| University Relations |
| UAH News Home |
| UAH in the News |
| UAH Background |
| UAH At-A-Glance |
| Expert Sources |
| Calendar |
| Campus Map |
| UAH Brief History |
| Contact Us |
1.800.UAH.CALL |
Taking Earth's temperature
In early November 1978 a microwave sensor aboard the National Oceanic and Atmospheric Administration's TIROS polar-orbiting satellite started scanning the Earth's atmosphere.It was looking at the intensity of microwaves emitted by oxygen molecules. Since the intensity of those microwaves relates directly to the temperature of the oxygen molecules, they provide an accurate "thermometer" for temperatures in the atmosphere. The National Weather Service hoped to use that nearly global temperature data to improve its weather forecasts.
Unfortunately, the computer forecasting models were designed to use precise temperature data from 14 designated altitudes. That data is collected by "radiosondes," instruments carried aloft by helium balloons.
The microwave sensors "see" huge volumes of atmosphere — about 50,000 cubic kilometers for each reading.
The data didn't fit the forecast models and, as day-to-day tools for forecasters, the microwave sensors weren't very useful. (The data has since been used to substantially improve the accuracy of weather forecasting models.)
The data the sensors collected, however, were dutifully recorded and stored. In a decade they created a tremendous backlog of data, including more than 500 million temperature readings.
In 1989 Dr. Roy Spencer, at that time a space scientist at NASA's Marshall Space Flight Center, proposed using the microwave data to look at global atmospheric temperatures.
An expert in microwave sensors, Spencer teamed with Dr. John Christy from The University of Alabama in Huntsville (UAH) to analyze the data. They concentrated on data from two altitude ranges: The lower troposphere, from sea level to about six miles high, and the lower stratosphere above 10 miles. In March 1990 they published their findings in Science: Although global climate models predicted global warming due to increased CO2 in the atmosphere should show up first and strongest in the troposphere, the first ten years of satellite data showed no sign of warming in that layer of the atmosphere.
Through the years, that result has changed.
The atmosphere is warming
From Nov. 16, 1978, through June 30, 2007, the global lower troposphere has warmed about 0.4 Celsius (about 0.72° Fahrenheit), or global warming at the rate of approximately 1.4 C (about 2.52° Fahrenheit) per century.Most of the warming that accounts for that trend, however, has happened since January 1998 in the northernmost third of the globe.
There has been little or no net warming in the tropics over the past 28+ years, while there is very slight warming in the southernmost third of the globe.
While this warming is within the range of natural climate variation, some of the warming is consistent with human effects — especially warming in the coldest air over the Northern Hemisphere, according to Christy. "That cold air has very little water vapor in it, so if you add another greenhouse gas you have an opportunity to trap more heat.
"When you go to the tropics, where there's lots of water vapor, the extra carbon dioxide doesn't have as much effect. As a greenhouse gas, carbon dioxide's greatest effect is in the driest, coldest places."
Global composite temperatures were driven by major climate events, including volcanic eruptions, El Niño Pacific Ocean warming events and La Niña Pacific Ocean cooling events.
Compared to seasonal norms, the hottest day in the past 25 years was April 6, 1998, when the global composite temperature climbed to 0.92 C (1.66° F) above normal.
April 1998 was also the warmest month, with an average global composite temperature that was 0.75 C (1.35° F) warmer than seasonal norms.
The 1997-1998 "El Niño of the century" made 1998 the hottest calendar year during the 28+ year record, with an annual average temperature that was 0.47 C (0.85° F) warmer than normal.
The hottest 12-month period, however, was from November 1997 through October 1998, with an average global composite temperature that was 0.473 C warmer than normal.
By contrast, the coolest 12-month period was from June 1992 through May 1993, when the eruption of the Mount Pinatubo volcano drove the average global composite temperature 0.28 C (0.5° F) below normal. The volcano erupted at the beginning of an El Niño warming event, which helped to offset the volcano's cooling effects.
The coolest calendar year was 1985, at 0.25 C (0.45° F) below normal.
The coolest month was September 1984, at 0.50 C (0.9° F) below normal.
The coldest day was September 19, 1984, when the global composite temperature dropped 0.67 C (1.21° F) below seasonal norms.
