How reliable are satellite temperature measurements?
"Bad astronomer" Phil Plait is on another of his deceptive rants below. He thinks the heavily manipulated and unevenly located thermometer data is accurate and the satellite readings are not. So just a few observations:
He says that accounts of the "pause" show a flat trend line only because the starting point of the graphs concerned is in the El Nino year of 1998. But that is not so. I trot out my favorite graph below.
The graph in fact starts from 1997. And think what would have happened if there had been no El Nino. Without that big spike, the trend would be down, showing cooling. Only that big 1998 upward spike cancels out the very low figures recorded in some subsequent years. So Warmists should be thankful of the 1998 El Nino spike rather than pooh-poohing it.
But the issue of a starting date is important. With many trend lines you can prove anything by cherrypicking particular starting and end points. It's classic chartmanship.
And when the egregious Michael Mann says at the beginning of the video that 2015 will show up as the warmest year as far back as we have data, it might seem that he is on the side of the angels in the matter. In assessing a trend, you should go back to the beginning.
But Mann does not in fact do that. He goes back only a little more than 100 years. If he had gone back a couple of thousand years -- to encapsulate both the Roman and the Medieval Warm periods, he would have to say that 2015 was among the cooler years in Earth's history. But "hockeystick" Mann has of course had a lot of issues with history.
So is the 18 year pause just cherrypicking? No. If we want to know if the present is like the past we have to assess the present. And the "hiatus" or plateau IS just simply there in the recent past.
Skeptics will have no problem with assertions below to the effect that there was warming over the 20th century. There WAS a slight warming, as far as we can tell. But by the same token Warmists need to accept that there has been no warming in the 21st century. Whatever produced the 20th century warming is no longer there in the 21st. Warmists postulate that there is a single systematic process going on that causes temperatures to rise. But that may not be so. We may simply be seeing random fluctuations. And no cherrypicking of your starting points will disprove that.
Amusing that Phil Plait too stresses below the importance of not cherrypicking your starting points, even though he and his friends themselves do exactly that. It's just another one of Phil's deceptions. James Taylor dissects another one of Phil's deceptions here. And Matt Ridley has a go at Phil here
I could go on but most of the rest is just old boilerplate Warmist assertions that conflict with the facts so I will leave any further debunking of the piece to others.
UPDATE: I do have a social life in addition to putting up six blogs daily so I sometimes have too little time to say all that I would like about a piece of Greenie nonsense. So after I have fired one bullet into its brain, I sometimes leave it for dead. And so it was with what I put up above. I have now however found a little more time to take on phony Phil so will make one additional point:
The satellite record is in fact validated by the terrestrial record. Even after all the manipulations that take place in the great temples of Warmism, the terrestrial temperature record of the 21st century has shown no statistically significant change from year to year. And to a scientist, no statistically significant difference means no difference. It means that the differences observed are so small that they have to be accepted as random and not the product of some systematic process. So from a scientific point of view the terrestrial and satellite records say the same thing. The terrestrial record confirms the accuracy of the satellite record. Both show no change. That Phil and Co. ignore statistical significance shows that their "science" is just a lot of phlogiston.
Warmists normally just keep shtumm about statistical significance though Rahmstorff did once endeavour to discredit it. I commented on that here. Even a layman can however see that the regular Warmist claims that some year is the warmest, 3rd warmest etc. are a lot of hokum if you look at the actual figures behind those assertions. The warming is not statistically significant because it is TRIVIAL. When you can show years differing only by hundredths and thousandths of one degree in temperature, you are showing warming that is for all practical purposes non-existent. The statistical significance is, in other words, telling us something important. We do well to heed it.
So it is not only phony Phil that is full of phlogiston. The people at NOAA and elsewhere who make these annual "warmest" announcements are also ignoring the science of the matter
In December, GOP senator, presidential hopeful, and outrageous science denier Ted Cruz held a Senate panel about climate change that could charitably be called a farce. He empaneled a series of people who ranged from lukewarmers (believing the Earth is warming, but it’s not dangerous, or not rapid enough to worry about now) to out-and-out head-in-the-sand deniers.
During the hearing, Cruz said a lot of completely false things, but two things he hammered over and again were the reliability of satellite data, and how those data don’t show any warming over the past 18 years* — the so-called pause.
As I’ve written many times, that "pause" doesn’t exist; we’re still getting warmer and have been for decades. He cherry-picked the data, looking only as far back as 1998, when a huge spike in temperature due to an El Niño event made it look like temperatures are flat since then (when you start high, it makes the rest of the graph look flatter). That’s hugely misleading, of course.
But it did make me wonder just how reliable the data are; I know that satellite measurements can be difficult to calibrate. Worse, satellites don’t actually measure temperature directly; they measure how much energy the Earth radiates, and that’s converted into a temperature. The conversion is dependent on a lot of theoretical models. How accurate are the models?
It turns out this is a good thing to wonder. Satellite measurements are not the most reliable method to get temperature. The folks at Yale Climate Connections have made a short video explaining this, and it’s very good. Watch!
The key thing to take away from this is that satellites measure radiance: energy radiated by the atmosphere as microwaves. They come from the air, but also from the surface, clouds, and more. Scientists then use models of what’s emitting these microwaves to disentangle all that and convert it to a temperature. But those models are sometimes not terribly accurate.
The best measurements have been and still are from thermometers in situ, at various stations across the globe, on land, over sea, and in the air. These data need adjusting sometimes too, but not nearly as much as satellite data. Thermometers more reliable.
I was also happy to see climatologist Andrew Dessler make a fantastic comment starting at the 4:00 minute mark, talking about the deniers who misrepresent or misinterpret the data:
The bottom line: You can’t cherry-pick when you start the temperature measurements, and you can’t cherry-pick the data sets themselves, even — especially — if they show what you want.
And remember, the satellite data are one small part of a vast amount of data that overwhelmingly show our planet is warming up: retreating glaciers, huge amounts of ice melting at both poles, the "death spiral" of Arctic ice every year at the summer minimum over time, earlier annual starts of warm weather and later starts of cold weather, warming oceans, rising sea levels, ocean acidification, more extreme weather, changing weather patterns overall, earlier snow melts, and lower snow cover in the spring …
Despite the claims of people like Cruz, Roy Spencer (yes, this Roy Spencer), Rep. Lamar Smith (R-Texas), and Sen. James Inhofe (R-Okla.), we know the Earth is warming up, and we know humans are the reason why.