December 2008 Update on Global Temperature – UAH
Posted by The Diatribe Guy on December 9, 2008
Congratulations, UAH, on a full 30 years of temperature anomaly data! We all knew you could do it!
OK, admittedly, it’s pretty hokey to be congratulating a data set, and really, to be getting excited about celebrating an anniversary of sorts for temperature anomaly data. But sometimes you just have to be hokey. Life’s too short.
With that, let us proceed to the wrap-up for the month, and a few charts…
- The UAH data set can be found here.
- The November 2008 anomaly was 0.254 (in degrees Celsius)
- November 2008 was 0.045 degrees warmer than November 2007, and had an anomaly of 0.088 degrees higher than October 2008.
- The November 2008 anomaly was the 65th highest anomaly of the 360 months in the data
- It was the 7th highest November of the 30 years in the data
One thing I was a little interested in, on the heels of my ENSO review last month, was how the UAH anomaly changes appear to lag the ENSO index. I did not do a long-term correlation analysis here (put it on my list!) but I just observed the last few months. THe ENSO data I used is from this source. The minimum ENSO value last occurred with the Feb/Mar average (-1.546). This was on the heels of a few anomalies that dipped below minus-one. The minimum anomaly in the UAH data was in May (-0.183). One data point does not a correlation make by any stretch, but this seems to at least fall in line with the 3-month lag I’ve heard bandied about.
Understanding that there are numerous factors that go into temperature, a cursory look at the ENSO index shows us that the maximum occurred with the May/June measure, though it was in the defined “neutral” territory through the Aug/Sept measure. And lo, we see the November anomaly took a little jump.
The last three ENSO readings average out at less than -0.5, which is some peoples’ definition of a La Niña. It can be reasonably expected that we’ve seen our highest anomaly for a little while, though who knows? It will be interesting to see how it plays out.
Another thing I like to look at occasionally are various trends in longer-term averages. This isn’t about fitting a trend line to raw data, it’s simply an exercise in seeing how the averages graph out as you collapse the data into longer periods. I present here a series of charts on the UAH temperatures:
We know that there is an overall increasing trend in temperatures over the last 30 years. Collapsing the temps into 6-month averages shows some relatively cyclical behavior, with an interesting jump in the chart around 1997.
The 12-month chart shows the same thing, with the cycles a little more defined. The 12-month averages have definitely been elevated since 1997, but the current 12-month average is currently approaching levels not seen since 2000.
We can start to see the upward trend in the troughs of the cycles a bit in the 24-month average chart. Interestingly, there was not a steady increase in the peaks of the cycles. Until 1998 the peaks are at an average around 10. In 1998 there is a sudden jump with the super El Niño that warmed us up pretty nicely. The peaks since then have hit near that level since then, and only recently did we reach levels prior to the mid-2000s El Niño activity. This helps support my idea of the El Niño persistency being as important as the intensity.
The 60 month average shows the more consistent trend upward over time. There is a current downturn, but in the context of some previous downturns, it doesn’t appear to be anything unusual.
The 120-month average is interesting. Until 1993, the 10-year average anomalies were relatively flat. From 1993 to 1999 there was a consistent upward trend. Then we see a 3-4 year period of flat anomalies again, before a significant and constinuous increase in anomalies until early 2008. This definitely corresponds with the increasing trend lines in global temperature that we see. The other interesting part of the chart is the way the chart has dipped over the last year. There is no other point in the graph where we see any reversal of the 10-year average of any significance close to the current dip in average anomaly. That being said, there is only 20 years to work with, so I won’t get too crazy with the sepculation about what that means. It’s just something worth pointing out.