Long climate records in the ocean are rare. In the California Current more than 50 year of observations obtained during the CalCoFI program afford a unique look at decadal climate variations in an eastern boundary region. Line 90 of the CalCoFI program extends several hundred kilometers offshore from the southern California bight, and has been sampled by several cruises annually from the early 1940s to the mid 1970s, and from the mid 1980s to present. In the 1970, date coverage was more intermittent, with several years without observations. In addition, historic data from the late 1930 and early 1940 have been added to extend the time coverage.

Salinity in the upper 150 m of the water column is dominated by decadal variability, with variations of the order of 0.15 psu. Salinities were low in the early 1950s, from the late 1960s through the early seventies, and in the early 1990s. Waters were anomalous salty in the late 1950s and early 1960, in the 1980s and late 1990s. The salinity anomalies are coherent across the entire line 90, with largest anomalies between 120W and 122W. The variations are associated with changes of the waters mass and are independent of the North Pacific Oscillation (NPO) index. In contrast, variations of temperature dominate anomalies of density, and result from a combination of ENSO and NPO signals.

Accordingly the spectrum of salinity, as measured by an index of the salinity variations, is red, while the spectra of temperature and geostrophic flow, to the extent that it can be estimated from the sparse temporal sampling, are white. This suggests that the lack of vigorous feedback with the air-sea fresh water flux leads to an accumulation of any imbalance of the salt budget, a random walk, with variance at decadal time scales dominant. Salinity variations before the 1980s are consistent with anomalous advection of the mean north-south salinity gradient. Estimates of the low frequency geostrophic flow and salinity changes are correlated, and imply a realistic, along-shore salinity gradient.

However, in the 1980s and 1990s, the warming of the California current, and its southward acceleration are not associated with a corresponding decrease of salinity. This indicates that other forcings are important. As yet, estimates of local forcing by the surface fresh water flux, across shore advection have not been shown to explain this signal, possibly reflecting the noisy nature of the data. We are exploring Pacific hindcasts and high resolution ocean models, to explore the salt dynamics of the upper ocean during the more recent decades.