Blog Archive

Wednesday, March 27, 2013

Jeff Masters: How to abuse statistics -- the false claim that global warming stopped in 1998

by Jeff Masters, Ph.D., wunderblog, Wunderground, March 27, 2013

One often hears the statement in the media that global warming stopped in 1998, or that there has been no global warming for the past 16 years. Why pick 16 years? Why not some nice round number like 20 years? Or better yet, 30 years, since the climate is generally defined as the average weather experienced over a period of 30 years or longer? Temperatures at Earth's surface undergo natural, decades-long warming and cooling trends, related to the La Niña/El Niño cycle and the 11-year sunspot cycle. The reason one often hears the year 1998 used as a base year to measure global temperature trends is that this is a cherry-picked year. An extraordinarily powerful El Niño event that was the strongest on record brought about a temporary increase in surface ocean temperatures over a vast area of the tropical Pacific that year, helping boost global surface temperatures to the highest levels on record (global temperatures were warmer in both 2005 and 2010, but not by much). But in the years from 2005 to 2012, La Niña events have been present for at least a portion of every single year, helping keep Earth's surface relatively cool. Thus, if one draws a straight-line fit of global surface temperatures from 1998 to 2012, a climate trend shows little global warming results. If one picks any year prior to 1998, or almost any year after 1998, a global warming trend does result. The choice of 1998 is a deliberate abuse of statistics in an attempt to manipulate people into drawing a false conclusion on global temperature trends. One of my favorite examples of this manipulation of statistics is shown an animated graph called "The Escalator" created by skepticalscience.com (Figure 1).




Figure 1. Average of NASA's GISS, NOAA's NCDC, and the UK Met Office's HadCRUT4 monthly global surface temperature departures from average, from January 1970 through November 2012 (blue), with linear trends applied to the time frames Jan 1970-Oct 1977, Apr 1977-Dec 1986, Sep 1987-Nov 1996, Jun 1997-Dec 2002, Nov 2002-Nov 2012. Climate change skeptics like to emphasize the shorter term fluctuations in global temperatures (blue lines) and ignore the long-term climate trend (red line). The global surface temperature trend from January 1970 through November 2012 (red line) is +0.16 °C (+0.29 °F) per decade. Image credit: skepticalscience.com.

Correcting for natural causes to find the human contribution to global temperature changes
We know that natural global warming or cooling on time scales of 1-11 years can be caused by changes in the El Niño-Southern Oscillation (ENSO) cycle, dust from volcanic eruptions, and changes in solar energy. For example, a study published in March 2013 in Geophysical Research Letters found that dust in the stratosphere has increased by 4-10% since 2000 due to volcanic eruptions, keeping the level of global warming up to 25% lower than might be expected. So, it is good to remove these natural causes of global temperature change over the past 34 years for which we have satellite data, to see what the human influence might have been during that time span. The three major surface temperature data sets (NCDC, GISS, and HadCRU) all show global temperatures have warmed by 0.16-0.17 °C (0.28-0.30 °F) per decade since satellite measurements began in 1979. The two satellite-based data sets of the lower atmosphere (UAH and RSS) give slightly less warming, about 0.14-0.15 °C (0.25-0.27 °F) per decade (keep in mind that satellite measurements of the lower atmosphere temperature are affected much more strongly by volcanic eruptions and the El Niño phenomena than are surface-based measurements taken by weather stations). A 2011 paper published by Grant Foster and Stefan Rahmstorf ("Global temperature evolution 1979- 2010) took the five major global temperature data sets and adjusted them to remove the influences of natural variations in sunlight, volcanic dust, and the El Niño/La Niña cycle. The researchers found that adjusting for these natural effects did not change the observed trend in global temperatures, which remained between 0.14-0.17 °C (0.25-0.31 °F) per decade in all five data sets. The warmest years since 1979 were 2010 and 2009 in all five adjusted data sets. Since the known natural causes of global warming have little to do with the observed increase in global temperatures over the past 34 years, either human activity or some unknown natural source is responsible for the global warming during that time period. 




