Sunday, August 10, 2014

IPCC models versus actual trends

This is an extension of my previous post comparing IPCC models and actual temperature data.  I had a request to directly compare the observed rates of temperature rise with the predicted rise from the average of the AR5 models.  First, my methods:  I averaged all 81 IPCC AR5 8.5 models.  I then calculated the rate of change for the average of the models as well as Berkeley Earth's Land + Ocean dataset, the new Cowtan-Way coverage-corrected version of HadCRUT4, and GISS.  All rates were calculated after compensating for autocorrelation.  With that out of the way, here's the rates of temperature rise for the last 30 full years (1984-2013) in three surface datasets that cover the entire globe versus the average of all 81 IPCC AR5 8.5 scenarios:

Rate of temperature increase between 1984 and 2013 for IPCC AR5 models (8.5 scenario) versus Berkeley Earth Land + Ocean data, Cowtan-Way

While the average rate from the IPCC AR5 8.5 models is higher than the observations, the observations are well within the 95% confidence interval.  The difference in the rates is not statistically significant.

The true test of the AR5 models, however, is their accuracy since January 2000.  That's the start of the actual predictions.  That also poses a problem for determining whether or not the models are accurate.  It's only been 14.58 years since January 2000.  You need over 17 years to reliably detect actual climate trends in temperature data (Santer et al. 2011), making comparisons since January 2000 largely meaningless as the following graph shows.

Notice the massive 95% confidence intervals around the observed trends?  The observed trends could be anything from less than -0.01ºC/year to over 0.03ºC/year.  That range easily contains the average trend calculated from the IPCC AR5 models.  So while it appears that the predicted average trend is far higher than the observed trends, the reality is that there's no statistical difference between the predicted and observed trends.  And while I'm at it, there's also no evidence that global warming has stopped or even really slowed—just look at that possible range around the observed trends.

The reality is that right now, there is no statistical basis to determine if the IPCC AR5 models are wrong, even without accounting for random climate events like ENSO, volcanic eruptions, or changes in solar output. Published research shows that just accounting for ENSO alone explains much of the discrepancy between the predicted trends and the observed trends (i.e. Kosaka and Xie 2013, Risbey et al. 2014).  The upshot of it all is that those proclaiming that the models are wrong are either greatly exaggerating or simply ignoring the evidence in favor of a simplistic and wrong view of how the world works.

Monday, July 21, 2014

Risbey et al. 2014

It seems the canard about how IPCC models are inaccurate just won't go away.  I've covered it before on this blog.  The newest incarnation of that canard revolves around a new paper by Risbey et al. (2014).  It seems that many just don't understand what Risbey et al. did and they definitely don't understand the results of that paper.

What Risbey et al. did was fairly simple.  They used a moving 15-year window and evaluated multiple climate models based on each model's ability to match the actual El Niño/Southern Oscillation (ENSO) state over that time period.  They took the models that best matched the actual ENSO state over that time period—regardless of how accurate anything else was about the selected models—calculated the predicted temperature trends from each selected model, and compared those predicted trends to the actual temperature trend over that same period.  Then they shifted the window and repeated the exercise.  One of the results was a graph (Figure 2) that looks similar to one that I created back in January to find the last time the Earth had a 15-year cooling trend

Figure 2 from Risbey et al. 2014 showing actual 15-year temperature trends versus raw IPCC models 15-year trends

What their main results show is the reason their paper is "controversial" in the denial bubble.  Risbey et al. showed that the mismatch between climate models and atmospheric temperatures in recent years is due to mainly to a mismatch between the predicted states of ENSO and the actual state of ENSO.  Not because the models are inherently wrong.  Not because scientists don't understand the climate system.  Not because the physics are wrong or any other standard denier canard about climate models or climate scientists.  It's because computer models are unable to predict, years in advance, exactly what one chaotic phenomenon will do.  If there's a mismatch between that predicted input and the actual input, then the model temperature predictions will appear to be off.  Once they used the match to ENSO as the selection criteria, the predicted temperature trends were very similar to the actual trends, as their Figure 3 shows.

