AMS State of the Climate Report for 2014

[donsutherland1]

This is hyperbolic nonsense, especially considering that the recent GISS update warmed 2014 by 0.093C, which is well above the stated error potential.

And Version 6.0 of UAH cooled 2011, 2012, 2013, and 2014 by 0.10°C or more.

[ORH_wxman]

And Version 6.0 of UAH cooled 2011, 2012, 2013, and 2014 by 0.10°C or more.

 

 

It also cooled 2008 and 2009 off by more than 0.10C.

[StudentOfClimatology]

After looking at the January 1998-December 2014 period (the timeframe where the new UAH is being touted to show a slight cooling trend, which is at odds with the surface datasets and RATPAC), comparing Versions 5.6 and 6.0 of the UAH, I have to state that I'm even more concerned about the reliability of UAH (V 6.0). I ran the numbers for annual Global anomalies and annual anomalies in the North Polar region for the timeframe.

Some highlights:

- In terms of annual Global anomalies, the 2001-2014 period saw V.6.0 come out with colder anomalies than V.5.6.

- The 1998-2014 mean Global anomaly was 0.46σ below the V.5.6 mean anomaly for the same timeframe.

- There was a defined trend in which the annual Global anomalies on V6.0 were growing colder by 0.104°C per decade relative to V.5.6 (R²: 0.877)

- In terms of annual North Polar anomalies, the entire 1998-2014 period saw V.6.0 come out with colder anomalies than V.5.6

- The 1998-2014 mean North Polar anomaly was 0.83σ below the V.5.6 mean anomaly for the same timeframe.

- There was a defined trend in which the annual North Polar anomalies on V.6.0 were growing colder by 0.172°C per decade relative to V.5.6 (R²:0.657)

By itself, the fact that V.6.0 is definitively colder than V.5.6 is one matter (affects estimates of global temperatures, but not trends). That there is a defined trend in which the data on V.6.0 is growing steadily colder relative to V.5.6--and with a high coefficient of correlation--suggests that V.6.0 may suffer from one or more calculation issues that make V.6.0 unreliable in terms of assessing temperature trends.

IMO, V.6.0 urgently needs peer review, especially if there is an embedded trend in which V.6.0 becomes increasingly cold relative to V.5.6. Given that issue, until there is a thorough vetting of UAH V.6.0, its utility for trend analysis is highly suspect.

Below is a chart showing the temperature anomalies (°C) on V.5.6 and V.6.0 (with 4/5 years beginning with the previous global temperature record producing a difference of 0.30°C or more in the North Polar region):

Well, UAHv6.0 was expected to come in cooler. UAHv5.6 was apparently depicting spurious warming due to radiometer decay and some diurnal bias resulting from drift.

My hunch (for now) is that the cooling depicted in UAHv6.0 relative to v5.6 is actually correct. The UAH and RSS datasets had been diverging from one another for awhile, and no one knew which one had it right. The two datasets are now in good agreement, and came to this conclusion using two very different interpolative processes.

There's really no reason at all to be "concerned" about the accuracy of UAHv6.0 just because it depicts cooling, especially before it has been peer reviewed. It is supposedly an upgrade, and UAHv5.6 was supposedly warm biased.

[donsutherland1]

My hunch (for now) is that the cooling depicted in UAHv6.0 relative to v5.6 is actually correct.

 

I'm not so sure. One is not dealing merely with cooling, but growing cooling relative to V.5.6.

 

When one is dealing with a revised data set that shows temperature trends that are diverging from all the surface temperature data sets, at least one atmospheric set (RATPAC), and its own earlier version (5.6), it may be more likely than not that the new version may be flawed. Considering that it hasn't even been peer-reviewed adds to the uncertainty.

[chubbs]

An update of a chart I posted when UAH6 came out, using the recent GISS update. One thing is clear, UAH6 is not a good proxy for surface temperature data since UAH6 and GISS vary differently on a range of timescales. One period of interest is roughly 2002 to 2010 when UAH6 cools at a relatively rapid rate vs GISS (in both GISS versions but somewhat larger in latest).

 

[StudentOfClimatology]

I'm not so sure. One is not dealing merely with cooling, but growing cooling relative to V.5.6.

When one is dealing with a revised data set that shows temperature trends that are diverging from all the surface temperature data sets, at least one atmospheric set (RATPAC), and its own earlier version (5.6), it may be more likely than not that the new version may be flawed. Considering that it hasn't even been peer-reviewed adds to the uncertainty.

