439
10-jj
Figure 13: Estimated spectrum of the residuum of equation (30) with 95% confidence
intervals.
of our model to timescales longer than the periods of typical barotropic
Rossby waves. Restricting the test to a low-frequency range (0 :S f :S 0.002
1/day) we accept the hypothesis for the equations (30) and (31) for the
(Y = 0.05 quantile. The test have also been performed for the baroclinic
equations with differing success: whenever the terms resulting from the
linear friction parametrization are explicitly included in the test, the error
bounds become large and the test is not very discriminating, i. e. the model
must be accepted.
5 Conclusions
Compared to the research on the variability of the atmospheric circulation
our knowledge of oceanic variability is still in its infancy, both in terms
of the observational background as well as understanding the mechanisms.
Direct observations of variations of the large-scale ocean circulation are
10-jj
Figure 13: Estimated spectrum of the residuum of equation (30) with 95% confidence
intervals.
of our model to timescales longer than the periods of typical barotropic
Rossby waves. Restricting the test to a low-frequency range (0 :S f :S 0.002
1/day) we accept the hypothesis for the equations (30) and (31) for the
(Y = 0.05 quantile. The test have also been performed for the baroclinic
equations with differing success: whenever the terms resulting from the
linear friction parametrization are explicitly included in the test, the error
bounds become large and the test is not very discriminating, i. e. the model
must be accepted.
5 Conclusions
Compared to the research on the variability of the atmospheric circulation
our knowledge of oceanic variability is still in its infancy, both in terms
of the observational background as well as understanding the mechanisms.
Direct observations of variations of the large-scale ocean circulation are