Global Temperature Report: December 1996
Vol. 6, No. 7
For Additional Information:
Dr. John Christy, UAH, (205) 922-5763
[email protected]
Dr. Roy Spencer, NASA, (205) 922-5960
[email protected]
Global temperature variation, in Celsius; trend since 1979, -0.035
degrees C per decade
DECEMBER TEMPERATURES
Global composite temp.: -0.11 degrees C (about 0.20 degrees Fahrenheit)
below 10-year average for December.
Northern Hemisphere: +0.01 degrees C (about 0.02 degrees Fahrenheit)
above 10-year average for December.
Southern Hemisphere: -0.23 degrees C (about 0.41 degrees Fahrenheit)
below 10-year average for December.
October Temperatures (revised):
Global Composite: -0.02 degrees C below 10-year average
Northern Hemisphere: +0.26 degrees C above 10-year average
Southern Hemisphere: -0.30 degrees C below 10-year average
(All temperature variations are based on a 10-year average (1982-1991)
for the month reported.)
Notes on data released January 8, 1997:
A slightly cooler than normal December ended a slightly cooler than
normal year, according to Dr. John Christy, an associate professor of
atmospheric science in the Earth System Science Laboratory at The
University of Alabama in Huntsville.
While temperatures as much as three degrees Celsius (5.4 degrees
Fahrenheit) below seasonal norms prevailed over an area stretching from
central Alaska southward through Washington and as far east as Minnesota,
the composite temperature for the Northern Hemisphere was balanced by an
adjacent region of warm air.
Temperatures that were warmer than seasonal norms were recorded over most
of Eastern Canada and the U.S. from Ohio to Chesapeake Bay. The region of
warmer than normal air extended eastward across the Atlantic almost to
Scotland.
The warmest anomalies in that region -- as much as four degrees Celsius
(7.2 degrees Fahrenheit) above seasonal norms -- were recorded over
Newfoundland, Baffin Island and the Baffin Sea, just west of Greenland.
The year-end average global temperature for 1996 is the seventh coolest
(12th warmest) in the 18 years since NOAA satellites started gathering
temperature data, Christy said. The composite global temperature for 1996
was 0.09 degrees Celsius (0.16 degrees Fahrenheit) below the annual norm.
Global temperatures throughout 1996 were influenced by a large RpoolS of
slightly cooler than normal water that formed in the Eastern Pacific
Ocean, Christy said.
In the 18-year record, there have been nine years with warmer than normal
composite temperatures and nine that were cooler than the ten-year
(1982-1991) norm.
As part of an ongoing NASA/UAH joint project, Christy and Dr. Roy
Spencer, a space scientist in the Earth Science Laboratory of the Global
Hydrology and Climate Center in NASAUs Marshall Space Flight Center, use
data gathered by microwave sounding units on NOAAUs TIROS-N satellites to
get accurate temperature readings for almost all regions of the Earth.
Once the monthly temperature data is collected and processed, it is
placed in a "public" computer file for immediate access by atmospheric
scientists in the U.S. and abroad.
SPECIAL ADVISORY
The information published in the December 1996 edition of the Global
Temperature Report reflects changes made after a recent analysis of
temperature data collected by NOAA satellites since 1979.
We have retested the data for the past 18 years to apply more stringent
standards for accuracy. Greater accuracy was achieved by reviewing the
comparisons of data collected by different satellites.
When a new satellite goes up, its instruments won't be precisely
calibrated with the older satellites. It may be a little warmer or a
little cooler than the old satellite. To remove any bias or errors, we
compare data from each satellite.
Our concern with the entire merging question relates to the need to have
confidence in the long-term stability of the time series.
The new analysis sets a higher standard for deciding which data are good
enough to use for that intercomparison.
Previously, we had used data from every day of common observations
between two satellites to calculate the bias. We found that there were
several days in which the data were in poor agreement, due mainly to the
fact that each satellite sampled slightly different geographical "swaths"
over the planet. The impact of high terrain, for instance, might be
observed by one satellite, but not the other.
We have identified these days of poor common coverage and did not use
them to compute the intersatellite biases.
To gauge the accuracy of the revised data, we compared the satellite
database to an "industry standard" -- temperatures recorded by
thermometers carried aloft daily by balloons.
While the unrevised data had a excellent correlation to the data gathered
by balloons (94 percent), the revised satellite data had a better
correlation (97 percent).
The revised data show 1984 to be slightly cooler than previously
calculated, causing the global temperature trend for the 18-year period
to be slightly warmer than before.
For the 18 years, however, there is no evidence of global warming in the
revised data. The temperature trend for that time is a cooling of 0.035
degrees Celsius (about 0.06 degrees Fahrenheit) per decade.
-- Dr. John Christy
# # #
RSS