[IGSMAIL-6287] CODE analysis model changes (VMF/HOI)

[Stefan Schaer]

Author: Stefan Schaer, Rolf Dach, Simon Lutz


Dear IGS colleagues,

Starting with GPS week 1604, the following model changes become
effective for CODE's IGS final analysis products:

- Use of VMF1 (instead of GMF) as troposphere mapping model.
- Consideration of three higher-order ionosphere correction terms:
  - second order term,
  - third order term,
  - ray bending/curvature term (according to IERS Conventions 2010).
  Note 1: A scaling parameter responding to each HOI correction term is 
set up
  (enabling it to turn on/off individually HOI correction terms at NEQ 
level).
  Note 2: Our IGS analysis results are referred to scaling parameter 
value 1.

See also IGS Report 19068:

http://igscb.jpl.nasa.gov/pipermail/igsreport/2010-October/019081.html


REMARKS concerning VMF:

The ECMWF-based hydrostatic part of the troposphere delay is mapped
using dry-VMF1 coefficients; the wet part is estimated using wet-VMF1
coefficients. The VMF1 information is interpolated on the basis of
6-hourly global grids provided by Johannes Boehm from TU Vienna. An
additional height correction is applied according to Kouba, JoG, 2008.

We intend to activate VMF for troposphere modeling for our rapid and
ultra-rapid GNSS analysis soon.


REMARKS concerning higher-order ionosphere (HOI) correction terms:

The HOI correction terms are computed on the basis of CODE's global
ionosphere map (GIM) information and geomagnetic field values from the
latest (11th) generation of International Geomagnetic Reference Field
(IGRF) series of models:

http://www.ngdc.noaa.gov/IAGA/vmod/
http://www.ngdc.noaa.gov/IAGA/vmod/igrf11.f

It should be mentioned that a two-term HOI correction is effective
already since 15-Feb-2010 for our rapid and ultra-rapid GNSS analysis
products (using CODE's rapid and predicted GIM products). Since
24-Sep-2010, all three HOI correction terms (including the ray bending
term) are considered in these analysis lines.

The ionosphere activity is still comparably moderate. At present, just
the second-order HOI term may be confirmed in a significant way (with
regard to the estimation of corresponding scaling parameters). The
following table quantifies the effect of the various HOI correction
terms (as computed for our 12-Oct-2010 rapid analysis results):

HOI terms    | RMS differences w.r.t. "no-HOI" reference solution (in mm)
             | Station coordinates (N/E/U/3D) | Orbit positions (R/A/O/3D)
-------------+--------------------------------+---------------------------
All          |  0.39   0.21   0.26   (0.30 )  |  1.0    2.5    1.5   (1.8)
Second order |  0.40   0.22   0.25   (0.30 )  |  1.0    2.5    1.5   (1.8)
Third order  |  0.04   0.03   0.08   (0.05 )  |  0.2    0.6    0.2   (0.4)
Ray bending  |  0.02   0.02   0.02   (0.02 )  |  0.0    0.2    0.1   (0.1)

Remark: The RMS values given above were computed compensating for a
translation vector. The most significant shift was found for the
Z-component, with 1.2 mm (stations) and 1.8 mm (orbits), respectively.


We will submit an updated version of the CODE analysis questionnaire
(CODE.ACN) to the IGS CB.

Best regards,

The CODE AC Team