The bulk sea surface temperature (SST) simulated by COAMPS/NCOM [Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPS) and the Navy Coastal Ocean Model (NCOM)] compares reasonably well with the bulk SST measured on the North Carolina shelf during the October of 2015 field campaign of the CASPER Project (Wang et al. in Bull Amer Met Soc 99:1449–1471, 2018). The model slightly overestimates the field data by about 0.15 °C; a linear regression is strong with a coefficient of determination r² > 0.9. In deep waters of the Gulf Stream core, COAMPS/NCOM typically underestimates the measured SST by less than 1 °C except in the cold lens or meander where the temperature gradients were smoothed by the model. It is evident that sub-mesoscale features like frontal eddies, meanders and filaments, require a resolution higher than the 3 km grid spacing adopted for modeling.

由 COAMPS/NCOM [耦合海洋/大气中尺度预测系统 (COAMPS) 和海军沿海海洋模型 (NCOM)] 模拟的整体海面温度 (SST) 与 10 月份在北卡罗来纳州大陆架测量的整体 SST 相当不错CASPER 项目的 2015 年实地活动（Wang et al. in Bull Amer Met Soc 99:1449–1471, 2018）。该模型略微高估了现场数据约 0.15 °C；线性回归很强，决定系数r ²  > 0.9。在墨西哥湾流核心的深水中，COAMPS/NCOM 通常会低估测量的 SST 小于 1°C，除了在冷透镜或曲流中，温度梯度被模型平滑。很明显，像正面涡流、曲折和细丝这样的亚中尺度特征需要比建模所采用的 3 公里网格间距更高的分辨率。