Satellite-Image-derived Gulf Stream Currents Compared with Numerical Model Results

Abstract

Sequential infrared satellite imagery is used to objectively compute surface currents in the Gulf Stream region using the maximum correlation (MCC) method. The infrared images, filtered for cloud cover, are used to find the displacement of surface temperature patterns by locating the maximum cross correlation in windowed portions of the image pair. Statistical significance and next-neighbor filter techniques are applied to remove fictitious surface current vectors due to the presence of residual cloud or other nonadvective processes. The core of the Gulf Stream is found to require special treatment due to the high local velocities and the weak sea surface temperature gradients. For the central Gulf Stream, currents are inferred by the MCC tracking of features along the northern edge of the stream. Other special MCC techniques are applied to the strong rotational motions in the Gulf Stream rings. To test the validity of the MCC technique in this geographic region where no in situ measurements were available, a quasigeostrophic numerical model was used to simulate ocean surface currents in the Gulf Stream region. A random surface tracer was introduced into the model field, tracked with the MCC method, and the resulting velocities were validated by comparisons with the model surface currents. Excellent agreement was found for those realizations less than 12 h apart in time, suggesting the reliability of MCC surface currents computed from sequential infrared images separated by less than 12 h.