Feb 08 2015

2014 the warmest year as global heating speeds up

2014 was the hottest year in the global surface temperature record, according to independent analyses by NASA’s Goddard Institute for Space Studies (GISS), US National Climatic Data Center (NCDC), UK’s Met Office, and even the team of skeptics at Berkeley Earth. The one exception is the University of York dataset, which has greater warming over the last two decades and a higher record which 2014 was not quite able to break. The former two records go back 135 years; the latter three cover 165 years.

NASA GISS director Gavin Schmidt said:

This is the latest in a series of warm years, in a series of warm decades. While the ranking of individual years can be affected by chaotic weather patterns, the long-term trends are attributable to drivers of climate change that right now are dominated by human emissions of greenhouse gases.

In the NCDC dataset, the global temperature during 2014 was 0.69°C above the 20th-century average. (Climatologists usually give temperatures as anomalies relative to an average, because they are easier to compare than absolute temperatures.) The rest of the ten warmest years are:

  • 2010 and 2005 (tied at 0.65°C above average)
  • 1998 (0.63°C)
  • 2013 and 2003 (tied at 0.62°C)
  • 2002 (0.61°C)
  • 2006 (0.60°C)
  • 2009 and 2007 (tied at 0.59°C)

2014 included the warmest April, May, June, August, September, October, and December; 4th warmest January, March, and July; 7th warmest November; and 21st warmest February. If you remove the seasonal cycle, December 2014 was the 8th warmest of all months, and monthly global ocean temperatures were at record levels from May to November 2014.

A record warm year was experienced in most of Europe, including the Central England Temperature record extending back 356 years. It was also a record warm year for northern Africa, Far East Russia, the west coast of North America, parts of inland South America, and parts of Australia’s west and east coasts. Australia as a whole had its warmest spring and third warmest year. About the only corner of the globe that was cooler than average was the central and eastern United States. Although global precipitation was near-average, extreme flooding and drought occurred in many parts of the world; California had its worst drought in 1,200 years.

You can read more about Australian weather in 2014 here and here.


2014 annual temperature anomalies around the globe (Source).

The surface global warming trend continues

Examples of noisy short-term trends (blue) contrasted with the far more reliable 45-year trend (red) in NASA GISS monthly global surface temperature anomaly data from January 1970 through December 2014. (Source)

Examples of noisy short-term trends (blue) contrasted with the far more reliable 45-year trend (red) in NASA GISS monthly global surface temperature anomaly data from January 1970 through December 2014. (Source)

In all the aforementioned datasets:

  • Every one of the 14 years so far in the 21st century is one of the 15 warmest.
  • As a 23-year-old, I have never experienced a cooler-than-average year.
  • The data show a warming trend since (very nearly) any start year you choose.
  • Each of the last few decades has been successively warmer.
  • The Earth has warmed ~0.85°C since the preindustrial era.
  • The Earth has warmed ~0.65°C in the last four decades (~0.75°C according to University of York, which finds greater recent warming because it has better coverage of the rapidly warming Arctic).

It’s disgraceful how deniers have manipulated the world into believing global warming has “paused”. Their latest talking point is that it is statistically uncertain whether 2014 is the warmest year; NCDC estimates it is 48% likely to be warmest while the University of York says it is second warmest. What the deniers ignore is that 2014 is more likely to be warmest than any other year, and in any case it is the long-term trend that is important. And the odds that the long-term trend would have occurred in the absence of human influence are at least 1 in 1.7 million.

Another denialist talking point is that since 1998 (a year which deniers like to cherry-pick because it was exceptionally hot) the warming trend is “not statistically significant” (ie. it is not distinguishable from random noise). In reality, this statement is meaningless because it is not uncommon for such a short-term trend to be statistically insignificant.

Moreover, the warming trend since 1998 is also not significantly less than it was before. Recently, Niamh Cahill from University College Dublin applied to NASA’s annual global temperature data a statistical technique called change point analysis, which is designed to identify objective points in time when a linear trend changes. He found only three change points: 1912, 1940, and 1970. The model refused to produce a fourth change point during the last 45 years.

So contrary to what the denialists claim, there is no discernible “slowdown” in global warming and certainly no “pause” or “cooling”. With that meme increasingly discredited, deniers are now falling back on the argument that the current global temperature is at the low end of climate model projections. There are several problems with this argument. Firstly, it’s true that the models predicted an acceleration, and in the surface temperature record we seem to be seeing a linear trend, but heating of the Earth as a whole is actually accelerating. Secondly, the extra heat is not showing up in surface warming because it is being masked by temporary factors. Thirdly, the latest science can in fact explain the observations. Finally, surface global warming will soon resume faster than ever.

