An investigation of summer sea surface temperature anomalies in the eastern North Pacific Ocean

Abstract

A study of summer sea surface temperature anomalies in the eastern North Pacific Ocean was undertaken to examine some of the processes that could affect their evolution and which may be important for their prediction. An empirical approach was utilized. The factors considered include sea surface temperature (SST) persistence (due to the relatively large heat capacity of the water), oceanic thermal advection, and wind mixing (which presumably acts through changes in the mixed layer depth of the ocean). The primary variable studied was the summer SST anomaly for each of 25, 5° latitude by 10° longitude boxes in the region 140°W-170°E, 30“55° N. Thirty years of data were used (1947—76).

Diagnostic analyses using data from four Ocean Weather Stations (C, D, P and V) were performed prior to the analyses described above in order to examine wind mixing more closely. Only at the OWS's are high frequency (3 hour), measured wind speeds available at the same location as SST observtions for a long period of time. These analyses indicated a statistically significant lag relationship between average April and May monthly wind speed and subsequent summer SST at Station P.

The monthly mean SLP gradient was used as an estimate of average monthly wind speed over the 25 box domain where measured winds are not routinely available. In addition, winds derived from daily sea level pressure (SLP) analyses were averaged to form monthly means. At most boxes, no significant lag relationship between summer SST anomalies and spring (April and May) winds was found using either wind data set.

The box-averaged data were employed to specify the summer SST from the components of the total horizontal thermal advection (based on the mean and anomalous components of the surface current and SST fields). The anomalous summer surface current advecting the mean summer SST field was found to be the dominant horizontal advective term affecting the local time rate of change of SST, particularly for 40—50° N. Substitution of derived May anomalous currents for those of summer greatly diminished the strength of the relationship.

In the predictive portion of the study, skill relative to persistence was marginal based upon dependent sample testing. Thus, in the predictive mode neither wind mixing (as parameterized here) nor oceanic thermal advection added very much information not contained in the initial SST anomaly field. However, persistence alone exhibited considerable skill, particularly in the northeastern portion of the domain.