Latest revision as of 20:25, 3 February 2021

Abstract

Thunderstorms are a leading cause of delay in the National Airspace System (NAS), and significant research has been conducted to predict the areas pilots will avoid during a storm. An example of such research is the Convective Weather Avoidance Model (CWAM), which provides the likelihood of pilot deviation due to convective weather in a given area. This paper extends the scope of CWAM to include low-altitude flights, which typically occur below the tops of convective weather and have slightly different operational constraints. In general, the set of low-altitude flights includes short-hop routes and low-altitude escape routes used to reduce the impact of convective weather in the terminal area. This paper will discuss the classification procedure, present the performance of low-altitude CWAM on observed and forecasted weather, analyze areas of poor performance, and suggest potential improvements to the model. I. Introduction ONVECTIVE weather is a significant impediment to effective and efficient Air Traffic Management (ATM) decisions, and sometimes results in unnecessary delays to the National Airspace System (NAS). In the NAS, 70% of delays are caused by weather, and of those delays, 60% are specifically accounted for by convective weather [1]. Currently, rerouting decisions made by air traffic managers are aided by weather products such as the Corridor Integrated Weather System (CIWS) and the National Convective Weather Forecast (NCWF) [2, 3]. In a Next Generation ATM system, decision support tools such as the Route Availability Planning Tool (RAPT) will mitigate weather-induced delays by supplementing the situational awareness of an air traffic manager with a forecast of the availability of specific flight routes [4]. RAPT is based on the Convective Weather Avoidance Model (CWAM), which is a probabilistic model of pilot decision making in the presence of convective weather [5]. CWAM is a tool originally developed for the en route flight regime to predict pilot deviation decisions by correlating in-flight deviations of aircraft to the weather features they encounter. The model is based on a database comprised of the deviation decision of each flight and weather statistics along each route, which are obtained from CIWS. Pattern classification experiments on the en route CWAM database show that the most descriptive predictors for deviation are related to echo top height, where the most descriptive is the difference in altitude between the aircraft and the echo top height [5]. In the terminal area, deviations are predicted with a different set of features. Several studies of the Dallas and Memphis areas using weather information from the Integrated Terminal Weather System (ITWS) show that deviation decisions are closely related to the radar intensity of the storm and the proximity of the aircraft to the airport [6, 7]. This paper presents the development of a low-altitude version of CWAM which is based on a database composed of weather encounters that occur during level flight between FL100 and FL240. This model is applicable to jet traffic that uses low altitude air routes to „escape‟ from terminal areas when weather or volume congestion impacts lead to constraints on high-altitude airspace, or to low-altitude flight by regional jets on „short hop‟ routes. Such traffic is common in major metroplex airspaces. In this analysis, flight trajectories are obtained from the Enhanced Traffic Management System (ETMS) database, and weather data are acquired from CIWS for 23 convective weather days across two geographical regions (Chicago and New York). A Gaussian classifier is used to determine a set of deviation predictors and the results are tested on observed and forecasted data. The predictor performance is compared to the existing terminal departure CWAM used in RAPT, and the differences are discussed.

Original document

The different versions of the original document can be found in:

• http://pdfs.semanticscholar.org/ece0/e9efc61dd044f296f67f2024e1388fedb4e1.pdf

• http://arc.aiaa.org/doi/pdf/10.2514/6.2011-6536,

http://dx.doi.org/10.2514/6.2011-6536

• https://arc.aiaa.org/doi/10.2514/6.2011-6536,

https://academic.microsoft.com/#/detail/2096272809