I don't think the facts support what you're speculating here, with all due respect. The observed radiometer decay and diurnal drift were leading contributors to what was/is supposedly exponentially spurious warming on UAHv5.6. When these issues were corrected for, UAHv6.0 fell into line with RSS.

The UAHv6.0 upgrade was designed to correct for the problems with v5.6.

[ORH_wxman]

I don't think the facts support what you're speculating here, with all due respect. The observed radiometer decay and diurnal drift were leading contributors to what was/is supposedly exponentially spurious warming on UAHv5.6. When these issues were corrected for, UAHv6.0 fell into line with RSS.

The UAHv6.0 upgrade was designed to correct for the problems with v5.6.

 

 

Given that 6.0 isn't peer reviewed yet...we don't know the detailed nuts and bolts of the corrections and how they stand up to closer scrutiny.

 

It's easy to be skeptical of this dataset.

[donsutherland1]

I don't think the facts support what you're speculating here, with all due respect. The observed radiometer decay and diurnal drift were leading contributors to what was/is supposedly exponentially spurious warming on UAHv5.6. When these issues were corrected for, UAHv6.0 fell into line with RSS.

The UAHv6.0 upgrade was designed to correct for the problems with v5.6.

Until it is peer reviewed, we don't know if it is an improvement over the earlier data set, much less reliable in representing recent temperature trends e.g., its slight negative trend since 1998 may well be spurious. My point is that V. 6.0's performance relative to other data raises some real questions.

 

Personally, I'm not sure why V. 6.0 was made public--even a beta version--until peer review had been completed.

[chrisf97212]

I don't think I accept the premise that surface temps and troposphere temps move in lockstep, implying that RATPAC is more accurate than UAH or RSS.  I fully expect the troposphere to spike later in the year, as a lagged response to record SSTs/Nino conditions.

[StudentOfClimatology]

Given that 6.0 isn't peer reviewed yet...we don't know the detailed nuts and bolts of the corrections and how they stand up to closer scrutiny.

It's easy to be skeptical of this dataset.

Of course I'm skeptical of it. I'm skeptical of every dataset. But some the suggestions being made here are based on nothing but pure speculation.

[StudentOfClimatology]

Until it is peer reviewed, we don't know if it is an improvement over the earlier data set, much less reliable in representing recent temperature trends e.g., it's slight negative trend since 1998 may well be spurious. My point is that V. 6.0's performance relative to other data raises some real questions.

Personally, I'm not sure why V. 6.0 was made public--even a beta version--until peer review had been completed.

I definitely agree with bolded.

[ORH_wxman]

I don't think I accept the premise that surface temps and troposphere temps move in lockstep, implying that RATPAC is more accurate than UAH or RSS.  I fully expect the troposphere to spike later in the year, as a lagged response to record SSTs/Nino conditions.

 

 

They should be fairly close over longer time periods. Before version 6.0, UAH was pretty close to surface trends over the whole period since 1979 (0.14C per decade)...now it is significantly lower at 0.11C per decade. RSS was pretty close too, but has begun to slip a bit in recent years down to 0.12C per decade.

[donsutherland1]

Of course I'm skeptical of it. I'm skeptical of every dataset. But some the suggestions being made here are based on nothing but pure speculation.

Use of data to point out possible issues is anything but speculation. That UAH 6.0 has a growing difference in temperatures from UAH 5.6 with a high coefficient of correlation raises questions. If UAH 5.6 had, for example, a warm bias of let's say 0.X°, such a trend wouldn't exist, even as minor year-to-year variability between the two data sets would be present.

 

It's entirely possible that some changes that addressed existing issues e.g., diurnal drift, led to new issues. One has seen such phenomena in the past when changes are made to models and data sets, so such an outcome wouldn't be unprecedented.

[StudentOfClimatology]

Use of data to point out possible issues is anything but speculation. That UAH 6.0 has a growing difference in temperatures from UAH 5.6 with a high coefficient of correlation raises questions. If UAH 5.6 had, for example, a warm bias of let's say 0.X°, such a trend wouldn't exist, even as minor year-to-year variability between the two data sets would be present.

It's speculation because there's no physical basis for the claim, currently speaking. Orbital/diurnal drift is exponential in nature, meaning it accelerates over time if it cannot be controlled (gravitational field decreases w/ distance relative to the mass of the satellite, which is obviously unchanged). That would explain why UAHv5.6 may have been running increasingly warm with time.

We'll have to see what the validation paper & subsequent literature say regarding the upgrade to v6.0, but as of right now the idea that it's flawed has no physical standing.

