ABSTRACT
Historically, the interior ocean has been mainly observed using instruments lowered from research ships or, later, suspended from moorings. Typical ship cruises last a month or two while moorings may last a year or two. The relatively high cost of these observation platforms has limited their number and, consequently, the spatial and temporal density at which the ocean has been observed. Initially this may not have seemed a serious hindrance because the ocean’s circulation was thought to be largely steady with broad spatial scales outside a few concentrated boundary currents. Over the last 30 years, however, satellite remote sensing and intensive experimental ocean observations have belied this view and shown that the ocean is highly variable on time scales that are somewhat longer than those of the atmosphere and space scales of tens of kilometers, much smaller than those of the atmosphere. Even before the fullness of ocean variability was known, Stommel (1955) likened the oceanographic observational approach to meteorologists observing the atmosphere using “half a dozen automobiles and kites to which air sounding instruments were attached and doing all their work on dark moonless nights when they could not see what was happening in their medium.”
