December 1996 — As with similar questions about a tree in the forest or a grain of sand on the beach, it may be hard to imagine that a few inches of rain matters to the deep ocean. After all, the ocean's average depth is around 4 kilometers and only 1 to 5 centimeters of water are held in the atmosphere at any one time. But it does matter, in part because the ocean is salty. The effect of rain diluting the salts in the ocean (or evaporation concentrating them) can be greater than the effect of heating (or cooling) on the density of seawater.
It also matters because rainfall and evaporation are not evenly distributed across and among ocean basins—some regions continuously gain water while others continuously lose it. This leads to ocean current systems that can be surprisingly strong. The processes of evaporation and precipitation over the ocean are a major part of what is called "the global water cycle;" indeed, by all estimates, they dominate the water cycle over land by factors of ten to a hundred. The addition of just one percent of Atlantic rainfall to the Mississippi River basin would more than double its discharge to the Gulf of Mexico.
As discussed previously in Oceanus, our knowledge of the water cycle over the ocean is extremely poor (see the Spring 1992 issue). Yet we now realize that it is one of the most important components of the climate system. One of the significant pieces of evidence for this comes from a description of the "Great Salinity Anomaly" put together by Robert Dickson (Fisheries Laboratory, Suffolk, England) with other European oceanographers. The Great Salinity Anomaly (GSA) can be characterized as a large, near-surface pool of fresher water that appeared off the east coast of Greenland in the late 1960s (see figure at upper right). It was carried around Greenland and into the Labrador Sea by the prevailing ocean currents, in the counterclockwise circulation known as the subpolar gyre. It hovered off Newfoundland in 1971-72 and was slowly carried back toward Europe in the North Atlantic Current, which is an extension of the Gulf Stream. It then completed its cycle and was back off the east coast of Greenland by the early 1980s, though reduced in size and intensity by mixing with surrounding waters. The origin of the Great Salinity Anomaly is thought to lie in an unusually large discharge of ice from the Arctic Ocean in 1967. Its climatic importance arises from the impact it had on ocean-atmosphere interaction in the areas it traversed.
The GSA derives its climate punch from the strong effect of salinity onseawater density, with salty water being considerably denser than fresh water. That is, these northern waters normally experience strong cooling in the winter, which causes the surface water to sink and mix with deeper waters. This process, called deep convection (see figure below right), is a way for the ocean to release heat to the atmosphere, heat that then helps to maintain a moderate winter climate for northern Europe. However, when the GSA passed through a region, the surface waters became so fresh and light that even strong cooling would not allow it to convect into the deeper waters. Thus, the deep water remained isolated from the atmosphere, which could not extract as much heat as usual from the ocean. The GSA acted as a sort of moving blanket, insulating different parts of the deep ocean from contact with the atmosphere as it moved around the gyre. Its impact in the Labrador Sea has been particularly well documented (see Curry/McCartney "Labrador Sea" article). When the surface waters were isolated from deep waters, they became cooler. Changing sea surface temperature patterns can affect atmospheric circulation, and may possibly reinforce a poorly understood, decades-long variation in North Atlantic meteorological conditions known as the North Atlantic Oscillation (see Deser article with McCartney "NAO" inset). For it is the ocean that contains the long-term memory of the climate system. By comparison, the atmosphere has hardly any thermal inertia. It is difficult to imagine how the atmosphere alone could develop a regular decadal oscillation, but the advection of freshwater anomalies by the ocean circulation could be an important key to this climate puzzle.
Unfortunately, we have no ready means of detecting freshwater pulses like the GSA. While surface temperature can be observed easily from space, surface salinity, so far, cannot. The salinity variations important for oceanography require high precision and accuracy, so there is no quick and inexpensive method of measurement. We have had to rely on careful analysis of sparse historical records from mostly random and unrelated surveys gleaned from several nations to piece the GSA's story together. But how many other "near-great" salinity anomalies have we missed because the signal was not quite large enough? Is there a systematic way to monitor salinity so that we know years in advance of another GSA's approach?
In addition to variability within an ocean basin, we would like to understand the large differences in salt concentration among ocean basins. (see global ocean figure below right) For example, the Pacific Ocean is significantly fresher than the Atlantic and, because it is lighter, stands about half a meter higher. This height difference drives the flow of Pacific water into the Arctic through the Bering Strait. The salinity difference between these two major oceans is thought to be caused by the transport of water vapor across Central America: The trade winds evaporate water from the surface of the Atlantic, carry it across Central America, and supply rainfall to the tropical Pacific. This water loss is the major cause of the Atlantic's greater saltiness and its propensity to form deep water. The extra rainfall on the Pacific makes it fresher and prevents deep convection. How does this atmospheric transport vary with time? Since salinity is a good indicator of the history of evaporation or precipitation, perhaps if we had sufficient data, we could see changes in the upper ocean salt content of the two oceans that reflect variations in atmospheric transports. How many years does it take for salinity anomalies in the tropical Atlantic to propagate to high-latitude convection regions and affect the sea-surface temperature there? What is the impact on the atmospheric circulation?
These and other climate problems will continue to perplex us until we make a serious attempt to monitor salinity on large space and time scales. One approach would be to maintain ships in certain places to sample the ocean continually. A modest effort along these lines was made after World War II when weather ships were maintained at specific sites by several nations (see Dinsmore article). The data they collected provide nearly the only long time-series measurements available from deep-ocean regions. However, the weather ships are all but gone; there is only one now, maintained seasonally by the Norwegians. Today's satellites provide information on approaching storm systems, but, unfortunately, they cannot tell us what we need to know about ocean salinity distributions.
It now appears that new technology will provide the key to the salinity monitoring problem, at a surprisingly modest cost. The Box on pages 6 and 7 describes how we might obtain temperature and salinity profiles from data collected by autonomous diving floats. It should be quite feasible to deploy an array of these station-keeping "Slocums" that would intercept and monitor the progress of the "Great Salinity Anomalies" of the future. In the next two years, a large number of profiling ALACE (precursor to the Slocum) floats will be deployed in the Atlantic in a preliminary test of the general concept. In addition to measuring temperature and salinity, Slocums might some day measure rain. It turns out that rain falling on the ocean does make a sound, and work is underway to record that sound with hydrophones and develop algorithms to convert the measured sound level to rain rates.The remaining technical obstacles to development of a globe-spanning array of station-keeping Slocums are small. The only thing lacking is a strong societal commitment to the support of such fundamental research on the climate system of the earth.
This research was sponsored by the National Science Foundation and the National Oceanic and Atmospheric Administration's Climate and Global Change Program. Most of Ray Schmitt's career has been focused on very small-scale processes in the ocean related to mixing by turbulence and "salt fingers." However, he has been driven toward studies of the global-scale hydrologic cycle by a desire to contribute to improved weather and climate prediction, so that he can better plan to take advantage of the rare good weather in Woods Hole.