Confirming the temperature data
The UAH dataset is the only satellite-based temperature dataset that has multiple, independent studies verifying its accuracy."Ours is the only dataset that has been compared to non-satellite data," said Christy. "This gives us confidence in its results. Several different radiosonde-based products have been compared to the satellite data and the results of those studies have been published."
In 1992, Christy and Spencer published a study in which they compared the satellite data to a set of U.S. radiosondes.
In 1997, the Hadley Center of the United Kingdom's Meteorology Office did an analysis using data from 400 radiosonde sites around the world. There was extremely close agreement between that radiosonde data and the UAH dataset.
Additional studies comparing the satellite and radiosonde data have appeared in reports published by the IPCC and the National Research Council.
Another comparison was published in 2003 in the Journal of Atmospheric and Oceanic Technology. In each case, the satellite data and the radiosonde data show a high level of agreement.
Each microwave sounding unit is also calibrated before launch, using heating elements warmed to precise temperatures.
In space, each microwave sounding unit self-calibrates every cycle. It views a warm target whose temperature is precisely monitored and then deep space, which has a temperature near absolute zero. Then it takes measurements of the Earth's atmosphere across a swath below the spacecraft.
Finding and correcting errors
As the satellite data become longer in extent, various issues became apparent that needed to be dealt with to ensure the data's long-term accuracy.Spencer and Christy discovered three of the four major problems that have been identified — orbital drift, instrument body warming and inter-instrument calibration — found solutions to those problems and published their results in peer-reviewed journals.
The fourth problem, orbital decay, was identified by Dr. Frank Wentz, et al., and a correction technique similar to one that he and his colleagues developed has been applied to the UAH dataset.
Orbital drift (or precession)
The orbital drift of NOAA's TIROS satellites causes two problems with the temperature data. A spacecraft launched to observe at 2 p.m. and 2 a.m. local time will drift to later local times through its operational lifetime, say to 5 p.m. and 5 a.m.
Because a typical location on Earth naturally cools between 2 (p.m. or a.m.) and 5 (p.m. or a.m.), a satellite observing this cooling over its lifetime would record a spurious long-term cooling trend.
To remove this diurnal drift effect UAH uses measurements taken by the satellites. Every 25 seconds the scanner sweeps west to east over a 2,000 km strip, observing the earth at different local times.
Using all of these measurements we are able to calculate an atmospheric temperature value vs. local time to use as a correction (see Christy et al. 2003 for details.)
By using these observations the subtle features of the diurnal cycle, such as stratospheric tides and latent heating cycles over oceans, are accounted for.
Separate adjustment values are found for land vs. ocean, each month of the year and each latitude band. This conservative approach is empirical in nature, relying on observed data.
Instrument body warming
Orbital drift also changes the angle at which sunlight strikes the instrument during its pass over the daylight side of the planet. Parts of the instrument differentially heat or cool due to changing shadows as the satellite's orbit drifts. The value of the measured atmospheric temperature tends to show a small spurious heating or cooling proportional to the temperature of the instrument components.
UAH discovered this effect (Christy et al. 2000) and developed a technique to remove it. A coefficient is calculated for each satellite which — when multiplied by the change in instrument temperature — determines the erroneous atmospheric temperature effect. This error can then be removed. UAH does this only when there is an obvious correlation between instrument temperatures and atmospheric temperature errors.
UAH also smoothes the data before calculating these coefficients. This produces higher correlations between instrument temperature and the component of erroneous atmospheric temperature. This gives higher confidence in the accuracy of the correction.
Orbital decay
Satellites lose altitude over time as their orbits decay. Wentz, et al, found that this loss of altitude changes the "footprint" seen by the microwave sound unit, introducing a false cooling signal into the data.
UAH developed a method for removing this false signal, which has been applied to the dataset.
Inter-instrument calibration
If a new satellite with a new microwave sensor is launched while one or more old TIROS satellites are still in orbit, that overlap gives Christy and Spencer an opportunity to check one instrument against the other.
This provides an additional layer of verification, while also helping identify potential instrument problems. Spencer and Christy use daily temperature readings from each satellite, corrected for known false signal effects, to intercalibrate the satellite instruments and produce a more homogeneous and accurate long-term temperature record.
