To claim global warming stopped in 1998 overlooks one simple physical reality - the land and atmosphere are just a small fraction of the Earth's climate (albeit the part we inhabit). The entire planet is accumulating heat due to an energy imbalance. The atmosphere is warming. Oceans are accumulating energy. Land absorbs energy and ice absorbs heat to melt. To get the full picture on global warming, you need to view the Earth's entire heat content.
This analysis is performed in An observationally based energy balance for the Earth since 1950 (Murphy 2009) which adds up heat content from the ocean, atmosphere, land and ice. To calculate the Earth's total heat content, the authors used data of ocean heat content from the upper 700 metres. They included heat content from deeper waters down to 3000 metres depth. They computed atmospheric heat content using the surface temperature record and the heat capacity of the troposphere. Land and ice heat content (the energy required to melt ice) were also included.
Figure 1: Total Earth Heat Content anomaly from 1950 (Murphy 2009). Ocean data taken from Domingues et al 2008. Land + Atmosphere includes the heat absorbed to melt ice.
Nuccitelli et al. (2012) arrived at a similar conclusion with more recent and updated data (Figure 2).
Figure 2: Land, atmosphere, and ice heating (red), 0-700 meter OHC increase (light blue), 700-2,000 meter OHC increase (dark blue). From Nuccitelli et al. (2012).
A look at the Earth's total heat content clearly shows global warming has continued past 1998. The planet is still accumulating heat. So why do surface temperature records show 1998 as the hottest year on record? We see in Figure 1 that the heat capacity of the land and atmosphere is small compared to the ocean. Hence, relatively small exchanges of heat between the atmosphere and ocean can cause significant changes in surface temperature.
In 1998, an abnormally strong El Nino caused heat transfer from the Pacific Ocean to the atmosphere. Consequently, we experienced above average surface temperatures. Conversely, the last few years have seen moderate La Nina conditions which had a cooling effect on global temperatures. And the last few months have swung back to warmer El Nino conditions. This has coincided with the warmest June-August sea surface temperatures on record. This internal variation where heat is shuffled around our climate is the reason why surface temperature is such a noisy signal.
With so much internal variability, scientists employ statistical methods to discern long-term trends in surface temperature. The easiest way to remove short-term variations, revealing any underlying trend, is to plot a moving average, performed in Waiting for Cooling (Fawcett & Jones 2008) . Figure 2 displays the 11-year moving average - an average calculated over the year itself and five years either side. They've used three different data-sets - NCDC, NASA GISS and the British HadCRUT3. In all three data-sets, the moving average shows no sign that the warming trend has reversed.
Figure 3: Globally-averaged annual mean temperature anomalies in degrees Celsius, together with 11-year unweighted moving averages (solid lines). Blue circles from the Hadley Centre (British). Red diamonds from NASA GISS. Green squares from NOAA NCDC. NASA GISS and NOAA NCDC are offset in vertical direction by increments of 0.5°C for visual clarity.
Next, Fawcett and Jones look for a cooling trend in the 10 years since 1998. They find the linear trend over 1998 to 2007 is a warming trend in all three data-sets. Note that HadCRUT3 displays less warming than NASA GISS and NCDC. This is most likely due to the fact that HadCRUT data doesn't cover parts of the Arctic where there has been strong warming in recent years.
Figure 4: Linear trends (solid lines) in the three global annual mean temperature anomaly time series over the decade 1998-2007.
Cowtan & Way (2013) also evaluates global surface warming across the globe by using a statistical method known as 'kriging' andby using satellite data to fill in the gaps where there are no temperature stations. Their study shows that the global surface warming trend for 1997–2012 is approximatley 0.11 to 0.12°C per decade.
The reason that 1998 was such an anomalously warm year was due to a strong El Niño that year. Fawcett and Jones remove the El Niño Southern Oscillation (ENSO) signal by calculating a linear regression of global temperatures against the Southern Oscillation Index. A detailed description of the process is found in Fawcett 2007. The result is shown in Figure 4.
Figure 5: Three time series of globally-averaged annual mean temperature anomalies (circles) in degrees Celsius, together with ENSO-adjusted versions (lines), for the period 1910-2007.
All 3 data sets demonstrate that the anomalously hot 1998 was due to the strong El Niño of 1997/98. When ENSO-adjusted, 1998 looks much less remarkable than it does in the original data. In all 3 ENSO-adjusted data-sets, 2006 is the hottest year on record and the trend from 1998 to 2007 is that of warming.
In addition to removing the ENSO signlal, Foster and Rahmstorf (2011) used multiple linear regression to remove the effects of solar and volcanic activity from the surface and lower troposphere temperature data. Their results are shown in Figure 6.
Figure 6: Average of all five data sets (GISS, NCDC, HadCRU, UAH, and RSS) with the effects of ENSO, solar irradiance, and volcanic emissions removed (Foster and Rahmstorf 2011)
When removing these short-term effects, the warming trend has barely even slowed since 1998 (0.163°C per decade from 1979 through 2010, vs. 0.155°C per decade from 1998 through 2010, and 0.187°C per decade for 2000 through 2010).
Of the three surface temperature records (HadCRUT3, NASA GISS, and NCDC), only HadCRUT3 actually shows 1998 as the hottest year on record. For NASA GISS and NCDC, the hottest year on record is 2005. A new independent analysis of the HadCRUT record sheds light on this discrepancy. The analysis is by the European Centre for Medium-Range Weather Forecasts (ECMWF) who calculated global temperature, utilizing a range of sources including surface temperature measurements, satellites, radiosondes, ships and buoys. They found warming has been higher than that shown by HadCRUT. This is because HadCRUT is sampling regions that have exhibited less change, on average, than the entire globe.
Figure 6 shows the regions that HadCRUT have sampled compared to the regions ECMWF included in their dataset. The ECMWF analysis shows that in data-sparse regions such as Russia, Africa and Canada, there is strong warming over land that is not included in the HadCRUT's sampling data. This leads the ECMWF to infer with high confidence that the HadCRUT record is at the lower end of likely warming.
Figure 7: Increase in mean near-surface temperature (°C) from (1989-98) to (1999-2008). Top figure shows HadCRUT sampling regions, lower figure shows ECMWF analysis (ECMWF 2009).
This result is not unexpected. NASA GISS find a major contributor to the record hot 2005 is the extreme warming in the Arctic (Hansen 2006). As there are few meteorological stations in the Arctic, NASA extrapolated temperature anomalies from the nearest measurement stations. They found the estimated strong Arctic warmth was consistent with infrared satellite measurements and record low sea ice concentrations.
Figure 8: Surface temperature anomaly for the first half-decade of the 21st century (Hansen 2006).
Intermediate rebuttal written by dana1981
Update July 2015:
Here is a related lecture-video from Denial101x - Making Sense of Climate Science Denial
|The Skeptical Science website by Skeptical Science is licensed under a Creative Commons Attribution 3.0 Unported License.|