This is an Advanced Very High Resolution Radiometer (also known as an AVHRR) thermal image of the Gulf of Mexico acquired in March, 1996. The dark regions near the top are land. Find Florida at the right edge and next to it, Louisiana and Texas. This is a composite of imagery acquired on two different days – March 13 and 14.
Let us compare these satellite-based temperature estimates with direct measurements shown in the table. This set of data was recorded by NDBC (National Data Buoy Center) buoy 42003 which is located at 26.03 degrees North latitude, 85.88 degrees W longitude. The position of this buoy is marked on the AVHRR image. ( white pointer)
The AVHRR images were acquired at approximately 19:25 Universal Time but we have shown buoy data for two hours before and after the satellite overpass. Here, the buoy measurements are the “ground truth” and they show air temperatures between 20 and 22 degrees; and water temperatures between 25 and 25.5 degrees. Notice that the satellite-based observations water termperature at that location appear to be far closer to the air temperatures than the water temperatures directly measured at the buoy!
The reason for this discrepancy is the buoy-based water temperatures were obtained at a depth of 1m below the water surface (this is known as the ‘bulk water temperature’), however, the satellite estimates the temperature at the sea surface (also known as the ‘water skin temperature’). The difference between the bulk and skin temperature can vary depending on atmospheric conditions, but tends to be on the order of one degree.
As you can see, a model that uses indirect measurements to estimate the sea surface termperature needs to be based on measurements of ocean temperature obtained close to the surface . Similarly, comparisons between direct and indirect measurements of ocean temperature must be for temperatures obtained at comparable depths.
