Myhrvold’s paper, which he made available online, is 110 pages long and has not yet gone through the peer review process – where a study is reviewed by experts unaffiliated with the research before to be published. He analyzed results from NASA’s Wide-Field Infrared Survey Explorer (WISE) spacecraft, which has been operating since 2009 and can read the heat signatures of asteroids that may approach Earth. A mission called NEOWISE translates this thermal data into size estimates using empirical models – mathematical models based on features that scientists have observed on individual asteroids.
Myhrvold says his study completely demystifies the statistical methods NASA scientists used to estimate the size of asteroids on the NEOWISE mission. They say he has no idea what he’s talking about. In fact, a NASA statement claims that Myhrvold made calculation errors so profound that he produced particularly incorrect size estimates for the asteroids, which have already been verified using other methods.
“For every error I found in his paper, if I got a bounty, I would be rich,” Ned Wright, WISE principal investigator at the University of California, told Science Magazine.
“If they really digest it all and say it’s not strong, I’m all ears for how I can make it stronger, but I guess that’s just dismissive,” a Myhrvold told the Post. Myhrvold views the field of astrophysics as dangerously closed to the outside world. He claims that some of the “highest officials working on asteroids” told him privately that they could not reproduce the results he criticizes, but that no one wanted to upset the status quo.
“The fact that I’m independent and outside the establishment makes it easier for me to challenge these things,” he said.
Amy Mainzer, principal investigator of NEOWISE, pointed out that many of the measurements from her projects have actually been replicated and confirmed. Japan recently operated an infrared telescope called Akari, which is similar to WISE and has measured many of the same asteroids.
“There is a very good paper that does a very good job of comparing these size estimates,” Mainzer told the Post. “Everyone does their own calculation and the diameters match.”
Not to mention the fact that his work was peer-reviewed, unlike Myhrvold’s. Experts have checked his work for obvious errors, and he can’t say the same.
“Science is a human process, and it’s true that human beings review these results, but peer review is a scientist’s best friend,” she said. “There are bumps and bumps, but science is interactive and peer review is what makes it strong. It’s the foundation of science.”
Myhrvold submitted his article for consideration in the journal Icarus, but it has not yet been accepted. He stands by his decision to put it online — and sit down with the New York Times — before getting approval from the scientific community.
“Some of the things I’ve heard from third parties are, ‘wow, we’ll see if this makes it through peer review,’ which makes me think they’re going to ambush it at the peer review,” Myhrvold said. He added that he tried for months to get WISE scientists to comment on his work, but they refused to respond until he made his work public. Mainzer disputes this, saying she proposed corrections to the document that Myhrvold ignored, including a problem he considers a possible sign of fraud.
“In more than 100 cases, the reported asteroid sizes matched those in previous papers exactly, that is, within a few meters. It is simply not possible for this to happen accidentally,” he said. he declared.
“Arguments about peer review of my paper are entirely valid where what I’m doing suggests something totally new,” he added, “but where I’m pointing out that they copied a lot data from other sources, it’s not about me, it’s about them.”
He said he had no way of knowing whether this was a sign of intentional fraud, but thought it was possible.
“We tried to explain this several times,” Mainzer said. These measurements, she continued, are intentionally taken from other sources: they are used to calibrate measurements made using the spacecraft with previously observed data. This is not an error or fraud, but a standard procedure.
She also challenges the criticism’s underlying thesis that NEOWISE’s calculations ignore Kirchhoff’s law of thermal radiation. The law states that brighter objects emit less infrared radiation, and Myhrvold was shocked to see this ignored in the modeling. From the New York Times:
“A simple demonstration of this, Dr. Myhrvold said, is the shiny chrome cooking surface of a restaurant hibachi grill. A dark grill surface would bathe diners in uncomfortable heat waves. NEOCAM models do not have “taken into account the effects of reflected sunlight,” he said.
“Of course, this model is not perfectly energy efficient, and no one ever said that was the case,” Mainzer said. The model, she explains, eliminates some of the complexities of an asteroid’s heat distribution to compensate for the fact that asteroids have incredibly complex surfaces. The scientists who use these models don’t have all the data points they would need to make a perfect calculation, and that’s the whole point of using the model in the first place.
“This article is fundamentally useless in the sense that it complains about a model that has always been clear does not satisfy the laws of physics,” Ned Wright told the Post. “So the question is does it work, is it a good approximation anyway, and the answer to that question seems to be generally yes.”
A model that takes into account all this additional, usually unknown, data would produce more accurate measurements, Wright said, but that’s not what Myhrvold’s paper does.
“What he did was change one of the ways that he doesn’t satisfy physical laws, and I’m not sure how much of that is an error in his computer program and how much of that is the result of the change he is implementing, but what is it? “He discovers that he is making the model worse. So he took a useful approximation and made it worse,” Wright said. He added that variations in reflectivity are probably much less of a problem than the new paper suggests.
“There are no shiny metallic asteroids,” he said. “Everything in the asteroid belt is covered in dirt, which seems to be what happens when you stay in space for a long time.”
Several researchers were quoted in the New York Times article praising Myhrvold for raising questions – an important aspect of the scientific method – but notably, none of them went so far as to agree with his conclusions. A glance at a popular Yahoo group for astronomy researchers shows many raised eyebrows.
But Myhrvold, who has already published work refuting paleontologists’ statistical methods, suspects that potential collaborators will soon emerge from the woodwork, regardless of the “mud” to which NASA subjects it.
“The path I took is certainly not the easiest,” he said.
Tech