It's entirely possible that some changes that addressed existing issues e.g., diurnal drift, led to new issues. One has seen such phenomena in the past when changes are made to models and data sets, so such an outcome wouldn't be unprecedented.

Anything is possible, but I think this is highly unlikely considering that orbital/diurnal drift and sensor degradation are the only major/recurring issues with the satellite data. Interpreting the O^2 microwave emission data is relatively easy and is based on elementary radiative transfer physics.

[donsutherland1]

It's speculation because there's no physical basis for the claim, currently speaking. Orbital/diurnal drift is exponential in nature, meaning it accelerates over time if it cannot be controlled (gravitational field decreases w/ distance relative to the mass of the satellite, which is obviously unchanged). That would explain why UAHv5.6 may have been running increasingly warm with time.

We'll have to see what the validation paper & subsequent literature say regarding the upgrade to v6.0, but as of right now the idea that it's flawed has no physical standing.

Anything is possible, but I think this is highly unlikely considering that orbital/diurnal drift and sensor degradation are the only major/recurring issues with the satellite data. Interpreting the O^2 microwave emission data is relatively easy and is based on elementary radiative transfer physics.

The same holds true with respect to defending the validity of the new version, because the adjustments have not been subjected to peer review. Some are now using it to claim a cooling trend since January 1998 when UAH 6.0 hasn't been vetted by peer review. That exercise is not good science. It amounts to drawing premature conclusions before experiments have been completed and the results analyzed and summarized. That no caveats or disclosures related to its not having been peer reviewed have been furnished with those premature conclusions compounds the matter. 

 

What's happening with UAH 6.0's divergence relative to the major surface data sets and RATPAC would be the equivalent of an alternative proxy for economic activity diverging from GDP at a growing rate. If personal consumption expenditures were rising, government spending were increasing, unemployment were falling, hours worked were increasing, factory utilization were rising, and corporate revenues were increasing, the divergent alternative measure would be an outlier. That's exactly the case with UAH version 6.0. It is providing an outcome that differs from not one, but all the major land data sets, along with a major atmospheric temperature data set, not to mention its earlier version. 

 

Given that situation, it is not unreasonable to ask whether there is a problem with UAH 6.0 and whether it reasonably represents recent temperature anomaly trends. It is far more likely that UAH 6.0 has issues (especially since it hasn't been peer-reviewed) than it is that all the major surface data sets and RATPAC have near simultaneously developed issues that compromise their validity.

[nflwxman]

Don, it's great to have you post here. Your contribution is always appreciated.

[StudentOfClimatology]

The same holds true with respect to defending the validity of the new version, because the adjustments have not been subjected to peer review. Some are now using it to claim a cooling trend since January 1998 when UAH 6.0 hasn't been vetted by peer review. That exercise is not good science. It amounts to drawing premature conclusions before experiments have been completed and the results analyzed and summarized. That no caveats or disclosures related to its not having been peer reviewed have been furnished with those premature conclusions compounds the matter.

Has anyone here done anything like that? I certainly haven't

Yes, the validation paper is still under construction, and I would rather have seen it published before the upgrade to UAHv6.0 took place. That said, the dataset is obviously intended to be an improvement and a replacement for UAHv5.6.

What's happening with UAH 6.0's divergence relative to the major surface data sets and RATPAC would be the equivalent of an alternative proxy for economic activity diverging from GDP at a growing rate.

I'm sorry, but this isn't accurate. The surface datasets measure in a completely different domain so they're not quantitatively relevant. The reason surface data is never used in validation studies within the TLT domain is because it's irrelevant. There are numerous natural processes that could theoretically force a divergence between the surface data and the satellite data.

There are only two mainstream TLT-based datasets, and these are UAH and RSS. The RATPAC aggregation is not widely used in peer review outside of the occasional validation study and was not used in AR5. It's also a much lower-resolution dataset and lacks coverage in a lot of remote areas.

That's exactly the case with UAH version 6.0. It is providing an outcome that differs from not one, but all the major land data sets, along with a major atmospheric temperature data set, not to mention its earlier version.

Again, the surface datasets do not measure in the lower troposphere..the satellite data is multi-domainal. There are numerous natural processes that could theoretically lead to a divergence between the surface data and the satellite data over an extended period of time. I don't understand why you keep bringing them up.

As far as the TLT datasets are concerned, there is only one outlier, and that is RATPAC. You can't even bring the surface datasets into the equation because they don't measure there.

