• DocumentCode
    340565
  • Title

    Precipitation measurements using 183-GHz AMSU satellite observations

  • Author

    Staelin, David H. ; Chen, Frederick W. ; Fuentes, Antonio

  • Author_Institution
    Res. Lab. of Electron., MIT, Cambridge, MA, USA
  • Volume
    4
  • fYear
    1999
  • fDate
    1999
  • Firstpage
    2069
  • Abstract
    The 183-GHz humidity channels of the Advanced Microwave Sounding Unit (AMSU) on the NOAA-15 operational weather satellite respond in part to precipitation. The 15-km resolution data at 183.31±7 and 183.31±1 GHz were combined to yield point precipitation estimates from a simple neural network trained using NEXRAD radar precipitation data. A computer used these estimates to locate all significant precipitation cells and to define extended nonoverlapping geographic regions centered on each. The total estimated precipitation rates (liters/sec) and areas for these regions were then fed to a second neural network trained using NEXRAD data. One two-day test involved the eastern United States during a major frontal passage and a major hurricane. NEXRAD missed regions where 2.3 percent of the AMSU-NEXRAD-estimated rain fell, and AMSU missed regions where 11.8 percent of the NEXRAD-estimated rain fell. For those 18 test regions detected by both systems, the precipitation-weighted rms difference between AMSU- and NEXRAD-based cell-wide precipitation estimates was 2.2 dB; these 18 test regions were different from the 18 used to train the second network. Percentage differences in precipitation were generally smaller for the larger rain cells
  • Keywords
    atmospheric precipitation; atmospheric techniques; geophysical signal processing; neural nets; radiometry; rain; remote sensing; 183 GHz; AMSU; Advanced Microwave Sounding Unit; atmosphere; measurement technique; meteorology; microwave radiometry; neural net; neural network; precipitation; precipitation cell; precipitation rate; rain; rainfall; satellite remote sensing; Acoustic scattering; Atmosphere; Brightness temperature; Electromagnetic heating; Frequency; Ice; Neural networks; Ocean temperature; Rain; Satellites;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Geoscience and Remote Sensing Symposium, 1999. IGARSS '99 Proceedings. IEEE 1999 International
  • Conference_Location
    Hamburg
  • Print_ISBN
    0-7803-5207-6
  • Type

    conf

  • DOI
    10.1109/IGARSS.1999.775034
  • Filename
    775034