The Washington PostDemocracy Dies in Darkness

Satellite temperature data, leaned on by climate change doubters, revised sharply upward

June 30, 2017 at 2:52 p.m. EDT
Animated map showing global warming in earlier and updated version of satellite-based temperature record. <br/> (Carl Mears, Remote Sensing Systems, adapted by CWG)

The world is getting warmer at a steady clip, it has not stopped in recent years, and it doesn’t matter whether you’re measuring the temperature from the surface of the Earth or space.

This conclusion is strongly supported by a new study that applied corrections to obtain a more accurate and, ultimately, 30 percent higher estimate of global warming from satellites orbiting the Earth.

In past decades, such a conclusion required more qualification and was vigorously challenged by those who doubt climate change. The amount of global warming estimated from satellite-based measurements exhibited was substantially less than surface-based sensors.

Climate change doubters argued the surface data shouldn’t be trusted. Scientists then conducted multiple studies attempting to account for and resolve this discrepancy.

This new study shows that after corrections to one of the two satellite temperature records are made, there is little discrepancy at all and that estimates of global warming from both Earth and space are consistent. In fact, the revised satellite-based temperature estimate shows even a little bit more warming since late 1978 than the surface-based record from NASA.The corrections to the satellite data were published in the Journal of Climate on Monday by Carl Mears and Frank Wentz of Remote Sensing Systems, who have maintained a record of satellite temperature estimates since 2003. They bring the satellite temperature record into close alignment with multiple surface-based records maintained by government centers in the United States (NOAA and NASA) and United Kingdom (Hadley Centre).

The corrections dramatically increased the amount of warming observed since the late 1990s, when a number of earlier studies suggested the warming had slowed down or that there was a “warming hiatus.” In recent years, some politicians, like Sen. Ted Cruz (R-Tex.) and Environmental Protection Agency Administrator Scott Pruitt, had pointed to the satellite data as evidence warming had leveled off.

But according to Zeke Hausfather, a climate scientist who posted an analysis of this study at the website Carbon Brief, the estimated warming since 1998 increased by 140 percent after the corrections were applied, indicating warming has continued unabated.

Hausfather also pointed out that corrections make 2016 stand out unambiguously as the warmest year on record in the revised satellite estimate.

Mears and Wentz said the corrections were necessary because earlier satellite-based estimates of temperature did not accurately account for satellites making measurements at different times of day. Satellite-based temperature records draw data from not just a single satellite but an entire constellation, which has evolved over time.

In an email, Mears said the revised satellite record is not only more consistent with surface-based temperature data but also with weather balloons, which take temperature measurements higher up in the atmosphere.

“The improved consistency leads us to have confidence that we are moving in the right direction, though the existence of further problems in the data set, which might result in further upgrades, is not in any way ruled out,” Mears said.

Analyzing satellite-based temperature data remains complex, and the results of Mears and Wentz are not unchallenged.

The other frequently cited satellite-based temperature record, published at the University of Alabama in Huntsville, continues to show less warming than surface-based records.

John Christy, one of two atmospheric scientists who maintain the Alabama record, said in an email that the temperature data in their record was better supported by weather balloon data.

As weather balloon data have their own uncertainty, Mears said “it is difficult to say what is right or wrong due to the lack of a comparison data set that is known to be right.”

Ben Santer, an atmospheric scientist at the Lawrence Livermore National Laboratory, said the Mears and Wentz study shows the critical importance of continually revisiting scientific methods to obtain improved results.

“The lesson is that this evolutionary process isn’t over,” Santer said. “This study shows again very graphically, very thoroughly and very rigorously that the true uncertainty in these satellite data records are large.”

Santer stressed the importance of more than one group conducting this work. “You need multiple groups looking at the same bits and bites,” he said. “I’m glad we live in a world where Remote Sensing Systems exists. I would not want to live in a world where one group was entrusted to do this work.”