New Paper finds Most of the Late-20th Century Warming was Naturally Induced

[ORH_wxman]

Adding them together doesn't give you the best fit. Adding them together doesn't make any sense anyway since they aren't on the same scale.

Weighting them respectively gives the best fit but its still not an amazing fit or anything. IIRC weighting them explains something like half (or a bit more) of the multi-decadal variability...but less than half in the more recent years. I think the previous paper about multi-decadal variability that I mentioned further up in the thread had a graph in there that showed this, but I need to find the paper...I can't remember who the author was.

The attribution or causation on such oscillations is not going to be rock solid due to their period. From a physical standpoint though it makes sense they would have a significant effect. More frequent and stronger positive ENSO events (during +PDO regime) would argue for higher global temperatures. More downwelling of cold North Atlantic water and upwelling (and transportation) of warm tropical water during a +AMO regime would also help argue for warmer global temperatures. Just on a smaller scale that can either be masked or enhanced by the PDO.

[skierinvermont]

Again, repeating my point above, just because it CAN explain half the multidecadal variability, doesn't mean that it actually DOES.

For example, if I threw global sulfur concentrations into the mix and did a linear regression on ENSO, TSI, Volcanoes, AMO, and Sulfur, a lot of the variability that the statistical model was giving to the AMO would now go to sulfur instead. And there is good reason to believe that sulfur has been much more causative than the AMO.

Heck, I could throw nearly any other variable into the model and much of the variability that was going to the AMO would go to the random variable.

[WeatherRusty]

We have little to no idea about the processes that were involved in the Earth's climate changes in the past, thus we have little to no idea about what the magnitude of those forcings were when they were impacting Earth's Climate during the Ice Ages.

That is also why many scientists get wildly different values when they try and extract Earth's Climate Sensitivity based off of paleoclimatological data. If GCRs were the primary driver of paleoclimate, then the sensitivity matches up with the paleoclimatological data. The problem is that we still do not have a good grasp with as to what the factors were that caused the Earth's Climate to change in the past, so this is still an area of uncertainty.

http://www.clim-past...8-4923-2012.pdf

http://www.uibk.ac.a...pdf/christl.pdf

http://stephenschnei...v-veizer-03.pdf

http://elpub.wdcb.ru...04163.htm#ref50

This is just not accurate. The orbital cycles very well explain the cycling of climate changes over the past 3 million years of ice ages. Your first link looks to data obtained from 33.4 million years ago. The world was a very different place back then, climate sensitivity may well have been significantly lower than the roughly 3C expected during the current time period in question for which we have analog climates. 33.4 million years ago there was no permanent northern ice and the Antarctic ice sheet was just beginning to form for goodness sake. CO2 was also much higher back then but continuing a general decline which commenced some 70 million years earlier. The Sun was a bit dimmer. Ocean circulations were different..No isthmus of Panama..etc.

[tacoman25]

Again, repeating my point above, just because it CAN explain half the multidecadal variability, doesn't mean that it actually DOES.

For example, if I threw global sulfur concentrations into the mix and did a linear regression on ENSO, TSI, Volcanoes, AMO, and Sulfur, a lot of the variability that the statistical model was giving to the AMO would now go to sulfur instead.

Heck, I could throw nearly any other variable into the model and much of the variability that was going to the AMO would go to the random variable.

In reference to your previous question, no I was not using the "denier" graph you show.

And which PDO source were you using? I use the UW source, as I believe the other one is tilted too cold for at least the past couple decades.

Finally, it doesn't make any sense to weight the PDO and AMO equally, as the Pacific is much larger and ENSO/PDO phases have a considerably larger effect on global temps. The correlation makes perfect sense, as we can see how global SST fluctuations on a yearly basis effect global temps, it's easy to see how they also effect trends on a longer time scale. The causation is there as well.

[skierinvermont]

In reference to your previous question, no I was not using the "denier" graph you show.

And which PDO source were you using? I use the UW source, as I believe the other one is tilted too cold for at least the past couple decades.

Finally, it doesn't make any sense to weight the PDO and AMO equally, as the Pacific is much larger and ENSO/PDO phases have a considerably larger effect on global temps. The correlation makes perfect sense, as we can see how global SST fluctuations on a yearly basis effect global temps, it's easy to see how they also effect trends on a longer time scale. The causation is there as well.

