May 2008 Update on Global Temperature, Part 3 – NCDC Data
Posted by The Diatribe Guy on May 28, 2008
Up until now, I have used the GISS data in the analyses I’ve presented. However, my intentions all along were to analyze all the different temperature measures. Since I have a life outside of this realm, that will be done as time allows.
In the meantime, I have expanded my model for the GISS Predicted Anomalies. You can find the update here.
Having crossed that bridge, my next task was moving on to work with a close cousin of the GISS, the NOAA (NCDC) data. The data set with the monthly anomaly information can be found here. This data set is a “close cousin” to GISS because they are determined largely the same way. Both measure temperature averages from surface stations and use much of the same readings. The difference will be in the final smoothing/estimation algorithms from the raw data.
Not unexpectedly, the results are similar. The overall slope since 1880 is 0.0426, which is just over a half-degree warming per century. This is in line with GISS (slightly lower in actual value). The latest flat/cooling period using NCDC goes back to February 2001, only a one-month difference from GISS, which dated back to March 2001. These charts are presented below.
I have run a smilar model with NCDC-specific weighting values to predict future anomalies, but that will be addressed in a separate post, later.
One additional chart was run, using the NCDC data, that I thought was interesting. It tracks the 60-month slope value over time. In the future this will be expanded to the running 120-month, 180-month, etc. charts, as well. I am not sure what it all tells me, exactly, but two things do jump out at me: (1) Warming and cooling cycles are very evident. A high trend in warming reverses itself quickly and eventually becomes a cooling trend. (2) Evident is the upward trend in the cycle minimums, without a corresponding increase in the maximums. This indicates that the warming has been more steady and less volatile since peak swings in the 1950s. It is a trend similar to the period fron the early 1900s to the 1940s. Eyeballing it without scientific or statistical analysis suggests we may be in a similar position to where we were around the mid 1940-s at the moment. But that’s conjecture.
Here is the overall trends chart using NCDC data:
Here is the latest cooling period, from February 2001:
Here is the chart of rolling 60-month slope values: