2014 Global Temperatures

[Jonger]

Yes, 0.2 on the CFSv2 generally means a GISS value between 0.7-0.8. Best guess is 0.75C.

Is the general rule for cfs to giss, add .45c?

[LithiaWx]

Here's a great unisys SST animation from July - Present. Note the significant drop in SSTs over the last few weeks. The NATL has almost reversed, probably due to the vortex present up there:

http://weather.unisys.com/archive/sst/sst_anom_loop.gif

 

The way those warm anomalies just vanish and evaporate is pretty epic stuff. 

[nflwxman]

Is the general rule for cfs to giss, add .45c?

.55.  The correlation is about a 0.93 R^2.  There are months that remain outliers, but it appears mostly related to ENSO and or steep changes in polar temperatures.  This is only usable after mid 2010, when the CFS went through some upgrades.

[StudentOfClimatology]

The way those warm anomalies just vanish and evaporate is pretty epic stuff.

Maybe the Atlantic has something to do with the recent volcanic activity in Iceland?

That heat isn't just vanishing, though. Over the last 4 weeks, the rate of decline has been unusually rapid, faster than 2013 at this time. Some of that heat is likely being advected into the lower troposohere, which should lead to warming in the satellite datasets.

Still am not a fan of the color scheme on these maps but shows the general progression nonetheless.

[nflwxman]

UAH at 0.29C for September.  Up from 0.20C in August.

 

[chubbs]

Updated table with August data included. Surface series maintained warmer pace relative to satellites in August. Both GISS and Hadcrut had some changes to historical data which increased record year slightly

 

[StudentOfClimatology]

Quite a rapid temperature drop with the development of a -AO and a +AAO:

[Msalgado]

Seems like that drop is entirely because of the southern hemisphere.  Wonder if that will last.

[StudentOfClimatology]

Seems like that drop is entirely because of the southern hemisphere. Wonder if that will last.

Yeah, about 80% of the drop can be attributed to the development of a +AAO, as I alluded to a few days ago. Doesn't look like it's going to change anytime soon, either. http://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao_index/aao/aao.sprd2.gif

The drop is also being influenced by the development of a -AO, which tends to increase mid-latitude cloud cover, cyclogenesis, meridional transport, etc..though the cooling associated with that tends to be slower.

[chubbs]

Yeah, about 80% of the drop can be attributed to the development of a +AAO, as I alluded to a few days ago. Doesn't look like it's going to change anytime soon, either. http://www.cpc.ncep.noaa.gov/products/precip/CWlink/daily_ao_index/aao/aao.sprd2.gif

The drop is also being influenced by the development of a -AO, which tends to increase mid-latitude cloud cover, cyclogenesis, meridional transport, etc..though the cooling associated with that tends to be slower.

NH + global temperatures will bounce back in a couple of days per climate reanalyzer

[StudentOfClimatology]

NH + global temperatures will bounce back in a couple of days per climate reanalyzer

Is this the site? http://cci-reanalyzer.org/Forecasts/

Comparing @ the 00z intervals, looks like a warming of about 0.2C over the next 3-4 days, so it should be picked up on today's CFS update if it's legit.

Short term variability aside, the OLR data alone over the last 4 weeks (good precursor to sfc temps) suggests October will probably run colder than recent months. There's a lot that goes into intra-seasonal fluctuations in global temperatures, so stability is never really achievable behind the antecedent forcings. Sometimes, lower-resolution smoothing is best at capturing the bigger picture.

[chubbs]

Is this the site? http://cci-reanalyzer.org/Forecasts/

Comparing @ the 00z intervals, looks like a warming of about 0.2C over the next 3-4 days, so it should be picked up on today's CFS update if it's legit.

Short term variability aside, the OLR data alone over the last 4 weeks (good precursor to sfc temps) suggests October will probably run colder than recent months. There's a lot that goes into intra-seasonal fluctuations in global temperatures, so stability is never really achievable behind the antecedent forcings. Sometimes, lower-resolution smoothing is best at capturing the bigger picture.

