WAKE FOREST UNIVERSITY
DEPARTMENT OF MATHEMATICS & STATISTICS
Wake Forest University
January 15, 2019 at 11:00am
Manchester Hall, Room 018
“ Estimating Atmospheric Motion Winds from Satellite Image Data
using Space-time Drift Models”
Geostationary satellites collect high-resolution weather data comprising a series of images which can be used to estimate wind speed and direction at< altitudes. The Derived Motion Winds (DMW) Algorithm estimates atmospheric winds by tracking features in images taken by the GOES-R series of the NOAA geostationary meteorological satellites. However, the wind estimates from the DMW Algorithm are sparse and it is not possible to quantify uncertainties associated with the process. This motivates us to statistically model wind motions using a spatial process drifting in time. We consider a covariance function depending on spatial and temporal lags and the drift parameter, which captures the wind speed and wind direction. We< estimate the parameters by maximizing the profile likelihood. Our method allows us to compute standard error of the estimates. We smooth the estimated fields using a Gaussian kernel, weighted by the inverses of estimated variances which gives us more accurate estimates of the wind field. We conduct extensive simulation studies to determine situations where our method should perform well. The proposed method is applied to the GOES-15 brightness temperature data over Colorado and reduces prediction error of brightness temperature compared to the DMW Algorithm.