Conflicting climate data
A common feature of climate model forecasts is that as carbon dioxide increases, the global surface temperature should rise along with an even more rapid warming in the troposphere — the atmosphere up to about 30,000 feet. This additional atmospheric warming would further promote warming at the surface — if models are correct.Surface temperature records indicate a long-term atmospheric warming trend of about 3° Fahrenheit per century.
Other research, however, finds the signs of major global warming more difficult to identify.
Long-term studies of El Niño Pacific Ocean warming events show that they are no more frequent now than in the past millennia. (1)(2)
Recent studies show sea conditions in the Arctic today are similar to conditions in the late 19th and early 20th centuries, while average Arctic temperatures are rising almost to their levels of the 1930s. (3)(4)(5)
Climate studies in the Antarctic report long-term cooling trends on scales ranging from 30 to more than 1,000 years, and that the ice cap there is growing.(6)(7)(8)
Studies of severe weather events in North America found no evidence that extreme weather events, including tornadoes, are more common or more violent now than they were in the late 1800s. (9)(10)
An analysis of hurricane and tropical cyclone data found those storms are not becoming either more frequent or more violent. (11)(12)
A study of mean global sea level found the approximately 3 mm/y rise of the past 150 years has not accelerated during the 20th century. (13)
And a Harvard-Smithsonian study (14) of more than 240 paleoclimate research papers published in the past four decades concluded that the 20th century was neither the warmest century nor the century with the most extreme weather of the past 1,000 years for specific regions.
The ongoing climate conundrum
One of the hottest controversies in climate science is the apparent disagreement between temperature data collected by thermometers at the surface and the satellite dataset."Global" surface thermometer networks show a warming trend of approximately 1.7 degrees Celsius per century — about 3° Fahrenheit.
The satellite data show a warming trend of 1.4 C or about 2.52° F per century.
Why do the two datasets disagree and which is the more accurate representation of the climate, especially since general circulation climate models used to predict global warming agree that anthropogenic warming should be seen first and strongest not at the surface but in the lower troposphere?
A recent analysis of the surface and satellite datasets hints that the apparent disagreement might have as much to do with coverage as with differing trends at different altitudes.
"In areas where you have high resolution, well- maintained scientific collection of temperature data, the satellites and the surface data show a high degree of agreement," said Christy. "Over North America, Europe, Russia, China and Australia, the agreement is basically one-to-one."
The satellite dataset provides tropospheric temperature information for more than 95 percent of the globe, excluding only the small portions of the poles not seen by the satellite instruments and those places where the troposphere is full of mountains or high plateaus.
The satellite-based microwave sounding units provide temperature data for many regions for which reliable climate data are not otherwise available, including many remote desert, jungle, ocean or mountain areas.
By comparison, surface temperature datasets have reliable data for much less of the globe. One widely cited dataset achieves significant global coverage only when each thermometer is assumed to provide temperature data within a 1,200 kilometer radius — equivalent to using a thermometer in Topeka, Kansas, to record temperatures from Brownsville, Texas, to Grand Forks, North Dakota.
The greatest disagreement between the surface and satellite datasets is in the tropics, which includes regions where weather stations are sparse (including central Africa and South America), and the three- fourths of the tropics that are covered by oceans, where proxy information such as sea surface temperatures has been used in lieu of actual atmospheric temperature data.
The value of sea surface temperatures as a proxy for air temperatures is in question after recent research found that sea surface temperatures and air temperatures as little as ten feet above the surface do not move in precise accord with one another.
There are other questions about the sea surface data. For decades, the majority of "sea surface" temperatures recorded were actually the temperatures of ocean water pumped onboard seagoing ships from as much as 30 feet below the surface.
In recent years much of the sea surface temperature data gathering has been done by a network of scientific buoys, which record temperatures at several depths, including the actual sea surface.
In most of the major surface temperature records, however, those two datasets were merged with little or no adjustment for the change in collection method or depth.
The spurious effects of these inconsistencies are thought to be small. The best conclusion may be that the surface and tropospheric temperatures are just going in slightly different directions.
Earth's climate is constantly changing
There is no scientific evidence to support the belief that Earth's climate is stable and will not change if human activity does not intervene.To the contrary, paleoclimate data indicates that Earth's climate is constantly changing and has never been stable.
