GSO Technical Report Number
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 License.
The objective analysis (OA) technique was adapted by Watts and Tracey in order to map the thermal frontal zone of the Gulf Stream. Here, we test the robustness of the adapted OA technique to the selection of four control parameters: mean field, standard deviation field, correlation function, and decimation time. Output OA maps of the thermocline depth are most affected by the choice of mean field, with the most realistic results produced using a time-averaged mean. The choice of the space-time correlation function has a large infuence on the size of the estimated error fields, which are associated with the OA maps. The smallest errors occur using the analytic function, ρWT, which is based on four years of inverted echo sounder data collected in the same region of the Gulf Stream. Variations in the selection of the standard deviation field and decimation time have little effect on the output OA maps.
We determine the accuracy of the output OA maps by comparing them with independent measurements of the thermal field. Two cases are evaluated: standard maps and high temporal resolution maps, with decimation times of 2 days and 1 day, respectively. Standard deviations (STD) between the standard maps at the 15% estimated error level and the XBTs (AXBTs) are determined to be 47- 53 m. The comparisons of the high temporal resolution maps at the 20% error level with the XBTs (AXBTs) give STD differences of 47 m.
Tracey, Karen L.; Friedlander, Amy I.; and Watts, D. Randolph, "Objective Analysis of the Gulf Stream Thermal Front: Methods and Accuracy" (1987). Physical Oceanography Technical Reports. Paper 34.