THE LINEAR PREDICTABILITY OF PRECIPITATION
Abstract
The problem of the short-term predictability of precipitation remained rather unamenable to investigation by dynamic methods. A diagnostic study of this problem was made using a linear prediction scheme based upon statistics of precipitation intensity. The data necessary for the determination of such statistics were obtained by the aid of a horizontally scanning radar with the signal returned by the precipitation being a measure of its intensity. It is found that (1) the linear predictability of precipitation is relatively insensitive to reasonable variations in the spatial density of the data, however, the spatial extent of the data may be a critical parameter; (2) the nonlinear behavior of precipitation elements whose sizes range from one mile up to 100 miles in order of magnitude is important for the short term predictability; (3) the space resolution with which a forecast is made will influence to some extent the level of predictability; and (4) the quality of linear prediction decreases with time as expected but not in a systematic manner because of the presence of different sizes of precipitating elements in the storms from which the data were obtained.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1962
- Accession Number
- AD0287465
Entities
People
- Thomas M. Noel
Organizations
- Massachusetts Institute of Technology