Figure 3 from Risbey et al. 2014 showing the match between model predictions and actual 15-year trends.

 To those who keep up with the scientific literature, this will not be a surprise.  Others (i.e.  Kosaka and Xie 2013) have shown before that the IPCC models accurately predict temperatures since 2000 if they are given actual ENSO values rather than predicted values.  Risbey et al. just adds to the evidence that the climate models are accurate, that the main problem lies in predicting ENSO values rather than any inherent problem in the climate models.  That should give us pause, as the effects of a chaotic oscillation will cancel out over time, leaving the trend unchanged.  And we know where that trend is headed, regardless of the short-term effects of ENSO.

So, what to make of the kerfuffle roiling the denialsphere over Risbey et al.?  Much of it is simple ignorance—they don't understand what Risbey et al. have done and what their paper shows.  The rest is simply willful ignorance from those who should know better.

Sunday, July 20, 2014

Seasonal trends by hemisphere

A reader raised a good question about my last post.  One of the confirmed predictions of climate change is that winters will warm faster than summers, yet my analysis showed that December-February warmed the least.  The question was why that would be.  The answer is simple: I used global temperature data.  December-February may be winter in the Northern Hemisphere but it is summer in the Southern.  The seasons largely cancel out.  The reason we see faster warming in the June-August  (Northern Hemisphere summer) versus December-February (Southern Hemisphere summer) is due to the position of land masses.  There's a greater proportion of ocean in the Southern Hemisphere and the ocean doesn't change temperature very readily compared to landmasses whereas the Northern Hemisphere has more landmass and consequently a larger response to seasonal changes in insolation.

To see whether or not winters are truly warming faster than the summers, you must use mid- to upper-latitude areas of each hemisphere.  The low-latitudes (i.e. tropics) don't experience much of a seasonal difference.  I'm going to use UAH satellite data, as it breaks down the monthly average temperatures of different regions of the planet, including the mid- to high-latitudes.  Looking at the Northern Hemisphere mid- and high-latitudes (labeled NoExt in the dataset), we see the following:

The trends for each season are as follows:

95% confidence interval
Winter0.281ºC/decade± 0.059ºC/decade
Spring0.306ºC/decade± 0.077ºC/decade
Summer0.176ºC/decade± 0.090ºC/decade
Autumn0.255ºC/decade± 0.063ºC/decade

Northern Hemisphere winters have warmed at a 0.281ºC/decade clip whereas summers have "only" warmed at a 0.176ºC/decade rate, the slowest warming rate of all the seasons. Also interesting is the warming rates of the spring and autumn, both very similar to that of winter.

In the Southern Hemisphere mid- to high-latitudes (SoExt in the UAH data), you find the following pattern:

The trends for each season are:

95% confidence interval
Winter0.115ºC/decade± 0.036ºC/decade
Spring0.177ºC/decade± 0.073ºC/decade
Summer0.053ºC/decade± 0.060ºC/decade
Autumn0.037ºC/decade± 0.055ºC/decade

Again, winters warm faster than the summers. In fact, the summers and autumns in the Southern Hemisphere haven't shown any statistically significant warming at all. In the Southern Hemisphere, only winter and spring show any statistically significant warming since 1979.

 So what to made of all this?  In both hemispheres, winter temperatures have warmed faster than summer temperatures, thereby confirming the prediction.  Spring temperatures have warmed the fastest in both hemispheres whereas autumn temperatures are mixed between the two hemispheres.

Saturday, July 19, 2014

Seasonal trends

It is almost comical how people will grasp at any straw they can come up with to claim that global warming isn't happening.  The most recent bit of hilarity?  A claim that the winter trend since 2002 is cooling, therefore we're in global cooling, not global warming.  Let's check that one out to see just how ridiculous it is.

First, here are the seasonal trends since 1979 using Berkeley Earth land + ocean data.  Why 1979?  Because that's the first full year of the satellite era, enabling readers to cross check me if they prefer satellite data instead of surface data.

