Sunday, September 28, 2014

Why is the Northwest U.S. warming? Natural variations or mankind's greenhouse gases?

During the past week, a major debate within the scientific community has reached the popular media: 

To what degree is the modest warming over the northwest U.S. during the last century caused by (1) natural variations or (2) by increases in greenhouse gases emitted by mankind?

This is a critical question with huge implications for public policy, climate adaptation, and scientific understanding of local climate change.

The two viewpoints

On one hand, there are some scientists saying that nearly all of the warming is due to mankind's emission of greenhouse gases.   For example, here is a quote (made to KUOW) by Dr. Phil Mote, a scientist at Oregon State and Oregon State climatologist:

“As far as the 1.3 degree warming over the last hundred years or so," Mote said, "that’s all because of human activity.”

He is claiming that all of the warming of the last 100 years is due to anthropogenic (human-related) forcing.  And  Dr. Mote  goes even farther than that, suggesting that the local warming has accelerated

"what is particularly significant is that the rate of warming is increasing"

Support for his claims is found in a paper he published in the Journal of Climate last April with John Abatzoglou (the lead author, University of Idaho) and David Rupp (Oregon State):

On the other hand, a recent paper by Drs. Jim Johnstone and Nate Mantua in the prestigious journal Preceedings of the National Academy of Sciences (PNAS) came to the OPPOSITE conclusion:

"these changes are not likely related to historical anthropogenic and natural radiative forcing"

As Dr. Mantua said to the Seattle Times
"We do not see a human hand in the warming of the West Coast,”

In their paper, Dr. Johnstone and Mantua demonstrate that natural variability explains virtually all of the temperature evolution of the past century over Northwest U.S.

So we have two groups of scientists, both with Ph.Ds and having published results in well-known academic journals, coming to the opposite conclusions about the origin of the roughly 1.3 degree F warming that has occurred over the Northwest in the last century.



.....................................................................................................................................



How can two groups of scientists looking at the same basic data come to radically different conclusions?

Both can not be correct.

As explained below, I believe the Johnstone/Mantua paper makes a far better case:  human-caused warming over the Northwest has been minimal, with natural variability dominating.

Both teams of atmospheric scientists agree that the earth and the Pacific Northwest will warm due to increases in greenhouse gases, what they disagree upon is the impact of greenhouse gases in the past compared to natural variability.

The warming of the Pacific Northwest

One things nearly everyone agrees upon is that the Pacific Northwest has warmed over the past century, but NOT in a continuous way.  Here are two plots of temperature change (from a base period) over the Northwest, one from the Abatzzoglou, Rupp, and Mote paper (upper figure) and the other from the summary report of the UW Climate Impacts Group (bottom).  The top figure has the temp variations from three different sources as well as lines that attempt to smooth out the yearly variations.  The second figure has a single trend line (which really doesn't make sense since the variation is so complex)



These plots can be summarized as follows:
  • Since 1900, the temperatures have warmed by about about 1.3 F
  • The temperature has NOT warmed in a steady way.
  • The largest warming was concentrated in two periods: 1915-1940 and 1975 to 1985
  • Temperature has hardly changed from roughly 1985, the last 30 years.
So it is immediately obvious that the claims of some local scientists that the temperature change in the Northwest is accelerating is simply not true.  I wish they would stop saying that; it completely undercuts their credibility and projects poorly on the rest of the scientific community.

Natural versus human-forced variability

So what is forcing the temperature changes shown above?

Let's begin by human-connected temperature changes forced by increasingly greenhouse gases (such as CO2).  The Intergovernmental Program for Climate Change (IPCC) has published a series of authoritative reports, including an estimate of the global impacts on radiation in the atmosphere due to mankind's influence on greenhouse gases.  Here is a plot of their estimates over time;  pay particularly attention to their estimates of the influence of Long Lived Greenhouse Gases (LLGHG, red line))
Such gases slowly increased from 1850 onward, but REALLY accelerated around 1960 as population and use of fossil fuels exploded.  So if the radiative effects from greenhouse gases was were only thing going on, we would have seen slow, steady warming until the mid-20th century, followed by sharp warming after 1960.  Some other factors have influenced the radiative balance of the earth as well, like more reflective particles from combustion, volcanic eruptions, and changes to the land surface.   Considering all those effects produces the black line, which suggests that the big warming due to greenhouse gases should have begun around 1965.

Clearly, the temperature change in the Northwest looks very different than this, so more must be going on!

For example, because of the non-uniform nature of the earth's surface and weather systems, some places would warm up more or less than others.  We can get a handle on that by looking at the average of many climate simulations for the upcoming century (as shown in the figure below, which shows the impact of  greenhouse gas increases on surface temperatures by the year 2100). The Arctic warms up the most, the continents warm more than the oceans, and the eastern oceans warm up less than the western oceans.  The Northwest is downstream of the eastern Pacific and thus would warm less than most places (the simulations below did not have the resolution to describe our cooling by the Pacific, higher-resolution simulations show this).