Glaciers and polar ice caps have been melting for much of the past 20,000 years as global average temperatures climbed approximately 8 C (more than 14° Fahrenheit). There is no scientific reason to believe this process will end in the next few decades.
Because glaciers and ice caps were melting, sea surface levels around the world rose at the rate of about one meter per century for about the first 6,000 of those 20,000 years.
Sea level changes at those rates are not expected in the near term. (Sea levels rose about six inches in the past century.)
"To match the last warm period of about 130,000 years ago, the sea needs to rise about six meters — almost 20 more feet," said Christy. "Until the next ice age, we should expect more land ice to melt and sea levels to rise."
The fossil and geological records indicate that some of these changes will happen rapidly, with major regional climate shifts occurring in time scales as brief as years or decades.
The 'Little Ice Age' that marked the end of the Medieval Warm Period and for several centuries brought a return of ice and snow to much of the world has ended.
Evidence from several sources indicates that Earth's rising temperatures are approaching the high levels of the Holocene Maximum more than 5,000 years ago.
While the approximately 0.14 C per decade of global warming seen in
the satellite data is minor compared to the scale of some past climate shifts, it reminds us that the natural processes of climate change have not stopped.
Looking at the history of paleoclimate data indicates that the climate is capable of significant changes for reasons that are not understood.
If Earth's climate follows the chaotic pattern of previous eons, our descendents will see climate changes that bring increased rain to some areas and decades-long droughts to others; cooling in Earth's climate, the preponderance of scientific evidence indicates that the climate will almost certainly change in the future.
The current level of knowledge about the climate doesn't provide the tools needed to predict when rapid natural climate changes will occur and what forms it might take. This makes it impossible to say with high confidence how much human factors might influence climate change.
Although both the forces that drive natural climate change and the events that trigger major climate events are too poorly understood to allow confident climate forecasts, the atmospheric science community is making major strides.
In the future, it may be possible for scientists to improve the reliability of climate predictions while at the same time accurately and comprehensively monitoring Earth's climate, and providing information about climate change as it occurs.
Climate and environmental priorities
If artificially-enhanced global warming at potentially damaging levels isn't happening, what might that mean in terms of environmental and conservation priorities?
At hearings before Congressional committees and in other settings, we have often been asked the hypothetical question: "If you were in charge, what would you do about climate change and the environment?"
The first thing is to do no harm. With the threat of catastrophic climate change, many proposals have been put forward to limit energy use.
A fundamental point that needs to be understood is that if any of these proposals (including the Kyoto protocol) are implemented, they will have an effect on the climate so small that it cannot be detected.
None of these proposals will change what the climate is doing enough to notice.
Those are good reasons not to artificially force energy prices up. While raising energy costs might damage the economy, it would disproportionately hurt the poor, especially those people living on the world's social and economic fringes.
While no direct evidence of ecological damage from carbon dioxide has been found, that is no excuse for reducing environmental protection.
We shouldn't undo the good things that have been done to clean the air and water. More should be done, especially in developing countries.
Beyond quality of life issues, human life itself is significantly more threatened by polluted water, polluted air, habitat destruction, unbridled population growth and a host of related ecological problems than it is by global climate change on the scales that we have seen in the past 28 years.
Millions of children around the world die every year due to water borne diseases. Tens of millions of people are forced to breathe air that is blackened and made toxic by fumes from leaded gasoline, industrial pollution and cooking fires.
Women and girls in some developing countries are forced to walk miles each day from their villages to the receding edges of the forests to harvest green wood and other low-energy biomass for the fires they use to cook their meals and heat their homes.
A U.N. report estimated that 1.6 million people — most of them women and children — die each year due to indoor pollution from cooking fires.
While the extent of human impacts on global climate change remains uncertain, research by our colleagues at UAH confirms that deforestation and land conversion are changing regional weather patterns and the local climate over some parts of the world.
We should also do what the U.S. does best: We should encourage and support the scientists and engineers who will develop new sources of low-cost energy. Just as transportation was "de-horsified" in the last century, we believe energy in the 21st century will continue to be de-carbonized.
Ironically, actions that artificially inflate the cost of energy might hamper those efforts, as healthy economies can better afford to find and develop alternative energy sources and cleaner energy technologies.
