Feasibility and error analysis of cloud motion wind extraction from near-simultaneous multiangle MISR measurements

Abstract

Satellite wind measurements represent an invaluable contribution to the description of the flow field over the oceans. Conventional cloud-tracking techniques suffer from the inability to simultaneously determine wind speed and height. Currently, the uncertainty in the independently calculated heights is the major factor limiting the accuracy of cloud motion winds. Near-simultaneous multiangle imagery from the multiangle imaging spectro-radiometer (MISR) forms the basis of a unique method able to simultaneously retrieve cloud motion and height. The coupled motion and height parallaxes can be unscrambled from three properly selected multiangle views through a purely geometric, stereoscopic approach. Results based on simulated data indicate that for a mesoscale domain the average along-track and cross-track horizontal wind components may be obtained with an accuracy as good as 3-4 m s-1, and 1-2 m s-1, respectively, and with a corresponding height error of 300-400 m. The technique also possesses a limited capability to distinguish between low and high features moving at different velocities in a multilayer cloud field.