So if you agree the effect of the PDO is greater you are essentially also agreeing that the conclusions of this study are mostly (or entirely) wrong. If the PDO effect is greater, the AMO effect would have to be pretty small.. much smaller than this study is claiming based on a loose correlation alone.

[tacoman25]

Adding them together doesn't give you the best fit. Adding them together doesn't make any sense anyway since they aren't on the same scale.

Weighting them respectively gives the best fit but its still not an amazing fit or anything. IIRC weighting them explains something like half (or a bit more) of the multi-decadal variability...but less than half in the more recent years. I think the previous paper about multi-decadal variability that I mentioned further up in the thread had a graph in there that showed this, but I need to find the paper...I can't remember who the author was.

The attribution or causation on such oscillations is not going to be rock solid due to their period. From a physical standpoint though it makes sense they would have a significant effect. More frequent and stronger positive ENSO events (during +PDO regime) would argue for higher global temperatures. More downwelling of cold North Atlantic water and upwelling (and transportation) of warm tropical water during a +AMO regime would also help argue for warmer global temperatures. Just on a smaller scale that can either be masked or enhanced by the PDO.

Exactly. Same points I'm making, but better said.

With a +PDO phase in effect from 1976 through the mid 2000s, and a +AMO phase in effect from the mid 1990s on, it's pretty easy to see how a temperature trend from the mid 1970s to the 2000s would be skewed towards greater warming by natural cycles.

I just don't understand the hesitancy of some to acknowledge the role these natural cycles play in multi-decadal temperature trends. It's not like it disproves AGW by any means to say that not all the warming from the 1970s to today was due to AGW.

[tacoman25]

So if you agree the effect of the PDO is greater you are essentially also agreeing that the conclusions of this study are mostly (or entirely) wrong. If the PDO effect is greater, the AMO effect would have to be pretty small.. much smaller than this study is claiming based on a loose correlation alone.

I'm not arguing in favor of this study. I'm arguing against discarding oceanic cycle correlations with global temperature trends, just because they have long cycles.

[skierinvermont]

I'm not arguing in favor of this study. I'm arguing against discarding oceanic cycle correlations with global temperature trends, just because they have long cycles.

And I'm saying you need more evidence than only 2 complete cycles with a fairly loose fit to temperature. The only acceptable evidence at this point would be causative. Which is why I accept some influence from the PDO because of its effect on ENSO. Maybe a bump of .05 during +PDO, and a dip of -.05 during -PDO.

I can form an equally powerful statistical model using sulfur instead of the PDO or AMO. There's no statistical reason to accept the AMO or the PDO over sulfur, or any of the three at all, without proof of causation. Unlike TSI or ENSO where the correlation is proof enough of causation. You can shout as loud as you want look at the correlation to the PDO!!!! look at the correlaton to the AMO!!!! and I will shout back the correlation to sulfur is just as good!!!!! and round and round we go... until proof of causation is offered.

And we know sulfur is causative but we don't know exactly how strong. Sulfur, based on what we know of its causative effects, probably explains most or all of the cooling/stall 1945-1970 depending on how strong it is.

[tacoman25]

And I'm saying you need more evidence than only 2 complete cycles with a fairly loose fit to temperature. The only acceptable evidence at this point would be causative. Which is why I accept some influence from the PDO because of its effect on ENSO. Maybe a bump of .05 during +PDO, and a dip of -.05 during -PDO.

I can form an equally powerful statistical model using sulfur instead of the PDO or AMO. There's no statistical reason to accept the AMO or the PDO over sulfur, or any of the three at all, without proof of causation. Unlike TSI or ENSO where the correlation is proof enough of causation. You can shout as loud as you want look at the correlation to the PDO!!!! look at the correlaton to the AMO!!!! and I will shout back the correlation to sulfur is just as good!!!!! and round and round we go... until proof of causation is offered.

And we know sulfur is causative but we don't know exactly how strong. Sulfur, based on what we know of its causative effects, probably explains most or all of the cooling/stall 1945-1970 depending on how strong it is.

