Analysis of normalized daily change of air temperature an S-shaped function to detect fog occurrence

摘要

A new method to evaluate the influence of fog on the pattern of normalized temperature from predawn to noon was developed. An S-shaped approximation function was used to characterize the temperature course. One parameter of this S-shaped function, a,can be used to detect places where the rise in air temperature is delayed. The relationship between a and fog occurrence was analyzed. Three findings were identified as follows. Foremost, a positive correlation was found between the number of days in November with fog and the values of a during each November. Secondly, a is strongly related to the visibility conditions after sunrise. Finally, a is large for observation points where fog occurs quite frequently during autumn, such as in Miyoshi, Hiroshima, Japan.