全球变化知识资源中心

Weather shocks frequently impair the smooth functioning of transportation systems. We use a dataset consisting of 2.14 million flight departures from ten large U.S. airports between January 2012 and September 2017, and estimate the effects sudden onset events, i.e., precipitation and wind, and slow onset events, i.e., temperature, have on departure delay. For sudden onset events, we apply a difference-in-difference framework that allows for inferences at the hourly level. The effects of slow onset events on departure delay are estimated based on a Prais Winstein estimator with panel-corrected standard errors. Our estimates show a significant increase in departure delay of up to 23 min depending on the weather type and intensity of the disturbance. Given the social costs of schedule delays, these results are of high economic importance.