The Utilization of Long Term Temperature Data in the Description of Forecast Temperatures

摘要

This effort was initiated to objectify the process of choosing descriptive adjectives for temperature forecasts. The initial approach to the problem was to look at the monthly means and the distribution of temperature values varies a great deal from season to season. The temperature range that could be considered the 'seasonal normal' has a much greater span in the winter months than during the summer season. In order to get a more positive picture of what should be considered a normal temperature range and what constitutes a significant deviation from the normal, the decision was made to look at an extensive time series of the daily maximum and minimum temperature data. A tape of daily Columbia, Missouri data for the period 1900-1979 was used to analyze the data.