Thats the site. It flagged the current dip  

[stadiumwave]

Two recent papers in "Nature" that could have big implications per findings:

 

"New research suggests that the upper layer of the ocean has warmed more than had been thought previously while the deeper ocean has cooled rather than warmed in recent years."

 

Good discussion by Dr. Judith Curry & also links to the two papers:

 

http://judithcurry.com/2014/10/05/evidence-of-deep-ocean-cooling/

[nflwxman]

Two recent papers in "Nature" that could have big implications per findings:

 

"New research suggests that the upper layer of the ocean has warmed more than had been thought previously while the deeper ocean has cooled rather than warmed in recent years."

 

Good discussion by Dr. Judith Curry & also links to the two papers:

 

http://judithcurry.com/2014/10/05/evidence-of-deep-ocean-cooling/

So it's using satellite data to infill the southern ocean?  I mean, the technique in theory is very similar to what Cowtan and Way used over the poles.  They are using ocean sfc temperatures as a proxy for 0-700m temperatures, which could be very uncertain (they mention this in the paper).

 

They also appear to only analyze on the 0-700 level.  On it's face, 0-700 would not be as closely correlated to sfc temperatures as 0-100m.  If OHC is rising faster than previously expected for 0-700m, does that same conclusion apply to the shallow layer of the ocean (0-100m)?  That would be more telling, IMO.  If not, it's just playing the depth game. 

 

When Cowtan and Way released their results about the poles warming quickly, I immediately thought, "well, of course that makes sense."   Personally, I think the answer lies mostly in the pacific with enhanced trades and ENSO negative conditions.

 

EDIT: looked for NASA for clarification- they are not invalidating any of the mid-ocean studies done recently.  They are merely talking about the bottom half of the ocean, which is under 2,000m.  Trenberth's "missing heat" has always referred to the 700-2000m level.  Thus, nothing in this study invalidates that theory.

 

"Coauthor Felix Landerer of JPL noted that during the same period, warming in the top half of the ocean continued unabated, an unequivocal sign that our planet is heating up. Some recent studies reporting deep-ocean warming were, in fact, referring to the warming in the upper half of the ocean but below the topmost layer, which ends about 0.4 mile (700 meters) down."

[stadiumwave]

So it's using satellite data to infill the southern ocean?  I mean, the technique in theory is very similar to what Cowtan and Way used over the poles.  They are using ocean sfc temperatures as a proxy for 0-700m temperatures, which could be very uncertain (they mention this in the paper).

 

They also appear to only analyze on the 0-700 level.  On it's face, 0-700 would not be as closely correlated to sfc temperatures as 0-100m.  If OHC is rising faster than previously expected for 0-700m, does that same conclusion apply to the shallow layer of the ocean (0-100m)?  That would be more telling, IMO.  If not, it's just playing the depth game. 

 

When Cowtan and Way released their results about the poles warming quickly, I immediately thought, "well, of course that makes sense."   Personally, I think the answer lies mostly in the pacific with enhanced trades and ENSO negative conditions.

 

EDIT: looked for NASA for clarification- they are not invalidating any of the mid-ocean studies done recently.  They are merely talking about the bottom half of the ocean, which is under 2,000m.  Trenberth's "missing heat" has always referred to the 700-2000m level.  Thus, nothing in this study invalidates that theory.

 

"Coauthor Felix Landerer of JPL noted that during the same period, warming in the top half of the ocean continued unabated, an unequivocal sign that our planet is heating up. Some recent studies reporting deep-ocean warming were, in fact, referring to the warming in the upper half of the ocean but below the topmost layer, which ends about 0.4 mile (700 meters) down."

 

 

 

 

 

The implication of this is that a build up of heat in the deep oceans is not the solution to the so called missing energy mystery that has puzzled climate scientists trying to match the observed heat build up on the planet with what the theory of global warming suggests should be happening. A number of climate scientists had previously suggested that heat is accumulating in the deep oceans and that this accounts for the missing energy.

