Sea surface temperature (SST) is an important parameter in climate studies. Traditionally, SST is measured by in-situ platforms (ships and moored and drifting buoys). Recent advancement in satellite technology has made space-based SST observations routine and operational. Satellite observations provide superior spatiotemporal coverage over in-situ measurements. However, satellite data may contain large biases (biases of 2oC have been found). These biases must be corrected using in-situ data for climate analysis. In this presentation, we discuss the strategy of designing an effective and efficient Buoy Need Network (BNN) in terms of satellite bias reduction rate, and to define the in-situ data density needed to reduce the biases to a required accuracy.

Specifically, a statistical method was used to extract the typical bias spatiotemporal patterns of the historical satellite SST data. Future satellite biases were simulated with the extracted bias patterns. Satellite bias reduction rate was examined as a function of buoy density. It was found that the bias reduction and the buoy density (BD) have a nearly exponential relationship, thus an optimal buoy density range can be defined. The biases decrease rapidly as BD on a 10o spatial grid increases from 0 to 3; beyond which the bias reduction levels off. To reduce a 2oC maximum bias to below 0.5oC, a BD of about 2 buoys/10º grid is required. The present in situ SST observing system was evaluated to define an equivalent buoy density (EBD), allowing ships to be used along with buoys according to their random errors. Seasonally averaged monthly EBD maps were computed to determine where additional buoys were needed for future buoy deployments. Additionally, a residual bias can be computed from the EBD of the current in situ observing system to assess the system’s adequacy to remove future potential satellite biases.