NOAA: Global temperature ties for warmest on record

Oct 18, 2010
Temperature Anomolies, September 2010 Credit: NOAA

The first nine months of 2010 tied with the same period in 1998 for the warmest combined land and ocean surface temperature on record. The global average land surface temperature for January-September was the second warmest on record, behind 2007. The global ocean surface temperature for January–September was also the second warmest on record, behind 1998.

The monthly analysis from NOAA’s National Climatic Data Center, which is based on records going back to 1880, is part of the suite of climate services NOAA provides government, business and community leaders, so they can make informed decisions.

Global Temperature Highlights

• For the year-to-date, the global combined land and ocean surface temperature of 58.67 F (14.75 C) tied with 1998 as the warmest January-September period on record. This value is 1.17 F (0.65 C) above the 20th century average.

• The combined global land and ocean average surface temperature for September 2010 tied with 1998 as the eighth warmest on record at 59.9 F (15.5 C), which is 0.90 F (0.50 C) above the 20th century average of 59.0 F (15.0 C).

• Separately, the September global land surface temperature was 1.19 F (0.66 C) above the 20th century average of 53.6 F (12.0 C) — the ninth warmest September on record. Warmer-than-average conditions dominated the world’s land areas. The most prominent warmth was in western Alaska, most of the contiguous United States, eastern Canada, Greenland, the Middle East, eastern and central Europe, western and far eastern Russia and northeastern Asia. Cooler-than-average regions included much of Australia, western Canada, parts of the northern United States, parts of western and central Europe, and central Russia.

• According to NOAA’s National Weather Service, Los Angeles set a new all-time maximum temperature on Sept. 27 when temperatures soared to 113 F (45 C), surpassing the previous record of 112 F (44.4 C) set in June 1990.

• According to Australia’s Bureau of Meteorology, the country had its coolest maximum temperatures since 1984. The Northern Territory had its coolest September since 1984, and Western Australia and Victoria each recorded their lowest maximum temperatures since 1992. South Australia had its second-lowest maximum temperatures on record for September. Overall, though, the nation had overnight minimum temperatures that were 1.62 F (0.90 C) above average.

• The worldwide ocean surface temperature was 0.79 F (0.44 C) above the 20th century average of 61.1 F (16.2 C) and the ninth warmest September on record. The warmth was most pronounced in the Atlantic Ocean and the western Pacific Ocean.

• La Niña’s magnitude strengthened to moderate in September, as sea surface temperatures continued to drop across the central and eastern equatorial Pacific . La Niña episodes are typically associated with global temperatures that are cooler than recent trends, and this was the case for September 2010. According to NOAA’s Climate Prediction Center, La Niña is expected to strengthen and last at least into the Northern Hemisphere spring of 2011.

Temperature Anomolies, January - September 2010. Credit: NOAA

Polar Sea Ice and Precipitation Highlights

• Arctic reached its annual minimum on Sept. 19, according to the National Snow and Ice Data Center. The average extent of 1.89 million square miles (4.90 million square kilometers) was the third lowest September sea ice extent on record (30.4 percent below average). The annual record was set in 2007 (38.9 percent below average). This year also marked the 14th consecutive September with below-average Arctic sea ice extent.

• Antarctic sea ice reached its annual maximum in September. September 2010 was the third largest sea ice extent on record (2.3 percent above average), behind 2006 (largest) and 2007 (second largest).

• According to Australia’s Bureau of Meteorology, the continent received an average precipitation of 1.91 inches (48.4 millimeters) during September — this is nearly double the 1961–1990 average and the highest September value on record.

Scientists, researchers and leaders in government and industry use NOAA’s monthly reports to help track trends and other changes in the world's climate. This climate service has a wide range of practical uses, from helping farmers know what and when to plant, to guiding resource managers with critical decisions about water, energy and other vital assets.

Explore further: Solar energy-driven process could revolutionize oil sands tailings reclamation

More information: September 2010 Global State of the Climate – Supplemental Figures & Information

add to favorites email to friend print save as pdf

Related Stories

June Earth's hottest ever: US monitors

Jul 15, 2010

Last month was the hottest June ever recorded on Earth, the US National Oceanic and Atmospheric Administration said Thursday, amid global climate warming worries.

Earth records 7th warmest July on record

Aug 15, 2007

Scientists said the month of July brought record and near-record warmth to the Western United States and was the seventh warmest July in recorded Earth history.

Worlds oceans warmest on record this summer

Sep 16, 2009

(AP) -- The world's in hot water. Sea-surface temperatures worldwide have been the hottest on record over the last three months, the National Oceanic and Atmospheric Administration said Wednesday.

Recommended for you

Big changes in the Sargasso Sea

11 hours ago

Over one thousand miles wide and three thousand miles long, the Sargasso Sea occupies almost two thirds of the North Atlantic Ocean. Within the sea, circling ocean currents accumulate mats of Sargassum seawee ...

Water-quality trading can reduce river pollution

11 hours ago

Allowing polluters to buy, sell or trade water-quality credits could significantly reduce pollution in river basins and estuaries faster and at lower cost than requiring the facilities to meet compliance costs on their own, ...

Managing land into the future

15 hours ago

Food production is the backbone of New Zealand's economy—and a computer modelling programme designed by a Victoria University of Wellington academic is helping ensure that farming practices here and overseas ...

User comments : 11

Adjust slider to filter visible comments by rank

Display comments: newest first

Shootist
2.7 / 5 (7) Oct 18, 2010
Global temperatures measured to within 1/100 of a degree? When the standard error of measure is greater than 1 degree?

How?