We should also enhance the national and international infrastructure for dealing with climate and weather events, including droughts, floods, hurricanes and tornadoes. We know these events will continue to happen whether the climate changes or not. Everyone would benefit if we were better prepared when they happen.
Finally, we recognize that climate change is real and that human activities are probably contributing to that change. We should continue to devote resources to monitoring and studying the climate system, so we can develop the systems that will let us know what the climate is doing and respond appropriately. Perhaps, at some point in the future, we might even be able to reliably forecast what the climate will do in future generations.
— Dr. John R. Christy & Dr. Roy Spencer
Earth System Science Center
The University of Alabama in Huntsville
Biographical Information
Dr. John Christy
Dr. John R. Christy is professor of atmospheric science and director of the Earth System Science Center at the University of Alabama in Huntsville, where he began studying global climate issues in 1987.
In 1989 Dr. Roy W. Spencer (then a space scientist at NASA's Marshall Space Flight Center and now a principle research scientist at UAH) and Christy developed a global temperature dataset from microwave data observed from satellites beginning in November 1978.
For this achievement, Spencer and Christy were awarded NASA's Medal for Exceptional Scientific Achievement in 1991. In 1996 they received a special award by the American Meteorological Society "for developing a global, precise record of earth's temperature from operational polar-orbiting satellites, fundamentally advancing our ability to monitor climate." In January 2002 Christy was inducted as a fellow of the American Meteorological Society.
Christy has served as a contributor (1992, 1994 and 1996) and lead author (2001) for the U.N. reports by the Intergovernmental Panel on Climate Change, in which the satellite temperatures were included as a high-quality data set for studying global climate change. He has or is serving on five National Research Council panels or committees, has performed research funded by NASA, NOAA, DOE, DOT and the State of Alabama and has published studies appearing in Science, Nature, the Journal of Climate and The Journal of Geophysical Research. He has testified before several congressional committees.
In November 2000 Gov. Don Siegelman appointed him to be Alabama's state climatologist.
Christy received M.S. and Ph.D. degrees in atmospheric sciences from the University of Illinois (1984, 1987). Prior to this career path he had graduated from the California State University in Fresno (B.A. Mathematics, 1973) and taught physics and chemistry for two years as a missionary teacher in Nyeri, Kenya.
After earning a Master of Divinity degree from Golden Gate Baptist Seminary (1978) he served four years as a bivocational mission-pastor in Vermillion, South Dakota, where he also taught college math.
Christy is married to the former Babs Joslin, a fellow missionary he met in Kenya. They have two children.
Christy's favorite hobby is panning for gold, a skill he learned as a teenager in California. He also runs races from two miles to extra-marathon length.
Dr. Roy Spencer
Dr. Roy Spencer is a principal research scientist in the Earth System Science Center at The University of Alabama in Huntsville.
He is the U.S. science team leader for the Advanced Microwave Scanning Radiometer flying on NASA's Aqua satellite.
He was a space scientist at NASA's Marshall Space Flight Center in 1989, when he teamed with Dr. John Christy to develop a system to measure global climate trends using microwave sensors aboard NOAA satellites.
Spencer and Christy were awarded NASA's Medal for Exceptional Scientific Achievement in 1991. In 1996 they received a special award by the American Meteorological Society "for developing a global, precise record of earth's temperature from operational polar-orbiting satellites, fundamentally advancing our ability to monitor climate."
At NASA Spencer directed research into developing and applying satellite microwave remote sensing systems for measuring global temperatures, water vapor and precipitation. In 1997 he was named senior scientist for climate studies at NASA/MSFC.
He earned M.S. and Ph.D. degrees in meteorology (1979, 1981) from the University of Wisconsin, Madison, and was an assistant scientist in the Space Science and Engineering Center in Madison, WI, for more than a year. He was a Universities' Space Research Association visiting scientist at NASA/MSFC from 1984 to 1987.
Spencer has testified in both the U.S. House of Representatives and the U.S. Senate on the subject of global warming.
Selected research publications
• Christy, J.R. and R.W. Spencer, "Reliability of satellite data sets." Science, 301, 1046-1047, 2003.
• Christy, J.R., 2003: "Climate Research: Response to 'Wanted Scientific Leadership on Climate' by Pielke, Jr. and Sarewitz." Issues in Science and Technology, Spring 2003, 9-10.
