Repository logo
 

Simple model of ocean-atmosphere interactions in the topical climate system

Date

1999-05

Authors

Kelly, Michael A., author

Journal Title

Journal ISSN

Volume Title

Abstract

The tropical sea surface temperature (SST) distribution strongly modulates the global atmospheric circulation. Although the mechanisms which generate SST anomalies have been the subject of intense scrutiny in recent years [see Neelin et al. (1998) for a review], the steady tropical climate has received much less attention. As a consequence, the dominant physical processes which maintain the steady tropical climate remain poorly underĀ­ stood. The goal ;)f this report is to construct and use a simple, mechanistic box model of the tropical ocean-atmosphere climate system to develop ideas about the interactions among various physical processes which can be tested against observations and results from more sophisticated models. Efforts to study the steady tropical climate have been aided by the recent emergence of box models. Pierrehumbert (1995) developed a two-box model of the tropical climate in which one box represents the ascending branch of the Hadley/Walker circulation, and the second box represents the subsiding branch. Pierrehumbert used the model to demonstrate the importance of a low-water-vapor region in exporting to space excess heat that is generated by the ascending branch. Later studies using box models have demonstrated the imĀ­portance of ocean dynamics (Sun and Liu 1996) and low-level stratus clouds (Miller 1997) for regulating ST and the SST gradient. Despite their success in simulating the tropical climate, simplifying assumptions in these box models make conclusions derived from them less than robust. None of these box models include a momentum budget for the ocean or atmosphere. Each of the box models emphasizes either the ocean or the atmosphere and settles for a highly simplified representation of the other. We have developed a simple coupled ocean-atmosphere model of the Walker circulation which has separate boxes for the ascending and descending branches of the atmospheric circulation and separate boxes for the Cold Pool, Warm Pool, and undercurrent. This is the first box model to include explicit momentum budgets for the atmosphere and ocean components and to calculate the fractional width of the Warm Pool. The atmospheric model contains an explicit hydrologic cycle, a simplified but physically based radiative transfer parameterization, and interactive clouds. We first explored the conditions under which the Warm Pool can establish a radiative-convective equilibrium. Under clear skies, quasi-tropical equilibria occur for realistic prescribed SSTs and wind speeds, but realistic clear-sky equilibria of the tropical ocean-atmosphere system do not occur. If the surface temperature is allowed to vary, the model runs away. When cloud radiative effects are incorporated, the model reaches an unrealistically warm, dry radiative-convective equilibrium. For simulations in which cloud radiative effects are incorporated and realistic, lateral transports of energy and moisture are specified, equilibrium of the ocean-atmosphere system occurs for an SST of 300 K and precipitable water of 40 kg m-2, which is quite realistic. We also demonstrated the sensitivity of the tropopause height and temperature to cloud radiative effects. The tropopause height and temperature are calculated based on the requirement of temperature continuity at the bottom of a two-layer stratosphere in radiative equilibrium. As the cloud optical depth or cloud fraction increase, the upward longwave flux across the tropopause decreases, and so the tropopause temperature decreases and tropopause height increases. Our results from the fully coupled model indicate that the intensity of the tropical circulation is crucially dependent on the specified cloud fraction in the Warm-Pool region and on the amount and distribution of water vapor above the Cold-Pool boundary layer (CPBL). In response to increasing the cloud fraction above the Warm Pool, a feedback involving the tropopause height slows the Walker circulation. As the cloud fraction over the Warm Pool increases, the altitude of the tropopause increases, and so air is advected to the Cold-Pool region from higher, drier altitudes. The effects of the drier air are to reduce the radiative cooling rate above the CPBL and, therefore, to reduce the subsidence rate. Since the width of the Cold Pool remains approximately constant, a decreased subsidence rate implies a weaker Walker circulation. In order to maintain energy balance in spite of a weaker circulation, the precipitable water over the Warm Pool must increase. The radiative effect of the precipitable water contributes to an increase of SST in the Warm Pool. Our "wet troposphere" experiment shows that the Walker circulation intensifies if air which is advected to the subsiding region originates from a lower altitude in the Warm-Pool region. Because the circulation is more intense, the SST and precipitable water of the Warm Pool must decrease in order to balance the energy and moisture budgets. Experiments using our ocean model reveal that cold-water upwelling is the dominant mechanism for regulating SST in the Cold Pool. Although the radiative effect of stratus clouds further depress SST, ocean dynamics prevent the mixed-layer temperature from warming by more than 4 K beyond the temperature of the undercurrent.

Description

May 1999.

Rights Access

Subject

Ocean-atmosphere interaction -- Tropics
Atmospheric circulation

Citation

Associated Publications