The Distribution of Clouds at Tucson, Arizona, with Respect to Type, Amount, and Time of Observation

Abstract

More than 25,000 Weather Bureau cloud observations for Tucson, Arizona, for the nine-year period from 1 July 1945 to 1 July 1954, were statistically analyzed. It was immediately obvious from this analysis that the cloud data for this period were not homogeneous. A study of the heterogeneities in the data indicated that they were probably caused by two changes which had been made in the Weather Bureau cloud observing procedures. These changes were: (1) In 1949 the descriptive material for cirrostratus clouds was changed when the revised version of the Manual of Cloud Forms and Codes for States of the Sky was published. !!his change is believed to have caused a large increase in the annual number of observations of cirrostratus clouds and a correspondingly large decrease in the annual number of observations of cirrus clouds. (2) On July 1, 1948 a new procedure for recording cloud observations on WBAN Form lOB was adopted. This new recording procedure resulted in cloud observations which were not as detailed as were the observations made before July 1, 1948. The change from the old "cloud family" recording procedure to the new "cloud layer" recording procedure resulted in a decreased number of recorded observations of cumulonimbus and altostratus clouds. The frequencies of occurrence were computed for eight types of clouds (stratocumulus, cumulus humilis, cumulus congestus, cumulonimbus, altostratus, altocumulus, cirrus and cirrostratus) and the annual and diurnal variations of these frequencies were determined. The average monthly tenths of cloud cover were also computed and the annual and diurnal variations of the amounts of cloud cover were determined. A comparison of the annual variations of these parameters showed that they were essentially the same for six of the cloud types but were notably different for the cumulus humilis and cumulus congestus cloud types. The annual variations of the frequencies of occurrence of the most common cloud types were also compared to the annual variations of three synoptic parameters (the number of frontal passages, the amount of precipitation and the amount of precipitable water). These comparisons seemed to indicate that there was a relationship between the general synoptic situation and the amount of cloudiness during some months. However, the number of synoptic parameters included in this part of the study need to be increased before any definite conclusions can be reached. For each month, the year-to-year variations in the amounts of high, middle and low cloudiness at Tucson were compared with the year-to-year variations of the average amount of precipitation for the Tucson area. In general, changes in the amount of cloudiness were of the same sign as changes in the amount of precipitation, but there were many noteworthy exceptions to this general pattern. In addition, the rank correlations -2- between the monthly amounts of precipitation and the monthly amounts of cloudiness were computed. In general, the precipitation was best correlated with the amount of low cloudiness and the highest of these correlations occurred during the winter months. However, there were many interesting deviations from this general pattern.