Validating time series of a combined GPS and MERIS Integrated Water Vapor product
Roderik Lindenbergh(1), Maxim Keshin(2), Hans Van der Marel(1) and Siebren De Haan(3)
(1) Delft University of Technology, P.O. Box 5058, 2600 GB Delft, Netherlands
(2) Finnish Geodetic Institute, Geodeetinrinne 2, FI-02431 Masala, Finland
(3) KNMI Royal Meteorological Institute, P.O.Box 201, 3730 AE De Bilt, Netherlands
Water vapor is the atmosphere's dominant greenhouse gas,
but it is still a challenge to determine its spatial-temporal
distribution at sufficient resolution from one single type of
contemporary meteorological instrument. It is possible
however to retrieve and consecutively combine water
vapor estimates from complementary satellite systems.
At ground stations from the world wide
Global Positioning System (GPS), the zenith Integrated Water
Vapor (IWV) is derived from estimates of GPS signal
travel time delay in troposphere. This derivation
results in relative good measurements with high temporal,
e.g. 1 hour, but low spatial resolution, e.g.
tenths of kilometers over Western Europe.
The Medium Resolution Imaging Spectrometer (MERIS) on the
Envisat satellite retrieves IWV by comparing radiances in two
spectral bands in the near infrared, with a
maximum spatial resolution of 300 m. Its temporal resolution
is restricted to 3 days. Moreover, MERIS only provides useful
IWV observations under clear sky conditions.
The topic of the proposed paper is the validation of a method
that creates time series of hourly IWV predictions
by combining complementary MERIS and GPS IWV observations, .
For this purpose a region of interest is considered of about
300 by 400 km covering The Netherlands. The method takes
GPS and MERIS observations of the period
March-May 2006 as input and creates a two months
time series of hourly IWV maps at about 10 km
spatial resolution. The quality of the integrated
water vapour product is assessed by cross-validation:
at each GPS ground station location two time series
of IWV values are compared. One time series consists
of the direct GPS IWV observations from that ground station;
the other time series is extracted from a combined
GPS-MERIS IWV product, created using all available observations,
except for the GPS IWV observations from the ground station at hand.
For the proposed data combination, GPS IWV estimates of
about 40 ground stations in the region of interest will be used.
These IWV estimates, available at at least hourly
intervals, were processed by the KNMI Royal Meteorological Institute.
The MERIS data product used here is MER_RRC_2P, the
Level 2 Reduced Resolution Cloud and Water vapor product.
There are about 60 MER_RRC_2P data sets available in the
period half March - half May 2006 that have at least
partial overlap with the region of interest.
The method of GPS and MERIS data fusion has been described
in detail in . In short, spatial and temporal
correlations between the different available IWV observations
are taken into account in a Co-Kriging approach. It has
been shown that such an approach not only leads
to a, in a specific mathematical sense, optimal
spatio-temporal interpolation result, but also to
a description of the quality of each interpolated value
in the space-time domain in the form a Kriging variance.
The proposed cross-validation method will also give insight in
the applicability of this Kriging variance as variance
values can be compared to differences between
interpolated and directly observed IWV estimates.
Finally, the additional value of incorporating MERIS
IWV will be quantified by comparing time series of interpolated
IWV estimated based on GPS IWV observations only to time series
that incorporate MERIS IWV as well.
 R. Lindenbergh, M. Keshin, H. van der Marel and S. de Haan, (2007).
Towards sequential water vapor predictions based on
time series of GPS and MERIS observations.
In: Proceedings ESA Envisat Symposium, Montreux, Switzerland, 2007.
 R. Lindenbergh, M. Keshin, H. van der Marel and R. Hanssen, (2008).
High resolution spatio-temporal water vapor mapping using
GPS and MERIS observations.
International Journal of Remote Sensing, 29(8), pp. 2393-2409.
Keywords: ESA European
Space Agency - Agence spatiale europeenne,
observation de la terre, earth observation,
satellite remote sensing,
teledetection, geophysique, altimetrie, radar,
chimique atmospherique, geophysics, altimetry, radar,