----------------------------------------------------------------

Summary of the first WP3 meeting

Date and Place: POL, Bidston, UK, 24-25 March 2003

----------------------------------------------------------------

I wish to thank you for a pleasant, positive and fruitful workshop last
week. Also many thanks to Lesley for hosting the workshop.

Based on the presentations and the discussion we agreed to modify the
strategy and methodology for the analyses of the tide gauge data, so that
we will
- use monthly averages - minimum 20 years of time series,
- remove trend
- remove seasonal variation using the 12 mean values for January, February,
March, etc... respectively.
- filling gaps if it can be done in a quite stable/comfortable way
- compute power spectrum using FFT and identify - if any - dominant
frequencies
- compute auto-correlation function and find times/lags of (first) 50%
correlation, first zero crossing. If any - identify peaks.
- for a group of stations an EOF analysis should be applied to identify
dominant modes.

According to the guidelines each partner should analyse
1) control stations
2) stations in an European region
3) stations in a global region

We also discussed the role of the web-pages and agreed that we should use
the web to distribute information about methodology, progress and results.
Hanne is willing to help and compile the information. She will distribute
guidelines for preparing inputs.

Hans-Peter showed some interesting results from earlier analyses where
running periodegrams were computed to evaluate the temporal changes in the
frequencies. Those results might be of help when the new results are
evaluated and may go on the web as well.

To reach the goals of Task 3.1 in time we agreed on the following action
for the analyses of time gauge data:

1) Revise and distribute the guidelines before 4/4 (Mikis, Per)
2) Describe methodology for the analyses and prepare it for the web before
30/4 (Each Partner)
3) Analyse control stations and prepare results for the web before 30/4
(Each Partner)
4) Analyse stations in a European region (assigned by Mikis) and prepare
results for the web before 30/6 (Each Partner)
5) Analyse stations in a global region (assigned by Mikis) and prepare
results for the web before 30/6 (Each Partner)


Ole Andersen presented the plan for the analyses of the altimetry. They
agreed to
- use Pathfinder data with NO IB-correction applied
- remove tilt at each location
- remove seasonal variation using 12 mean values for Jan, Feb, Mar......
respectively
- compute power spectrum and detect dominant frequencies
- compute and evaluate auto-correlation function - 50% correlation time and
time of first zero X-ing
- carry out EOF analyses in regions, in European seas and globally.

Those analyses should be accomplished before 30/6  and the results should
go on the web.

I guess that we are busy..... Good luck

Best regards - Per