The ocean has an important role in climate variability and change. The ocean’s heat capacity is about 1,000 times larger than that of the atmosphere, and the oceans net heat uptake since 1960 is around 20 times greater than that of the atmosphere (Levitus et al., 2005a). This large amount of heat, which has been mainly stored in the upper layers of the ocean, plays a crucial role in climate change, in particular variations on seasonal to decadal time scales. The transport of heat and freshwater by ocean currents can have an important effect on regional climates, and the large-scale Meridional Overturning Circulation (MOC; also referred to as thermohaline circulation) influences the climate on a global scale (e.g., Vellinga and Wood, 2002). Life in the sea is dependent on the biogeochemical status of the ocean and is influenced by changes in the physical state and circulation. Changes in ocean biogeochemistry can directly feed back to the climate system, for example, through changes in the uptake or release of radiatively active gases such as carbon dioxide. Changes in sea level are also important for human society, and are linked to changes in ocean circulation. Finally, oceanic parameters can be useful for detecting climate change, in particular temperature and salinity changes in the deeper layers and in different regions where the short-term variability is smaller and the signal-to-noise ratio is higher.
The large-scale, three-dimensional ocean circulation and the formation of water masses that ventilate the main thermocline together create pathways for the transport of heat, freshwater and dissolved gases such as carbon dioxide from the surface ocean into the density-stratified deeper ocean, thereby isolating them from further interaction with the atmosphere. These pathways are also important for the transport of anomalies in these parameters caused by changes in the surface conditions. Furthermore, changes in the storage of heat and in the distribution of ocean salinity cause the ocean to expand or contract and hence change the sea level both regionally and globally.
The ocean varies over a broad range of time scales, from seasonal (e.g., in the surface mixed layer) to decadal (e.g., circulation in the main subtropical gyres) to centennial and longer (associated with the MOC). The main modes of climate variability, which are described in Chapter 3, are the El Niño-Southern Oscillation (ENSO), the Pacific Decadal Oscillation (PDO), the Northern Annular Mode (NAM), which is related to the North Atlantic Oscillation (NAO), and the Southern Annular Mode (SAM). Forcing of the oceans is often related to these modes, which cause changes in ocean circulation through changed patterns of winds and changes in surface ocean density.
The Third Assessment Report (TAR) discussed some aspects of the ocean’s role. Folland et al. (2001) concluded that the global ocean has significantly warmed since the late 1950’s. This assessment provides updated estimates of temperature changes for the oceans. Furthermore, it discusses new evidence for changes in the ocean freshwater budget and the ocean circulation. The TAR estimate of the total inorganic carbon increase in the ocean (Prentice et al., 2001) was based entirely on indirect evidence. This assessment provides updated indirect estimates and reports on new and direct evidence for changes in total carbon increase and for changes in ocean biogeochemistry (including pH and oxygen). Church et al. (2001) determined a range of 1 to 2 mm yr–1 for the observed global average sea level rise in the 20th century. This assessment provides new estimates for sea level change and the climate-related contributions to sea level change from thermal expansion and melting of ice sheets, glaciers and ice caps. The focus of this chapter is on observed changes in the global ocean basins, however some regional changes in the ocean state are also considered.
Many ocean observations are poorly sampled in space and time, and regional distributions often are quite heterogeneous. Furthermore, the observational records only cover a relatively short period of time (e.g., the 1950s to the present). Many of the observed changes have significant decadal variability associated with them, and in some cases decadal variability and/or poor sampling may prevent detection of long-term trends. When time series of oceanic parameters are considered, linear trends are often computed in order to quantify the observed long-term changes; however, this does not imply that the original signal is best represented by a linear increase in time. For plotting time series, this chapter generally uses the difference (anomaly) from the average value for the years 1961 to 1990. Wherever possible, error bars are provided to quantify the uncertainty of the observations. As in other parts of this report, 90% confidence intervals are used throughout. If not otherwise stated, values with error bars given as x ± e should hence be interpreted as a 90% chance that the true value is in the range x – e to x + e.