Digital Diatribes

A presentation of data on climate and other stuff

September 2008 Update on Global Temperature – NCDC

Posted by The Diatribe Guy on September 19, 2008

And the beat goes on… NCDC released the August anomaly a few days ago, and it came in at a value of 0.4425. The predicted anomaly was about a tenth of a degree higher, so this came in quite a bit lower than the model suggested.

The information is found here..

The August anomaly is 0.4425 (in terms of 1 degree Celsius).

*It is 0.0503 degrees colder than July 2008
*It is 0.0293 degrees colder than August 2007
*It is the coldest August anomaly since 2000
*The furthest back we can go to find an anomaly at least as large is December 1987. There are no anomalies previous to that where the value is at least 0.4425.

*11th warmest (119th coldest) August anomaly out of 129 data points since 1880
*110th warmest (1,435th coldest) anomaly out of the total 1,544 monthly observations
*More recently, it is the 5th coldest of the last 31 anomalies. The four colder anomalies are all since December 2008.

*The latest 12-month average is now 0.4563, wich is the coldest 12-month stretch since the period ending October 2001.

*This month was cooler than previous year, after an increase last month.
*12 of the last 14 months show a year-over-year anomaly decline

NCDC is demonstrating similar trends as the other measures. I will present the updated overall trends and the updated cooling trends.

I mostly want to present and discuss the updated forecast model I put together. Last month it was based on the 60-month trend model, because that produced the minimum least squares estimate of all the other methods. However, this month I added an additional estimate based strictly on weighting past anomalies, and in addition, added weighting factors across all the different trend models to produce a single, best, least squares estimate of anomalies. Adding these additional parameters, and extending the time out through 2017, produces the chart below.

Based on this chart, the anomalies trend down only slightly through 2012. They vary from a low of 28 to a high of 54. The higher anomalies are within the next year, and it bounces around up and down with a slight downward trend. The model, though, predicts a substantial shift downward in the first two months of 2013 to a lower level, and it hovers around that area until mid-2015, at which point we see a continued descent into negative anomalies into 2017.

Let me reiterate that this model is completely objective. All my weights are strictly determined through least squares estimation, and strictly from the data. There is no massaging of the information and no assumptions made at all.

This projection will continue to be updated as new anomalies come in. We’ll see what happens.


3 Responses to “September 2008 Update on Global Temperature – NCDC”

  1. jnicklin said

    I suppose it is reasonable to assume that, all things being equal, current trends will continue into the future. That is what the pundits assumed in the 80s and 90s. Of course, all things aren’t equal, hence the cessation of warming in the 21st century.

    Only time will tell. Thanks for the glimpse of the future.

  2. Diatribical Idiot said

    Yes, it’s just a model. We know how those go…

    But I guess what I’ve tried to get across is that, while the weighting of anomalies is determined by different trending models, the predicted anomalies themselves are not linearly trended. It’s too simplistic and a proven fallacy to try and extend trends out in a linear fashion. The model above attempts to capture cyclic elements in the data, and points of reversal, through a weighting scheme of the previous 132 anomalies. This is done in seven different ways, and then those seven ways are further weighted across each other to produce the final minimum least squares estimate. There are further refinements I think I can make to the model yet to reduce that squared error even further, and I’ll continue to explore those things.

    I just didn’t want people to think the above is based on linear trending. It’s a little tough to explain in full detail without boring people to tears.

  3. John Nicklin said

    No critism intended. Your work, as always is much appreciated.

Leave a 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: