Characteristics of 3D Atmospheric Motion Vectors (AMV) from Water Vapor Feature-Tracking Technique

Atmospheric motion vectors (AMVs) can be derived from tracking clouds or water vapor features in consecutive satellite images. Considering future space-borne lidar can produce 3D winds but in very narrow slices, AMVs from constellations of passive sensors on LEO or GEO orbits can provide 3D distributions of horizontal winds in broad areas complementary to lidar wind measurements. Accurate determination of the uncertainties of AMVs is crucial if the observations are to be properly assessed in the context of Observing System Simulation Experiments (OSSE). In this study, we derive AMVs by water vapor feature-tracking technique based on maximum correlation between two consecutive water vapor fields. Ten WRF simulations of an extratropical cyclone off the US east coast are used as ensemble nature runs. The sensitivity of the derived AMVs to temporal gap of the water vapor fields, tracking box size, water vapor and its gradient, wind speed and direction will be examined. We also coarse out the high-resolution model outputs to match the proposed horizontal and vertical resolutions of a CubeSat infrared sounder to estimate its 3D wind measurement uncertainties. Hypothetical orbital simulations will be performed to estimate possible spatial coverage of 3D winds by CubeSat constellations of IR sounders.