co2regressAdjAnnotatedreseultsLeif

a1 = .006 (.006)
a2 = .059 (.051)
g1 = .041 (.018)
g2 = .061 (.055)
d1 = 3 years (37 years)
d2 = 73 years (84 years)
Ci = -.041 degs

ResPSD
Residual power spectral density (peak is at 57.9 years)

detrendStats4

detrendStats4

 

Reverse TSI Bootstrap

Ran the sim with the reversed time series below

reverse

Using the parameters from the prior optimization correlation is negative as expected.

Start

First step of the optimization (match the acfs) works because it is insensitive to scaling, including a scaling of -1

 

ReverseACF

The fit though is not too is not too hot 🙂

acfMatch

Skip the first step and give the optimizer control of all  parameters, best fit (not guaranteed global minimum)

acfNoMatch

‘course now the acf is horked

acfbestfit

Leif’s magic series

comp

Optimized Results

The results at the top of this page were manually optimized. Running the optimizer  does much better.

LeifResults

LeifDetail.png

Run the sim to 1976

short1976

 

firstdiff.png

Advertisements