THE QUALITY OF TABULATED DECK LOG SWELL OBSERVATIONS

Abstract

To determine the areas of data deficiencies, the totals of sea and swell observations were summarized by unit Marsden squares for the world. An interesting result of this summary was a comparison of the totals of sea and swell observations. The totals of swell observations averaged two-thirds those of sea observations. Although this ratio was fairly representative of the ocean basins, considerable variation occurred in semi-enclosed seas where swell observations averaged only one-tenth those of sea observations. A few ocean basin locations had totals of swell observations that approached those of sea observations, but no Marsden square had more swell observations than sea observations. Strangely enough, swell tabulations, in areas where the ratio of swell to sea observations were smallest (semi-enclosed basins), appeared least reliable although it was suspected at the time that this was probably due to smallness of the sample. Subsequent additions of card decks have neither changed these ratios appreciably nor the suspect reliability. Thus, due to the relative sparseness of observations in some areas and the inexplainable ratios of swell to sea observations in others, the quality of observations contained in IBM listings have for some time been questioned.