To date, very little attention has been given to quantifying the effects of thunderstorms on air traffic in enroute airspace. What types of storms cause pilots to deviate from their nominal flight routes? What types of storms do pilots fly through? Around? Over? When thunderstorms are forecast to affect a particular region, how many planes will need to be rerouted? Which ones? Which aspects of the storm need to be accurately forecast in order to answer those questions? How does the forecast accuracy affect the quality of airspace capacity predictions? Quantitative answers to these questions would contribute to the design of useful decision support tools. Federal Aviation Administration decision support tools are being equipped with the ability for air traffic managers to define dynamic "flow constrained areas" (FCAs). Each FCA will be a polygon in latitude/longitude space with ceiling and floor altitudes and a motion vector. One primary use for FCAs will be to define regions that do, or probably will, contain convective thunderstorm activity. These tools will help air traffic managers decide which planes to re-route around the weather and which planes have a reasonable chance of flying through, between, or over the storms. Although it will be helpful to have the ability to manually define FCAs in the traffic managers' tools, the efficiency of the solutions that will be worked out with those tools would be greatly enhanced by answers to the questions posed above. In our prior work we have attempted to quantify the behavior of pilots who encounter thunderstorms in terminal airspace during the final 60 nautical miles of flight. In this study we compare the storm avoidance behavior of pilots in enroute airspace with that of pilots who encountered the very same storms at lower altitudes, in terminal airspace. The study is preliminary, but it complements the terminal work, affords some insight into pilot behavior, and raises questions that should be addressed in a larger study.