Separating Climate Fact From Fiction
Testimony of March 13, 2002 by Dr Sallie Baliunas provided to the Senate Committee on Environment and Public Works, chaired by Sen. James M. Jeffords.

Fossil fuels currently provide around 84% of energy consumed in the United States, and roughly 80% of the energy produced worldwide. Those energy resources are key to improving the human condition and the environment.

Human use of fossil fuels has increased the amount of greenhouse gases, in particular, carbon dioxide, in the atmosphere. Carbon dioxide is essential to life on earth. Moreover, the greenhouse effect is important to life on earth in that the greenhouse gases help retain energy near the surface that would otherwise escape to space. Based on ideas about how climate works, the small additional energy resulting from the air's increased carbon dioxide content should warm the planet.

Projections of future energy use, applied to the scientifically most sophisticated computer simulations of climate, have yielded wide-ranging forecasts of future temperature increases from a continued increase of carbon dioxide concentration in the air. These have been compiled by the United Nations' Intergovernmental Panel on Climate Change (IPCC). The middle range forecast of their estimates of future warming, based on expected growth in fossil fuel use without any curbs, is for a 1 degree Celsius increase between now and 2050. A simulation counting in the effect of the as yet unimplemented Kyoto Protocol, negotiated in 1997 and calling for a worldwide 5% cut in carbon dioxide emissions from 1990 levels, would reduce that increase to 0.94 C - an insignificant 0.06 C cut (Figure 1). That means if increased atmospheric concentrations of carbon dioxide are a major problem, then much steeper cuts than those outlined in the Kyoto Protocol are warranted.


Figure 1 -- Forecast of year-to-year temperature rise from years 2000 to 2050 C.E. (thin line) assuming an increase in the air's greenhouse gas concentration from human activities, based on the Hadley Center's model (UKMO HADCM3 IS92A version). The upper line (labeled "Without Kyoto") is the linear trend fit to the model's forecast temperature rise, without implementation of the Kyoto Protocol. The lower line is the estimate of the impact on temperature with the implementation of the Kyoto Protocol. By the year 2050, around 0.06 C global warming is averted by the implementation of the Kyoto Protocol.

One key scientific question is: What has been the response of the climate thus far to the small amount of energy added by humans from increased greenhouse gases in the air? To prove the reliability of their future forecasts, computer simulations need verification by testing past, well-documented temperature fluctuations. New federal investment in technology, especially that of space-based instrumentation, has helped address the issue of observed response of the climate to the air's increased greenhouse gas concentration. Two capitol tests of the reliability of the computer simulations are the past decades of surface temperature and lower troposphere change.

Record of surface temperature

In the 20th Century the global average surface temperature (Figure 2) rose about 0.5 C, after a five hundred year cool period called the Little Ice Age. The uncharacteristic cold had followed a widespread warm interval, called the Medieval Warm Period (ca. 800 - 1200 C.E.). The 20th Century warming trend may have a human component attributable to fossil fuel use, which increased sharply in the 20th Century. But a closer look at the 20th century temperature shows three distinct trends:


Figure 2 -- Surface temperature changes sampled worldwide and analyzed by Cambridge Research Unit (CRU) and NASA-Goddard Institute of Space Studies (GISS). The pattern of 20th century temperature change has three distinct phases: an early 20th-century warming, a mid-century cooling, and a late 20th-century warming.

First, a strong warming trend of about 0.5 C began in the late 19th century and peaked around 1940. Next, the temperature decreased from 1940 until the late 1970s. Recently, a third trend has emerged -- a modest warming from the late 1970s to the present.

Because about 80% of the carbon dioxide from human activities was added to the air after 1940, the early 20th Century warming trend had to be largely natural. Human effects from increased concentrations of greenhouse gases amount to at most 0.1 C per decade - the maximum amount of the surface warming trend seen since the late 1970s. This surface warming would suggest a temperature trend of about 1 C per century, which is less than that predicted by the computer simulations of the air's increased human-made greenhouse gas content. Accumulated over a century, civilization will readily adapt to such a modest warming trend. However, the recent trend in surface warming may not be primarily attributable to human-made greenhouse gases.

Record of lower troposphere temperature

Computer simulations of climate in which the air's greenhouse gas concentrations increase owing to human activities predict detectable warming not only near the surface but also in the layer of air above the surface, the lower troposphere, which rises in altitude from roughly two to eight kilometers. Records from NASA's Microwave Sounder Units aboard satellites extend back 21 years and cover most of the globe (Figure 3). The satellite-derived record is validated independently by measurements from NOAA balloon radiosonde instruments, and those records extend back over 40 years (Figure 4). Those records show that the temperature of the lower troposphere does vary, e.g., the strong El Niño warming pulse of 1997-98 is obvious. However, no meaningful human warming trend, as forecast by the computer simulations, can be found.