Not sure you are listening completely to what I'm saying. The causation is there: it's oceanic SSTs. With +PDO, they tend to be higher, heating the atmosphere more. Same with +AMO, to a lesser degree.

[skierinvermont]

Not sure you are listening completely to what I'm saying. The causation is there: it's global SSTs. With +PDO, they tend to be higher, heating the atmosphere more. Same with +AMO, to a lesser degree.

Ok glad you agree to stop relying on these inconclusive correlations. I only have been pointing out how inconclusive they are because you brought them up.

Actually a +PDO indicates colder SSTs but because of the effect on ENSO you get less trade winds in the equatorial pacific, which reduces the equatorial pacific heat sink (the equatorial pacific is a huge heat sink in both Ninos and Ninas - just a bigger one in Ninas) allowing heat to build up in the atmosphere instead... which you are correct does tend to eventually distribute to the global oceans later in the ENSO event. Just saying +PDO --> warm SSTs --> warm atmosphere is an inaccurate description of the flow of causation. It's more like +PDO --> reduced trade winds and reduced equatorial pacific heat sink --- > warmer atmosphere ---> warmer SSTs

So yes, by affecting ENSO frequency there is a causative mechanism. But there's really no good way of estimating how much. Just because there is some causation, doesn't mean that all of the correlation is causative. Especially when there is an equally strong correlation to sulfur with a stronger causative mechanism as well.

Same with the AMO- no way of knowing how much.

Also there are other ocean oscillations around the world which get completely ignored at the expense of the AMO and PDO. The antarctic ocean has been very cold for a while now. Some chance that the southern ocean works in reverse of the northern ocean.

One way to possibly isolate causation would be to take out the ENSO, TSI, volcano and GHG components using best estimates of the relationships. Also take out the effect of sulfur within the bounds of uncertainty. You'd probably have to do 3 different models - weak sulfur, best estimate sulfur, and strong sulfur. Taking out the GHG component would actually make the long-term trend negative. Taking out sulfur would return it to neutral and also eliminate much or all of the cooling 1945-1970.

Then assume that any of the remaining variability that can be explained by the PDO or AMO is causative. Even that though may be a big (false) assumption.

[tacoman25]

Ok glad you agree to stop relying on these inconclusive correlations. I only have been pointing out how inconclusive they are because you brought them up.

Actually a +PDO indicates colder SSTs but because of the effect on ENSO you get less trade winds in the equatorial pacific, which reduces the equatorial pacific heat sink (the equatorial pacific is a huge heat sink in both Ninos and Ninas - just a bigger one in Ninas) allowing heat to build up in the atmosphere instead... which you are correct does tend to eventually distribute to the global oceans later in the ENSO event.

So yes, by affecting ENSO frequency there is a causative mechanism. But there's really no good way of estimating how much. Just because there is some causation, doesn't mean that all of the correlation is causative. Especially when there is an equally strong correlation to sulfur with a stronger causative mechanism as well.

Same with the AMO- no way of knowing how much.

Also there are other ocean oscillations around the world which get completely ignored at the expense of the AMO and PDO. The antarctic ocean has been very cold for a while now. Some chance this is even related to the northern hemisphere warmth and the +AMO even.

I've never relied only on correlations. I like to incorporate some common sense as well.

I agree that it is difficult to know exactly how much oceanic cycles affect global temperature trends. And of course there are other cycles besides the PDO/AMO, but the Pacific and Atlantic are by far the biggest oceans. Regardless, climate scientists still don't understand much about how they work, but as the oceans are the clearest sources of internal climate variations (outside of volcanoes), it makes perfect sense that they have a signficant effect as they go through various cycles.

I do not see how sulphur is as strong a correlation over the period of record.

[skierinvermont]

I've never relied only on correlations. I like to incorporate some common sense as well.

I agree that it is difficult to know exactly how much oceanic cycles affect global temperature trends. And of course there are other cycles besides the PDO/AMO, but the Pacific and Atlantic are by far the biggest oceans. Regardless, climate scientists still don't understand much about how they work, but as the oceans are the clearest sources of internal climate variations (outside of volcanoes), it makes perfect sense that they have a signficant effect as they go through various cycles.

I do not see how sulphur is as strong a correlation over the period of record.

