Basic concepts of the optimal interpolation method for basin-scale subsurface thermal field mapping are reviewed. The signal to noise ratio and the autocorrelation function in the analyzed hydrographic field are used as known parameters in this method. Since the accuracy of estimation of these statistical parameters has improved remarkably owing to recent international efforts to accumulate observation data, the optimal interpolation method is recently well applied to oceanographic analysis. It is emphasized that the definition of signal depends upon the temporal and spatial scales of phenomena that one interests. Distinction of signal and noise is explained by power spectrum schematically, and the methods of estimation of signal to noise ratio are demonstrated by using autocorrelation function and structure function. The calculation of the weights of the data in the interpolation is explained. And the change of weights in terms of the sampling density is interpreted. It is also shown that the interpolation error mainly depends upon sampling density, and that the optimal interpolation method is useful to design the future observation network. Advantages and characteristics of this method are discussed by compering with the other interpolation methods. Recent advanced applications of the optimal interpolation method to the thermal field mapping are reviewed.
No takes yet. Share an insight, caveat, or question.
Keisuke Mizuno (1995) studied this question.