Figure 3 -- Monthly averaged temperatures sampled nearly globally for the lower troposphere (roughly 5,000 to 28,000 feet altitude) from Microwave Sounder Unit (MSU) instruments onboard NASA satellites. The large spike of warmth resulted from the temporary natural warming of the Pacific Ocean by the 1997 - 1998 El Niño event. The linear trend is +0.04 C per decade
(data are from http://wwwghcc.msfc.nasa.gov/temperature/)



Figure 4 -- The seasonal average temperature anomaly sampled worldwide for the lower troposphere as measured by radiosonde instruments carried aboard balloons. Although a linear trend of +0.09C per decade is present if fitted across the entire period of the record, that trend is affected by the presence of the abrupt warming that occurred in 1976-1977, owing to the action of the Pacific Decadal Oscillation (PDO). The trends before and after the 1976-1977 Great Pacific Climate Shift indicate no evidence of a significant human-made warming trend
(source of data http://cdiac.esd.ornl.gov/ftp/trends/temp/angell/glob.dat)

The radiosonde record from balloons confirms the results of the satellites. Although the radiosonde record lacks the dense spatial coverage from satellites, the radiosonde record extends back to 1957, a period that includes the recent rapid rise in the air's carbon dioxide concentration. The balloon record shows no warming trend in global average temperature prior to the dramatic shift in 1976-77. That warming, known as the Great Pacific Climate Shift of 1976 - 1977, is not attributable to human causes but is a natural, shift in the Pacific that occurs every 20 to 30 years, and can affect global average temperatures.

When compared to the observed response of the climate system, the computer simulations all have forecast warming trends much steeper over the last several decades than measured. The forecasts exaggerate to some degree the warming at the surface, and profoundly in the lower troposphere.

The complexity of the computer simulations of climate is one reason the forecasts are unreliable. The simulations must track over 5 million parameters. To simulate climate change for a period of several decades is a computational task that requires 10,000,000,000,000,000,000 degrees of freedom. To improve the forecasts, much better information is required, including accurate understanding of the two major, natural greenhouse gas effects - water vapor and clouds.

Natural climate variability: The sun's influence

Given the lack of an observed warming trend in the lower troposphere, the result is that most of the surface warming in recent decades cannot owe to a human-caused enhanced greenhouse effect. What might cause the surface warming, especially in the early 20th century when greenhouse gases from human activities had not significantly increased in concentration in the atmosphere? The 20th Century temperature pattern shows a strong correlation to energy output of the sun (Figure 5). Although the causes of the changing sun's particle, magnetic and energy outputs are uncertain, as are the responses of the climate to the Sun's various changes, the correlation is pronounced. It explains especially well the early 20th Century warming trend, which cannot have much human contribution.


Figure 5 -- Changes in the sun's magnetism (as evidenced by the changing length of the 22-year, or Hale Polarity Cycle, dotted line) and changes in Northern Hemisphere land temperature (solid line) are closely correlated. The sun's shorter magnetic cycles are more intense, suggesting periods of a brighter sun, then a fainter sun during longer cycles. Lags or leads between the two curves that are shorter than twenty years are not significant, owing to the 22-year time frame of the proxy for brightness change. The record of reconstructed Northern Hemisphere land temperature substitutes for global temperature, which is unavailable back to 1700 (S. Baliunas and W. Soon, 1995, Astrophysical Journal, 450, 896).

Based on the key temperature measurements of the last several decades, the actual response of the climate to the increased concentration of carbon dioxide and other human-made greenhouse gases content in the air has shown no significant man-made global warming trend. The magnitude of expected human change is especially constrained by the observed temperature trends of the lower troposphere.

This means that the human global warming effect, if present, is small and slow to develop. That creates a window of time and opportunity to continue and improve observations and computer simulations of climate to better define the magnitude of human-made warming. Proposals like the Kyoto agreement to sharply cut greenhouse gas emissions are estimated in most economic studies to have enormous economic, social and environmental costs. The cost estimates for the U.S. alone amount to $100 billion to $400 billion per year. Those costs would fall disproportionately on America's and the world's elderly and poor.


Sallie Baliunas, Ph.D. is an Astrophysicist at the Harvard-Smithsonian Center for Astrophysics and Deputy Director of Mount Wilson Observatory. Dr. Baliunas serves as Senior Scientist at the George C. Marshall Institute in Washington, DC, and chairs the Institute's Science Advisory Board. She is also Visiting Professor at Brigham Young University, Adjunct Professor at Tennessee State University and past contributing editor to the World Climate Report. Her awards include the Newton-Lacy-Pierce Prize of the American Astronomical Society, the Petr Beckmann Award for Scientific Freedom and the Bok Prize from Harvard University. She has written over 200 scientific research articles. She received her M.A. (1975) and Ph.D. (1980) degrees in Astrophysics from Harvard University.

The remarks herein are personal views and imply no institutional endorsement by any of her affiliations.