 

An analysis of ocean data together with satellite measurements suggests that the warming rate for the top 700m of ocean in the southern hemisphere has been underestimated – at least from 1970 until the early 2000s when an array of measurement buoys, known as Argo, began to collect data.

 

Separately, an analysis of satellite measurements and ocean temperature data has revealed that that the deeper half of the ocean (below 2 km depth) has, on average, not warmed from 2005 to 2013 and may have cooled – in contrast to the prevailing view, based on sparse ship-based measurements, that had suggested deep ocean warming between the 1990s and 2005.

 

[skierinvermont]

I think you both have it a bit wrong.

 

The Llovel paper doesn't suggest a difference in the total energy budget from previously thought. It just uses the sea level budget and the better constrained upper ocean (above 2000m) thermal component to estimate the rate of warming/cooling in the deep ocean. They find that the deep ocean might have cooled. But the error bars are huge becaise the sea level budget is poorly constrained. It relies upon estimates of freshwater contribution to sea level rise which is only a fairly rough estimate. The implication, to me, is that the sea level budget has always slightly disagreed with the OHC measurements and suggested a slightly lower imbalance over the last 8 years. But this study doesn't change that picture. It only points out that, if the sea level budget is correct, and the upper ocean OHC measurements are correct, then the deep ocean is cooling slightly. But this is fairly speculative because the sea level budget is only roughly constrained due to uncertainties in the freshwater runoff component.

 

The other study suggests that the earth's energy imbalance is larger than previously thought because a poorly sampled region of the oceans (the southern ocean) has been warming quickly.

 

 

So one study has no effect on the net energy budget (only on where within the budget) and the other suggest the imbalance is larger than thought.

[StudentOfClimatology]

Wouldn't surprise me if the abyssal oceans (as said, below 2km in depth) were still cooling in equilibration to the previous LIA...you're talking about a fluid body with gargantuan thermal capacity essentially out of direct thermodynamic exchange with the surface/atmosphere boundary layer...this part of the ocean would take several millennia to equilibriate to any forcing.

[skierinvermont]

Wouldn't surprise me if the abyssal oceans (as said, below 2km in depth) were still cooling in equilibration to the previous LIA...you're talking about a fluid body with gargantuan thermal capacity essentially out of direct thermodynamic exchange with the surface/atmosphere boundary layer...this part of the ocean would take several millennia to equilibriate to any forcing.

 

It would kind of surprise me. The LIA (if you're talking about the full 1500-1900 cool period) isn't that much longer than our present warm period or that much cooler than the warm period before it. If you're talking about the brief cool period in the late 19th century that period is actually shorter than our present warm period so I highly doubt that deep ocean is cooling from that alone.

 

If mixing is very slow then I could see a scenario in which the 1000-2000m layer is still very cold (due to the cold earth 1500-1900) although starting to respond and warm to the current warm period. The deep >2000m layer could still be cooling as it never fully equilibriated to the coolness of the layer above it.

 

However, this would require very slow mixing and I don't think mixing is that slow. If warmth has just started to penetrate below 1000m, it would imply that mixing on 30+ year timescales only occurs down to 700m and mostly only down to 300m or less. The earth's large energy imbalance shows that mixing is faster than that and prevents skin temperatures from warming up immediately in response to forcing.

 

And I find the scenario where the late 1800s cold period is still causing cooling of the deep ocean highly implausible because that cold period was shorter than our current warm period and our current warm period is much warmer than the cold period was cold. The length and intensity of our current warm period should have totally overwhelmed the effects of a more brief less intense cool period. Unless somehow cold and warm periods creat very symmetrical warm and cold pulses that propagate downwards very slowly without ever mixing with each other. Somehow I doubt the system is that simple.

[skierinvermont]

This is some high-caliber nonsense, even by your standards. Why the hell would I use the 1951-1980 GISS baseline when I've been openly opposed to it since joining here?

