Climate Change 2001:
Working Group I: The Scientific Basis
Other reports in this collection

7.6.3 Monsoons and Teleconnections

The weather and climate around the world in one place is generally strongly linked to that in other places through atmospheric linkages. In the tropics and sub-tropics, large-scale overturning in the atmosphere, which is manifested as the seasonal monsoon variations, links the wet summer monsoons to the dry subsiding regions usually in the tropics and sub-tropics of the winter hemisphere. Throughout the world, teleconnections link neighbouring regions mainly through large-scale, quasi-stationary atmospheric Rossby waves. A direct consequence of these linkages is that some regions are wetter and/or hotter than the prevailing global scale changes, while half a wavelength away the regions may be dryer and/or cooler than the global pattern. Moreover, because of the way these patterns set up relative to land and ocean, they can alter the global mean changes. In addition, errors in models (such as in convection in the tropical Pacific) can be manifested non-locally through teleconnections, e.g., in the North Pacific SSTs (see Figure 8.1a), although other processes are also involved.

The term "monsoon" is now generally applied to tropical and sub-tropical seasonal reversals in both the atmospheric circulation and associated precipitation. These changes arise from reversals in temperature gradients between continental regions and the adjacent oceans with the progression of the seasons. The dominant monsoon systems in the world are the Asian-Australian, African and the American monsoons. As land heats faster than ocean in summer, heated air rises and draws moist low-level maritime air inland where convection and release of latent heat fuel the monsoon circulation. For the Asian monsoon, a regional meridional temperature gradient extending from the tropical Indian Ocean north to mid-latitude Asia develops prior to the monsoon through a considerable depth of the troposphere (Webster et al., 1998). To a first order, the stronger this meridional temperature gradient, the stronger the monsoon. Thus land-surface processes, such as soil moisture and snow cover in Asia can influence the monsoon and, along with SST variations, may induce quasi-biennial variability (Meehl, 1997). Additionally, large-scale forcing associated with tropical Pacific SSTs influences monsoon strength through the large-scale east-west overturning in the atmosphere. Anomalously cold (warm) Pacific SSTs often are associated with a strong (weak) monsoon, though these connections are somewhat intermittent.

Some teleconnections arise simply from natural preferred modes of the atmosphere associated with the mean climate state and the land-sea distribution. Several are directly linked to SST changes (Trenberth et al., 1998). The most prominent are the Pacific-North American (PNA) and the North Atlantic Oscillation (NAO; see Section 7.6.4) in the Northern Hemisphere, and both account for a substantial part of the pattern of northern hemispheric temperature change, especially in winter (Hurrell, 1996), in part through the "cold ocean warm land" (COWL) pattern (Wallace et al., 1995; Hurrell and Trenberth, 1996) (see Chapter 2). Although evidently a prominent mode of the atmosphere alone, the PNA is also influenced by changes in ENSO (see Section 7.6.5). Thompson and Wallace (1998, 2000) suggest that the NAO may be the regional manifestation of an annular (zonally symmetric) hemispheric mode of variability characterised by a seesaw of atmospheric mass between the polar cap and the middle latitudes in both the Atlantic and Pacific Ocean basins and they call this the Arctic Oscillation (AO). A similar, even more zonal structure is dominant in the Southern Hemisphere (Trenberth et al., 1998) (the Southern Annular Mode, sometimes called the Antarctic Oscillation, AAO). The vertical structure of both AO and AAO extends well into the stratosphere (Perlwitz and Graf, 1995; Thompson and Wallace, 1998).

In the Atlantic, an important emerging coupled mode of variability is the so-called tropical Atlantic dipole, which involves variations of opposite sign in the sea level pressure field across the equatorial Atlantic and corresponding variations in the ITCZ location. Given the background SST and wind fields, anomalous SSTs of opposite sign across the equatorial region are apt to alter the surface winds in such a way as to enhance or reduce evaporative cooling of the ocean and reinforce the original SST pattern (Carton et al., 1996; Chang et al., 1997). The ocean provides a decadal time-scale to the coupled interactions.

Dominant large-scale patterns of ocean-atmosphere interactions are also found in the tropical Indian Ocean (Saji et al., 1999; Webster et al., 1999). They have characteristics similar to El Niño and are associated with large east-west SST changes and a switch of the major tropical convection areas from Africa to Indonesia. There are indications that this atmosphere-ocean process is somewhat independent of ENSO and represents a natural mode of the tropical Indian Ocean.

The vital processes for improved monsoon simulation in models are those associated with the hydrological cycle, especially in the tropics. These include convection, precipitation and other atmospheric processes (see Section 7.2) as well as land surface processes (see Section 7.4), and interactions of the atmosphere with complex topography and with the ocean. The difficulties in assembling all of these elements together has led to problems in simulating mean precipitation as well as interannual monsoon variability, although improvements are evident (Webster et al., 1998, and see Chapter 8, Section 8.7.3).

For teleconnections and regional climate patterns, not only are there demanding requirements on simulating the variations in tropical SSTs that drive many of the interannual and longer-term fluctuations through latent heating in the associated tropical precipitation, but results also depend on the mean state of the atmosphere through which Rossby waves propagate. Feedbacks from changes in storm tracks, and thus momentum and heat transports by transient atmospheric disturbances, as well as interactions with changed land-surface soil moisture from precipitation changes and interactions with extra-tropical oceans are critical. While there is scope for further improvements, great strides have been made in modelling all these aspects in the recent years.



Other reports in this collection