Generation of Hourly Ambient Temperature from Generalized Cumulative Frequency Distributions

The present article is devoted to make available a method to generate hourly ambient temperature values when the monthly average hourly ambient temperature, $T¯ah$ is known. Cumulative frequency distributions for hourly ambient temperature have been developed employing normalized variables. The normalized variables used in the present study contain not only the monthly average hourly ambient temperature but also the maximum, $Tah,max,$ and minimum, $Tah,min.$ In order that the frequency distributions developed in the present study become available for the purpose of calculating hourly ambient temperature when $T¯ah$ only is known, correlations to estimate $Tah,max$ and $Tah,min$ from known parameters have been presented. The ability of the present distributions to generate hourly ambient temperatures during a day has also been established. A large data base of 239 locations has been employed in developing and validating the distributions and the needed additional correlations. The predicted hourly ambient temperature values have been found to be in agreement with data values within a % rms difference of 0.30 (when $Tah$ is in K) and the standard deviation has been found to be 0.85 K. The agreement between the predicted hourly ambient temperatures during a day employing the present distributions and the data values has been excellent.