Don't you realize that CBS's crowd-counting methodology was the only reliable estimate of the 8/28 crowd? They told us so here:
To calculate attendance at the Beck rally, AirPhotosLive.com used what is called a surveillance aerostat balloon to take pictures from both above the event and closer to the ground. In the video above, which was provided by the company, you can see some of the images used to come up with a figure.
And don't even think about questioning how they can count the people between the trees:
The balloon, Westergard said, gave the company the capacity to move up and down, allowing it to photograph people who were standing beneath trees in addition to taking photos from high above.
Heck, CBS was even generous enough to give a plus or minus of 9,000 people. So yes, there could have been more than 87,000--it might have been 96,000!
And remember, CBS's number "was the only scientific [emphasis mine] estimate made of the number of people at the rally."
(P.S. I hope you don't read the article by the Christian Science Monitor and fire me. You know, the one that says using aerial photo estimates "reminds us of the old intelligence agency practice of 'shedology,' in which the Central Intelligence Agency and others estimated Soviet weapons numbers by looking at overhead photos of military sheds and calculating how many tanks they might hold." Yeah, don't read that.)
A possible sighting of the scientific estimation in progress: