# Differences in 2019 temperatures

HadCRUT4.6 has released their annual temperature result for 2019 as 0.74C. They grid the station data (CRUTEM4) with SST data (HADSST3) in 5×5 degree bins and perform an area weighted: $\cos (lat)$ global average. CRUTEM4 has 7680 stations which contribute to calculating this average. Although GHCN-V4 has more stations (17280) it is not clear to me that the coverage is really all that much better. CRUTEM4 is similar to the coverage of V3 but with some additional stations.

I downloaded the CRUTEM4 station data and calculated the global average with HadSST3 using my 3D averaging method (Spherical Triangulation). This is exactly the same data as that used by HadCRUT4.  Both results are compared below together with those using GHCN-V4 and HadSST3..

Comparison of methodology and station data used to calculate annual global temperatures.

The largest difference is that between spatial integration techniques. Exactly the same data produces a difference of 0.07C in the result for 2019. The reason for this is simply because spherical triangulation makes an implicit interpolation over both poles, whereas the traditional 5×5 lat/lon grid averages over just the occupied cells. This difference will  depend from year by year on just how much warmer high latitude stations warm as compared to those at lower latitudes. So in 2004 and 2015 there was little difference between the two as compared to say 2016.

Here is the 3D grid used to calculate the anomaly for December 2019.

The spherical triangulation grid for December 2019.

This shows how the triangulation connects together all station locations covering all the earth’s surface and as a result interpolates the average temperature from the 3 vertices across each triangular area. The coloured triangles show the relative increase in temperature anomalies relative to 1961-1990..

This procedure also gives very similar results to those of Cowtan & Way who instead use a kriging technique to interpolate into polar regions.

PhD High Energy Physics Worked at CERN, Rutherford Lab, JET, JRC, OSVision
This entry was posted in AGW, Climate Change and tagged , . Bookmark the permalink.

### 7 Responses to Differences in 2019 temperatures

1. Bindidon says:

Clive

“The reason for this is simply because spherical triangulation makes an implicit interpolation over both poles, whereas the traditional 5×5 lat/lon grid averages over just the occupied cells.”

Suppose we would transform UAH6.0’s monthly 2.5 degree grid data (which is free of latitude weighting) into a set of 9504 virtual ‘stations’ (the 3 southernmost and northernmost bands contain no valid data).

Each virtual station then would be located at the centre of a grid cell; each would have the same monthly time series as its ‘surrounding’ cell, with the same data as is visible in the files stored in

https://www.nsstc.uah.edu/data/msu/v6.0/tlt/

e.g., for 2019:

https://www.nsstc.uah.edu/data/msu/v6.0/tlt/tltmonamg.2019_6.0

It would be very easy to transform this format, reminding us the good ol’ IBM punching card devices, in something more appropriate.

How would be your estimate for the effect of applying your spherical triangulation on this flat data set?

Regards
J.-P.

• Clive Best says:

If UAH evenly samples the globe then it should produce a set of equal triangles on the surface of a sphere. In this case there is no need to use spherical triangulation. I haven’t really ever looked properly into the satellite data. The usual criticism of the surface people is that satellites measure the average temperature of a height range in the troposphere rather than the surface temperature. I don’t know enough about it to say whether that is true or not.

• Bindidon says:

1. Of course it is evenly sampled, that should be evident to us.

But my question was not about triangulation between 82.5N and 82.5S, but rather about the rest, i.e. 82.5N-90N resp. 82.5S-90S, for which there is no data available.

This is insofar a really interesting point because UAH6.0’s predecessor UAH5.6 had data for 90N-90S, as you can see when looking at

https://www.nsstc.uah.edu/data/msu/t2lt/tltmonamg.2017_5.6

and compare this file with the UAH6.0 file for 2019 I pointed to in the comment above.

2. “The usual criticism of the surface people is that satellites measure the average temperature of a height range in the troposphere rather than the surface temperature.”

This is correct but it never prevented me from looking up satellite data.

Using anomalies wrt the same reference period, you may easily compare them with surface data or, of course, with satellite data processed by other ‘vendor’s.

The comparison of UAH6.0 data for the lower troposphere with NOAA STAR’s data for the mid troposphere is ‘interesting’.

2. Gerald Machnee says:

RE:CRUTEM4 has 7680 stations which contribute to calculating this average. Although GHCN-V4 has more stations (17280) it is not clear to me that the coverage is really all that much better. CRUTEM4 is similar to the coverage of V3 but with some additional stations.
RE: 7680
RE: 17280
How many in each group are actual stations with instruments?

• Bindidon says:

Gerald Machnee

I never processed CRUTEM on station basis, and didn’t start that for GHCN V4 yet.

Here is a chart showing you, for the entire GHCN daily data set (about 40,000 stations measuring temperature / 50 % of that in the US), the temporal station distribution (Globe, CONUS) together with that of the 2.5 lat/lon degree grid cells encompassing them: