Precipitation Mapping

Abstract

Techniques have been explored for the automated detection and characterization of precipitation areas in weather radar data. The purpose of this effort is to develop guidance for the operational forecaster for the NOWCASTing of start and stop of precipitation. Precipitation intensity is not considered in this effort. Two techniques have been adopted to extract precipitation areas from the radar reflectivity fields: contour extraction and edge detection. The first method allows the selection of regions within a storm that the forecaster considers to be significant. An efficient technique for the extraction and characterization of the contours based on the Freeman Chain Code has been adopted. The other technique is an edge detection technique that is based on gradient of reflectivity factors. This has the capability of detecting more of the internal structure of the storm and to characterize the regions along the outer edge where the weather might be more intense. This latter technique produces lines like the contour method that can be characterized by the Freeman Chain Code. Efficient techniques for the characterization of these lines or areas encompassed by these lines have been developed based on the chain code. It has been found that a combination of the contour and edge detection is a powerful tool that permits interpretation of the location and subsequent movement of precipitation areas and also an indication of changes in intensity.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Feb 15, 1991
Accession Number
ADA246724

Entities

People

  • Teresa M. Bals

Tags

Communities of Interest

  • Space

DTIC Thesaurus Topics

  • Air Force
  • Atmospheric Sciences
  • Cartesian Coordinates
  • Change Detection
  • Cold Fronts
  • Computations
  • Contracts
  • Coordinate Systems
  • Corporations
  • Detection
  • Filtration
  • Grids
  • Leading Edges
  • Meteorological Radar
  • Military Operations
  • New England
  • Radar

Readers

  • Atmospheric Science/Meteorology
  • Computational Fluid Dynamics (CFD)
  • Computer Vision.