Sulfur is as strong or stronger of a correlation. It can explain why we cooled 1945-1970 (when sulfur increased the fastest) and it explains why the GHG signal reemerged post 1970 when sulfur stabilized. A linear regression that included ENSO, TSI, volcanoes, GHGs, and sulfur would be able to explain nearly all the variation in the temperature record.

[skierinvermont]

By reducing the long-term trend on sulfur (IE subtracting some GHGs from it) and then flipping it over, it looks just like the AMO/PDO.

Cold early century, peak 1940s warming, cooling 1945-1970, and enhanced warming post 1970.

Here's the graph showing it. Line A is actually sulfur. Line B is if we reduce the long-term trend (by subtracting some GHG warming from it.

Line C is if we flip it over. As you can see, it looks very similar to the AMO/PDO line now. Up to 1940s, down mid century, back up late century.

[tacoman25]

Sulfur is as strong or stronger of a correlation. It can explain why we cooled 1945-1970 (when sulfur increased the fastest) and it explains why the GHG signal reemerged post 1970 when sulfur stabilized. A linear regression that included ENSO, TSI, volcanoes, GHGs, and sulfur would be able to explain nearly all the variation in the temperature record.

I would like to see how sulphur correlates through the full period of record. We see a correlation between oceanic cyles and global temperatures all the way back to the early 1900s, when data then begins to get sketchy.

As far as the GHG signal "re-emerging" post-1970, we didn't really begin to see warming until later in the 1970s and then much stronger in the 1980s...after the PDO was in its positive phase.

[skierinvermont]

I would like to see how sulphur correlates through the full period of record. We see a correlation between oceanic cyles and global temperatures all the way back to the early 1900s, when data then begins to get sketchy.

As far as the GHG signal "re-emerging" post-1970, we didn't really begin to see warming until later in the 1970s and then much stronger in the 1980s...after the PDO was in its positive phase.

Check out above. I hope you like the sketch. By changing the long-term slope, and flipping the line over (since the effect is inverse), sulfur concentration looks very much like the AMO/PDO

[tacoman25]

Check out above. I hope you like the sketch. By changing the long-term slope, and flipping the line over (since the effect is inverse), sulfur concentration looks very much like the AMO/PDO

I appreciate the effort, but you do have any actual graphs of sulfur concentration? Also, what would cause the sulphur cycles?

Again, we can see clearly on a physical basis how oceanic variations can effect global temperature, so why are you trying to disprove the influence of oceanic cycles on temperature trends? It appears you are saying sulphur is actually responsible for the same trends that are correlated to oceanic cycles.

[skierinvermont]

I appreciate the effort, but you do have any actual graphs of sulfur concentration? Also, what would cause the sulphur cycles?

Again, we can see clearly on a physical basis how oceanic variations can effect global temperature, so why are you trying to disprove the influence of oceanic cycles on temperature trends? It appears you are saying sulphur is actually responsible for the same trends that are correlated to oceanic cycles.

But you have no idea how strong the causation is. Which leaves a lot of room for other factors - like sulfur. The effect of sulfur, while the magnitude is not certain, is probably better estimated than it is for ocean oscillations. We know sulfur has a moderate to strong cooling effect.

Also Sulfur is not cycling on its face... It went up really fast 1945-1975. But if you think about it, that is a cycle of sorts. Instead of up down up (AMO PDO) it's flat up flat. Now change the long-term trend by subtracting some GHGs from sulfur and its down up down. Now flip the sign and its up down up... same as the AMO/PDO... in a very general loose sense. But that's all the correlation for the AMO/PDO was too.. a loose fit from the general up down up theme.

Line A on my graph is supposed to represent actual sulfur. I sketched it based on this concentration graph. The GISP2 concentration data ends in 1970.. but emissions have been dropping since then so I estiimated that concentrations have stabilized.

[skierinvermont]

Point is, if you feed sulfur through a linear regression (along with ENSO, TSI, volcanoes, and GHGs) I can guarantee you it will be as strong or stronger of a correlation as the AMO or PDO.

And the causation is much better estimated as well.

[skierinvermont]

Also just throwing it out there the peak in sulfur in GISP2 during the 1910s and 20s may be because of the proximity of coal and wood burning in Europe to Greenland (I think GISP2 is from Greenland). The emissions tell a different story. I would venture a guess that the mini-peak in the 10s and 20s was not global. It could be real though.. it doesn't matter to my point ... just thought I'd throw that out there.

