Photos: the crowd at Donald Trump’s inauguration vs. Barack Obama’s

President Donald Trump boasted his inauguration would have an “unbelievable, perhaps record-setting turnout.“

But aerial shots of the National Mall from President Barack Obama’s 2009 inauguration and today show that isn’t likely. Here’s an image taken at about 11:30 AM ET in 2009:

And here’s what it looks like today as of 11:04 AM ET.

And here they are side-by-side:

Federal and local agencies have estimated that anywhere from 700,000 to 900,000 people will be in Washington, DC, today for Trump’s inauguration. That’s roughly half the number of people who attended Obama’s inauguration in 2009.

It’s also less than the turnout for Obama’s 2013 inauguration, which drew 1 million people.

But experts caution it will be weeks before we know the final count; in the meantime, take all estimates with a grain of salt.

Allison Puccioni, an imagery analyst, told me she helped estimate the number of people in attendance at Obama’s inauguration in 2009. To measure the density of people, she used satellite imagery, making a 90-square grid area of the Capitol and National Mall.

A few days later, the DC government, along with several federal agencies and media organizations, released official estimates that weren’t too far off from Puccioni’s 1.8 million using ticket information from turnstiles and other sources for their count.

Charles Seife, a professor of journalism at New York University and mathematician, has written about the perils of counting crowds. He says that using satellite imagery is one of the best ways to measure the size of a crowd if it’s not a strictly ticketed event, but warned it’s not a perfect tool.

“The best way is by aerial photography ... take pictures and count heads, but even that is fraught,” said Seife. “Pictures are taken at particular times, and [crowd] density may change over time. You’re not going to see everybody. People will be popping in and out of buildings; there are people under trees — you can get a pretty sizable estimate, but it’s not perfectly accurate.”

Estimates of political events can be particularly contentious, cautioned Steve Doig, a professor of journalism at Arizona State University who has been estimating crowd sizes for many years. “No one should be astonished when their claim is ignored, reviled, and claimed to be fake news,” he said.