RAINFALL PROBABILITIES DURING THE CROP SEASON IN SOUTHERN LOWER MICHIGAN

View More View Less
  • 1 U.S. Weather Bureau, East Lansing, Mich.
  • | 2 Michigan State University, East Lansing, Mich.
© Get Permissions Rent on DeepDyve
Restricted access

Abstract

Daily precipitation records for 91 years are analyzed to determine the probability of rain tomorrow (1) regard-less of whether or not it is raining today, (2) given the condition that it is raining today, and (3) given the condition that it is not raining today. Considerable short-period variation was found along with a seasonal trend. A curve for seasonal trend was obtained by the least squares method. Daily variations from this curve tested by the Chi-Square method indicated that these short-term variations were not real, but due to insufficient size of sample. Also the number of n consecutive dates with probability values above or below the smoothed curve were found through Chi-Square tests to conform in frequency to the formula. 0.5n.

Abstract

Daily precipitation records for 91 years are analyzed to determine the probability of rain tomorrow (1) regard-less of whether or not it is raining today, (2) given the condition that it is raining today, and (3) given the condition that it is not raining today. Considerable short-period variation was found along with a seasonal trend. A curve for seasonal trend was obtained by the least squares method. Daily variations from this curve tested by the Chi-Square method indicated that these short-term variations were not real, but due to insufficient size of sample. Also the number of n consecutive dates with probability values above or below the smoothed curve were found through Chi-Square tests to conform in frequency to the formula. 0.5n.

Save