• Christy, J.R., R.W. Spencer, W.B. Norris, W.D. Braswell and D.E. Parker, "Error estimates of Version 5.0 of MSU/AMSU bulk atmospheric temperatures." Journal of Atmospheric and Oceanic Technology, 20, 613-629, 2003.
• Christy, J.R., "When was the hottest summer? A state climatologist struggles for an answer." Bulletin of the American Meteorological Society, 83, 723-734, 2002.
• Christy, J.R., D.E. Parker, S.J. Brown, I. Macadam, M. Stendel and W.B. Norris, "Differential trends in tropical sea surface and atmospheric temperatures." Geophysical Research Letters, 28, 183-186, 2001
• Spencer, R.W., and W.D. Braswell, "Atlantic tropical cyclone monitoring with AMSU-A: Estimation of maximum sustained wind speeds." Monthly Weather Review, 129, 1518-1532, 2001.
• Robertson, R.W., R.W. Spencer and D.E. Fitzjarrald, "A new satellite deep convective ice index for tropical climate monitoring: Possible implications for existing oceanic precipitation datasets." Geophysical Research Letters, 28-2, 251-254, 2001.
• Winter, A., O. Tadamichi, H. Ishioroshi, T. Watanabe and J. R. Christy, "A two-to-three degree cooling of Caribbean sea surface temperatures during the Little Ice Age." Geophysical Research Letters, 27, 3365-3368, 2000
• Hurrell, J., S.J. Brown, K.E. Trenberth and J.R. Christy, "Comparison of tropospheric temperatures from radiosondes and satellites: 1979-1998." Bulletin of the American Meteorological Society, 81, 2165-2177, 2000.
• Imaoka, K., and R.W. Spencer, "Diurnal variation of precipitation over the tropical oceans observed with TRMM/TMI combined with SSM/I." Journal of Climate, 13, 4149-4158, 2000.
• Gaffen, D.J., B.D. Santer, J.S. Boyle, J.R. Christy, N.E. Graham, R.J. Ross, "Multidecadal changes in the vertical structure of the tropical troposphere." Science, 287, 1242-1245, 2000.
• Christy, J.R., "Global Climate Change: Scientific and Social Impacts." Bridges. 7, 39-57, 2000.
• Christy, J.R., R.W. Spencer, and W.D. Braswell, "MSU Tropospheric temperatures: Data set construction and radiosonde comparisons." Journal of Atmospheric and Oceanic Technology. 17,1153-1170, 2000.
• Stendel, M., J.R. Christy and L. Bengtsson, "Assessing levels of uncertainty in recent temperature time series." Climate Dynamics, 16(8), 587-601, 2000.
• Christy, J.R., R.W. Spencer, and E. Lobl, "Analysis of the merging procedure for the MSU daily temperature time series." Journal of Climate, 11, 2016-2041, 1998.
• Spencer, R.W., F.J. Fontaine, T. DeFelice and F.J. Wentz, "Tropical oceanic precipitation changes after the 1991 Pinatubo eruption." Journal of Atmospheric Science, 55, 1707-1713, 1998.
• Christy, J.R., R.W. Spencer, and W.D. Braswell, "How accurate are satellite 'thermometers?'" Nature, 389, 342-3, 1997.
• Spencer, R.W., and W.D. Braswell, "How dry is the tropical troposphere? Implications for global warming theory." Bulletin of the American Meteorological Society, 78, 1097-1106, 1997.
• Christy, J.R., R.W. Spencer and R.T. McNider, "Reducing noise in the MSU daily lower tropospheric global temperature data set." Journal of Climate, 8, 888-896, 1995.
• Christy, J.R. and J. Goodridge, "Precision global temperatures from satellites and urban warming effects of non-satellite data." Atmospheric Environment, 29, 1957-1961, 1995.
• Christy, J.R. and R. T. McNider, "Satellite greenhouse signal." Nature, 367, 325, 1994.
• Christy, J.R. and S.J. Drouilhet, "Variability in daily, zonal mean lower-stratospheric temperatures." Journal of Climate, 7, 106-120, 1994.
• Spencer, R.W., and J.R. Christy, "Precise monitoring of global temperature trends from satellites." Science, 247, 1558-1562, 1990.