Figure 5: A snapshot of a Kuroshio model run.
3.3 Applications to Assimilation and a Kuroshio computation
The new assimilation scheme will be tested on a Kuroshio ocean model. At this point we are not concerned with errors in the model, but only computational errors which can be attributed to the numerical scheme, as described above.
We include here one snapshot from the Kuroshio model run, see Figure (5), along with the associated wavelet analysis of the snapshot, see Figure (6). It is quite clear from the wavelet-error detection mechanism, that the wavelet analysis has detected the regions of the domain where the computational errors will be greatest.
4 Wavelet Analysis of SST Data: Jameson and Kazmin
In this section we give a brief example of the application of wavelet analysis to signal detection. In this example we are searching for the ENSO signal in the North Pacific from 25 degres north to 45 degress north. In Figure (7) have many combinations of wavelet filtering. For the moment these details are not important. However, the presence of the ENSO signal is detected in the lower two figures in the right hand column of the figure.
