An Empirical Method of Determining Summer Wet-Bulb Design Temperatures

Abstract

An empirical method is developed for determining monthly cumulative frequency distributions of the hourly wet-bulb temperatures. The monthly distributions for the four warmest months can be combined into a four-month cumulative frequency distribution to determine summer wet-bulb design temperatures. The method is based on the fact that these monthly distributions closely approximate normal distributions. The standard deviations of theft monthly distributions (Sw) vary little over large areas, especially in tropical regions below 30 degrees latitude. The only parameters required to determine a particular wet-bulb temperature distribution are the monthly means of dry-bulb temperature and relative humidity plus the monthly values of Sw for that area. Values of Sw are given for many locations throughout the world. The monthly distributions are slightly platykurtic over regions with a continental temperature regime, and a correction factor is developed to adjust for this deviation from a normal distribution. The method was evaluated for 10 stations throughout the world with a wide variety of climates. The average absolute error of the estimate design temperatures for these stations was less than 1.0F.