A new trial on the harmonic analysis for short period observation of tide and tidal current, using the least square method.
In a harmonic analysis for tide and tidal current, the number of calculable harmonic constituents is limited by the duration of observation. And, in order to utilize the result from a short period observation, for a long range prediction, it is necessary to compensate such a short period result, taking into account some effects of harmonic constituents which can not be separated or determined directly from the short observation. For this purpose, it is thus customary to calculate more constituents than are determined in the conventional methods, on basis of an assumption of the relation in amplitude and phase lag between the constituents. The essential theme in the harmonic analysis for short period observation lies in how such an assumption is established and expressed in the harmoic analysis. In this report, the newly developed analyzing method is compared with two conventional methods, in the aspects of the compatibility between observed values and predicted ones and the reproducibility of original spectrum. The method is characterized by the assumed relations which are expressed directly by normal equations for the least square method. Application to 15 day tidal current data shows that the new method is better than the two conventional methods, with respect to the fitness between the observed value and the predicted one and the reproducibility of the original spectrum.
| Main Author: | |
|---|---|
| Format: | Journal Contribution biblioteca |
| Language: | Japanese |
| Published: |
1981
|
| Online Access: | http://hdl.handle.net/1834/16248 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|