Interview: Farouk El-Baz Discusses the Varying Estimates of Protesters in Washington over the Weekend Demonstrating Against a Possible War with Iraq

All Things Considered: January 20, 2003

War Protest Numbers

ROBERT SIEGEL, host:

Saturday's anti-war protest in Washington, DC, drew an awful lot of people to the National Mall. `An awful lot' was what passed for a crowd estimate by the District of Columbia police. The US Capitol Police put the number at 30 to 50,000. Some organizers said it was a half a million. Many news organizations, including this one, reported that tens of thousands of demonstrators took part, a phrase so vague that it seems merely to quantify that police estimate of `an awful lot' of demonstrators.

The US Park Police have been barred by Congress from issuing crowd estimates ever since they said that the Million Man March drew about 400,000. In that controversy, Professor Farouk El-Baz of Boston University stepped in and applied satellite remote sensing techniques. He and a team at BU came up with a crowd estimate for the Million Man March of 878,587, give or take 25 percent.

Now, Professor El-Baz, once again we have a demonstration in Washington and widely varying estimates of how many people took place. Why is it so complicated to come up with a number that reasonably estimates how many people were in one place at one time?

Professor FAROUK EL-BAZ (Boston University): It is really very difficult to estimate a crowd if you are at the ground level or in helicopters flying away from the crowd and just eyeing them in and giving a kind of a feel of how many people. So these are really guesstimates. They are not even estimates.

SIEGEL: Well, what is the ideal perspective from which to count the people on, say, the National Mall?

Prof. EL-BAZ: For a given time, a very specific time, you can have a plane flying overhead, so it will have to be a fixed-wing aircraft, and take photographs all at one time and you can apply the methodologies that we apply in understanding things in satellite images. For instance, I developed a methodology to count sand dunes in the desert by that, by taking just a satellite image and applying a grid and figuring out how many units per grid and then figuring out very swiftly how many dunes are in a given square mileage, like a hundred miles by a hundred miles.

SIEGEL: Let's say that the Associated Press, which is a widely subscribed to news service, decided to undertake the responsibility of coming up with a crowd estimate. How expensive a proposition would it be for, say, a news service to accurately count the number of people in Washington last Saturday?

Prof. EL-BAZ: It really wouldn't be very expensive if one can rent a fixed-wing aircraft for a short period time, only a couple of hours, and come up with a real number. I think it would be a very great service. The controversy of the Million Man March resulted in the National Park Service getting out of that. After we did our count and our numbers were more than twice the National Park Service estimate, the National Park Service asked me to write a white paper, so to speak, to tell them, how would they do it if they can do it well? And so I did, and they actually took this white paper to Congress, and Congress at the time elected not to give them the money, so they scratched the responsibility of counting people in crowds from the job description of the National Park Service.

SIEGEL: Your estimate was plus or minus 25 percent. That's a pretty big margin of error.

SOUNDBITE OF LAUGHTER

Prof. EL-BAZ: Yes. And most people ask what this huge margin of error was due to, and it really was only because of the obliquity of the photographs, meaning that buildings covered areas where people were and trees covered where people stood and behind them, and so we took these segments within each photograph and it came out to be like this, 25 percent.

SIEGEL: You're saying that your margin of error was plus or minus 25 percent because you were working with the pictures as they were presented to you, not as they might have been taken by a fixed-wing plane going over ideally, as you would describe it.

Prof. EL-BAZ: That's correct. If we had been able to see the ground in its totality, then we would have come to a number that is very close to reality within a fraction of a percent.

SIEGEL: Well, Professor Farouk El-Baz of Boston University, thank you very much for talking with us today.

Prof. EL-BAZ: Thank you.

Copyright 2003 National Public Radio®. All rights reserved. No quotes from the materials contained herein may be used in any media without attribution to National Public Radio. This transcript may not be reproduced in whole or in part without prior written permission. For further information, please contact NPR's Permissions Coordinator at (202) 513-2000.

This transcript was created by a contractor for NPR, and NPR has not verified its accuracy. For all NPR programs, the broadcast audio should be considered the authoritative version.