Return to Website

Number Watch Web Forum

This forum is about wrong numbers in science, politics and the media. It respects good science and good English.

Number Watch Web Forum
Start a New Topic 
Author
Comment
World Temps 1900 - 2011

The CRU released their raw data (sort of raw, it is monthly average data).

I took that data and made this

This is all the temperatures experienced in any given month. The Y axis is in Kelvin.

Each point is colored based on how many times that temperature appeared in the data set. The pinkish color represents the highest frequency for the month.

Black marks the extreme for the month (high or low).

Right now the data has been rounded to the nearest kelvin. After I get all the kinks worked out I will do it to tenths. There are 78,000 points on this chart right now.

I am working finding a better color scheme. I need to get more qualifying information into this chart, but it is all being done in PHP, point by point, so I am taking it one step at a time.

Re: World Temps 1900 - 2011

A different view of the same data -- I overlaid the GISS temp anomaly and suppressed the zero..



(clicking on image this time will take you to the image on its own page.)

Re: World Temps 1900 - 2011

Yet another view with other information.

The gray points are raw data from CRU.
The Red line is the GISS anomaly data.
The orange dots are simply the arithmetic mean for each month of the data recorded.
The Black dots on the bottom represent the total number observations in the data set (multiply the chart reading by 20).




I am definitely not qualified in the eyes of "real" scientists to make assessments of what the raw data says. Looking at averages is a useful technique. Ignoring the raw data altogether though can cause you to forget about magnitudes. The GISS Anomaly temperatures seem consistent with the feel of the raw data and the simple mean. The magnitude of the changes in the GISS Anomaly seems insignificant when presented with background of the Raw Data.

(Once again, clicking on the image will take you to the full size image which is rather large. The total number of points in the data set is approx 4.8M. I attempt to show the number of times a temperature appeared in the dataset with the darkness of the point representing it. )