PowerPointPrsentation

1
Three-dimensional ionospheric imaging using
ground-based and CHAMP GPS observations
validations and results
C. Stolle1, S. Schlüter2, S. Heise3, Ch.
Jacobi1, N. Jakowski2
Contact stolle_at_uni-leipzig.de
1 Institute for Meteorology, University Leipzig
2 DLR Neustrelitz, Institute of Communications
and Navigation 3 GeoForschungsZentrum, Potsdam
2
Outline
1. Assimilation process 2. Validations ?
Comparisons with PLP data ? Comparisons with
ionosonde data 3. Results ? Polar plasma
convection at Oct. 2001 ? The evening of Oct.
30, 2003 4. Conclusions and Outlook
3
1. Assimilation process
Three-dimensional electron density fields in the
ionosphere
Data processing

• Ground-based GPS TEC observations from IGS and
  SOPAC and CHAMP occultation data
• Calibration of GPS TEC observations

• Correction for cycle slips and ambiguity
  Blewitt, 1990
• Instrumental receiver and transmitter bias
• ground-based and transmitter Sardon et al.,
  1994
• occultation Heise et al., 2004

• Correction for plasmaspheric contribution to the
  GPS TEC observations by use of the IRI/GCPM model

4
CHAMP occultation bias
CHAMP zenith antenna bias
Heise Stefan, Three-dimensional reconstruction of
electron density fields based on CHAMP-GPS
observations (Rekonstruktion dreidimensionaler
Elektronendichteverteilung basierend auf
CHAMP-GPS-Messungen), PhD thesis Freie
Universität Berlin, 2002.
CHAMP occultation bias
Heise, S., Stolle, S., Schlüter, S. and Jakowski,
N., Differential Code Bias of GPS receivers in
Low Earth Orbit An Assessment for CHAMP and
SAC-C, CHAMP Mission Results for Gravity and
Magnetic Field Mapping, and GPS Atmosphere
Sounding, Springer-Verlag Berlin Heidelberg
NewYork, Eds Ch. Reigber and H. Lühr, P.
Schwintzer and J. Wickert, 2004.
5
CHAMP occultation bias
What is the instrumental bias?
6
CHAMP occultation bias

• Model-assisted bias estimation
• Select for high elevation angles and polar night
  time regions the absolute error by modeling
  (PIM) is reduced

Condition the
occultation bias is constant during one
day Improvement by using zenith TEC assimilation
results instead of pure modelling
7
CHAMP occultation bias
CHAMP occultation biases for a period of 2002
using CHAMP zenith assimilations into PIM as
background model
Mean bias of the period 14.2 TECU (rms 4TECU)
8
CHAMP occultation bias
Simultaneous TEC observations from the same GPS
satellite signal - zenith bias estimates are
more stable (rms?0.9TECU) - requires
occultation rays of positive elevation angle
9
CHAMP occultation bias
2001 initial phase of CHAMP 2003
occultation bias campaign
Mean bias of the period
13.3 TECU
15.2 TECU
Daily rms ? 1 TECU
10
1. Assimilation process
Three-dimensional electron density fields in the
ionosphere
Assimilation

• TEC assimilation into the background model IRI
  which is combined with the GCPM in the topside
  ionosphere
• Discretization of model space and GPS rays
• Use of the iterative MART algorithm

11

1. Assimilation process

• Initialization with IRI/GCPM

I number of rays J number of
voxels k iteration step
12
1. Assimilation process
13
1. Assimilation process
Application to
Europe North polar region
1000km
80km
14
2. Validations
Comparison to PLP electron density records
15
2. Validations
Comparison to PLP electron density records
4491 samples
Initial model (IRI/GCPM) 104.1
Pure IGS data assimilations 98.4
16
2. Validations
Comparison to PLP electron density records
4491 samples
Pure IGS data assimilations 98.4
IGS and CHAMP data assimilations 86.3
17
2. Validations
Comparison to ionosonde data

• Midlatitude NmF2/HmF2 electron density and layer
  height value of the F2-peak
• Available by the SPIDR network
  (http//spidr.ngdc.noaa.gov/ spidr/index.html)

18
2. Validations
Comparison to ionosonde data
47 samples
IGS and CHAMP data assimilations 22.3 mean
quatr. dev. 1.61011 m-3
Initial model (IRI/GCPM) 27.5 mean quatr.
dev. 2.31011 m-3
19
2. Validations
Comparison to ionosonde data
38 samples
IGS and CHAMP data assimilations 17.6 median
diff. -5.5 km (rms25km)
Initial model (IRI/GCPM) 18.9 median diff.
-17.5 km (rms23km)
20
3. Results
1) Night of Oct. 31 Nov.1, 2001 observation
of antisunward polar plasma convection 2) The
evening of Oct. 30, 2003 included in the large
magnetic storm period last fall
21
3. Results Night of Oct. 31 Nov.1, 2001

