Fróðskaparrit - 01.01.2007, Blaðsíða 141

SAMANBERING AV ALDUFORSØGNUM FRÁ ECMWF VIÐ ALDUMÁTINGAR
GJØRDAR Á FØROYALEIÐINI
139
21 22 23 24 25 26 27 28 29 30 31 1/9 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1D wave-spectra from ECMWF-WAM in event 1 (20/8-20/9 2000)
Figure 6. All recorded 1D wave spectra at WVD-4 in Event 1 vs. lD-wave spectra from ECMWF. y-axis
gives frequency, x-axis gives time and colours give the logarithm of the spectral energy density. The spurious
spectral feature occurring in the WVD-4 spectra on the 8/9 is due to data error.
Southeast of the Faroe Islands, and sec-
ondly the EW4 model most likely does not
resolve the bathymetric features that cause
negative shoaling at WV-2. The effect of
negative shoaling is significant on low fre-
quencies at the WV-2 site, according to
idealized tests, some of which are reported
in Niclasen and Simonsen (2005c).
The statistical parameters derived from
WV-2, WVD-4 and K7 are given in Table
4, and the time series of EW4 and meas-
ured parameters at WVD-4 are given in
Figures 4 and 5. In order to get a general
comparison of the spectral shape of the
EW4 model vs. observed spectra at WVD-
4, the lD-spectra time series for Event 1
are displayed in Figure 6. No effort is
made here to investigate the fit between
observed and modelled 2D-spectra as
pitch/roll buoys have intrinsically poor di-
rectional resolving power (Young, 1994;
Arntsen and Tørum, 2005).
One clear feature is the underestimation
of the wave height in the peak of the
storms in all events, and general underesti-
mation of the wave height in the winter