Archive for the ‘GIStemp’ Category

GISTEMP: Build Notes

2011 March 7 3 comments
I received an email request to assist in a linux build of GISTEMP. Looking at my old notes and directories, I realized that I did not have an end-to-end instruction set. And as far as I can tell, no one else does either. So here they are. The build has become much easier, primarily due to bug fixes over the last 2 years or so, many of them initiated by findings from Barnes and Jones at Clear Climate Code.

Read more…

Lines, Sines, and Curve Fittings 17 – More Morlets

2011 February 22 4 comments

Natural Variability 1 – One World, Two Cuts

2011 February 15 2 comments

This is a sparse post. It just a test spin on my way to looking for regional trends. Lots of tables, little code.

Read more…

Lines, Sines, and Curve Fittings 14 – Two Halves

2011 February 14 Comments off

Assume a spherical cow. Slice in two. Broil one half. Bake the other. Bones go into the stew.

Read more…

Lines, Sines, and Curve Fittings 13 – Fourier

2011 February 13 Comments off

To date, I have used two methods to examine the apparent sine wave in the annualized surface temperature date. I looked for a best fit to the residuals of detrended (linear, exponential) data. I have also just a priori added a sine component to the best fits of the data (manual fits, nls). Also, while working through this, I have at times withheld some of the data to check the fits a posteriori. This has yielded a range of possible periodicities:

56: lines-sines-and-curve-fitting-1-oh-my
50: lines-sines-and-curve-fitting-2-r
50 (second 140): lines-sines-and-curve-fitting-3-double-down
61,71 (second 141): lines-sines-and-curve-fittings-4-walk-and-chew-gum
50,56,61: lines-sines-and-curve-fitting-5-a-growth
58,59,59: lines-sines-and-curve-fitting-11-more-extrapolation

Its time to bring a new tool to the problem: Fast Fourier Transform. Specifically, we will use the spectrum function in R which is wrapper for spec.pgram. The latter is “calculates the periodogram using a fast Fourier transform, and optionally smooths the result with a series of modified Daniell smoothers (moving averages giving half weight to the end values).”

Read more…

Lines, Sines, and Curve Fittings 12 – heteroskedasticity 1

2011 January 21 3 comments

In statistics, a sequence of random variables is heteroscedastic, or heteroskedastic, if the random variables have different variances. The term means “differing variance” and comes from the Greek “hetero” (‘different’) and “skedasis” (‘dispersion’). In contrast, a sequence of random variables is called homoscedastic if it has constant variance.

Read more…

Lines, Sines, and Curve Fitting 11 – more extrapolation

2011 January 19 6 comments

I’ve mostly been working through GISTEMP in this series, but the exp+sine results were interesting enough that I wanted to pause and look at both line+sine and exp+sine in all three data sets.

Read more…

Lines, Sines, and Curve Fitting 10 – nls

2011 January 18 19 comments

The “nonlinear least squares” (nls) function is part of the core of R. John Fox wrote an introduction to it: Nonlinear Regression and Nonlinear Least Squares. This function will in a few dozen iterations return a better fit than my brain-dead looping around parameter space a few tens of thousands of times.

Read more…

Lines, Sines, and Curve Fitting 8 – D'Agostino

2011 January 16 Comments off

The eyeball and quick sigma population checks in the previous post provided some confidence that the global temperature anomalies are normally distributed over the mean. But there are more formal tests, including D’Agostino normality test.

Read more…

Lines, Sines, and Curve Fitting 7 – normal

2011 January 15 6 comments

Zeke Hausfather has tendered a challenge to Joe Bastardi regarding future warming -v- cooling. A bet similar to the “Did Global Warming Stop …” series I ran through earlier this month.

Zeke describes a portion of the bet as follows:

The graph below shows the trend in annual (Jan-December) temperatures from 1970 to 2010, with two standard deviations of the detrended residuals around the trend to show expected confidence intervals of variability. This means that on average, we would expect only 2.5% of observations to exceed the red upper dotted line and 2.5% of observations to fall below the lower dotted in any given year. The linear trend and confidence intervals for the 1970 to 2010 data are extended up to 2030 to provide a testable projection.

I’ve been meaning to test this. Is the distribution of the detrended global temperature anomalies ‘normal‘? Which is to say, do the residuals around the OLS trend assume a gaussian distribution?

Read more…