deep convection and subduction, and perhaps could provide an early warning of changes in circulation that could lead to an abrupt climate change. If we are to develop predictive capabilities regarding the thermohaline circulation, we must observe its strength and structure. To date, however, no observational network exists to observe the thermohaline circulation on a continuous basis. Systematic, long-term observations of the heat and water fluxes influencing the thermohaline circulation are needed, especially in the areas of Northern Hemisphere deepwater formation. Moreover, remote influences on the thermohaline circulation must be monitored, particularly the low-latitude atmospheric water-vapor transport from the Atlantic to the Pacific and the influence of Southern Ocean changes.
The meaning of trends in sea-ice volume and extent and associated freshwater fluxes and how these will affect the potential for abrupt climate change is still debated. The debate is due, in part, to limitations in the data especially on ice thickness. Data collected from submarines provide some insight into changes in Arctic ice, but generation of long-term data is uncertain because of the planned cessation of submarine science cruises. Advancing the science of abrupt climate change requires improved observations of sea-ice extent, thickness, and fractional coverage.
The volume and extent of portions of the Greenland and Antarctic ice sheets are known to be changing rapidly, but full coverage by altimetry and interferometric synthetic-aperture radar used to measure thickness and velocity change of ice flow is not yet available, and near-polar holes in other remotely sensed fields miss key parts, especially in Antarctica. Basal conditions of the large ice sheets, and thus their potential for crossing thresholds and leading to rapid climate changes, are known at only a handful of points and with less confidence along limited aerogeophysical flight lines; most of the ice-sheet beds are uncharacterized. Processes beneath the floating extensions called ice shelves, where ice-sheet stability meets deepwater formation, are poorly known. In light of the clear paleoclimatic evidence of abrupt ice-sheet changes affecting global climate and sea level, enhanced emphasis on ice-sheet characterization over time is essential. In addition to the major ice sheets of Greenland and Antarctica, mountain glaciers also offer research opportunities: they are sensitive to climate change and can serve as sentinels of change, so improved monitoring and understanding of these systems is also needed.
Land hydrology links the atmosphere to oceans and ecosystems. Most terrestrial rainfall and snowmelt are used by plants, and much of the remainder runs off in rivers, which locally freshen seawater. Terrestrial pre-