我，本作品著作权人，特此采用以下许可协议发表本作品：

本文件采用知识共享署名 3.0 未本地化版本许可协议授权。

您可以自由地：

• 共享 – 复制、发行并传播本作品
• 修改 – 改编作品

惟须遵守下列条件：

• 署名 – 您必须对作品进行署名，提供授权条款的链接，并说明是否对原始内容进行了更改。您可以用任何合理的方式来署名，但不得以任何方式表明许可人认可您或您的使用。

https://creativecommons.org/licenses/by/3.0CC BY 3.0 Creative Commons Attribution 3.0 truetrue

The graph shows global monthly near-surface air temperature anomaly estimates since 1850 from eight sources spanning three different estimation approaches. The plotted values are temperature anomalies — the difference between a month’s temperature from a given series and the mean for that month of the year for that series across a reference interval, here taken as 1951 – 1980. The overlaid trend is a 4th order polynomial fit to the mean of the plotted series.

1. NASA GISTEMP global temperature series^[1]

Data from: http://data.giss.nasa.gov/gistemp/tabledata_v3/GLB.Ts+dSST.txt.

2. HadCRUT4 global temperature series^[2]

Data from: http://www.cru.uea.ac.uk/cru/data/temperature/HadCRUT4-gl.dat.

3. HadCRUT + Cowtan & Way global temperature series (version 2)^[3]

Data from: http://www-users.york.ac.uk/~kdc3/papers/coverage2013/had4_krig_v2_0_0.txt.

4. NOAA National Climate Data Center global temperature series^[4]

Data from: ftp://ftp.ncdc.noaa.gov/pub/data/anomalies/monthly.land_ocean.90S.90N.df_1901-2000mean.dat.

5. Berkeley Earth land + ocean global temperature series^[5]

Data from: http://berkeleyearth.lbl.gov/auto/Global/Land_and_Ocean_complete.txt.

6. NCEP / NCAR reanalysis (version 1) global average surface air temperatures^[6][7]

Data from: http://www.esrl.noaa.gov/psd/cgi-bin/data/timeseries/timeseries.pl?ntype=1&var=Air+Temperature&level=2000&lat1=-90&lat2=90&lon1=0&lon2=360&iseas=0&mon1=0&mon2=0&iarea=1&typeout=1&Submit=Create+Timeseries.

Satellite MSU remote sensing cannot measure meteorological surface air temperature (typically defined as that at 2 m height). The best that can be obtained is a temperature estimate across the lower troposphere,^[8] about the bottom 5000 m of atmosphere^[8] That temperature is of course much colder than the surface temperature and varies somewhat differently with climatic influences. Nevertheless it is often compared with surface temperatures.^[9] The satellite series do not extend across the reference interval adopted for the graph. Instead they are each rebased so that their mean across the 30 year interval 1979 – 2008 equals the mean of the five instrumental series across the same interval.

7. RSS temperature lower troposphere global temperature series (RSS TLT version 3.3)^[10]

Data from: http://www.remss.com/data/msu/monthly_time_series/RSS_Monthly_MSU_AMSU_Channel_TLT_Anomalies_Land_and_Ocean_v03_3.txt.

8. UAH temperature lower troposphere global temperature series (UAH TLT version 5.6)^[11]

Data from: http://vortex.nsstc.uah.edu/data/msu/t2lt/tltglhmam_5.6.txt.

For comparison, the Intergovernmental Panel on Climate Change Fifth Assessment Report^[12] central temperature estimates for 2050 for representative concentration pathways RCP6 (mid range) and RCP8.5 (high) are also shown.

The Microsoft Excel spreadsheet that produced this graph can be downloaded here: http://gergs.net/?attachment_id=1246.