[skierinvermont]

You do see how and why Line A becomes Line C in my awesome graphic don't you?

[tacoman25]

But you have no idea how strong the causation is. Which leaves a lot of room for other factors - like sulfur. The effect of sulfur, while the magnitude is not certain, is probably better estimated than it is for ocean oscillations. We know sulfur has a moderate to strong cooling effect.

Also Sulfur is not cycling on its face... It went up really fast 1945-1975. But if you think about it, that is a cycle of sorts. Instead of up down up (AMO PDO) it's flat up flat. Now change the long-term trend by subtracting some GHGs from sulfur and its down up down. Now flip the sign and its up down up... same as the AMO/PDO... in a very general loose sense. But that's all the correlation for the AMO/PDO was too.. a loose fit from the general up down up theme.

Line A on my graph is supposed to represent actual sulfur. I sketched it based on this concentration graph. The GISP2 concentration data ends in 1970.. but emissions have been dropping since then so I estiimated that concentrations have stabilized.

Thanks for the graph. Yeah, I can definitely see where the correlation would be. The only problem is that the timing seems a little off compared to oceanic cycles. As I pointed out before, global temperatures bottomed out from the mid 1960s to mid 1970s (this was the coolest ten year stretch, also with both -PDO/-AMO phases), and then began rebounding rapidly in the late 1970s, accelerating through the 1980s and 1990s (with +PDO taking hold, then later +AMO in the 1990s).

In contrast, despite sulfates continuing to fall through the 2000s, global temperature rise has slowed considerably over the last decade or so as the PDO turned negative. This is why I see a stronger correlation with oceanic cycles.

[Jonger]

I would like to see how sulphur correlates through the full period of record. We see a correlation between oceanic cyles and global temperatures all the way back to the early 1900s, when data then begins to get sketchy.

As far as the GHG signal "re-emerging" post-1970, we didn't really begin to see warming until later in the 1970s and then much stronger in the 1980s...after the PDO was in its positive phase.

sulphur.. *cough Cop-out cough*

[WeatherRusty]

sulphur.. *cough Cop-out cough*

We are talking aerosols here. As in light scattering aerosols. They represent an important and not so well understood radiative forcing.

No cop-out at all. Rather an important piece of the radiative budget which affects everything from direct scattering of shortwave light back to space to the size of cloud droplets and cloud albedo. There is nothing trivial about anthropogenic sulfur based aerosols.

[skierinvermont]

Thanks for the graph. Yeah, I can definitely see where the correlation would be. The only problem is that the timing seems a little off compared to oceanic cycles. As I pointed out before, global temperatures bottomed out from the mid 1960s to mid 1970s (this was the coolest ten year stretch, also with both -PDO/-AMO phases), and then began rebounding rapidly in the late 1970s, accelerating through the 1980s and 1990s (with +PDO taking hold, then later +AMO in the 1990s).

In contrast, despite sulfates continuing to fall through the 2000s, global temperature rise has slowed considerably over the last decade or so as the PDO turned negative. This is why I see a stronger correlation with oceanic cycles.

ENSO and TSI already explain the cooling in the 2000s. As I said before, I promise you that if you run sulfur through the same model they did in this study it will explain just as much if not more of the variation as ocean cycles.

[ORH_wxman]

ENSO and TSI already explain the cooling in the 2000s. As I said before, I promise you that if you run sulfur through the same model they did in this study it will explain just as much if not more of the variation as ocean cycles.

ENSO is linked with the PDO though....they aren't mutually exclusive. ENSO (the PDO) is likely the primary cause of the cooling in the 1940s-1970s time period as well. That was a particularly deep -PDO period.

[tacoman25]

ENSO and TSI already explain the cooling in the 2000s. As I said before, I promise you that if you run sulfur through the same model they did in this study it will explain just as much if not more of the variation as ocean cycles.

ENSO trends are directly tied to the -PDO phase setting in.

[tacoman25]

ENSO is linked with the PDO though....they aren't mutually exclusive. ENSO (the PDO) is likely the primary cause of the cooling in the 1940s-1970s time period as well. That was a particularly deep -PDO period.

