Research
)
"It is important to keep the Paris agreement in mind in a similar manner to a doctor’s health warning. We must limit climate change down to levels that are as low as possible, in order to minimise its effects locally and globally."
The 2015 Paris Agreement aims to limit the average global temperature to no more than 1.5°C above pre-industrial levels; however, as temperatures continue to rise, we are very close to exceeding this level.
In order to understand the science behind the calculation of the 1.5°C figure that underlies climate change negotiations across the planet, we spoke with Prof Tim Osborn (Director of UEA’s Climatic Research Unit [CRU] and Professor of Climate Science, School of Environmental Science, University of East Anglia [UEA]), Prof Phil Jones (Emeritus Professor, CRU, School of Environmental Sciences, UEA) and Prof Manoj Joshi (Professor of Climate Dynamics, CRU, School of Environmental Sciences, UEA). We questioned how we can scientifically define when the 1.5°C level has been reached and what the global implications of this will be.
How do we calculate global temperature averages?
The first person to estimate a large-scale temperature average was Vladimir Köppen in the 1880s. Later in 1938, Guy Stewart Callendar calculated an average from 60 degrees north to 60 degrees south, and estimated the pre-industrial concentration of carbon dioxide at 280 ppmv. Although both Köppen and Callendar made use of a relatively limited number of instrumental weather stations from the terrestrial regions of our planet, Callendar’s work in 1938, and later in 1961, compares very favourably with modern land-based estimates that use many more stations.
"In 1986, it was CRU/UEA who first combined land and sea data to produce a global picture of temperature averages."
In the mid-1980s, digitisation of ship logbook data (through, what is now known as, the International Comprehensive Ocean Atmosphere Dataset, or ‘ICOADS’ project) allowed incorporation of data from the ocean – which covers 70% of the Earth’s surface – and, in 1986, it was in fact CRU/UEA who first combined land and sea data to produce a global picture of temperature averages. Since 1990, drifting and fixed buoys across the world’s oceans have added to that data. More recent estimates of global temperature carried out this decade incorporate over 8000 weather station time series and about 260 million individual observations of sea surface and marine air temperatures from ships.
Are there uncertainties in the global temperature record?
Time series of monthly global temperature estimates extend back to 1850. Uncertainties for annual averages are larger in the nineteenth century (±0.15°C in the 1850s) than in more recent decades (±0.05°C since the 1960s). Such uncertainties arise from relatively sparse coverage of data in the nineteenth century, as well as from changes in instrumentation, sites being moved (often from town/city centres to airports), and changes in the environment around stations, particularly urbanisation. Marine records have specific sources of biases such as changes in bucket design for sampling sea surface temperature (or SST) from the 1850s to the 1940s, followed by a completely new technique of using thermometers installed in the engine-intake pipes of steamships.
CRU/UEA isn’t the only group estimating global temperatures. Three groups in the United States (National Aeronautics and Space Administration [NASA]/Goddard Institute for Space Studies [GISS], National Oceanic and Atmospheric Administration [NOAA] and Berkeley Earth), as well as groups in Japan (Japan Meteorological Agency [JMA]) and China (China Meteorological Administration [CMA]) all produce very similar estimates of global temperature. Excellent agreement is also evident with reanalyses, which use complex numerical techniques to combine surface weather observations with measurements from radiosondes as well as satellites to produce global gridded data.
How do recent changes in global temperature compare to those in the past 2000 years?
"The difference between the warmest and coldest centuries between AD 1 and AD 1900 was at most 0.5°C; since then we have experienced approximately 1.2–1.3°C of warming, much of it in the last 50 years."
Estimates of temperatures from further back in time can be derived from natural proxies such as tree-rings, ice cores, corals and marine and lake sediments – and even documentary records where direct archival evidence was written down. The latest compilations of data indicate that the warmest century of the last 2000 years was the twentieth century, and that the difference between the warmest and coldest centuries between AD 1 and AD 1900 was at most 0.5°C; since then we have experienced approximately 1.2–1.3°C of warming, much of it in the last 50 years.
)
How do we scientifically calculate when 1.5°C has been reached?
There is no set definition for calculating exactly when the global temperature, when averaged over a number of years, will exceed 1.5°C above pre-industrial levels. The effects of natural short-term climatic variability, such as the El Niño Southern Oscillation (ENSO), can cause the global temperature to rise or fall by a fraction of a degree, meaning that while an individual year may exceed an increase of 1.5°C, it could well be followed by several years that do not, even in a world that is warming at about 0.2°C per decade.
In addition, it is not possible to pin-point the exact month, or year, when global temperatures will exceed the 1.5°C threshold due to the uncertainties in the global temperature record discussed above. These imply that there is an uncertainty of about 0.1°C in the rise of global temperature; this translates to an uncertainty of approximately five years in when global temperatures exceed 1.5°C above pre-industrial values. This uncertainty is unlikely to be significantly reduced because it mostly comes from the limited data available for defining the pre-industrial baseline (and 1850-1900 is often used in lieu of an earlier and even more poorly observed period).
"Some parts of the world have already warmed by significantly more than 1.5°C."
It must also be noted that while climate change is global, the world warms unevenly: the land warms more than the oceans, while the Arctic is warming much faster than the Antarctic. This means that different regions will reach an increase of 1.5°C at different times. Indeed, some parts of the world have already warmed by significantly more than 1.5°C.
)
An important point to note is that the 2015 Paris Agreement goals (1.5 to 2°C for global temperature) take into account that warming is uneven across the globe, that there is short-term natural variability in individual yearly values, and that global warming is associated with changes in rainfall, drought, ice, snow and extreme weather. The global temperature is simply a convenient metric that represents all of these changes in a single measure of climate change. Therefore, when global temperature rise reaches 1.5°C it is implicit that the land will have warmed more than this and the oceans less than this, that the Arctic will have warmed more than 1.5°C, and that some years will have global temperatures slightly above 1.5°C and some will still be slightly below. The level will not have been breached if some parts of the world have already warmed more than 1.5°C (as they already have) or if global temperatures more than 1.5°C above pre-industrial levels are observed in some individual years (or in individual months, as they already have). It is the global average temperature averaged over multiple years that determines if and when we reach a 1.5°C increase.
What does all this mean for the 1.5°C objective?
One of the 2015 Paris Agreement’s principle objectives is to keep global temperature rise this century to well below 2°C above pre-industrial levels, and to pursue efforts to limit the temperature increase to 1.5°C. This does not mean that 1.5°C and 2°C are specific thresholds above which physical impacts suddenly start in a given year or location, rather they are expressions of the fact that the more the world warms, the worse the impacts and risks of climate change will be. Such impacts will be significantly worse under a warming of 2°C rather than 1.5°C.
"We must limit climate change down to levels that are as low as possible, in order to minimise its effects locally and globally."
Many parts of the world are already seeing extreme events such as heatwaves, flooding or forest fires. The use of state-of-the-art models has shown that the intensities or impacts of some of these events have significantly worsened because of climate change. Not only this, but the more the world warms, the greater the potential for surprises in the climate and Earth systems which have unexpectedly high impacts.
For all these reasons, it is therefore important to keep the Paris agreement in mind in a similar manner to a doctor’s health warning. We must limit climate change down to levels that are as low as possible, in order to minimise its effects locally and globally.