So what should you conclude?    The radiative effects of greenhouse gases were relatively small before roughly 1965 and that our area should experience less warming than most places.  

There is nothing controversial in this statement.  Thus, one would certainly not expect the warming in the early part of the century over our regon to be forced by mankind's greenhouse gas emissions.

But although the Northwest should not have been highly impacted by global warming from greenhouse gases, we ARE highly impacted by natural variability, changes in atmospheric circulations that occur due to complex interactions within the atmosphere and with the oceans and land surface.

One example many of you know:  El Nino and La Nina, in which sloshing water in the tropical Pacific causes an oscillation of warming and cooling of equatorial waters over a period of 4-7 years.  El Nino/La Nina, often called ENSO (El Nino Southern Oscillation), has weather impacts all over the world.

But there is another type of natural variability that has a huge impact on the weather/climate of the Pacific basin:  The Pacific Decadal Oscillation.   This climate feature, discussed in depth for the first time by Dr. Nate Mantua (yes the same Mantua of the paper noted above), has a period of around 50 years, oscillating between warm and cold cycles (see figure).  It was in a cool cycle between roughly 1950 and 1977, then a warm cycle until around 2005, and more recently looks to be in a cooler cycle.


If you compare the variations of the PDO with the Northwest temperature traces shown above, it is very obvious that the variations of NW temps seems to closely follow the PDO changes, suggesting our temperatures are highly controlled by this mode of natural variability.

Thus, when some local scientists say that the temperature changes experienced here in the Northwest are mainly due to greenhouse gas emissions they are certainly incorrect.  We live in an area where the greenhouse gas signal is small and where natural variability (as forced by the PDO and ENSO) are quite large (and there are other modes of natural variability I have not even discussed).

The Controversy

As noted earlier, the media has been full of reports and stories about this controversy of natural versus greenhouse gas forcing here in the Northwest.  There have been literally hundreds of stories in national/international media outlets and the Seattle Times even wrote an editorial about the subject.

Many of the media reports were stimulated by the Johnstone/Mantua paper in PNAS.    This paper presented stunning results.   They showed that one could explain the sea surface temperature pattern off the West Coast if one knows the pressure/wind field.  Amazingly, they could produce a nearly perfect evolution of the temperatures over the past 100 years using pressures/winds (a.k.a., the circulation).

 Here is an example of their results from this paper.  SLP1 is their pressure/circulation index, SSTarc is the sea surface temperatures off the West Coast, and SATarc are air temperatures over land along the West Coast.  Folks, this is amazing correspondence.

Then they plotted the observed temperature change over the last century (left panel below) and took out the contribution of changes in circulation (wind/pressure)--right panel (residual trends).  The temperature change over the past hundred years goes away for most locations (right side) when you take our trends in atmospheric circulation.

You might ask, could the changes in the winds/pressure that explain the temperature changes be explained by greenhouse gas increases?   At this point in time, there is no theoretical or observational evidence to suggest so, and tree ring data suggests long term circulation changes have occurred for thousands of years in the past.  Furthermore, when Johnstone/Mantua examined the pressure/circulations changes produced by global climate models over the next century, they did NOT find any suggestions of changes in natural variability resulting from greenhouse gas increases.

So the bottom line of the Johnstone/Mantua paper is that natural variability has dominated temperature changes during the past century over the Northwest.   I believe their  arguments are very strong.



In contrast, there is the paper of Abatzoglou, Rupp, and Mote (2014) that suggests greenhouse gases are the dominant source of warming.  Specifically, they stated that anthropogenic forcing was "the leading contributor to long term warming."  In their paper, they used a technique called multiple linear regression to determine the forcing of temperature by several forcing mechanism (natural variability, volcanoes, solar variability, and human greenhouse gas forcing).  Their results suggested that anthropogenic contributions (greenhouse gases) were dominant.  But unfortunately, their approach has some critical problems that make their conclusions insupportable.  Problems so severe I am surprised they got through the review process.

The biggest deficiency--and this is going to get technical-- is that they detrended (over time) their index of natural variability (they detrended the 500-hPa height anomalies they used).  In doing so, they removed the ability of natural variability to explain the long-term temperature trends.  In other words, they threw away EXACTLY what Johnstone/Manuta showed to be important.

And the Abatzoglou paper had other problems, like using a poor index of natural variability (they did not use the PDO or anything like it).

I know this has been a long blog, but the issues are very important.  What should you come away with?

  • Global warming is not globally uniform, some places will warm more or less quickly.  The Northwest is a regional where warming will be slow due to our proximity of the eastern Pacific.
  • Natural variability in climate can be quite large and in some locations it is currently far larger than the contribution of mankind from greenhouse gas emissions  and other impacts.
  • For the West Coast, natural variability has been far larger than greenhouse gas warming.