[StudentOfClimatology]

The same misunderstandings have been leading to the same debate(s) in here for an eternity now. The surface data is two dimensional and confined to the planetary skin/surface boundary. The TLT date is three dimensional and multi-domainal.

There are a slew of processes that could theoretically force an extended divergence between the surface at the TLT. A slowdown in global wind speeds, for example, will reduce evaporative cooling at the sea surface as well as latent heat release in the troposphere. In response, SSTs/surface temps will warm while the TLT later will cool. A change in the atomic processes that govern aerosol nucleation/cloud formation could do the same.

The speculative who's right who's wrong crap is tiring to read. It's been going on forever and has no basis in physical reality.

[donsutherland1]

Has anyone here done anything like that? I certainly haven't

No. You haven't. The name at the chart using Version 6.0 sans relevant disclaimers in post #2 is an example. I'm not surprised by various bloggers who are using the new version to make claims about the recent temperature trend (confirmation bias). I would expect a little more from scientists (and Watts is a blogger who also falls into the scientist category).

 

We simply disagree regarding UAH 6.0. I'm not sure that the new version is, in fact, an improvement over 5.6, due to the divergence of that data set from others. Of course, I could be wrong and peer review will sustain the revisions.

 

 

 

Again, the surface datasets do not measure in the lower troposphere..the satellite data is multi-domainal. There are numerous natural processes that could theoretically lead to a divergence between the surface data and the satellite data over an extended period of time.

 

I bring the other data sets in precisely because UAH 6.0 is being used by some e.g., the author of the chart, to assert that global temperatures have been declining somewhat since January 1998. As it is being used to make claims about global surface temperatures (and to undercut climate model forecasts for surface temperatures), it makes sense to see whether it is representative of such temperatures. Had the author of the chart used the data in a proper sense (strictly tying it to atmospheric temperatures), this would not be a big issue. But, that's not what happened. His chart made a claim about global temperatures and from the CMIP5 trend, it is clear that the author was making a claim about surface temperatures.

[StudentOfClimatology]

No. You haven't. The name at the chart using Version 6.0 sans relevant disclaimers in post #2 is an example. I'm not surprised by various bloggers who are using the new version to make claims about the recent temperature trend (confirmation bias). I would expect a little more from scientists (and Watts is a blogger who also falls into the scientist category).

I didn't mean to sound rude in my previous post, so I'm sorry if I came off that way.

I agree, and try to avoid blog science for this very reason. Most of these bloggers appear to be political extremists as well.

We simply disagree regarding UAH 6.0. I'm not sure that the new version is, in fact, an improvement over 5.6, due to the divergence of that data set from others. Of course, I could be wrong and peer review will sustain the revisions.

I guess we'll see what the literature has to say about it. I'm fairly certain it'll hold up, but I've been wrong before.

I bring the other data sets in precisely because UAH 6.0 is being used by some e.g., the author of the chart, to assert that global temperatures have been declining somewhat since January 1998. As it is being used to make claims about global surface temperatures (and to undercut climate model forecasts for surface temperatures), it makes sense to see whether it is representative of such temperatures. Had the author of the chart used the data in a proper sense (strictly tying it to atmospheric temperatures), this would not be a big issue. But, that's not what happened. His chart made a claim about global temperatures and from the CMIP5 trend, it is clear that the author was making a claim about surface temperatures.

I think it's obvious the globe has continued accumulating heat during the hiatus..see the OHC data. Whether or not this added heat is being effectively transported into the lower troposphere is another story, though. If wind speeds over the tropical/subtropical oceans have declined (broad/weak Hadley Cells due to AGW?), that would effectively inhibit evaporation at the sea surface and reduce latent heat transport into the lower troposphere, relatively speaking.

Theoretically, this would explain the hiatus in the TLT data.

[donsutherland1]

I didn't mean to sound rude in my previous post, so I'm sorry if I came off that way.

I agree, and try to avoid blog science for this very reason. Most of these bloggers appear to be political extremists as well.

I guess we'll see what the literature has to say about it. I'm fairly certain it'll hold up, but I've been wrong before.

I think it's obvious the globe has continued accumulating heat during the hiatus..see the OHC data. Whether or not this added heat is being effectively transported into the lower troposphere is another story, though. If wind speeds over the tropical/subtropical oceans have declined (broad/weak Hadley Cells due to AGW?), that would effectively inhibit evaporation at the sea surface and reduce latent heat transport into the lower troposphere, relatively speaking.

Theoretically, this would explain the hiatus in the TLT data.

No offense whatsoever taken. I didn't think you sounded rude. I think it's fine to raise questions and to disagree from time to time. I don't take such matters personally.

