeorological Institute (CHMI)
• pressure: 0.3 – 0.5 hPa
• temperature: 0.3 – 0.5 0C
• humidity: 3 - 5%
– meteo packs mounted close to GPS antenna
Meteorological parameters: Sources
correcting 1
hour
observations
with mean day
difference
compensate
for sensor
displacement
interpolate to
similiar
resolution (1 h)
Meteo labs on
GPS stations
calculating
corrections
24h
average
calculate GPT
model
values
(p,T)
Model
„mean sea level”
plotting
position of synoptic station
compensate
for synop
station vertical
distance
compensate for
downscalling to
mean sea level
on some synop
stations
IMGW CHMU
synoptic stations
60
300
30
13 m
124 m
position of pressure sensor
16 m
SNEC
GPSanntena
anntena
position
GPS
BISK
30 position
GOPE
WROC
Station
meteo files
time
resolution [s]
data interpolation comparison and calibration
Meteorological parameters:
0.94
-0.53
-0,07
24.37
GOPE
SNEC
BISK
6.14
0.40
5.02
0.40
-1.03
0.21
-2.98
0.74
4.20
0.38
2.80
0.60
error
bias
bias
error
temperature [0C]
pressure [hPa]
WROC
Station
Meteorological parameters: Results
Berg equation
5.225
h ≈ 400[m]
p = p + [1 − 0.0000226 ⋅ (h − ht )]
••Synoptic
Lack
data
stationssensor
heightslarge
not defined
well
•BISKofpressure
bias
[hPa]
Meteorological parameters: Problems
-2.64
0.14
0.32
-48.88
GOPE
SNEC
BISK
bias
9.14
6.58
4.02
7.68
error
pressure [hPa]
WROC
Stations
-2.22
-1.85
0.74
-1.36
bias
4.32
3.92
6.94
3.75
error
temperature [0C]
GPT accuracy assasement
Meteorological parameters:
Zenith delay corrections (1h each
station), wet-Niell MF. HG (tilting)
for 24 hours.
ZTD = ZWD + ZHD

 1255
ZWD = 0.002277 ⋅  o + 0.05  ⋅ e0 [hPa]

 T0 [ K ]
ZHD =
 1 
1 − 0.00266 ⋅ cos 2ϕ − 0.00028 ⋅   ⋅ he
 km 
 m 
0.0022767 ⋅ 
⋅ Po

hPa


• Tropospheric delay - Saastamoinen eqations:
Dry Niell
ZWD (each 1h) + Wet Niell MF
Dry Saastamoinen + dryNiell MF
BKG since 1320
BKG since 1400
ZTD (each 1h). Wet Niell MF, HG
(tiliting) for 24 hours
Estimates
Dry Niell + MF
Apriori model
OLG since 1397
Local Analysis Center
\examples\
• Euref product means weekly from all LAC’s:
Tropospheric Delay: Sources
•• Station
Station GOPE
WROC
BISK
SNEC
Tropospheric Delay: Comparison
•
•
•
•
BISK station pressure data failure,
GOPE too few reference data,
GPT reliable global model,
ZTD from meteo data shows general
agreement with EUREF mean solution.
Conclusion
Thank you for your attention!