Large scale experiments planned for the future as part of the Global Atmospheric Research Project (GARP) will call upon measurements taken from all categories of ocean vessels.
During the 1968 Barbados Experiment meteorological measurements were made aboard the USCGSS Discoverer using a variety of sensors and methods typical of many ocean vessels. A comparison of the Discoverer data systems provides an estimate of the inherent error in some shipboard measurements. It is demonstrated that under unstable stratification, wind speed differences from anemometers mounted at separate locations and heights are small while under neutral conditions large differences appear. Unless special precautions are taken, heating of sensors due to solar radiation and ship effects may be as large as 2C. Such an error in dry bulb and wet bulb temperatures with a mean wind speed of 6 m sec−1 yields a 100% error in the sensible heat flux and a 50% error in the latent heat flux calculations. Nomograms of sensible and latent heat flux are constructed to demonstrate the manner in which such errors influence these quantities.