the Air Vent

Because the world needs another opinion

Tropospheric Amplification Using GISS

Posted by Jeff Id on January 31, 2009

This is a continuation of my previous post on temperature  amplification.

Tropospheric Temperature Trend Amplification

Where I used a frequency sorting method to plot the amplification factor against time.  The following graph was a result of that calculation as shown in the previous post.  I put it here for comparison.


The next plot is exactly the same as above, using the GISS data.


Note the difference in Y axis, and the extension of the amplification into the negative Y axis.   Another thing I see is that the GISS amplificatoin of 1.2 for the global signal looks pretty good while the tropcial amplification extends as high as 1.5.  In the GISS version the 17 year trend even extends into the negative for the tropical UAHTMT curve.  I plotted the negative point curves as generated in R and the graphs of the two lines look like they could be negative.


I did a quick arma check on the data with trend and found the Hadcrut data has  an AR 1 value of 0.87 vs AR1 of 0.73 for GISS.   I think this indicates Hadcrut employs an algorithm which generates an increased time dependence in the data.  The filtering would make it less useful for this type of analysis since it suppresses the high frequency trends.

It makes me wonder what would happen if we balanced the AR coefficients between satellite and ground and redid the analysis.  Even if my frequency weighting of my filter was different, we should get closer to the actual result.

4 Responses to “Tropospheric Amplification Using GISS”

  1. Layman Lurker said

    Jeff, I think the brief dip for trop UAH TMT into negative territory around 16 or 17 year scale may be another artifact from the 1992 discontinuity in UAH. In fact it may verify your use of GISS as a correction to both sat metrics. What are your thoughts on the 24 – 26 year scale?

  2. Layman Lurker said


    Oops. Sorry I goofed. Wrong direction for it to be the step.

  3. Have you tried applying your filtering to random brown or ARMA noise?
    After all, we wouldn’t want you to have built a hokey-stick algorithm!

    Fascinating stuff as ever Jeff.

  4. Jeff Id said

    #3 I did run an arma analysis in one of my first posts.

    #1 I want to run the model data next, then maybe I’ll know something new. There seems like so many possible explanations now I don’t know what to do with it.

    It’s good to have some smart people helping interpret these curves though. Otherwise, what’s the point.

Leave a Reply to Jeff Id Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: