WG-ROUND and DAQUAMAP report
Inga Groehn 1) , Christian Häberli 1),2), and Reinhold Steinacker 1)
1) Dept. of Meteorology and Geophysics, University of Vienna, 1090 Vienna, Austria
2) MeteoSwiss, 8044 Zurich, Switzerland
Introduction and Motivation
Instead of organising a WGROUND-meeting during
the last MAP-Meeting in Schliersee, Germany,
a second workshop on quality control of
meteorological data in MAP was planned and
was finally held in Vienna, Austria on 21st -22nd
of February 2002. 19 persons from 9 institutions
in 6 countries participated.
As the DAQUAMAP-project is still going on until
the end of this year, data providers and users of
the quality control results had the opportunity
during this workshop to express their needs and
make comments. But also the possibility to exchange
experience about the use of various
quality control tools has been used.
In the following a brief summary of the scientific
presentations and the main results of the Working
Group discussions will be presented.
Short summary of scientific presentations (in chronological order)
The first scientific presentation about MAP-SOP
Alpine Precipitation Analysis (see for more information:
FreiHaeller.pdf) was given by Esther Häller from
the Institute for Atmospheric and Climate Science at the ETH
in Zurich, Switzerland (MAP Data Centre). The
daily precipitation observations 06-06 UTC from
the MAP database in a high-resolution rain-gauge
network (about 5000 stations) over the
SOP were analysed and the results outlined. A
new version of MAP-SOP Alpine Precipitation
Analysis was announced and will be done again
in the future by Christoph Frei and Esther
Häller. Then e.g. ITAMAP2.0 1 and HZB 2 (about
880 stations more than in the present version)
will be included.
An area quality control was presented by Radim
Tolasz from the Czech Hydrometeorological Institute
in Ostrava – Poruba. The tool used is CLIDATA
and results are visualised on a GIS. The
method itself (catchword: empirical probability)
and its advantages were presented and examples
were shown.
The spatial quality control method for temperature
data in the Czech Republic 1961-2000 was
described by Vít Kveton, Prague.
The focus of the whole afternoon session was on
DAQUAMAP. At first, Christian Häberli gave a
presentation about the method which is applied
in DAQUAMAP (and VERA). Then, the two DAQUAMAP-
homepages were presented. The two
homepages are identical with two exceptions:
The homepage at http://www.univie.ac.at/IMG-Wien/
daquamap/ includes information for the
data providers. There they have the possibility
to check the DAQUAMAP-results of their institute
before they are published on the other
DAQUAMAP-homepage in the password protected
MAP area on the MDC (http://www.map.meteoswiss.ch/mm-doc/daquamap/daquamain.html).
In some cases very helpful additional
comments from the data providers are included.
In the second half of the workshop, the
participants checked out the DAQUAMAP-results
of their institution. This was a good occasion
to ask questions and discuss the results
thoroughly.
In the following discussion it was proposed to
have a special workshop about the interpretation
of the DAQUAMAP results of the SOP for
the regional weather services in Northern Italy.
Esther Häller from the MAP Data Centre gave a
lecture on the status of the SOP data base. (For
information see http://www.map.meteoswiss.ch/sop-doc/sop_info/sop_info.htm; ->Surface-Data).
The data base contains a very large and heterogeneous
collection of data from many different
data providers. As a consequence there exists a
high variation in the quality of the data. But
only a superficial and formal quality check of
the data is done by the staff at MDC. It was
pointed out that it is not the objective of the
MAP Data Centre to perform a quality control.
For routine network data this is rather the task
of DAQUAMAP. If a database user detects suspicious
data she or he should contact the MDC
rather than the data provider directly. The staff
at MDC will forward the trouble report to the
data provider and ask for remedy action. If a database
user contacts the data provider directly,
it is not assured that detected data problems
will also be corrected in the MDC. Details of the
procedure for error reporting will be documented
by the MDC during the following weeks
and will be made available on the internet.
As a preparatory work for the DAQUAMAP, Inga
Groehn has performed a “data availability
check” in January 2002. For each station the
mean value of the SOP for station pressure,
mean sea level pressure, station temperature
and dew point was calculated. Based on these
results, a “gross error test” revealed that e.g. for
a couple of stations geopotential heights are
stored in the mean sea level pressure field of the
data base. Further on, a “number-of-measurements-“
and a “double-datasets“-test were performed.
There are some stations which have
only very few measurement data in the data
base. In the meantime most of the errors have
already been solved including a reload of GTS
data from the MARS archive at ECMWF. Important
to mention is that now all data of the SOP
are in one single data base table which was not
the case before.
A still open task of WGROUND is that the data
providers check whether all data they delivered
to the MDC are stored correctly in the MAP database.
This work will be carried out during the
next few months as soon as the “data availability
check” is redone. For this purpose it is essential
that data providers apply for a MAP password
from the MAP program office (/map-doc/DataAccessPolicy.htm).
Reinhold Steinacker gave a short overview of the
latest developments of the applied methods in
DAQUAMAP and VERA 3 . A replacement for the
operational finite element 2D version for scalar
and vector quantities by a regular grid version
was outlined.
Summary of Working Group
Discussion
One important objective of WGROUND is to
maintain and complete the inventory of hydro-meteorological
stations in the Alpine Region. A
lot of work and time have already been into the
inventory of hydrometeorological stations in the
Alpine Region. Nevertheless, work is still ongoing
(never ending?), e.g. missing heights, duplicates,…
The suggestion was made to add information
about measuring systems during SOP, e.g. for
rain gauges because this would be helpful for
the interpretation of the analyses.
This would be possible for the German Weather
Service (DWD), Environmental Agency of the Republic
of Slovenia (HMIS), and the Austrian
Central Institute of Meteorology and Geodynamics
(ZAMG). For Italy it is only possible for the
GTS stations. For the Czech Hydrometeorological
Institute (CHMI) and for MeteoSwiss it is
easy feasible. For the HZB 2 (Austria) the measurement
systems are already known. As nobody
from France attended this workshop it was not
possible to make a statement for the French
Weather Service.
Another topic was the quality control of rain
gauges and how to mark suspicious data. There
was agreement that it would be very valuable to
have more information about the checking and
correction practices. Before contacting the data
providers again, it should be checked out, how
much of this information is already available
with other projects (e.g. GPCC).
It was part of the objectives of this workshop to
discuss the future of WGROUND. The participants
agreed that it will be helpful also in the
future to have a platform to exchange experience
and methods for quality control. On the
WGROUND-Homepage there is an (incomplete)
overview of all different quality control checks of
the members (http://www.univie.ac.at/IMG-Wien/daquamap/qualres.htm)
It was recognized, that the results and experiences
of this working group might be also helpful
for future activities or projects. (e.g. MAC 4 ,
UNIDART 5 , ECSN 6 , …). For this purpose a representative
of the WGROUND should act as a
rapporteur between the members of WGROUND
and the members of e.g. ECSN. It was pointed
out that it is not necessary that all institutions
participating in WGROUND are members of
ECSN or EUMETNET 7 .
The project DAQUAMAP will end 2002. The final
results will be:
- A summary for each station (already available);
- A list containing the bias for each station detected
by the DAQUAMAP algorithm;
- Flags for all measurements in the MDB detected
as ‘gross error’ by the DAQUAMAP algorithm.
A project DAQUAMAP II was created in order to
correct the dry bias in the humidity measurements
of the radiosoundings performed during
the SOP.
In the next weeks a more detailed report of the
workshop will be put on the WGROUND-Homepage
http://www.univie.ac.at/IMG-Wien/daquamap/wgroundmain.html.
1 The data set of non-GTS surface hourly gathered during the
MAP field phase
2 Hydrographisches Zentralbüro, Austria
3 Vienna Enhanced Resolution Analysis: http://www.univie.ac.at/IMG-Wien/vera/
4 Mesoscale Alpine Climate
5 A Uniform Data Request Interface
6 European Climate Support Network
7 The network of European Meteorological Services
© MAP Data Centre - May '06 - MAP WebMaster