September could very well pull a +0.7C on HADCRUT4 under optimal conditions...as you said, global SSTs are/were smashing all-time records, and September happens to hold the highest correlation coefficient w/ SSTs outside ENSO.

Of all people, you're usually the one hyping this stuff. Get it together, man. There's no conspiracy here.

 

 

Why would you say 'another' .7C if you were referring to Hadley which hasn't had any .7Cs?

 

To me, it sounds as if you were referring to GISS.

 

And if that's the case saying .7C might occur if max potential is reached was a bit conservative. It looks like it might come in at .75C.

 

Also why would you 'oppose' the 1951-1980 baseline? Baselines don't make any difference to temperature change. Opposing a baseline is like opposing math.

[StudentOfClimatology]

Why would you say 'another' .7C if you were referring to Hadley which hasn't had any .7Cs?

Because HADCRUT4 has achieved .7C...not sure why you think otherwise? It's an infrequent event on that 1961-1990 baseline, hence my usage of the word "another".

Also why would you 'oppose' the 1951-1980 baseline? Baselines don't make any difference to temperature change. Opposing a baseline is like opposing math.

I oppose that baseline as a reflection of what is "normal", on a decadal scale. It's clearly not "normal" anymore, as AGW has progressed.

[StudentOfClimatology]

It would kind of surprise me. The LIA (if you're talking about the full 1500-1900 cool period) isn't that much longer than our present warm period or that much cooler than the warm period before it. If you're talking about the brief cool period in the late 19th century that period is actually shorter than our present warm period so I highly doubt that deep ocean is cooling from that alone.

If mixing is very slow then I could see a scenario in which the 1000-2000m layer is still very cold (due to the cold earth 1500-1900) although starting to respond and warm to the current warm period. The deep >2000m layer could still be cooling as it never fully equilibriated to the coolness of the layer above it.

However, this would require very slow mixing and I don't think mixing is that slow. If warmth has just started to penetrate below 1000m, it would imply that mixing on 30+ year timescales only occurs down to 700m and mostly only down to 300m or less. The earth's large energy imbalance shows that mixing is faster than that and prevents skin temperatures from warming up immediately in response to forcing.

And I find the scenario where the late 1800s cold period is still causing cooling of the deep ocean highly implausible because that cold period was shorter than our current warm period and our current warm period is much warmer than the cold period was cold. The length and intensity of our current warm period should have totally overwhelmed the effects of a more brief less intense cool period. Unless somehow cold and warm periods creat very symmetrical warm and cold pulses that propagate downwards very slowly without ever mixing with each other. Somehow I doubt the system is that simple.

Technically AGW may be allowing residual cooling to persist/continue at depth. As the upper oceans warm, convective/kinematic overturning at depth will actually slow due to an increasingly stable (sterically forced) pressure gradient...so diffusion would be the sole make-up mechanism. The observed AGW has rapidly thrown the system out of the LIA climate, however, we're looking at 75-100yrs here. Is that window long enough to allow the upper-oceanic warming to diffuse through the column? I could be wrong, but I suspect we'll need to wait 50-150yrs to see warming show up below 2000m.

[skierinvermont]

Except that quote doesn't state what you claim it does, and I was referring to the global aggregate (satellite + surface), which you would have known if you'd read a bit deeper into that discussion. I have a huge pet peeve when it comes to people taking me out of context.

I specifically referred to O/N/D, as well, and did not mention September. You're really pulling teeth here.

 

You specifically said August was probably the peak. And you suggested that the max potential for September might be .70C (in my opinion you were clearly referring to GISS).

 

August was not the peak. September will be warmer on surface and satelite and the aggregate. And GISS will almost certainly be over .70C.

 

I tend to agree with your hypothesis that extreme warmth in the NPAC will not be maintained as the Hadley cells shrink. However, you appear to have overly relied upon it and perhaps rushed the timing. With extremely high global SSTs, even once SSTs fall land temperatures may stay very warm or warm further. There tends to be a lag with peak surface temp occurring after peak SSTs.