Pournelle has been asking the same questions.

http://jerrypourn...Saturday
mongander
3.7 / 5 (3) Oct 18, 2010
Drop dead. We're sick of you BS.
thermodynamics
4 / 5 (4) Oct 19, 2010
Shootist: It is called "statistics." I am sure you have never heard of it. You might tell Jerry that he should buy a good book on statistical analysis. I recommend "Probability and Statistics" by DeGroot and Schervish. The book will teach you how to tease signal out of noise. Another book you should have is "The American National Standard for Calibration: U.S. Guide to the Expression of Uncertainty in Measurement." That will tell you how to assess the certainty of a measurement and determine the total uncertainty in a measurement. If you had taken a course in metrology or engineering analysis you would not even be asking that question. Since you are I have to assume you are ignorant of the subject (I am using the formal meaning of the term ignorant - meaning you have no knowledge of the subject and the term does not reflect on intelligence). Please read some books on statistics, engineering measurements, understand calibration, and validation.
Yvan_Dutil
5 / 5 (2) Oct 19, 2010
First, thermometer "precision" is better than 0.1 C. Second, if you have hundred of those thermometer error average out. By the way, this is first year college scienceé
omatumr
1 / 5 (2) Oct 20, 2010
First, thermometer "precision" is better than 0.1 C. Second, if you have hundred of those thermometer error average out. By the way, this is first year college science(?)


Please explain in more detail.

Since thermometers actually measure expansion rather than temperature, the "hundreds of thermometers" need to all be calibrated to the same standard.

If you had thousands of thermometers carefully calibrated to (+/- 1 degree) and laid out across the entire continent, would it be correct to average out all of the readings and report the average temperature of the continent with an error of (+/- 0.01) degree?

With kind regards,
Oliver K. Manuel
thermodynamics
3 / 5 (2) Oct 20, 2010
omatumr:

You said: "Since thermometers actually measure expansion rather than temperature, the "hundreds of thermometers" need to all be calibrated to the same standard."

Expansion thermometers (glass bulb, twisted metal, or bimetal) haven't been used for decades. Have you been away from measurement for that long?

Weather stations are usually set up with two PRTs (platinum resistance thermometers) for redundancy (sometimes only one).

They are calibrated against the triple point of water using triple point cells and the triple point cells are good to +/- 0.001C. With rapid sampling and data collection they form grids with uncertainties on the order of +/- 0.01C. It is not the thermometers that are the problem. It is adjusting them for the microclimate. However, that is also very carefully done if you read the papers on the methods. You seem to owe Yvan_Dutil an apology for pretending you know how they measure temperature in this decade.
CarolinaScotsman
5 / 5 (1) Oct 24, 2010
Shootist: It is called "statistics." ... Please read some books on statistics, engineering measurements, understand calibration, and validation.


The best book on statistics that I ever read was Darrell Huff's 1954 "How To Lie With Statistics". Truly educational and essential if one wishes to navigate the current glut of statistiacl analyses that are used to "prove" virtually everything.
GSwift7
1 / 5 (1) Oct 25, 2010
"Expansion thermometers (glass bulb, twisted metal, or bimetal) haven't been used for decades"

This article is based on data from 1880 to present, from which they are basing the claim that a new record has been set. I refute any claim you could possible make that global temperature records over the span of 1880 till the past few decades are that accurate. Official records of that data have error bars in excess of the limit needed to claim a new record for the whole of that time period without also stating a confidence estimate (which I'll be willing to assume the original source did). That confidence could be anything, but it is a non-zero number. This article does not present the information as an estimate, but as a scientific fact with 100% certainty. I do not agree with that.
thermodynamics
not rated yet Oct 25, 2010
GSwift7: Since omatumr used the present tense: "Since thermometers actually measure expansion rather than temperature, the "hundreds of thermometers" need to all be calibrated to the same standard." instead of the past tense.

I directly used his text instead of the article. I was replying to his assertion that temperature is always a measure of expansion. Thanks for pointing out that my reply could be taken as responding that temperature had never been measured by expansion. That would clearly be wrong. My comment was aimed at his apparent assertion that expansion is the only way to measure temperature.
thermodynamics
not rated yet Oct 25, 2010
GSwift7: You said: "Official records of that data have error bars in excess of the limit needed to claim a new record for the whole of that time period without also stating a confidence estimate (which I'll be willing to assume the original source did)."

Here is a link to the site of anomalies with error bars to give you a view of the estimated error in the data set:

http://www.ncdc.n...-bar.gif

You would have to go back to the data set to tell what uncertainty they would have to attach to the claim, but it appears it would be relatively low uncertainty. Naturally, the uncertainty declines as the technology has improved, but the error bars do not overlap for the periods they are claiming.

You are right they should have been more clear but it appears the claim has merit and the uncertainty bars are not overlapping. If you have access to a source that shows that the uncertainty overlaps, I would like to see. Thanks
GSwift7
1 / 5 (1) Oct 26, 2010
First, common meteorlogical platinum resistance thermometers with digital display are only certified to +/- .1 degree accuracy. Second, standards have changed over time (especially prior to 1930 when absolute zero was adjusted by .07 kelvin, though that has nothing to do with global temp records). Third, mercury in glass thermometers are still commonly used for meteorology. Fourth, callibration of common thermometers used for meteorology in the 19th century varied widely, and many open air thermometers were actually still in use (Fahrenheit didn't even discover the effect of air pressure on the boiling point of water until around 1700, and they use the boiling point of water to callibrate the high side temp on most thermometers. Many common people still don't understand that effect today, lol). Fifth, Statistical means with statistically significant estimates of global temperature and associated statistical error bounds are not the same as the physical reliability of the instruments.