University of Colorado, Boulder, CO 80309-0391
Sea surface temperature (SST) regulation in the tropical oceans is an important aspect of global climate change. It has been observed that SST in the equatorial zone has not exceeded 304K over, at least, the past 10,000 years, and probably longer. Furthermore, recent satellite observations from the Earth Radiation Budget Experiment (ERBE) suggest that the greenhouse effect associated with mesoscale organized convection increases with increasing SST at a rate faster than this energy can be re-radiated to space. This suggests that a runaway greenhouse effect is possible in those parts of the tropical oceans where mesoscale convective systems (MCS) are prevalent. However, this is not observed.
Thus, a search for mechanism(s) which can account for SST regulation is underway. The " thermostat hypothesis" of Ramanathan and Collins conjectures that surface shading (relative cooling) by cirrus anvils associated with MCS effectively balances the enhanced greenhouse effect (relative warming) caused by the increase in tropospheric moisture also associated with these cloud systems. The Thermostat Hypothesis related both of these effects to a single parameter, SST. Because of data obtained from two major field expeditions to the equatorial Pacific, COARE and CEPEX, it may be possible to test this hypothesis.
Recent work on the global heat budget by Trenberth and Solomon shows the Eastern Pacific to be a maximum of ocean heat transport while the Western Pacific is a maximum in atmospheric heat transport. OLR and HRC atlases show the E. Pacific to be a region of low frequency of occurrence of MCS while the W. Pacific is a well-known maximum. This causes the E. Pacific to be characterized by high insolation while, by comparison, the W. Pacific has relatively low insolation. Further, the E. Pacific has relatively steady strong easterlies compared to the W. Pacific, which is an area of relatively low winds associated with a planetary scale maximum in low-level convergence there; in fact, westerly winds, known as westerly wind bursts (WWB) are intermittently observed in this area. The strong easterlies are associated with basin-scale upwelling in the E. Pacific while the W. Pacific has little upwelling. In the E. Pacific, the SST regulation is a balance between high insolation, relatively high evaporation, and, importantly, the upwelling. In the W. Pacific, Grossman conjectures that the balance is probably between net radiation, evaporation, and ocean mixing (as opposed to upwelling).
Observational and theoretical evidence exists to suggest the importance of other feedback mechanisms as opposed to the cirrus shading and "super greenhouse effect" supported by the thermostat hypothesis. These complementary mechanisms are: evaporation, near-surface and near-thermocline ocean mixing, and sensible heat transfer. Using a variety of data sources from COARE and CEPEX ‹ ship, buoy, and aircraft ‹ Grossman proposes that any model which includes these effects must take into account these recurring sequences of discrete events. One could view these sequences as coherent structures on the climate scale. This research emphasizes four events: fair weather (FW), meso- scale convection (MCS), westerly winds (WWB), and disturbed weather (DW) as important to SST regulation in the W. Pacific warm pool. All of these events are associated with large-scale dynamical patterns in the atmosphere.
MCS is important both to the thermostat hypothesis and to this work. The Gray criteria for the formation and maintenance of MCS were first proposed by Gray in 1968. These are: near-surface convergence, near-tropopause divergence, SST in excess of 301šK, and low shear between the surface and upper troposphere. For example, FW is associated with no near-surface convergence or near- tropopause divergence (often the opposite). The SST criteria are a necessary, but not sufficient criteria for MCS; for example, there are parts of the E. Pacific where SST is greater than 28šC but little MCS; the same is true of the Arabian Sea during pre-monsoon conditions. Large scale dynamics seems to play a part.
A simple model of an ocean column, including the surface, shows the effects of the different events. Grossman has used such a model to analyze several events observed in COARE and CEPEX. Each of the components of the columnar heat budget (net insolation, net IR cooling, evaporation, and ocean mixing) had different values for each event. In summary, FW was a heating event (1-7 days), WWB was a cooling event (4-10 days), MCS was a cooling event (2-4 days), and DW (1-7 days) was borderline, either little or no heating/cooling.
The sequence of these events and how they could account for the observed SST variability was discussed. The frequency of occurrence of these events is important to the problem. For instance, the strongest cooling was associated with a MCS-WWB couplet, which is consistently observed. However, the MCS- WWB couplet (strong cooling) is very intermittent, as is a long period of FW (strong warming). In between, short periods of FW and DW are often observed followed by MCS. It therefore appears that a strong warming event can follow a strong cooling event, with shorter periods of modest heating and cooling in between. It is relatively easy to build sequences which can explain current observations of W. Pacific SST variability. This "extended" period of modest heating and cooling would be associated with MCS which would be necessary to account for the maximum in atmospheric transport noted by Trenberth and Solomon. Grossman's current research is directed to continuing this investigation using observations from COARE and CEPEX.
In Tropical Ocean Global Atmosphere (TOGA) COARE, measurements of atmospheric and oceanic variables in the Western Tropical Pacific were made for a period of four months. Figure 6.1 shows SST and wind speed variations during this period. It appears that SST shows a cycle of 3- 4 months but clearly, a longer observation period would be necessary to confirm this. There are indications that at least some of the time, warm SSTs are associated with low wind speeds, and low SSTs follow periods of high wind speed. Figure 6.2 is a phase diagram showing sequence of heating and cooling of ocean surface and upper ocean (0- 200 m) as a function of wind speed.
