All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 0 0 0
PDF Downloads 1 1 1

Climatic Presentations for Short-Range Forecasting Based on Event Occurrence and Reoccurrence Profiles

View More View Less
  • 1 Dept. of Earth and Atmospheric Sciences, Saint Louis University, St. Louis, Mo. 63156
Full access

Abstract

The climatology of persistency has been found to provide useful information to the short-range forecaster. This Information can be obtained directly for any given station by computer-tracing initially observed events throughout the historical records for specified time intervals and tabulating the respective probability of their reoccurrence statistics. Several investigators have suggested ways of generating conditional probabilities of an event reoccurring without resorting to such extensive computer processing. Most of these approximation procedures incorporate the assumption that the conditional-unconditional relationships are independent of the initial forecast time, ceiling category, cloud type, and station location. Research conducted on cloud ceilings at Saint Louis University has shown that these assumptions break down during instances of rare events or when the forecast ceilings under consideration are below 1000 ft. Hence, data from 15 climatically non-homogeneous stations were processed to devise “Universal” analytic functions which obviate them. The resulting empirical formulations permit the calculation of meaningful reoccurrence probabilities for stratified subsets of a station's data base.

The fundamental fact that the climatological statistics, for any one station when properly arranged, can separate certain informations which are geographically or topographically dependent, from others which are not, is discussed. A possible means of augmenting the information content of the climatic data base for any one location through the transferring of pertinent relationships contained in the data base of another is presented.

Abstract

The climatology of persistency has been found to provide useful information to the short-range forecaster. This Information can be obtained directly for any given station by computer-tracing initially observed events throughout the historical records for specified time intervals and tabulating the respective probability of their reoccurrence statistics. Several investigators have suggested ways of generating conditional probabilities of an event reoccurring without resorting to such extensive computer processing. Most of these approximation procedures incorporate the assumption that the conditional-unconditional relationships are independent of the initial forecast time, ceiling category, cloud type, and station location. Research conducted on cloud ceilings at Saint Louis University has shown that these assumptions break down during instances of rare events or when the forecast ceilings under consideration are below 1000 ft. Hence, data from 15 climatically non-homogeneous stations were processed to devise “Universal” analytic functions which obviate them. The resulting empirical formulations permit the calculation of meaningful reoccurrence probabilities for stratified subsets of a station's data base.

The fundamental fact that the climatological statistics, for any one station when properly arranged, can separate certain informations which are geographically or topographically dependent, from others which are not, is discussed. A possible means of augmenting the information content of the climatic data base for any one location through the transferring of pertinent relationships contained in the data base of another is presented.

Save