An Objective Method with a Continuity Constraint for Improving Surface Velocity Estimates from the Geostationary Ocean Color Imager

Zifeng Hu,Lan Li,Jun Zhao,Dongxiao Wang

doi:10.3390/rs14010014

Abstract

Mapping surface currents with high spatiotemporal resolution over a wide coverage is crucial for understanding ocean dynamics and associated biogeochemical processes. The most widely used algorithm for estimating surface velocities from sequential satellite observations is the maximum cross-correlation (MCC) method. However, many unrealistic vectors still exist, despite the utilization of various filtering techniques. In this study, an objective method has been developed through the combination of MCC and multivariate optimum interpolation (MOI) analysis under a continuity constraint. The MCC method, with and without MOI, is applied to sequences of simulated sea surface temperature (SST) fields with a 1/48° spatial resolution over the East China Sea continental shelf. Integration of MOI into MCC reduces the average absolute differences between the model’s ‘actual’ velocity and the SST-derived velocity by 19% in relative magnitude and 22% in direction, respectively. Application of the proposed method to Geostationary Ocean Color Imager (GOCI) satellite observations produces good agreement between derived surface velocities and the Oregon State University (OSU) regional tidal model outputs. Our results demonstrate that the incorporation of MOI into MCC can provide a significant improvement in the reliability and accuracy of satellite-derived velocity fields.

Highlights

Ocean surface current observations have critical importance in understanding ocean dynamics, air–sea interactions, and biogeochemical processes
Our results demonstrate that the incorporation of multivariate optimum interpolation (MOI) into maximum cross-correlation (MCC) can provide a significant improvement in the reliability and accuracy of satellitederived velocity fields
The histograms for MCC+MOI are closer to zero (∆V = ∆θ = 0). These results indicate that the incorporation of MOI into MCC can greatly improve the accuracy of surface velocity estimates, together with no data loss

Summary

Introduction

Ocean surface current observations have critical importance in understanding ocean dynamics, air–sea interactions, and biogeochemical processes. In coastal waters, real-time current observations with a high spatiotemporal resolution take much more significance through their impacts on search and rescue missions, environmental monitoring, and optimal ship routing. The prospect of observing nearly instantaneous surface velocity fields without temporal aliasing over a large area has prompted numerous attempts to infer advective surface currents directly from sequential satellite imagery [1,2,3,4,5,6,7]. A well-known method called the maximum cross-correlation (MCC) method is by far the most widely used to obtain high spatial resolution surface currents in coastal areas from thermal infrared and ocean color satellite observations since it is less sensitive to errors and does not need precise values [2,8,9,10]. By combing chlorophyll a with SST images off the California coast, Crocker et al [8] demonstrated

Objectives

Methods

Conclusion