• Vertical integrated elec-tron density from
  assi-milation ? vertical TEC
• IRI/GCPM predictions
• Similar for same local times during the night
• White cross geomag-netic pole

22
3. Results Night of Oct. 31 Nov.1, 2001
23
3. Results Night of Oct. 31 Nov.1, 2001
PLP electron density along the CHAMP orbit at
about 440km altitude
24
3. Results Night of Oct. 31 Nov.1, 2001

• Plasma drift measurements by the SuperDARN radar
  network
• Indication of the 23 TECU isoline

25
3. Results The evening of Oct. 30, 2003
Assimilation results (IGS /SOPAC)

• Vertical integrated TEC maps
• High ionospheric variability 10 min resolution
• Red cross CHAMP data included

Animated GIF soon available at http//www.uni-lei
pzig.de/stolle/301003/
26
3. Results The evening of Oct. 30, 2003
EISCAT campaign

• EISCAT group of the Laboratoire de Planétologie
  de Grenoble
• Oct. 29 Nov. 2, 2003
• First presented by Barthélemy et al., 2003
• Electron density profiling along radar beams
• Svalbard 16.5 E / 78.15N Tromsø 19.2E
  / 69.6N

10 min resolution ? Take care of the F region
variability (seen by GPS) !!!
27
3. Results The evening of Oct. 30, 2003
Svalbard 2010 UT
Tromsø
28
3. Results The evening of Oct. 30, 2003
Svalbard 2040 UT
2100 UT
Tromsø
21 UT The plasma accumulation reaches northern
Europe
29
3. Results The evening of Oct. 30, 2003
Svalbard 2110 UT
Tromsø
30
3. Results The evening of Oct. 30, 2003
Svalbard 2140 UT
2200 UT
Tromsø
22 UT low TEC values above northern Europe (IMF
Bz north)
31
3. Results The evening of Oct. 30, 2003
Svalbard 2210 UT
Tromsø
32
3. Results The evening of Oct. 30, 2003
Svalbard 2240 UT
Tromsø
33
3. Results The evening of Oct. 30, 2003
Svalbard 2310 UT
2330 UT
Tromsø
2310UT High electron density values in auroral
latitudes but low TEC values above 70
34
3. Conclusion and Outlook

• The assimilation of GPS TEC measurements improves
  the electron density background by about 20 in
  average.
• Assimilation results provide good potential to
  image large scale plasma distributions, e.g.
  ionospheric irregularities which are not
  represented by climatologic models

• Promising issue with increasing number of GPS
  observations (additional LEO satellite missions
  like COSMIC, ... and increasing number of GPS
  groundbased receivers the assimilation of
  further ionospheric data (ionosondes, in situ
  satellite data, e.g., PLP )
• Sophisticated models, developed computer
  techniques and new algorithms enhance the results
  of the assimilation process

35
References Blewitt, G., An automatic editing
algorithm for GPS data, Geophys. Res. Lett.,
17(3), 199-202, 1990. Barthélemy, M., Culot, F.,
Kofman, W., Lathuillére, C., Lilensten, J.,
Perrin, J.-M., Pibaret, B., Simon, C., Thuillier,
G. and Van Eyken, T., French EISCAT campaign
from October 29 to November 2, 2003 Observations
from EISCAT UHF at Tromsø, and ESR and EPIS
interferometer at Svalbard, AGU Fall Meeting, San
Fransisco, USA, Dec., 2003. Heise, S., Stolle,
C., Schlüter, S. and Jakowski, N., Differential
Code Bias of GPS receivers in Low Earth Orbit An
Assessment for CHAMP and SAC-C, CHAMP Mission
Results for Gravity and Magnetic Field Mapping,
and GPS Atmosphere Sounding, Springer-Verlag
Berlin Heidelberg NewYork, Eds Ch. Reigber and
H. Lühr, P. Schwintzer and J. Wickert,
2004. Jakowski, N., Heise, S., Wehrenpfennig, A.,
Schlüter, S. and Reimer. R., GPS/ GLONASS-based
TEC measurements as a contributor for space
weather forecast, J. Atmos. Solar-Terr. Phys.,
64729-735, 2002. Sardón, E., Ruis, A. and
Zarraoa, N., Estimation of the transmitter and
receiver differential biases and the ionospheric
total electron content from Global Positioning
Systems observations, Radio Sci., 29(3), 577-586,
1994.
Thanks M. Förster, H. Lühr, EISCAT group of the
Laboratoire de Planétologie de Grenoble, IGS, GFZ
Potsdam, SOPAC, NCAR for the opportunity to this
talk !!