by Roy Spencer, October 4, 2019 in WUWT
While the vast majority of our monthly global temperature updates are pretty routine, September 2019 is proving to be a unique exception. The bottom line is that there is nothing wrong with the UAH temperatures we originally reported. But what I discovered about last month is pretty unusual.
It all started when our global lower tropospheric (LT) temperature came in at an unexpectedly high +0.61 deg. C above the 1981-2010 average. I say “unexpected” because, as WeatherBell’s Joe Bastardi has pointed out, the global average surface temperature from NOAA’s CFS model had been running about 0.3 C above normal, and our numbers are usually not that different from that model product.
[By way of review, the three basic layers we compute average temperatures from the satellites are, in increasing altitude, the mid-troposphere (MT), tropopause region (TP), and lower stratosphere (LS). From these three deep layer temperatures, we compute the lower tropospheric (LT) product using a linear combination of the three main channels, LT = 1.548MT – 0.538TP +0.01LS.]
Yesterday, John Christy noticed that the Southern Hemisphere was unusually warm in our lower stratosphere (LS) temperature product, while the Northern Hemisphere was unusually cool. This led me to look at the tropical results for our mid-troposphere (MT) and ‘tropopause’ (TP) products, which in the tropics usually track each other. A scatterplot of them revealed September 2019 to be a clear outlier, that is, the TP temperature anomaly was too cool for the MT temperature anomaly.
So, John put a notice on his monthly global temperature update report, and I added a notice to the top of my monthly blog post, that we suspected maybe one of the two satellites we are currently using (NOAA-19 and Metop-B) had problems.
As it turns out, there were no problems with the data. Just an unusual regional weather event that produced an unusual global response.