Comparisons of Satellite-Derived Atmospheric Motion Vectors, Rawinsondes, and NOAA Wind Profiler Observations

• Published in: Journal of applied meteorology and climatology Vol. 48; no. 8; pp. 1542 - 1561
• Main Authors: Bedka, Kristopher M., Velden, Christopher S., Petersen, Ralph A., Feltz, Wayne F., Mecikalski, John R.
• Format: Journal Article
• Language: English
• Published: Boston, MA American Meteorological Society 01.08.2009
• Subjects: Accuracy; Agreements; Air flow; Algorithms; Atmospheric models; Atmospheric motion; Atmospherics; Automation; Boundary layers; Clouds; Confluence; Control algorithms; Convection clouds; Data assimilation; Data processing; Datasets; Diagnosis; Earth surface; Earth, ocean, space; Exact sciences and technology; External geophysics; Flow distribution; Flow pattern; Geostationary satellites; GOES satellites; Information processing; Information services; Low-level jets; Lower stratosphere; Mathematical vectors; Mesoscale phenomena; Meteorological satellites; Meteorology; Nonfiction; Numerical prediction; Numerical weather forecasting; Planetary boundary layer; Prediction models; Rawinsondes; Satellite imagery; Satellite observation; Satellites; Storms; Stratosphere; Synchronous satellites; Systems development; Temporal resolution; Thunderstorm development; Thunderstorms; Troposphere; Validation studies; Vectors; Weather; Weather forecasting; Wind; Wind velocity; Narrows; New York; United States; Wisconsin; United States > US; Validation; algorithms; Mid latitude; GOES satellites; NOAA satellite; Weather forecast; Space remote sensing; Satellite observation; Geostationary satellite; digital simulation; Rawinsonde; Low level jet; Forecast model; troposphere; North America; Polar orbiting satellite; Mesoscale; Meteorological satellite; Confluence; Wind satellite; winds; Numerical forecast; stratosphere
• ISSN: 1558-8424; 1558-8432
• DOI: 10.1175/2009jamc1867.1

Geostationary satellite-derived atmospheric motion vectors (AMVs) have been used over several decades in a wide variety of meteorological applications. The ever-increasing horizontal and vertical resolution of numerical weather prediction models puts a greater demand on satellite-derived wind products to monitor flow accurately at smaller scales and higher temporal resolution. The focus of this paper is to evaluate the accuracy and potential applications of a newly developed experimental mesoscale AMV product derived from Geostationary Operational Environmental Satellite (GOES) imagery. The mesoscale AMV product is derived through a variant on processing methods used within the University of Wisconsin—Madison Cooperative Institute for Meteorological Satellite Studies (UW-CIMSS) AMV algorithm and features a significant increase in vector density throughout the troposphere and lower stratosphere over current NOAA/National Environmental Satellite, Data, and Information Service (NESDIS) processing methods forGOES-12Imager data. The primary objectives of this paper are to 1) highlight applications of experimental GOES mesoscale AMVs toward weather diagnosis and forecasting, 2) compare the coverage and accuracy of mesoscale AMVs with the NOAA/NESDIS operational AMV product, and 3) demonstrate the utility of 6-min NOAA Wind Profiler Network observations for satellite-derived AMV validation. Although the more conservative NOAA/NESDIS AMV product exhibits closer statistical agreement to rawinsonde and wind profiler observations than do the experimental mesoscale AMVs, a comparison of these two products for selected events shows that the mesoscale product better depicts the circulation center of a midlatitude cyclone, boundary layer confluence patterns, and a narrow low-level jet that is well correlated with subsequent severe thunderstorm development. Thus, while the individual experimental mesoscale AMVs may sacrifice some absolute accuracy, they show promise in providing greater temporal and spatial flow detail that can benefit diagnosis of upper-air flow patterns in near–real time. The results also show good agreement between 6-min wind profiler and rawinsonde observations within the 700–200-hPa layer, with larger differences in the stratosphere, near the mean top of the planetary boundary layer, and just above the earth’s surface. Despite these larger differences within select layers, the stability of the difference profile with height builds confidence in the use of 6-min, ~404-MHz NOAA Wind Profiler Network observations to evaluate and better understand satellite AMV error characteristics.