Bathymetric prediction from dense satellite altimetry and sparse shipboard bathymetry

摘要: The southern oceans (south of 30°S) are densely covered with satellite-derived gravity data (track spacing 2–4 km) and sparsely shipboard depth soundings (hundreds kilometers between tracks in some areas). Flexural isostatic compensation theory suggests that bathymetry downward continued may show linear correlation a band wave-lengths 15–160 km, if sediment cover is thin seafloor relief moderate. At shorter wave-lengths, the field insensitive to topography because upward continuation from sea surface; at longer wavelengths, cancels out most due topography. We combine this Wiener optimization empirical evidence for noise-to-signal ratios design low-pass band-pass filters use predicting gravity. prediction combines long wavelengths (>160 low-pass-filtered an intermediate-wavelength solution obtained multiplying continued, band-pass-filtered (15–160 by scaling factor S. S empirically determined existing km robust regression varies wave-lengths. find areas less than 200 m significant 99% level, be related density materials these areas. has horizontal resolution limit 5–10 position within 100 actual 50% grid points 240 80% these. In very rugged underestimates peak amplitudes features. Images reveal many tectonic features not seen on any bathymetrie charts. Because relies <160 it errors navigation sounding lines but also cannot completely reproduce them. Therefore used locate should assess hazards navigation. available National Geophysical Data Center both digital printed form.