Thursday, January 30, 2014

The problem with "All of the above"

President Obama in his State of the Union address praised his "all of the above" energy strategy.  What he left out is that "all of the above" is a recipe for disastrous climate change.

The 12-month running mean for Mauna Loa CO2 levels is currently at 396.18 ppmv.


With a climate sensitivity of 0.809ºC/W/m2 (3ºC per doubling CO2), that translates to warming of
ΔT = λΔF = λ*[5.35 W/m2 * ln(C/C0)]
ΔT = 0.809ºC/W/m2 * [5.35 W/m2 * ln(396.18 ppmv/280 ppmv)]
ΔT = 1.50ºC above pre-industrial levels
with just the CO2 levels of today.  However, that climate sensitivity value is just the 100-year value.  At longer time spans (i.e. 1,000 years), sensitivity is actually closer 1.618ºC/W/m2.  That means that at today's CO2 levels, we're already committed to 3ºC of warming over the next millennium.  The key phrase?  "At today's CO2 levels."  And that's the main problem I have with how President Obama is approaching the whole issue.

The "all of the above" approach sounds good—and partially insulates the president from right-wing attacks—but it does little to stabilize CO2 levels.  We're still pumping out fossil fuels, still burning them for energy, still dumping the waste CO2 into the atmosphere.  CO2 levels have been climbing at an average +1.484 ppmv per year since 1958 and that rate has been accelerating by an average of +0.01196 ppmv per year2.  If we continue accumulating CO2 at the same rate, CO2 levels by AD 2100 will be 670.81 ppmv and we'll be locked into 3.78ºC of warming by AD 2200 and 7.56ºC of warming over the next millennium.  Considering the observed consequences of just the 0.8ºC of warming we've already experienced and the projected consequences of future warming (i.e. Sherwood and Huber. 2010; Chapter 14 of the IPCC AR5 report; Levermann et al. 2013), to say nothing of methylmercury emissions, landuse changes, etc associated with fossil fuels, any energy plan that does not phase out fossil fuels (yes, ALL fossil fuels) and quickly is simply irresponsible to future generations.  That is the main issue I have with the president's approach—it does little to phase out fossil fuels.  While I realize that nothing will get through the current do-nothing Congress, we need to at least start a serious conversation on phasing out fossil fuels by whatever means necessary.  And the longer we wait to have that national conversation and phase out fossil fuels, the more expensive it will be.  The science on climate change is clear.  The Earth is warming and technology is the reason.  Now we need to ignore the squawks of the flat-earth science deniers and act on the science.

So, what could we do to phase out fossil fuels?  The policy I favor is to put a price on carbon emissions via a tax-and-rebate system on fossil fuels but I know that isn't everyone's preference.

Tuesday, January 28, 2014

The last time the Earth had a 15-year cooling trend of any kind...

I was asked by a friend to identify the last time the Earth experienced a 15-year cooling trend.  The way I answered this was to use a rolling regression on GISS surface data (R code at the bottom).  Turns out that the last time was before I was born.  The period from February 1958-January 1973 (cooling of -0.00188ºC per decade) was the last 15-year cooling trend in GISS surface data.  Every 15-year period since has shown a warming trend of some magnitude—and yes, that even includes trends starting in 1998.
 
Figure 1.  Graph of temperature trends over each 15-year period.  Each point represents the trend over the preceding 15-year period.

Now as to the last time the Earth showed a statistically significant 15-year cooling trend, well, that's a bit tougher to answer.  The code I'm using cannot account for autocorrelation, which means that it is biased toward showing significant time series trends when in reality the trends are not significant.  The original code also didn't compensate for multiple comparisons.  Even using that extremely lenient standard of significance, however, the last time the Earth experienced a statistically significant cooling trend was from August 1957 to July 1972 (trend: -0.00424ºC per year, p-value = 0.0425).  So at the very least, it's been 41 years since the Earth last had a 15-year cooling trend of any sort—and 41.5 years since the Earth had a statistically significant 15-year cooling trend.

Update
 I realized that I made a mistake in not compensating for multiple comparisons.  Since I was computing every possible 15-year trend, 5% of the trends would be statistically significant by chance alone.  With 1,416 computed trends, that means around 71 of them would be statistically significant by chance.  To compensate for multiple comparisons, I used the p.adjust function in R to use the Benjamini & Yekutieli (2001) method for compensating for fake positives.  When I did that, the last statistically significant cooling trend was pushed back in time to the 15 years between February 1943 and January 1958 (trend: -0.06929ºC per decade, p = 0.04).  Again, this is without autocorrelation, so it's very likely that even that cooling was not actually statistically significant.  But it's been at least 56 years since the last statistically significant 15-year cooling trend.  And my guess is that if I could incorporate autocorrelation in to my code, it would be far longer than that.  The current warming really started in 1910, so the safe bet would be 1910 or before.

If any of my readers know how to incorporate ARMA into a rolling regression in R, please leave a comment with the code and I'll update my analysis.