[StudentOfClimatology]

You specifically said August was probably the peak. And you suggested that the max potential for September might be .70C (in my opinion you were clearly referring to GISS).

August was not the peak. September will be warmer on surface and satelite and the aggregate. And GISS will almost certainly be over .70C.

I tend to agree with your hypothesis that extreme warmth in the NPAC will not be maintained as the Hadley cells shrink. However, you appear to have overly relied upon it and perhaps rushed the timing. With extremely high global SSTs, even once SSTs fall land temperatures may stay very warm or warm further. There tends to be a lag with peak surface temp occurring after peak SSTs.

I suspected August would run slightly warmer than September, yes. Problem was I didn't consider the AAO in my prognostication..so while the NH technically followed my prediction, the SH had other ideas:

[skierinvermont]

Technically AGW may be allowing residual cooling to persist/continue at depth. As the upper oceans warm, convective/kinematic overturning at depth will actually slow due to an increasingly stable (sterically forced) pressure gradient...so diffusion would be the sole make-up mechanism. The observed AGW has rapidly thrown the system out of the LIA climate, however, we're looking at 75-100yrs here. Is that window long enough to allow the upper-oceanic warming to diffuse through the column? I could be wrong, but I suspect we'll need to wait 50-150yrs to see warming show up below 2000m.

 

That might be true in a bathtube or very uniform geomtric system. Even if it were true, it would mean AGW were slightly slowing mixing not halting it completely. If you have any mixing at all, and the upper layers are very warm and rapidly getting warmer, then deep layers will warm too if there is any significant amount of mixing. Your hypothesis holds only if mixing basically occurs only in the top 500-700m with very slight mixing below that such that the 1000-2000m layer, though beginning to warm, is somehow still colder than it was 100 years ago. I highly doubt that the 1000-2000m layer is colder than it was 100 years ago. I think there are probably even good enough measurements to show substantial warming of that layer. And if that layer is warmer than it was 100 years ago, and if there is any mixing to the >2000m layer, then it would be warming the 0-2000m layer as well (albeit very slowly). Your hypothesis means that basically all of the 8-10" of sea level rise in the last century, all of that accumulation of heat occurred almost entirely in the 0-700m layer, with a slight amount beginning to penetrate the 1000-2000m layer and cooling still occurring below 2000m.

 

 

I find that implausible. The oceans have currents and movements and chaos much too great.

 

For one thing, the Atlantic Meridional Overturning Current has a northward warm current that occupies the top 1000m of the ocean. This warm water then cools feshens and sinks and flows southward as the cold "North Atlantic Deep Water" or NADW which exists at depths of 1,500m to 4,500m. This means there is direct mixing between the warm top 1000m and the deep 1,500-4,500m layer. If the top 1000m warms, which it clearly has, then heat transfer to the NADW would be greater.

 

 

Something similar exists in the southern atlantic as well.

 

In addition, mixing in the tropical pacific can occur so aggressively that anomalies related to ENSO of over 1C can persist down to 400 or 500m. 1C anomalies are huge for OHC changes which are on the order of hundredths of a degree. If you can mix 1C anomalies down 500m, surely you can mix .01C anomalies down 2000m.

[skierinvermont]

Because HADCRUT4 has achieved .7C...not sure why you think otherwise? It's an infrequent event on that 1961-1990 baseline, hence my usage of the word "another".

I oppose that baseline as a reflection of what is "normal", on a decadal scale. It's clearly not "normal" anymore, as AGW has progressed.

 

You can wriggle all you want. In context, you were clearly referring to GISS.

 

Why would you say you expect September to be cooler than August but then say it might reach .70C on HadCRUT4 which would make warmer than August.

 

You would be contradicting yourself. It doesn't make sense. I don't have a stick in this at all but you appear to be trying to wriggle out of this. You thought September would be cooler, you thought it would come in at maybe .70C on GISS, and it appears to have come in warmer than both those predictions. It's fine. Swallow and move on. I agree with your general hypothesis you just relied on it too heavily.