Global heating is speeding up

The vast majority of heat from global warming goes into the oceans, so ocean heat content is a more reliable indicator of climate than surface or atmospheric temperature. This data (consistent with satellite measurements at the top of the atmosphere) shows global warming has accelerated in the last 15 years. In 2014 the Earth’s oceans accumulated energy at a rate of 7 Hiroshima atomic bombs per second, according to global ocean heat content records from the US National Oceanographic Data Center (NODC). This rapid heating in 2014 compares to an average of 4 Hiroshima bombs per second since 1998, and 2 bombs per second since records began in 1955.


Global ocean heat content 1955-2014. (Source)

We are observing an acceleration in other climate indicators. Arctic sea ice melt is proceeding much faster than predicted, threatening to set off a chain reaction of tipping points including ice sheet melt and release of carbon from permafrost. The ice sheets of Antarctica and Greenland are now losing 500 cubic km of ice per year, with Western Antarctica melting three times faster than five years ago and five times faster than the mid-1990s, and Greenland two times faster than five years ago and ten times faster than the mid-1990s. A 2014 study found the West Antarctic melt appears “unstoppable”, implying an eventual sea level rise of 1-5 metres. Another recent study discovered sea level rise is accelerating faster than previously thought. And in 2014 permafrost thaw caused huge craters to appear in Siberia’s Yamal Peninsula, which fittingly means “end of the world”.

Several studies show climate scientists have systematically underestimated the impacts of global warming. New results are more than 20 times more likely to be worse than predicted than they are to be better. It appears scientists have been overcorrecting in response to the deniers’ accusations of alarmism.

Intensifying heatwaves, floods, droughts, and fires are already costing lives and reducing crop yields. Global warming has already reached a dangerous level; future warming will take us further into the danger zone.

Temporary factors are masking some surface warming

Ocean cycles

So why is not all of this extra heat showing up in the surface temperature record? The answer lies in ocean cycles that periodically redistribute heat between the Earth’s oceans and atmosphere, explaining most of the year-to-year variation in surface temperature. These cycles are unrelated to long-term climate change caused by heat entering and leaving the Earth’s climate system. This internal variability is the reason we should focus on long-term trends instead of short-term ones, and the total amount of heat building up rather than the rate of surface warming.

The most important cycle is the Southern Oscillation, which alternates between El Niño (warm), La Niña (cool), and neutral. In an El Niño phase, the trade winds slow and the surface of the tropical Pacific Ocean is warmer than usual. The anomalous warmth of 1998 was due to a super El Niño. In a La Niña phase, the trade winds accelerate and warm water is pushed to the western Pacific, causing the ocean to absorb more heat and bringing cold water to the surface behind it. 2011 and 2012 were cooler than the surrounding years because they occurred during a La Niña. 2014’s record warmth occurred during a neutral phase of the El Niño Southern Oscillation (ENSO), unlike most record warm years which tend to occur in an El Niño. When you compare years with comparable Southern Oscillation conditions, the ongoing warming trend becomes much clearer (see this animated graph, which annoyingly won’t work on my website).

Recently the trade winds have surged to an unprecedented speed, shifting the relative frequency of Southern Oscillation phases toward La Niña and temporarily burying much of the incoming heat from global warming beneath the ocean surface. The increase in the trade winds appears to be driven by a longer cycle, the Interdecadal Pacific Oscillation (IPO), which has been in a cool phase since 2001. Some scientists argue more frequent La Niñas could be a negative feedback on global warming, but others argue global warming will make El Niños more frequent.

Sun, volcanoes, and air pollution

The buildup of heat has continued despite several influences acting to temporarily decrease the amount of heat coming in, reducing the human-caused warming that otherwise would have occurred:

Science can explain the apparent discrepancy

Let’s put all the pieces of the puzzle together. The median climate model projection for global warming during 1998-2012 was 0.3°C. Met Office surface temperature data for the same period show only 0.06°C warming. After accounting for Arctic warmth, the University of York finds an extra 0.11°C. A study analysing all the natural cooling factors estimates that ocean cycles contributed 0.06°C cooling, the Sun 0.04°C, and volcanoes 0.035°C. If these numbers are correct, this puts climate models right on target.

