Please reference this work if used elsewhere

Summary of Work
Strong disagreements exist regarding the mechanisms which regulate Sea Surface Temperature (SST) over the Western Pacific Warm Pool. As these waters have a critical impact on convection in the western Pacific which in turn influences the planetary scale Walker and Hadley circulations, the climatic implications of the warm pool are extensive (Webster and Lukas, 1992) .

Here we examine fluctations in Warm Pool SST as it relates to shifts in the the surface energy balance during strong intraseasonal variations in SST in an effort to understand the warming and cooling of the oceanic mixed layer. Portions of both the Indian and Pacific Ocean Warm Pools are considered. The reanalysis products (NCEP and ECMWF) are used to assess the surface turbulent flux over an extended climatology. Recognizing the potential presence of significant error in these reanlaysis fields, TOGA COARE flux data are cited for error estimation. Large errors are found in the surface shortwave flux from the reanlaysis products and alternative methods for diagnosing surface shortwave radiation (Bishop et al., 1997; Shinoda and Hendon, 1997) are also used.

Four results from this study are worthy of mention.
  1. Scatter plots of SST vs Evaporation can be misleading regarding the nature of SST regulation mechanisms when used as they have by previous authors. First, they often mix spatial and temporal variability (as in Ramanathan's response to Wallace in Nature, 1992) and they can be influenced by differential rates of heating and cooling.
  2. Based on the observed surface energy balance, anomalies in solar and evaporative fluxes are responsible for anomalous energy input to the mixed layer during its warming and cooling phase, often to near equal extent. During the cooling phase, both solar and evaporation anomalies perturb the mean surface energy balance most significantly, and often to near equal extent, followed by and longwave radiation, and sensible heat.
  3. The Indian and Pacific Ocean warm pools exhibit behavior which suggests quite different ocean characteristics such as the depth of the mixed layer and role of oceanic heat advection.
  4. Variations in SST near its peak are modified by large-scale variations in the atmosphere. The onset of the southeast Asian monsoon is critical for the Indian warm pool while the propagation of the MJO throughout the year is associated with fluctuations in Pacific warm pool SST near 29.5degC.

.....now for the details


The Surface Energy Balance


The surface energy balance over the western Pacific Warm Pool is near zero in the net over seasonal time scales. Its variation on shorter times scales can be large however and it is this variability which we investigate here.
Several portions of the Indian and Pacific Warm Pool are analized in this study. A summary of the ocean domains considered can be found here.
The mean NCEP Surface Energy Balance can be found here with comparisons to other estimates.
Also shown here is the standard deviation of flux values over time. Because of the large variability in surface shortwave flux and evaporative flux, the two fluxes are the primary focus of this analysis.


Some have deduced from scatter plots of Latent Heat vs SST that evaporation acts as a positive feedback mechanism for warming SST. That is, as SST warms, evaporation decreases. We maintain that this is not the case. For example, using fictional data, we show how scatter plots can be misleading. In the time evolution of SST from this plot, evaporation and solar flux, it is clear that evaporation is responsible for the cooling of SST yet the scatter plot suggests that it is.
A TOGA CASE STUDY
Measurements from the Moana Wave ship during the TOGA COARE Intensive Operating Period Capture a period of exceptional SST variation from Nov 1, 1992 to Jan 11, 1993.

SST

SST
A comparison between Reynolds SST and ship measurements is generally good. Note that the first half of our period corresponds to an SST warming phase and the second half corresponds to SST cooling. Also note that the Reynolds measurements fail to capture the large diurnal variation in SST (blue dashed line) and thus does not capture the variations in surface evaporation or longwave emission. The longwave and evaporative cooling contributions calculated in the NCEP model are lower than they otherwise should be because of this oversight.

Evap

Evaporation
Excellent agreement with NCEP in general because winds are good. This is true despite the large scale discrepency of the measurements. Lower caption is a comparison of NCEP with a satellite derivation method over a large scale again with excellent agreement.

Wind

Wind
Fluxes compare well because wind compares well.

Radiation

Radiation
NCEP does a generally poor job based on this comparison, however
a technique developed by Shinoda and Hendon (1997) from OLR
regressions does substantially better. In the comparison above,
both NCEP and OLR shortwave radiation are shown and compared.
Sui and Lau (1997, J. Climate) found near equal roles for evaporation and solar flux anomalies in cooling the ocean during intraseasonal variations.
The representativeness of their results are uncertain however.

With the techniques for diagnosing surface fluxes discussed above,
attention can be broadened to other temporal and spatial domains.
  1. Evaporation is large during the cooling phase and is not a positive feedback mechanism for SST.
  2. Shortwave anomalies are positive during the cooling phase but are not long lasting relative to evaporative anomalies.
  3. Longwave and sensible (not shown) flux anomalies are small relative to contributions from evaporation and shortwave fluxes.
A CLIMATOLOGY OF SURFACE FLUX ANOMALIES

COMPOSITE RESULTS


identified warm events over the small domain.
identified warm events over the medium (Pacific) domain. Indian Medium Domain
identified warm events over the large (Pacific) domain. Indian Large Domain
Net Shortwave Flux
Latent Flux
Net Shortwave (Med Scale)
Latent Flux (Med Scale)
Net Shortwave (Large Scale)
Latent Flux (Large Scale)
Integrated Flux Anomalies
Integrated Flux Anomalies (Med Scale)
Integrated Flux Anomalies (Large Scale)
THE PUNCH LINE

Warming of SST is primarily associated with below normal evaporation and, to a slightly lesser extent, enhanced surface solar radiation. Fighting against the warming are longwave and sensible fluxes which act to cool the mixed layer during this stage. There is no evidence of a "super greenhouse" in the NCEP product...ie: The enhanced downwelling flux from the atmosphere (at the surface) is less than the increased upwelling flux associated with a warming surface.

Cooling of SST is largely associated with both stronger than normal evaporation and reduced solar flux anomalies. Over the Pacific, these flux anomalies are often nearly equal. Over the Indian Ocean warm pool, the flux anomalies are often dominated by solar flux anomalies. In both cases, there exist important couplings to the large-scale atmosphere and the variability of SST appears to be largely DRIVEN BY THE ATMOSPHERE rather than the reverse. Thus in the variability observed during in-situ experiments, there is little evidence of SST regulation via an internal thermodynamic "threshold". In press, Journal of Climate.


----- Your are visitor #