A method developed in the former Soviet Union for predicting cloud amounts is supplemented with a new method of determining the base and tops of clouds. Criteria for predicting a cloud layer are 0 ≤ T″(z) and R″(z) ≤ 0, where T″ is the second derivative of the vertical profile of temperature and R″ is the second derivative of the relative humidity. This test was found from an analyses of United States radiosonde data.
Cloud amount (sky cover) is predicted from a relationship between cloud amount and dewpoint depression within the predicted cloud layer and the temperature at that level. This relationship is based on data from the former Soviet Union and data from the Indian 0cean and divides cloud amount into four categories: 0%20%, 20%60%, 60%80%, and 80%100% coverage.
The new composite method is evaluated using data from several United States radiosonde stations within different climates. Evaluation data was selected to include only situations in which the observer (providing the “truth”) could me only one cloud layer. Consequently, the evaluation is biased toward stratified cloud conditions. The method will provide cloud information that can be used in models of radiosonde sensors to adjusted temperature data.