VARIABILITY IN A MIXED-LAYER OCEAN MODEL-DRIVEN BY STOCHASTIC ATMOSPHERIC FORCING

A stochastic model of atmospheric surface conditions, developed from 3 0 years of data at Ocean Weather Station P in the northeast Pacific, i s used to drive a mixed layer model of the upper ocean. The spectral c haracteristics of anomalies in the four atmospheric variables; air and dewpoint temperature, wind speed and solar radiation, and many ocean features, including the seasonal cycle are reasonably well reproduced in a 500-year model simulation. However, the ocean model slightly unde restimates the range of the mean and standard deviation of both temper ature and mixed layer depth over the course of the year. The spectrum of the monthly SST anomalies from the model simulation are in close ag reement with observations, especially when atmospheric forcing associa ted with El Nino is included. The spectral characteristics of the midl atitude SST anomalies is consistent with stochastic climate theory pro posed by Frankignoul and Hasselmann (1977) for periods up to similar t o 6 months. Lead/lag correlations and composites indicate a clear conn ection between the observed SST anomalies in spring and the following fall, as anomalous warm or cold water created in the deep mixed layer during winter/spring remain below the shallow mixed layer in summer an d is then reentrained into the surface layer in the following fall and winter. This re-emergence mechanism also occurs in the model but the temperature anomaly pattern is more diffuse and influences the surface layer over a longer period compared with observations. A detailed ana lysis of the simulated mixed layer temperature tendency indicates that the anomalous net surface heat flux plays an important role in the gr owth of SST anomalies throughout the year and is the dominant term dur ing winter. Entrainment of water into the mixed layer from below stron gly influences SST anomalies in fall when the mixed layer is relativel y shallow and thus has little thermal inertia. Mixed layer depth anoma lies are highly correlated with the anomalous surface mechanical mixin g in summer and surface buoyancy forcing in winter.