Prediction of the flight path was done using the CUSF Landing Predictor. This software was able to run predictions using upper level wind forcasts up to 200 hours in advance of the flight. This was critical, as we are nearly surrounded by water in New Brunswick. The projected flight path dictated when we could and could not fly, as well as where we needed to take off from.

Flight Prediction for October 24th.2010: Red dot indicates launch location, green dot: projected landing. Note the projected balloon burst location just North of Hampton.
Oct_24.JPG
As you can see, the apex of the flight would occur just north of Hampton, with recovery south of Petitcodiac. Another possible launch location would be in McAdam, which would keep the baloon further north of the Bay of Fundy, but then we have to start looking out for the Northumberland Strait if winds carry us further west than expected.

CFB Gagetown might also be a concern, but only if we landed in the middle of it. Crossing it in the air should not be a problem, especially above 60 000' which is where I'm predicting we would be by that point.

Interestingly, here is a prediction for a flight launched on October 25th, 2010 at 4PM, you can see it's a much shorter flight in terms of Easterly movement.
Oct_25.JPG

The length of the flight path is highly dependent on the ascent rate and on the decent rate.

It was found that the flight path predictions were very accurate for the ascent. However, the descent was poorly predicted. The prediction was that the payload would travel south as it fell, when in fact in continued to travel east as it had during its ascent. The result was that the landing was nearly 30 km from the predicted zone.

Comparison of predicted and actual flight paths for April 6th, 2011 flight.

PredictedLarger_Flight_Prediction.jpg

Actual. (Note that the launch site was moved about 4 km south prior to launch in order to avoid landing near the water in Hatfield point. In the end, this did not turn out to be an issue)
True_Flight_Path.jpg