Temperature, Pressure and Potential Vorticity Fields at Theta Surfaces from the European Centre for Medium-Range Weather Forecasting

The pressure, temperature and potential vorticity (PV) fields are used for
European stratospheric ozone research projects, such as EASOE and SESAME
as well as other EU funded projects.

The PV, pressure and temp. on the different isentropic levels are
calculated from the ECMWF T213 initialized analysis extracted at a 1.125
degree lat-lon grid.

The trajectories are 10 day backward analyses ending at
a large number of stations and grid points in Europe,
Asia and the Arctic. The trajectory data contain information
on the geographical position, temperature and PV along
the trajectory. The time step is 2 hours.

For both the fields and trajectories there are data for
12 UT every day from early November through April starting Nov. 1991.

The PV and temperature on the different isentropic levels are calculated
from the ECMWF T213 initialized analysis extracted as a 1.25 degree
lat-lon grid.

PV is calculated using a centered difference scheme except between the two
upper levels (10 and 30hPa) of the ECMWF model. There are substantial
errors on PV. This is primarily due to the coarse vertical interpolations
and unreal small-scale structures in

The only difference is that the trajectory calculation is done in a 1.5
degree distance grid, so the difference between them is small.

The data are, however, only available in a 2.5 degree distance grid at
NILU, but PV is calculated for every station using the finer grid and are
available in the directory:

/nadir/data/ecmwf/pvatstat/YYYY/MM

The isentropic data are organised into year(YYYY) and month(MM) in the
subdirectories:

/nadir/data/ecmwf/isentrop/YYYY/MM

The data are stored in files with names:
atYYMMDD.HHH

where YYMMDD gives the date of the analyses and HHH is the forecast hour.

HHH equals 000 for the 12 UT analysis. HHH equals 12, 36 and 60 hoursfor
the forecast based on on the 0UT analyses, while HHH equals 24, 48, 72,
96, 120 and 192 for the forecast based on the 12 UT analyses.

Thus, at951128.012 is the 12 hour forecast from 0 UT analysis on November
28, 1995, which is valid at 12 UT on November 28, 1995. The forecast in
at951127.24 is also valid at this time, but since it is a longer forecast,
it should not be used. However, the forecasts based on the 0 UT analyses,
will arrive later at NILU (at about 6 UT).

Note, that on the first day of a month you have to go to the directory of
the previous month, to get the forecast based on the 12 UT analysis.