Which surface atmospheric variable drives the seasonal cycle of sea surface temperature over the global ocean?

Abstract

[1] The impact of near-surface atmospheric variables used in driving the seasonal cycle of climatological mean sea surface temperature (SST) is quantified over the global ocean. The six atmospheric variables are air temperature, vapor mixing ratio, wind speed, precipitation, net shortwave radiation, and net longwave radiation, the first (last) three just above (at) the sea surface. Atmospherically forced ocean general circulation model (OGCM) simulations with no data assimilation are performed using monthly and annual means of those variables under the assumption that variations in climatological monthly SSTs are driven by atmospheric variables. SSTs resulting from these simulations are compared with those from a satellite-based field to determine the impact of each atmospheric variable. Large spatial variability is found in the order of impact (most to least) of six atmospheric variables. In general, the SST seasonal cycle is driven primarily by shortwave radiation at midlatitudes, but wind speed is the major controlling variable in the Indian Ocean. Precipitation has almost no significant influence on monthly SST. Overall, shortwave radiation is the most influential variable controlling the seasonal cycle of SST over 34.3% of the global ocean. Wind speed is the second most important variable (27.2%). In tropical regions and the Arabian Sea, sources other than the atmospheric thermal forcing are found to play a significant role in regulating the SST seasonal cycle.

Can an atmospherically forced ocean model accurately simulate sea surface temperature during ENSO events?

“…Here, we explore contributions to those biases from atmospheric forcing variables. Additional factors, such as physical parametrizations, mixing schemes in the model, etc., can also contribute to model bias, but here we only focus on a contribution from the surface forcing used in the model simulation, specifically net short-wave radiation at the sea surface, which is one of the most important variables that drives the seasonal cycle of SST from the ocean model (Kara et al, 2009b). Simulating SST from an ocean model is particularly challenging during the 1998 ENSO transition period because of the exceptionally large temperature drop of almost 7 • C in just one month, from May to June.…”

Tree-Structured Methods for Prediction and Data Visualization

“…The collaboration allowed the PI to test the GUIDE computer program on large data sets of the order of several hundred megabytes (the meteorological measurements were recorded at one-degree intervals over the earth's oceans monthly for several years). Two papers have been published (Kara et al, 2007(Kara et al, , 2009b, one accepted for publication (Kara et al, 2009a), and and two submitted for publication (Barron et al, 2008;Gunduz et al, 2009).…”