DJF = December, January, February, MAM = March, April, May, JJA = June, July, August, SON = September, October, November
Looking at the trend lines, it's obvious that SON (September, October, November) has experienced the largest increase, DJF (December, January, February) the least.  The calculated trends bear this out:

95% Confidence
DJF0.137ºC/decade± 0.043ºC per decade
MAM0.155ºC/decade± 0.038ºC per decade
JJA0.160ºC/decade± 0.036ºC per decade
SON0.193ºC/decade± 0.036ºC per decade

The overall trend?  0.161ºC per decade ± 0.034ºC per decade.  Yep.  Sure sounds like global cooling to me.  /sarcasm.

Now for a closer look at the data since 2002, which is far short of the 17 year minimum needed to reliably detect climate trends.

DJF = December, January, February, MAM = March, April, May, JJA = June, July, August, SON = September, October, November
It does indeed look like there's been quite a bit of cooling during the DJF quarter.  However, take a look at the 95% confidence interval, as appearances can be deceiving:

95% Confidence
DJF-0.090ºC/decade± 0.189ºC/decade
MAM-0.020ºC/decade± 0.159ºC/decade
JJA0.085ºC/decade± 0.123ºC/decade
SON0.045ºC/decade± 0.104ºC/decade

The data shows a minor trend in each quarter but the confidence intervals tell us that those trends are not statistically significant from no trend at all.  The overall temperature change since 2002 using all the data? 0.000872ºC per decade ± 0.107839ºc per decade. In other words, as flat as the myth of global cooling since 2002.

Friday, July 11, 2014

What will summers be in AD 2100?

While I'm working on a longer post, here is an interesting interactive graphic from Climate Central.  Just type in the city you are interested in and see what the summers will be for that city in AD 2100. It shows which US city (or world city) currently experiences average summer temperatures as hot as those predicted for selected US cities and gives you an idea of how far north the climate bands will have shifted over the next 86 years.

Average summer highs in my current hometown are predicted to warm up by nearly 6.5ºC from the current average high of 29.4ºC to a predicted average high of 35.9ºC.  How much will your closest city warm?

Thursday, July 3, 2014

NewsMax, Ambler, and BS about Antarctic sea ice printed a poorly writen story about Antarctic sea ice intended to sow confusion amongst its readers which was unfortunately shared on my Facebook timeline by an acquaintance.  However, the article is even worse than being merely poorly written.  The "author," Sandy Fitzgerald, extensively plagiarizes a blog post by Harold Ambler while leaving out nearly everything Ambler wrote about why Antarctic sea ice is increasing.  Fitzgerald manages to make it sound like a new record extent in Antarctic sea ice somehow contradicts global warming.  Ambler's original is a far more nuanced argument.

Ambler begins by noting that the sea ice anomaly for Antarctica hit a record high on either June 26 or June 27, as shown by the Cryosphere Today website.  And yes, the daily anomaly has been generally increasing since around 2010.  However, saying that the anomaly hit a new high on any particular date is somewhat misleading.  That graph is of the daily anomaly.  It simply take the ice extent of a particular day relative to the average for that day.  Ice extent changes on a daily basis due to the actual amount of ice and how storms or prevailing winds shift the ice around, as anyone who examines the daily ice extent data for either the Arctic or Antarctic will know.  What is more meaningful is the overall trend in ice extent, not the ice extent anomaly on a particular day.

Ambler's main argument is that global daily sea ice anomalies (Arctic + Antarctic) do not show much of a decline over the past 13 years with some growth over the past 18 months.  Therefore, global warming isn't that bad, because the loss of sea ice in the Arctic is being offset by sea ice growth in the Antarctic.  Quite frankly, I'm not impressed with either his method (adding the daily anomaly for the Arctic with that of the Antarctic) or his argument.

