Skip to content

A recent paper in nature by Allen et al (2012) in Nature has received some attention because explains the northern expansion of the tropics seen  since 1979 as a result of pollution by black carbon and ozone. The southern expansion of the tropics has already, it seems, been attributed to ozone.

Look a little further than the abstract however and things do not seem quite so clear cut. The paper’s Fig 2 shows observed and modeled Northern Hemisphere tropical expansion based on five metrics involving combinations of ozone and black carbon. Click on image to enlarge.


I wonder if, like me, you find this figure very unimpressive. Following each measurement, and observing the very wide error bars in relation to the changes I don’t have any confidence that the data as presented this way shows an northernward expansion of the tropics.

Allen et al (2012) state: “Recent observational analyses show that the tropics have widened by 2 deg –5 deg latitude since 1979 (ref. 1). This evidence is based on several metrics, including a poleward shift of the Hadley cell, subtropical dry zones, and extratropical storm tracks. A more recent estimate of the tropospheric jet shift, based on satellite-derived temperatures, suggests a smaller rate of expansion of 1.6 deg.”

Ref 1 is Seidel at al (2008) “Widening of the tropical belt in a changing climate.” Its figure 2 shows changes in several estimates of the width of the tropical belt since 1979. Does this figure show what it is said to show? Click on the image to enlarge.


This is a problem in science publications, and especially in climate change. It only takes a single overly optimistic analysis of a data set to appear in print to be taken up in a subsequent paper and portrayed as being scientifically more solid than it really is. Then of course in a subsequent paper it becomes an undisputed fact. Finally it becomes institutionalized and appears in a pamphlet or on a minister’s or science advisor’s powerpoint presentation.