Although Met Office observations (solid black line) seem inconsistent with earlier model projections (dotted orange line), University of York’s corrected observations (dashed black line) are consistent with models corrected to include observed cooling influences (solid orange line) (Source).

Although Met Office observations (solid black line) seem inconsistent with earlier model projections (dotted orange line), University of York’s corrected observations (dashed black line) are consistent with models corrected to include observed cooling influences (solid orange line) (Source).

Other studies claim a larger cooling effect from ocean cycles (0.1-0.2°C) or volcanoes (0.05-0.12°C), in which case the underlying human-caused warming trend might even be greater than the models predicted. In other words, the climate might be more sensitive than the median modelled value of 3°C per doubling of CO2. Then there is the question of how anthropogenic aerosols fit into the picture.

In a sense, even the old climate projections can explain observed surface temperatures when correctly understood. The models were never designed to predict 15-year trends because the timing of ocean cycles is unpredictable. Instead they ran many simulations with random ocean fluctuations, and some simulations indeed predicted periods of up to 21 years where surface warming appeared to pause. The simulations which came closest to real-world ocean cycles accurately predicted the observed surface temperatures.

Rapid surface warming will resume

Ocean cycles are just that: cycles. Although currently they are counteracting the underlying greenhouse warming trend, sooner or later the cycles will turn around and reinforce it, rapidly warming the surface until it catches up to the temperature it would otherwise be. The Earth continues to accumulate heat, which will be released back into the atmosphere when the IPO shifts back into a warm phase and the trade winds return to normal speed, expected to happen within the next decade. Even if increased trade winds turn out to be permanent, the effect will be overwhelmed by rising greenhouse gases as emissions continue (and warming oceans have other impacts such as sea level rise). Natural cycles are now merely waves on the rising tide of greenhouse warming.

Temperatures in 2015 will be largely determined by the Southern Oscillation. Most Southern Oscillation models predict an El Niño in the second half of the year, though some predict conditions will remain neutral. If the former occurs, it is likely to be a super El Niño similar to 1998, which would make 2015 much warmer than 2014 since it was a neutral year.

The total cooling effect of all anthropogenic aerosols could be as large as 1°C (some of which climate models don’t account for). Unfortunately, this can’t last because when we cut fossil fuel use that will also cut aerosol emissions, and aerosols remain in the atmosphere for a much shorter timeframe than CO2. Similarly volcanic particles will fade away within a few years unless there are more eruptions. Even if we deliberately pollute the atmosphere with new aerosols, that won’t stop CO2 acidifying the oceans.

Much like the passengers on the Titanic, we’ve been lulled into a false sense of security. Atmospheric CO2 is currently at 400 ppm and rising (its highest level in at least 13 million years), and global emissions continue to increase. Including other greenhouse gases brings the level to 480 ppm CO2-equivalent, enough to send temperatures soaring over 2°C above preindustrial temperature (1°C above today’s, which you will recall is already dangerous – also see my discussion on impacts of various global temperature scenarios).

A recent climate modelling study by Mike Mann examines how long it will take global warming to reach 2°C. Based on 3°C climate sensitivity, if CO2 emissions continue to grow exponentially while anthropogenic aerosols decrease exponentially, global warming will reach 2°C in just two decades and 4°C by the end of the century. If it turns out the last 15 years have been masking a more sensitive climate, then warming might reach 2°C even sooner. If the last 15 years turn out to be representative of a 2.5°C sensitivity, then warming will still reach 2°C in three decades.

Total cumulative CO2 emissions will be the main factor in the magnitude of long-term global warming. Mann says avoiding 2°C would require limiting CO2 to 405 ppm, which would require ceasing all fossil fuel use. Long-term feedbacks excluded from Mann’s analysis (but which are already being triggered by tipping points we’re crossing now) lower the estimated safe level of CO2 to below 335 ppm.

All this means that it is nonsense to talk about any further amount of fossil fuels being safely “burnable”. To preserve a stable climate we need to approach zero global emissions yesterday, leaving in the ground as much fossil fuel as possible, and then we will probably need to use risky geoengineering measures to remove carbon from the atmosphere and cool the planet. Most importantly, we must phase out the largest and longest-lived cause of global warming, fossil fuel CO2 emissions.

If we want global warming to truly pause, we must hit the pause button. There is no time to lose.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>