[skierinvermont]

I suspected August would run slightly warmer than September, yes. Problem was I didn't consider the AAO in my prognostication..so while the NH technically followed my prediction, the SH had other ideas:

 

 

OK but when nflwxman said that you predicted a drop off in September you said "I absolutely did not."

 

Now you acknowledge this is what you were thinking.

 

Again, you are contradicting yourself.

 

The prediction was technically wrong but your reasoning was probably sound. The NPAC did indeed cool and probably held down the surface anomalies a bit. At worst, you were a little too overly reliant on one variable.

[skierinvermont]

I oppose that baseline as a reflection of what is "normal", on a decadal scale. It's clearly not "normal" anymore, as AGW has progressed.

 

1961-1990 isn't a good baseline for depicting 'normal' either. If you really want a 'normal' baseline it would be 1981-2010 but even this probably doesn't fully reflect that abnormal rapidity of warming since then.

 

Besides, why would we want to have a baseline that represents the present state when researching and discussing climate change. If you're discussing change you need an older baseline so you can quantify how much change has occurred.

 

But in the end, it really doesn't matter. Baselines are baselines but all that really matters are trendlines.

[skierinvermont]

For the record, I think October will probably be colder at the surface because of cooling NPAC SSTs.

[StudentOfClimatology]

That might be true in a bathtube or very uniform geomtric system. Even if it were true, it would mean AGW were slightly slowing mixing not halting it completely. If you have any mixing at all, and the upper layers are very warm and rapidly getting warmer, then deep layers will warm too if there is any significant amount of mixing. Your hypothesis holds only if mixing basically occurs only in the top 500-700m with very slight mixing below that such that the 1000-2000m layer, though beginning to warm, is somehow still colder than it was 100 years ago. I highly doubt that the 1000-2000m layer is colder than it was 100 years ago. I think there are probably even good enough measurements to show substantial warming of that layer. And if that layer is warmer than it was 100 years ago, and if there is any mixing to the >2000m layer, then it would be warming the 0-2000m layer as well (albeit very slowly). Your hypothesis means that basically all of the 8-10" of sea level rise in the last century, all of that accumulation of heat occurred almost entirely in the 0-700m layer, with a slight amount beginning to penetrate the 1000-2000m layer and cooling still occurring below 2000m.

Well that's the thing, it's a matter of macroscale equilibrium which is set by direct thermal transport. The abyssal oceans (below 2000m) rely solely on diffusion and vertical mixing for grounding in the boundaries of equilibrium set by the surface-atmosphere domain. Whether or not this layer is initially cooling in response to the LIA (or perhaps was stable before AGW initiated), what happens when steric-gradient forcing via AGW progressively slows vertical mixing? This domain will be forced to cool as vertical overturning accounts for the large majority (60-75%) of thermal transport...you'd be relying a lot more heavily on diffusion which is much less efficient at these depths. Doesn't matter if the upper ocean is warming.

In addition, mixing in the tropical pacific can occur so aggressively that anomalies related to ENSO of over 1C can persist down to 400 or 500m. 1C anomalies are huge for OHC changes which are on the order of hundredths of a degree. If you can mix 1C anomalies down 500m, surely you can mix .01C anomalies down 2000m.

It's important to distinguish between the upper oceanic mixing layer and deep ocean overturning. The upper oceanic mixing layer will always be vigorous as you've got constant kinematic (wind) forcing driving these eddies. The rate of parcel motion declines almost exponentially with depth.

[StudentOfClimatology]

You can wriggle all you want. In context, you were clearly referring to GISS.

Why would you say you expect September to be cooler than August but then say it might reach .70C on HadCRUT4 which would make warmer than August.

How far apart in time were those two quotes spaced? I eventually changed my tune on September when the AAO excursion was becoming obvious. If I posted them together, then it was an unforced error.