1880 - Measured Temperature Averages - 2019

SIMULATION 1880 - 1979: The measurements of 1880 to 1980 are compared with a computer calculation shown as the small red triangles (see figure). This simulation takes the atmospheric increase of CO₂ from the measured values of 280 to 300 parts per million in 1880 to the values obtained by Keeling of 335 to 340 ppm in 1979 as given. An optimal solution for the observed temperatures was found by varying the climate sensitivity, the heat uptake of the oceans,

PawPlot - P.Duinker

Clicking on a Hansen et al. reference leads to the original publication (the prediction for the 2010's is from their 2007 publication). The text "Combined Decadal Averages", the data points themselves and the references to the Data Centers are clicable and lead to some details of the determination of the averages and the published yearly temperature measurements by the various Centers respectively. The average of the last eight years of the present decennium is also shown. For a pdf-file of the figure click here.

volcanic effects and for allowing a small increase in the solar luminosity up to 1940. The best value for the climate sensitivity was found to be ΔT=2.8° Celsius. The first decade with the Krakatau explosion of 1883 shows the largest deviaton, however the agreement overall is quite reasonable - certainly if one takes into account the early stage of the complicated climate model calculations at that point in time [1]. The authors summarise this part of their publication with the remark:

Key consequences are: (i) emperical evidence that much of the global climate variability on time scales of decades to centuries is deterministic and (ii) improved confidence in the ability of models to predict future CO2 climate effects.

PREDICTIONS 1981 - 2020: If we compare the measurements after 1980 with the model calculations the conclusion is obvious: The predicted rise is clearly seen. The causes of the observed deviations - too low in 1981 (red curves) and too high in 1988 (light blue triangles) - are welll known [2]. The prediction for the present decade (red rectangle) is taken from a publication from 2007 and a more recent simulation for the years 1880-2003 can be found in [3]. For a more detailed comparison of the data with the natural varibility region see the next page.

Comments
1. In the late sixties a pioneering climate calculation was published by Syukuro Manabe and Kirk Bryan both of the Geophysical Fluid Dynamics Laboratory at Princeton (USA). For a short historical overview of these type of calculations see a webpage of the NOAA.
2. The prediction of 1981 includes only the anticipated increase of CO₂ and not the other greenhouse gasses. The climate sensitivities used for calculating the dashed, the drawn and the dashed-dotted curves were ΔT=1.4 - 2.8 - 5.6 ° C respectively. For the 1988 prediction the contribution of all greenhouse gasses to the temperature rise was estimated rightly, however the climate sensitivity itself was taken at the too high value of ΔT=4.2° C, instead of the presently best estimate of ΔT=3.0±1.0° C.
3. https://pubs.giss.nasa.gov/docs/2007/2007_Hansen_ha00210r.pdf.