Thanks.

Updated R code used for this analysis
Climate=read.csv("monthly.csv", header=T)
install.packages("zoo")
library(zoo)
Climate.1880=subset(Climate, Year>=1880)
GISS=ts(Climate.1880$GISS, start=c(1880,1), frequency=12)
GISS=as.zoo(GISS)
GISS=merge(GISS=GISS, time=time(GISS))
func=function(z) {
    co=coef(summary(lm(GISS~time, data=as.data.frame(z))))
    c(Est=co[,1], SE=co[,2], t=co[,3], p=co[,4])
    }
rr=rollapply(GISS, width=180, FUN=func, by.column=FALSE, align="right")
rr$p.time.adj=p.adjust(rr$p.time, method="BY", n=length(rr$p.time))
rr
plot(rr[,2], type="l", xlab="Years", ylab="15-year trend (ºC/year)", main="15-year trends in GISS surface temperature data")
curve(0+x*0, add=T)
text(1950, 0.03, "Warming Trend", col="Red", lwd=2)
text(1950, -0.02, "Cooling Trend", col="Blue", lwd=2)

Saturday, January 25, 2014

Ignorance, blindness, and outright misinformation

A childhood acquaintance on Facebook posted a link to a CBN News article that serves as a good example of what is wrong with conservative news media.  Written by Dale Hurd, it is a mash-up of science denier canards with little in the way of actual science or evidence.

The first paragraph sets the tone, making fun of the scientists who were trapped in Antarctic ice.  Nowhere does Hurd mention HOW the scientists were trapped: A storm packed sea ice into the bay their ship was in.  Now this can happen at any time of the year but it doesn't mean that the ice is growing.  It's summer down in Antarctica, a time when Antarctic sea ice melts from ~19 million km2 in September down to ~2-3 million km2 by March.  That leaves plenty of broken pack ice for winds to blow around, especially in the early part of the melt season such as December when the ship was trapped.  So the irony is that those scientists were trapped by MELTING sea ice, an irony completely lost on Hurd.

Hurd's claim: "Ice is not only growing in the South Pole, but in parts of the North Pole, too."

While satellite data shows that Antarctic sea ice is growing, there's reason to believe that some of that growth is not real.  Eisenman et al. (2014) showed that part of the reported growth in Antarctic sea ice is due to how the satellite data is processed rather than actual ice growth.  The most likely source of that error is a change in the type of sensor used to measure ice extent in December 1991 which made the ice appear more extensive than the previous sensor (see their figure 2).  Eisenman and his co-authors suggest subtracting 150,000 km2 from all monthly Antarctic sea ice averages after December 1991 to calibrate the data and factor out the shift due to the change in sensors.  When I do that, the trend in Antarctic sea ice changes +17,961 km2/year (± 2,427 km2/year 1σ standard error) down to +11,819 km2/year (± 2,689 km2/year 1σ standard error).  Still statistically significant but 1/3 less than the trend of the uncalibrated data.  Hurd also ignores the research that shows Antarctica lost an average of -71 billion metric tons of land ice per year for the past 20 years (Shepherd et al. 2012) and that Antarctic sea ice is growing because of the melt water flowing off the continent diluting the top layer of the Southern Ocean and making it fresher (Bintanja et al. 2013).

The only way his claim about growing ice in parts of the North Pole works is if you start your trend in September 2012, the lowest point on record, and completely ignore everything else.  Yes, Arctic sea ice "rebounded" in 2013—to the 6th lowest extent on record.  And Arctic sea ice in December 2013 was "only" the 4th lowest on record.  One years worth of random fluctuation doesn't change the long-term trend, a statistical concept that is apparently lost on Hurd.


Hurd's claim: "And the coldest arctic temperatures in decades have descended upon North America."

Just because it's cold in your backyard doesn't mean that it's colder everywhere.  Look beyond your nose:

Average surface temperature anomaly December 2013
 Notice anything?  And you can't claim that it's just GISS, either.  UAH satellite temperature anomaly maps for December 2013 shows the same pattern:



Hurd's claim: "There are signs that the Earth is entering a very unpleasant cooling period. Sunspot activity remains very low."

Sunspot activity has been falling since 1957.  If global temperatures depended solely on sunspot activity, then global temperatures would have peaked in the late 1950s/early 1960s.  As it is, there's plenty of research available that shows that solar activity has had little to no impact on global temperatures over the past 40+ years (i.e. Usoskin et al. 2005; Foster and Rahmstorf 2011; Huber and Knutti 2011).  Scientists have also examined the impact of a new Maunder Minimum (the lowest solar activity on record) and found that it will have minimal impact, with global temperatures rising by "only" 3.7ºC by AD 2100 rather than the 4ºC rise currently forecast (Feulner and Rahmstorf 2010).

Hurd's claim: 'The last time the sun was this quiet, North America and Europe suffered through a weather event from the 1600s to the 1800s known as "Little Ice Age..."'

