Derived Motion Winds
Background
The Derived Motion Winds product is derived from using a sequence of visible or IR spectral bands to track the motion of cloud features and water vapor gradients. The resulting estimates of atmospheric motion are assigned heights by using the Cloud Height product. The Derived Motion Wind product provides vital tropospheric wind information over expansive regions of the earth devoid of in-situ wind observations that include oceans and Southern Hemisphere land masses. This product provides key wind observations to operational NWP data assimilation systems where their use has been demonstrated to improved numerical weather prediction forecasts including tropical cyclones. In addition, this product provides improved guidance for NWS field forecasters.
Product Description
This product retrieves atmospheric winds by tracking features in satellite water vapor and longwave and shortwave IR measurements. These are designated as 'water vapor' and 'cloud drift' (or 'cloud motion vector') winds respectively. The GOES-R ABI Derived Motion Winds Algorithm (DMWA) employs a sequence of images to estimate atmospheric motion and assign this motion to a representative height in the atmosphere.
Improvements and Benefits
GOES-R will provide high spatial and temporal resolutions which will improve the accuracy of the output of the winds given the ability to track smaller gradients at a very high temporal resolution. In addition, GOES-R will provide improved geolocated satellite data, this will reduce the errors and biases of the algorithm output. Satellite "wobble" can cause major issues with the accuracy of the output, and GOES-R will be improved in that aspect.
How does it work? - Algorithm
The DMWA uses the ABI visible and infrared observations to extract atmospheric motion. The choice of spectral band will determine the intended target (cloud or moisture gradient) to be tracked, its height in the atmosphere, as well as the scale of its motion. Historically, the coverage of operational GOES DMWs is diurnally consistent in the mid- to upper tropospheric levels (100-600 hPa) through the use of the mid-wave (6.7μm - 7.3μm) water vapor channels and longwave (10.7μm) infrared (LWIR) channel for deriving vectors. In the lower levels (600-950 hPa), DMWs are provided by a combination of the visible (VIS) and IR channels, depending on the time of day. During daylight imaging periods, the VIS channel usually provides superior low-level tracer detection than the LWIR channel due to its finer spatial resolution and decreased susceptibility to attenuation by low-level moisture. During night-time imaging periods, the shortwave (3.9μm) infrared (SWIR) channel compliments the LWIR channel to derive DMWs. The SWIR channel is a slightly "cleaner" window channel than the LWIR (less WV attenuation), making it more sensitive to warmer (lower tropospheric) temperature features (Dunion and Velden, 2002). The SWIR channel is also not as sensitive as the LWIR channel to cirrus clouds that may obscure low-level cloud tracers. These two characteristics make it a superior channel for producing low level DMWs at night.
Example of the Derived Winds product as derived from the Derived Winds algorithm using full disk 15-minute Meteosat-8 10.8μm SEVIRI data for 12 UTC on 01 February 2007. These winds are derived from tracking cloud features using the 10.8μm channel. High level (100-400 hPa) winds are shown in violet; mid-level (400-700 hPa) are shown in cyan; and low levels (below 700 hPa) are shown in yellow.
See the GOES-R ATBD page for all ATBDs.
How are the results compared to existing data? - Calibration and Validation
The DMW product is validated in comparison with reference wind data from radiosondes and wind profilers for wind values over land, from the National Centers for Environmental Prediction (NCEP) Global Forecast System (GFS) model analysis wind fields over oceans, and from other (independent) satellite-based wind products for land and ocean. Metrics are computed that characterize the level of agreement between the satellite retrievals and reference values. The evaluations/comparisons should be performed over reasonably long (preferably continuous) time period that cover different seasons of the year.
A more technical validation presentation, (PDF, 31.02 MB) is also available.