 

I don't disagree with you that there could be explanations for the "hiatus." That's why I'm looking forward to seeing what some of the other major climate centers find regarding the ERSSTv4 data set. It leads to a reduction in overall warming during the period of record and also suggests that the recent slowdown in warming was less pronounced than initially believed. Given the importance of the matter, I'm looking forward to further examination of the Karl paper. Review is healthy.

[stadiumwave]

Main purpose appears to be for JB and others  to make spurious cooling claims. We've been over this many times on this forum in the past couple of years. This series has problems in 2010 when there was an upgrade. WxBell is the only site I am aware of that stitches the two CFS versions together to make a long-term record.

 

No...I'm serious. What is it for? It's an NCEP product...what do they use it for?  Yes, Dr. Maue graphs it out over the years but it's the actual data from the NCEP reanalysis.  So...again, what does NCEP use it for? It has to have some purpose.

[donsutherland1]

No...I'm serious. What is it for? It's an NCEP product...what do they use it for?  Yes, Dr. Maue graphs it out over the years but it's the actual data from the NCEP reanalysis.  So...again, what does NCEP use it for? It has to have some purpose.

To provide representations of the atmosphere, of which temperatures is one variable. For a full discussion of CFSR:

 

http://journals.ametsoc.org/doi/pdf/10.1175/2010BAMS3001.1

 

CFSR is not intended to be a climate dataset along the lines of Hadcrut, GISS, NCDC, etc.

[Isotherm]

The same misunderstandings have been leading to the same debate(s) in here for an eternity now. The surface data is two dimensional and confined to the planetary skin/surface boundary. The TLT date is three dimensional and multi-domainal.

There are a slew of processes that could theoretically force an extended divergence between the surface at the TLT. A slowdown in global wind speeds, for example, will reduce evaporative cooling at the sea surface as well as latent heat release in the troposphere. In response, SSTs/surface temps will warm while the TLT later will cool. A change in the atomic processes that govern aerosol nucleation/cloud formation could do the same.

The speculative who's right who's wrong crap is tiring to read. It's been going on forever and has no basis in physical reality.

 

 

Strongly agree; well said. I'm not sure why there's such surprise over the divergence b/t the sfc and satellite datasets given historical progression in El Nino events. UAH and RSS may very well be approaching record levels by the end of this year or early in 2016, depending upon the eventual magnitude of the Nino event. If the two datasets do not register a significant temperature anomaly increase through the winter, then that would be disconcerting. However, it's very early relatively speaking in the development of the El Nino. The very warm SST's globally coupled with weaker global winds makes the divergence in datasets understandable.

[Isotherm]

The flaws are of sufficient magnitude that even as the surface data sets and RATPAC (surface to 500 mb) are running their highest warm anomalies on record during the January-June 2015 period, RSS and UAH are showing an entirely different outcome.

 

 

The time period of comparison is far too short to draw any conclusions regarding the accuracy of RSS/UAH. The lagged atmospheric response can often allow a significant divergence for 6+ months before the datasets begin to equalize. The mechanism(s) by which the surface heat is transferred into the lower atmosphere have not been activated up to this point. Until that occurs, the divergence will remain. There must be a force to induce the latent heat release. Typically, the Nino convection energizes in the autumn.

[stadiumwave]

To provide representations of the atmosphere, of which temperatures is one variable. For a full discussion of CFSR:

 

http://journals.ametsoc.org/doi/pdf/10.1175/2010BAMS3001.1

 

CFSR is not intended to be a climate dataset along the lines of Hadcrut, GISS, NCDC, etc.

Thanks for the link Don!

I'm still somewhat puzzled by the cooling trend even on the CFSR. I understand that it's not meant to be a dataset like GISS, UAH, etc...but if it's designed to analyze the current state of the atmosphere as the paper says why the cooling trend?

[chubbs]

Thanks for the link Don!

I'm still somewhat puzzled by the cooling trend even on the CFSR. I understand that it's not meant to be a dataset like GISS, UAH, etc...but if it's designed to analyze the current state of the atmosphere as the paper says why the cooling trend?

 

WxBell CFS doesn't match any other dataset in 2010/11 - when it was updated to a new version and all the "cooling" occurred. Its an artifact not a cooling trend

[wolfpackmet]

I don't follow JB on twitter anymore but see his rants on weatherbell.  He has been hitting the CFS hard over the last couple of weeks.  Drives me crazy that they stitch together 2 different versions with no acknowledgment.  If it went the other way imagine his reaction.