First, Hurd might want to check his dates.  The Little Ice Age began around AD 1300 with a series of volcanic eruptions (Miller et al. 2012), NOT in the 1600s.  Second, once again, just because it's colder in your backyard does not mean that it's colder everywhere.

Average temperature anomaly (1961-1990 baseline) during the Little Ice Age.  Taken from Mann et al. 2009.
Hurd's claim: "Pedersen said climate scientists know the Earth stopped warming 15 years ago."

If this statement is true, Jens Pedersen should be ashamed of himself for spouting a blatant lie and for botching his calender to boot (1998 was 16 years ago, not 15).  I've dealt with this one extensively before.  There's no evidence that the Earth stopped warming in 1998.  In fact, as I've already shown here, you can't even conclude that the rate of temperature rise has even changed.  Most of the apparent "pause" in global warming is due to problems with the current surface data sets (HadCRUT4 doesn't include the polar regions, GISS doesn't calibrate buoy-based sea surface temperature measurements to ship-based measurements).  As Cowtan and Way (2013) showed, once you correct those problems, warming since 1998 continues at a 0.1ºC per decade clip—and the underlying rate of rise after factoring out ENSO, volcanic activity, and the solar cycle is running at 0.18ºC per decade.

Hurd's claim: '"In particular one of the issues has been why global warming has stopped during the last 15 years, and climate scientists were very frank that the climate models do not match the climate we observe," Pedersen said.'

Leaving out the fact that Pedersen repeats his lie about global warming stopping 15 years ago, the more interesting question is WHY climate models do not match observations.  I've already considered this one as well and shown that much of the mismatch is due to a combination of errors in the observations and changes in ENSO, volcanic activity, and solar activity.

Hurd's claim: "It has become a political movement, a cash cow for climate scientists and environmental groups, and a way for world leaders to control economies and people."

Ah yes, when all else fails, pull out the conspiracy theories.  Does Hurd also doubt the moon landing as well?

Hurd's claim: "Climate change skeptics have been censored and compared to Holocaust deniers and even child molesters."

 Actually, it was science deniers who compared a climate scientist to a child molester.

In short, Hurd's article is very short on actual facts and analysis, long on debunked talking points, and really just serves to keep his intended audience ignorant while thinking that they know far more than they actually do.  Which appears to be the entire point.

Tuesday, January 7, 2014

Polar vortex and global warming

I've had several people to use the current cold weather in the US as "proof" that global warming isn't happening. Unfortunately for those arguments, they're pure bunk.

First, the Eastern US and Canada do not represent the entire planet. The December 2013 global map of average temperature anomalies show this:


While the eastern US and Canada are cold, the rest of the planet is relatively warm, with only a few areas (e.g. the Middle East) colder than the 1951-1980 average.  This same pattern has been in place since November 2013 and I expect that the January 2014 map won't change much.   [As expected, it did not.  Here's the January 2014 map:]



Again, the Eastern US was far below average whereas the rest of the planet, especially much of the Arctic was far above average.]  This demonstrates the importance of looking at the global temperature data, rather than just the weather in your backyard when contemplating global warming.


Second, the current weather pattern fits a trend that has been tied to global warming.  The rapid warming in the Arctic and loss of Arctic sea ice has been tied to negative phases of  the Arctic Oscillation (i.e. Francis and Vavrus 2012, Jaiser et al. 2012, Liu et al. 2012, Tang et al. 2013).  The Arctic has warmed at a rate of 0.4345ºC per decade since 1980 whereas September sea ice has melted at a rate of -850,560 km2 per decade.


The Arctic Oscillation (AO) measures the pressure difference between the Azores High and the Icelandic Low.  When the AO is negative, the pressure difference is low.  That low pressure difference weakens and slows the polar jet stream, allowing waves to develop.  The more negative the AO, the larger those waves become, allowing cold polar air to spill southward in the troughs and drawing warm air northward along the peaks.  When the AO is positive, the polar jet stream is faster and straighter, bottling cold polar air further north.  Large waves in the jet stream have become more frequent as the ice has melted (Cohen et al. 2013, Francis and Vavrus 2015), making the wintertime pattern of a warm Arctic coupled with cold temperatures over the continents much more frequent.  What does that mean for us?  Toss in the increase in water vapor in the atmosphere (i.e. Santer et al. 2007) and we get colder, snowier winters if in a trough of the jet stream and warm, snow-less winters if in a peak.  Looking at the temperature anomalies map I posted above, it's pretty obvious where the trough in the jet stream is this winter.

One of the other effects of a warming Actic is also evident.  Because the jet stream move slower, storm systems and weather patterns move slower.  Once a particular pattern is in place, it tends to stay that way.  And the more extreme the pattern, the more likely it is to get stuck in place (see Petoukhov et al. 2013).  So buckle up, USA.  It's going to stay colder than normal for a while this winter.  While we're shivering in -20ºC temperatures, just think of Siberia and Europe enjoying their warmer-than-average winter.