Monday, July 04, 2016

Lower Troposphere Sees Warmest 12 Months Ever

That's according to RSS. Their calculation of the temperature of the lower troposphere for June shows that its annualized temperature (12-month moving average) is at a record high:

UAH's calculations, which just moved to a new subversion, v6beta6, have the 12-month moving average at 0.02°C below the record high period of Jan-Dec 1998. But the endpoints of a "year" are arbitrary, right?


Unknown said...

Thanks for posting the link to Roy Spencer's blog, which shows that UAH temperatures are plummeting like a stone. And yes, the end points of a year are arbitrary, which is why Roy Spencer plots a centred 13 month running average. I expect that will soon turn downward also.

Layzej said...

Say David,

Any idea why the satellites have an exaggerated response to ENSO in 1998, 2008, and 2010, but not this year?

David Appell said...

Layzej: No, but that's a good question. I'll try to find out more.

I don't really trust the satellite measurements anyway. Too many large alterations in (esp) UAH's model over the years. The leader of RSS thinks satellite data aren't as reliable as surface stations:

from Carl Mears, Senior Research Scientist, Remote Sensing Systems (RSS)

"A similar, but stronger case can be made using surface temperature datasets, which I consider to be more reliable than satellite datasets…."

Layzej said...

Thanks! Fair points about the satellite measurements. I made a bet that 2016 would be the warmest year. I used UAH 5.6 as the reference in part to get in front of any excuses once I won but also because I figured it would shoot up higher than the land based measurements. That has not turned out how I'd hoped, though I'm still winning so far...

1998 0.471 2016 0.616
1998.08 0.651 2016.08 0.898
1998.17 0.421 2016.17 0.85
1998.25 0.663 2016.25 0.761
1998.33 0.562 2016.33 0.551
1998.42 0.507
1998.5 0.44
1998.58 0.445
1998.67 0.329
1998.75 0.291
1998.83 0.08
1998.92 0.19

Unknown said...

The link provided by David Appell points to which shows the 2015-16 El Nino peaking about 0.1 °C higher than 1997-98, so I don't understand why you are saying "Any idea why the satellites have an exaggerated response to ENSO in 1998, 2008, and 2010, but not this year?"

David Appell said...

Layjez means why do satellites show peak temperatures for the 1997-1998 and 2010-2011 El Ninos about 0.15-0.2 C above peak surface temperatures, but not for the 2015-16 El Nino.

Unknown said...

Ah so he's comparing UAH with NASA GISS ('peak surface temperatures') and asking why the current version of UAH is not higher than NASA GISS when the previous versions were ? The answer is presumably that the current version of UAH is generally closer to NASA GISS (and presumably NOAA NCDC and HadCRUT4) than the previous versions were.

That would also be the reason why UAH was closer to NASA GISS at the minimum in 2012 than it was at the minimum in 2008 and in 1999-2000.

UAH is constantly being updated, just as the IPCC climate models are constantly being updated.

David Appell said...

No, that's not it. The plot of the UAH line is all for the same version, not one version for 1998 and another version for 2015.

Why should satellite temperatures be equal to surface temperatures?

"UAH is constantly being updated, just as the IPCC climate models are constantly being updated."

UAH has had the largest swings, historically, of any temperature dataset. It took them several years to admit to a sign error, that changed their results from cooling to warming. Most people think their skepticism/denialism is reflected in their numbers. The fact that they won't share their code doesn't help.

Layzej said...


I'm only looking at UAH 5.6. Nothing to do with version changes and not specific to UAH. In the past stratospheric measurements typically drop lower than surface measurements during La Nino and jump higher than surface measurements during El Nino. I'm curious why the troposphere doesn't appear to be reacting the same way to this El Nino.

David Appell said...

Layzej: In awhile I'll post a graph that compares UAH LT v6beta6 to GISS. It shows a similar gap for the 1997-98 El Nino, but not for the 2010 El Nino.

Gotta finish some work first.

Layzej said...

Interesting. You may need to detrend them both if you are interested in comparing the variability rather than the trend.