Figure 2. Tavurvur volcano in New Guinea erupting on March 7, 2009. According to the 2011 study, "Major influence of tropical volcanic eruptions on the stratospheric aerosol layer during the last decade," an October 7, 2006, eruption of this volcano, in combination with a May 20, 2006, eruption of the Soufrière Hills volcano on Montserrat Island in the Caribbean, hurled a significant amount of sulfur into the stratosphere, helping reduce global temperatures. Image credit: Taro Taylor.


Figure 3. Departure from average of annual global temperatures between 1979 and 2012, adjusted to remove natural variations due to the El Niño-Southern Oscillation (ENSO) cycle, dust from volcanic eruptions, and changes in solar energy. The five most frequently cited global temperature records are presented: surface temperature estimates by NASA's GISS, HadCRU from the UK Met Office, and NOAA's NCDC, and satellite-based lower-atmosphere estimates from Remote Sensing Systems, Inc. (RSS) and the University of Alabama-Huntsville (UAH). Image is an update (via realclimate.org) of one from a 2011 study, Global temperature evolution 1979-2010, by Grant Foster and Stefan Rahmstorf (Environ. Res. Lett., 6 (2011) 044022; doi: 10.1088/1748-9326/6/4/044022).




Figure 4. Average of NASA's GISS, NOAA's NCDC, and the UK Met Office's HadCRUT4 monthly global surface temperature departures from average, from January 1970 through November 2012 (blue), with linear trends applied to the time frames Jan 1970-Oct 1977, Apr 1977-Dec 1986, Sep 1987-Nov 1996, Jun 1997-Dec 2002, Nov 2002-Nov 2012. Climate change skeptics like to emphasize the shorter term fluctuations in global temperatures (blue lines) and ignore the long-term climate trend (red line). The global surface temperature trend from January 1970 through November 2012 (red line) is +0.16 °C (+0.29 °F) per decade. Image credit: skepticalscience.com.


Video 1. An animated description of how correcting for El Niño-Southern Oscillation (ENSO) cycle, dust from volcanic eruptions, and changes in solar energy shows that global warming has continued. Video credit: skepticalscience.com.

Where is the missing heat going? Into the oceans
The preponderance of La Niña events in recent years has caused a large amount of heat from global warming to be transferred to the deep oceans, according to a journal article published earlier this week by Balmaseda et al."Distinctive climate signals in reanalysis of global ocean heat content". The warming at the surface has slowed down in recent years, but the total amount of heat going in the atmosphere, oceans, and surface has continued unabated. The next big El Niño event will be able to liberate some of this stored heat back to the surface, but much of the new deep ocean heat will stay down there for hundreds of years. As far as civilization is concerned, that is a good thing, though the extra heat energy does make ocean waters expand, raising sea levels.


Figure 5. Land, atmosphere, and ice heating (red), 0-700 meter ocean heat content (OHC) increase (light blue), and 700-2,000 meter OHC increase (dark blue).  From Nuccitelli et al. (2012), described at skepticalscience.com.

Links
In October 2012 Carbonbrief.org published a list of six blogs and videos done to debunk the claim that the Earth hasn't warmed since 1998.

Balmaseda et al., 2013, "Distinctive climate signals in reanalysis of global ocean heat content," Geophysical Research Letters, doi: 10.1002/grl.50382.

The 2013 realclimate.org annual adjusted global temperature analysis, for 1979-2012, concludes: "the models are on the low side of some changes, and on the high side of others, but despite short-term ups and downs, global warming continues much as predicted."

To answer frequently cited challenges to climate change science, see the wunderground.com webpage, Top Ten Skeptic Arguments, as debunked by skepticalscience.com.


http://www.wunderground.com/blog/JeffMasters/comment.html?entrynum=2374

No comments: