An Empirical Model to Predict Widespread Occurrences of Contrails

Abstract

The increases in total cloud amount documented for large regions during the latter half of the twentieth century have focused attention on the potential contribution from jet condensation trails (contrails). The environmental conditions that favor contrail formation and persistence are not well understood primarily due to the limited number of empirical studies. This study presents an empirical model to predict widespread occurrences of contrails (outbreaks), which was developed from a combination of rawinsonde temperature and GOES water vapor information. Environments containing persisting contrails were first identified on Defense Meteorological Satellite Program satellite imagery for the United States for January and April 1987 and then analyzed in more detail using Advanced Very High Resolution Radiometer (AVHRR) satellite digital data. Adjacent clear and cloudy environments not containing contrails were identified to compare with the conditions favorable for contrail persistence. For this purpose, a predictive logistic model was developed through multiple regression analysis. The model performance was evaluated through goodness-of-fit methods and found to be statistically significant across a range of atmospheric conditions. To further evaluate the model and to demonstrate its application on a real-time basis, predictions of the probability of persisting contrails were made for a case day. Comparisons of the predictions to satellite observations of the existing conditions (using AVHRR data) demonstrate good model performance and suggest the utility of this approach for predicting persisting contrail occurrence. Implementation of this model should allow climate researchers to better quantify the influence of contrails on surface climate and natural cloud formation.