First, he completely misses the key point that explains why the loss of the Arctic ice cap is more important to global climate than the growth in wintertime Antarctic sea ice.  The reason Antarctic sea ice is not as important to the climate system is that it forms in the polar winter and then melts almost completely during the polar spring and summer.  It's dark at the poles in the winter.  Not much solar energy to reflect back into space.  All that Antarctic sea ice?  Doesn't do much to the energy balance of the planet.  The Arctic ice cap, on the other hand, lasts throughout the polar summer.  Plenty of sunlight up at there during the summer, energy which would normally get reflected back into space.  Losing that ice cap means more energy gets absorbed by the Arctic Ocean and turned into heat.  It doesn't take a genius to figure out why the loss of the permanent Arctic ice cap is more important than a new peak in wintertime-only Antarctic sea ice.

Second, his preferred method of measuring the state of the sea ice is to simply add Arctic ice extent with Antarctic ice extent, ignoring the offset in the seasons between the poles.  Arctic sea ice reaches its winter maximum in March, its summer minimum in September.  Antarctic sea ice reaches its summer minimum in February, its winter maximum in September.  That offset means that simply adding the two masks any decline in the permanent sea ice cap, hiding it behind the temporary seasonal sea ice of the respective polar winters.  If Ambler wants to really compare apples to apples and if he really wanted to focus on the sea ice that actually impacts global climate by sticking around in the summer, he should use the extent of multiyear sea ice which is revealed by the yearly sea ice minimums for his calculations.

Minimum global sea ice.  Trend calculated using loess regression and shown with 95% confidence intervals.

I rather suspect that Ambler won't like that sort of analysis, as it doesn't quite fit with his narrative.  Neither does analyzing the actual extent of sea ice at each pole (not daily anomalies) or the actual global sea ice extent.  I used a 12-month running mean to compensate for the yearly cycle along with loess trend lines to show highlight the trends.

The results of analyzing the actual extent rather than daily anomalies reveal that while sea ice may have increased over the past 18 months, it hasn't been enough to change the overall trend.  That 18 month growth is based on a temporary growth in Arctic sea ice (note that the 12-month average for the Arctic has already started declining again) combined with an increased growth in Antarctic seasonal ice.

Third, while Ambler acknowledges that the surface of the Southern Ocean is freshening which would lead to greater ice growth in the winter (as shown by Bintanja et al. 2013), he fails to mention why the surface of the Southern Ocean is freshening.  As Shepherd et al. (2012) showed, the reason is the -71 billion metric tons of ice that has been disappearing from the Antarctic continent every year on average since 1992.  Even that average understates the issue, as Shepherd et al. showed that the amount of ice lost has accelerated from a net of -48 billion metric tons per year between 1992 and 2000 to a net of -81 billion metric tons per year between 2005 and 2010.  Of course, acknowledging the loss of Antarctic continental ice would put a damper on his argument that Antarctic sea ice growth offsets the loss of Arctic sea ice as it would mean that ice at both poles is melting.  Again, it doesn't fit in with his narrative.

So, what do I make of his article?  He relies on short time frames, leaves out the  inconvenient fact about Antarctic continental ice, glosses over the reasons that show why the growth in Antarctic wintertime ice matters less than the loss in the permanent Arctic ice cap, and prefers a type of graph that hides the decline.  In short, other than the quotes from various climate scientists, I'm not very impressed. 

Tuesday, July 1, 2014

How low will it go?

This is the time of year all eyes start turning toward the Arctic, specifically how rapidly and/or much the multiyear sea ice will melt this particular year.  Here is the average sea ice extent for September since 1979:

I've added a loess regression line along with 95% confidence intervals to highlight the trend.  Despite last year's rebound from the 2012 low, the overall trend is decidedly down, with less multiyear ice remaining as time goes by.

Now on to my prediction of what the ice will do this year.  What I did was simply to extrapolate based on the loess regression, which yielded a prediction of 4.135 million km2.  I also fitted a polynomial regression to the data, which produced a prediction of 4.085 million km2.  Both are well above the record low in 2012 (3.58 million km2), showing how anomalous the 2012 low was.  If the loess curve holds, I wouldn't expect the trend to fall below the 2012 low until 2018.

So, what is your prediction of what the September low will hold for Arctic sea ice?