We will address the much debated issue concerning the regulation of the time mean maximum SST (about 303 K) in the tropical oceans. In particular, we use available field data, including CEPEX and TOGA, to assess the importance of various mechanisms that have been proposed. Our analysis does not deal with tropical averaged SSTs; but is restricted to SSTs in the warmest oceanic regions such as the Pacific warm pool. Its focus is on one fundamental question: Why do the time mean (seasonal to longer) maximum SSTs in warm oceans (length scales of 103 km or larger) rarely exceed about 303 K? From voluminous measurements conducted at the sea surface, as well as inside and on top of the atmosphere (TOA) during CEPEX (1993), and from buoy and satellite data (TOGA, ERBE, GMS), the magnitude of the various feedback terms and their time constants was determined. The analysis combines the data with an ocean mixed layer energy balance model, that simulates the time mean as well as the weekly evolution of SSTs over several buoys sites in the western tropical Pacific, for the year 1993.
We first demonstrate that, the spatial patterns of SSTs and the water vapor super greenhouse effect, require a thermostat type control of the SSTs. We show that the typical e-folding time constant of this thermostat regulation process is about 2 months and the asymptotic value of maximum SST is about 302.5 K. The relative rapid time scale is because of the shallow salinity mixed layer (Å 30 m) and the large value (Å 20 W.m-2 K-1) for surface-atmosphere feedback.
We then use the time dependent energy balance equation for the mixed layer, to examine the role of various feedbacks: evaporation, clouds, and ocean transport. The daily fluctuations in surface heat fluxes are inferred from Buoy and satellite data. The model is able to account for weekly to longer time scale fluctuations in observed SSTs over several buoy sites. This successful simulation of shorter to longer term SSTs enables us to arrive at the following conclusions: I) Unaccounted for atmospheric dynamical feedback effects (in the model) are not playing an important role in regulating maximum SSTs; and 2) weather events modulate, but not disrupt, the thermostat process. In this way, weather events may raise maximum daily SSTÕs occasionally to values near 304 K.
The analysis gives a rough indication of the relative importance of various feedbacks: The cumulative effect of all terms is a (slightly non-linear) negative feedback in the range of 19 to 30 W m-2K-1. The important term is the short wave cloud forcing, i.e., the reduction of solar radiation by increasing convective cloudiness above the threshold SST of about 299 to 300 K. Evaporative cooling contributes to the negative feedback. However, this contribution vanishes above 301 K owing to the behavior of the wind field. Ocean flux feedbacks appear to be quite small, at least in the Western Pacific WP, particularly because of the faster adjustment time scales (Å 2 months) of the atmospheric response of clouds and convection to SST changes.