As the century progresses, the human-caused global warming signal will increase while natural variability should remain about the same. Thus,the lack of human-caused warming on the West Coast does not imply that anthropogenically forced global warming is unimportant.  It is a serious issue mankind must deal with.

This situation also shows you some of the strengths and weaknesses of the scientific process.   Some papers get published that have problematic results.   Some scientists say things that are not supported by rigorous research.   But the system has a way of righting itself and self correcting over a period of time.  The Johnstone/Mantua paper has made a substantial contribution to recentering the discussions regarding Northwest climate change.

23 comments:

Westside guy said...

An interesting post, Cliff. I appreciate the amount of detail you provided.

There's nothing inherently wrong with long blog posts, as long as the subject matter requires it - and this topic certainly requires attention to the details!

Brad Koehler said...

I have been following this story for a while, and this is another great perspective on the issue. Cliff thanks for taking the (controversial) position based on objective evidence. BTW, check the 2040 date, might be 1940.

Scott Souchock said...

Thanks! I was wondering what to think about all of the news reports and very much appreciate your explanation.

Dave Less said...

You state that "These plots can be summarized as follows: The temperature has NOT warmed in a steady way." True. The temperature rises during times of large-scale warfare (WW1 and WW2) and increases of gas/oil drilling/consumption (Reagan/Cheney) and falls during times of recessions and oil embargoes and cessation of world wars. All this in addition to natural variations. The variation in temperatures in the Pacific NW are driven by changes in wind patterns - which makes deciding what causes weather variations here difficult to track down to their cause.

"The largest warming was concentrated in two periods: 1915-1940 and 1980 to 1990. Temperature has hardly changed from roughly 1980, the last 35 years." These two statements directly contradict each other. Perhaps you meant to say that temperature has hardly changed from early 1990s, the past 20 years (Clinton years and Great Recession).

"The temperature change in the Northwest is accelerating is simply not true" is a statement that can be made only if one denies the observation that temperatures fluctuate in the short term, and that the latest increase (19080 - 1995) is much larger than the two previous increases.

The Radiative Forcing plot that you show is an estimate made by a model. Looking at the link you provided it is not clear if they considered that man's contribution varied over the years (can one imagine a time during which there was more human-made pollution than at the height of WWII?)

"Tree ring data suggests long term circulation changes have occurred for thousands of years in the past." True, but that's akin to implying that humans do not wreck cars because there is historical data that proves that cars have gotten wrecked by natural forces (e.g. winds) ever since the invention of the automobile.

I don't disagree with all of your conclusions, but many of your proofs and arguments are as suspect as the reports that you criticize.

Jonn-E said...

Cliff,
This post really shows that our understanding of oceanic oscillation forces on our local climate has increased quite a bit in the last 10 years. Taking a step back though, are the ENSO/PDO SSTs influenced by anthropongenic forces?

I'd read the paper myself for answers, but as usual it is behind a not-insignificant paywall.

Cliff Mass said...

dave less,
there was a typo in that line..fixed. Thanks, cliff

PS: I suspect the effects of WWII is he opposite of what you are suggesting. Economic activity increased rapidly after the war. During the war, just the opposite...

Ace said...

Great post.

Is there any hint of the Urban Heat Index in the 'residual trends' map, or did the paper account for that?

beerfish said...

Just found your blog last week and really enjoy it.

Question regarding the PDO: Given that we are in the cold cycle of the PDO, would it be logical to expect that temps would be cooler than long-term average, and the fact that we are still above agerage might be due to anthropogenic factors? Yes, we do not have significant warming over the past 20+ years, but long term trends would lead us to expect cooling.

Another question: I am guessing that both of these studies are looking at annual or even seasonal averages. Is it possible that on a smaller time scale we would see a greater influence of greenhouse gas induced climate change? I am specifically thinking of the "Blocking Patterns" that have been linked to climate change in some studies. In the Northwest it seems to lead to either long stretches of great weather (suck it, Midwest!), or storm after storm rolling through, which through the year will average out to about the same as historical. We just end up concentrating our heat/cold and dry/wet. Has anyone looked at that for the Pacific Northwest?

Thanks!

Jack Bloss said...

I don't completely disagree with your rebuttals but let me try and answer a few things. First off Cliff said that the greater rise in temp was 1975-1985 not 80-90. So Cliff's statement that temp has hardly increased starting 1980 is untrue but notice that the temp goes back down after 2000, though not dramatically it does decrease. Looking at that graph it appears then that the temperature has not increased since a little before 1990.
Also, although it would seem as though there was a lot of production during WW2 here is a website that shows that that production increase was minimal in comparison to present day industrial production http://research.stlouisfed.org/fred2/graph/?g=Klm
In fact this graph looks quite similar to Cliff's. I am intrigued by what Cliff is saying and don't hold much opposition because I don't have enough knowledge on the subject to do so but I think he makes a very good case for his argument

mjgrota said...