It's very frustrating. A rather obvious (now) cause/correlation that we can see played out as well on a yearly basis with ENSO, etc, and yet there is still so much reluctance to accept the oceanic phases tie to global temperature trends.

I honestly think the preference towards aerosols as an explanation is because it's a man-made cause...like AGW. Not saying that's necessarily why skiier is pushing it, but I do believe that is part of the reason many AGW proponents have leaned towards aerosols. And because so little is known about oceanic cycles still, and they only became accepted in climate science well after other forcings were already incorporated into forcing models.

[The_Global_Warmer]

Thank you skier for arguing these points far better than I am capable of. This is exactly how I see it in my head but am not knowledgeable enough to argue it or present it like that.

taking this from Wikipedia:

This figure, based on Meehl et al. (2004), shows the ability with which a global climate model (the DOE PCM [1]) is able to reconstruct the historical temperature record and the degree to which the associated temperature changes can be decomposed into various forcing factors. The top part of the figure compares a five year average of global temperature measurements (Jones and Moberg 2001) to the Meehl et al. results incorporating the effects of five predetermined forcing factors: greenhouse gases, man-made sulfate emissions, solar variability, ozone changes (both stratospheric and tropospheric), and volcanic emissions (including natural sulfates). The time history and radiative forcing qualities for each of these factors was specified in advance and was not adjusted to specifically match the temperature record.

Also shown are grey bands indicating the 68% and 95% range for natural variability in temperature relative to the climatic expectation as determined from multiple simulations with different initial conditions. In other words, they indicate the estimated size of variations that are expected to occur due to fluctuation in weather rather than changes in climate. Ideally the model should be able to reconstruct temperature variations to within about the tolerance specified by these bands. Some of the remaining misfit may be accounted for by the ~0.05 °C uncertainty in the temperature reconstruction. However, though the model captures the gross features of twentieth century climate change, it remains likely that some of the differences between model and observation still reflect the limitations of the model and/or our understanding of the histories of the observed forcing factors.

In the lower portion of the figure are the results of additional simulations in which the model was operated with only one forcing factor used at a time. A key conclusion of the Meehl et al. (2004) work is that the model response to all factors combined is to a good approximation equal to the sum of the responses to each factor taken individually. This means it is reasonable to talk about the temperature change due to individual aspects of the evolving man-made and natural influences on climate. The zeros on both plots are set equal to 1900 temperatures, and it is apparent that most of the 0.52 °C global warming between 1900 and 1994 should be attributed to a 0.69 °C temperature forcing from greenhouse gases partially offset by a 0.27 °C cooling due to man-made sulfate emissions and with other factors contributing the balance. This contrasts with the warming from 1900 to 1940 for which the model only attributes a net increases of 0.06 °C to the combined effects of greenhouse gases and sulfate emissions.

Here is another link with a lot of information on climate forcing.

http://climatechange.umaine.edu/icecores/IceCore/References.html

[tacoman25]

Thank you skier for arguing these points far better than I am capable of. This is exactly how I see it in my head but am not knowledgeable enough to argue it or present it like that.

taking this from Wikipedia:

Here is another link with a lot of information on climate forcing.

http://climatechange...References.html

Outdated. They didn't even try to factor in oceanic cycles into the temperature trend in that graph.

And the sulfate trend there looks considerably different than what skiier illustrated.

[nflwxman]

It's very frustrating. A rather obvious (now) cause/correlation that we can see played out as well on a yearly basis with ENSO, etc, and yet there is still so much reluctance to accept the oceanic phases tie to global temperature trends.

I honestly think the preference towards aerosols as an explanation is because it's a man-made cause...like AGW. Not saying that's necessarily why skiier is pushing it, but I do believe that is part of the reason many AGW proponents have leaned towards aerosols. And because so little is known about oceanic cycles still, and they only became accepted in climate science well after other forcings were already incorporated into forcing models.

We know the PDO affects decadal ENSO and that is reflected in the ONI values provided by NOAA. I'm not sure what the issue here is? The AMO is less understood, but it is clear that it has a much less significant impact to global temperature trends than the PDO/ENSO. Why have the climate models been able to replicate the past 100 years with relative precision if this was not the case?