On the accuracy of sea‐surface temperatures and salinities observed in the northeast pacifie ocean

Observations of sea‐surface temperatures and salinities, made by a variety of methods during August and September 1975 in the northeast Pacifie Ocean, are examined to evaluate the quality of surface data. The bucket method is capable of providing sea‐surface temperatures to an accuracy (standard deviation) of ±0.15°C. The thermograph/salinograph when corrected by applying a “field‐calibration” value, gives temperatures with a standard deviation one half that obtained by the bucket method. Expendable bathythermograph temperatures were, on the average, 0.3°C urate as the true values. Were it not for this offset they would have been as accurate as those obtained with bucket thermometers. Engine‐intake temperatures observed by the engine‐room crew were, on the average, 0.3°C larger than the true values, but were characterized by large inaccuracies, with a standard deviation about an order of magnitude greater than those found for other methods. These variations are believed to be due to reading errors. Sea‐surface salinities observed with the bucket could be, with reasonable care, accurate within the limitation of the salinometer method used aboard ships. The quality of data has been found to vary significantly between observers. Results obtained from this cruise and from weathership data (1956–1976) suggest that the surface temperatures and salinities observed during the past, 1956–1962, in the northeast Pacific Ocean have generally been overestimated.