Cliff,
I feel this may be your best and most meaningful blog sine you started down this path. I for some time have felt that mans contributions to climate are real and important but that they are not match for power of the natural environment. Imagine man attempting to change the track of a wintertime cyclone approaching the north west. Man has not ability to do that (outside of a silly Hollywood production- "Transformers 9 The WAR on Weather). Please.....

Again I appreciate your even and thoughtful analysis. I know you have skin thick enough to take the heat( pun intended).
Mike Gilroy

Puffin said...

On the other hand, in Australia...

goingforthegold said...

How would one tell if the anthropogenic effects were indeed causing a rise, and the natural effects were mitigating or minimizing them?

Pete said...

Thanks Cliff! I have tremendous respect for you following the empirical facts rather than the shoddy work that much of climate science represents. Earth's climate is far too complex for the kind of definitive predictions being made in the media to have any validity. Natural variation has dominated CO2's effects throughout Earth's geologic history. I suspect that the physics principle of equal and opposite reaction results in a negative system response to the extra retained heat that CO2 produces. Any thoughts on that?

NateLMP said...

I am curious about your thoughts on this, cliff: http://www.weatherwest.com/archives/1797

He is linking the persistent ridging pattern with climate change.

Alf said...

Interesting analysis and I tend to agree with most of your assumptions. What I don't understand (so I don't disagree or agree) is that you state when talking about the two temp charts: "The top figure has the temp variations from three different sources as well as lines that attempt to smooth out the yearly variations. The second figure has a single trend line (which really doesn't make sense since the variation is so complex)."

i've heard it argued as to tidal variation that these wide variations are what is the most disastrous. I.e. if "king" tides are having higher mean tides over time, the real threat to low lying communities is the *variation* peaks...which do the most damage. The peaks in the charts show me that the highs are higher, which are almost 2 degrees over the low. That could have real significant impacts in crop yield, etc. But I don't understand why you do not accept their trend line, since it seems to be a median to the temp ranges. That seems a valid way to chart it. And the trend is up. What am I missing in that pov?

William H. Calvin said...

This leaves open, however, the possibility that global-scale overheating has modified the PDO. There must be intermediate-scale levels of explanation between increases in mean global surface temperature and regional temperature changes.

Remember the history of evolutionary explanations. Darwin's 1859 natural selection was said to have been not the real story by 1900 when mutations were all the rage as an evolutionary explanation. Took 40 years to sort that out: that it was both in combination.

Patrick said...

Pete, I believe Cliff addressed your question in one of his closing paragraphs:

"As the century progresses, the human-caused global warming signal will increase while natural variability should remain about the same. Thus, the lack of human-caused warming on the West Coast does not imply that anthropogenically forced global warming is unimportant. It is a serious issue mankind must deal with."

Cliff Mass said...

Bill Calvin,
Global climate model simulations under various degrees of warming do not suggest a change in PDO amplitude...clif

Terri Jackson MSc MPhil MInstP said...

recent paper from D Bozhinoval et al Meteorology Group Wageningen University netherlands Institute for Marine and Atmospheric Research Utrecht(published in Atmospheric Chemistry and Physics 14, 7273-7290, 2014), they found that only 3.75%(about 15ppm) of atmospheric CO2 is man made! The remainder of the 400ppm is from land use changes and natural sources such as plant respiration and some ocean de-gassing!

Pete said...

Patrick,

Actually Cliff didn't respond to my question. The ice core record shows that CO2 is always at peak values when global cooling begins. There is some system response that negates CO2's amplification. Ice ages have started with CO2 at 10x current levels. The geologic record clearly shows that natural variation is more powerful than CO2.

I don't agree with Cliff that natural variation will remain about the same. NOBODY knows what the future holds but the consensus from scientists who study solar cycles is that we are entering a grand solar minimum similar to the Maunder that produced the little ice age. If that happens we will get a much clearer idea of what CO2's true climate sensitivity is. The trend in climate research is for much lower values that the IPCC and media would have you believe.

Pete said...

Terry Jackson,

I haven't read the paper but I believe they are saying humans are contributing 3.75%/year not total. Burning fossil fuels produces a different isotope than natural sources so we know that humans are increasing the total amount.

At current levels this seems to be beneficial with a measurable increase in Earth's biomass and a greening of the planet.

Unknown said...

A response in the form of a new letter by Abatzoglou, Rupp and Mote has appeared in the Dec 30th 2014 issue of PNAS.
http://www.pnas.org/content/111/52/E5605

Cliff Mass said...

Unknown,
There response is very weak. I still have real problems with their work..cliff