Credit: NASA

The use of planetary boundaries to estimate a safe operating space for humanity is a very interesting and useful concept. In this week's issue of Nature, Rockström et al. (Nature 461, 472–475; 2009) define acceptable limits for Earth-system processes in such a way that crossing a boundary would risk triggering abrupt or irreversible environmental changes that would be very damaging or even catastrophic for society.

As a boundary for stratospheric ozone depletion, they choose a five-per-cent decrease in column ozone levels — that is, in the total amount of ozone in the atmospheric column — for any latitude, with respect to 1964–1980 levels. Their choice is reasonable, but a bit arbitrary. Although Rockström et al. also identify the appearance of the Antarctic ozone hole as a tipping point, it is not connected to this five-per-cent boundary, which is still well within the bounds of linear behaviour for global ozone loss.

Arguably, a more relevant tipping point is reached when certain substances containing chlorine and bromine trigger massive ozone depletion at all latitudes. This abrupt change results from the same non-linear behaviour of ozone-depleting chemical reactions that causes the Antarctic ozone hole. Such potential change was referred to early on as the 'chlorine catastrophe' and has been more recently analyzed by Newman et al. (Atmos. Chem. Phys. Discuss. 8, 20565–20606; 2008). They show that if chlorofluorocarbons (CFCs) had not been regulated by the Montreal Protocol, ozone-hole chemistry would appear in the tropical lower stratosphere in about 2052, leading to complete lower-stratospheric ozone loss by 2058, assuming growth of three per cent per year in the manufacture of CFCs. This corresponds to about a 60-per-cent decrease in column ozone levels, triggered by an atmospheric concentration of effective equivalent stratospheric chlorine (EESC) of about 30 parts per billion (p.p.b.). EESC is calculated by summing total stratospheric chlorine and bromine levels, and it quantifies their combined effect on ozone depletion in the stratosphere. The Montreal Protocol limited EESC to about 4 p.p.b., leading to a maximum total ozone loss of roughly five to six per cent.

So while the choice of a five-per-cent decrease in column ozone levels as the boundary for stratospheric ozone depletion appears reasonable, one could argue that a more realistic boundary is 10 or even 20 p.p.b. of EESC. Either of these boundaries would still maintain a safe distance from the 30-p.p.b. tipping point that would lead to massive ozone loss; a 10-p.p.b. EESC boundary, for example, would lead to about 15 per cent total stratospheric ozone loss.

World leaders decided to ban the industrial production of CFCs early enough that the decrease in stratospheric ozone was limited to about five per cent. Although the non-linear behaviour of lower-stratospheric ozone loss was not even a consideration in the discussions that led to the CFC ban, the decision was well-justified in light of the potential damage to human health and to ecological systems from an ozone loss greater than five per cent. It also made sense because of the CFC ban's relatively small cost to society, given that replacement compounds could be developed.

In summary, the planetary boundary concept is a very important one, and its proposal should now be followed by discussions of the connections between the various boundaries and of their association with other concepts such as the 'limits to growth'. Importantly, this novel concept highlights the risk of reaching thresholds or tipping points for non-linear or abrupt changes in Earth-system processes. As such, it can help society to reach the agreements required for dealing effectively with existing global environmental threats, such as climate change. Stratospheric ozone depletion was properly dealt with well before crossing the boundary that would trigger an abrupt change of global proportions, but well after reaching the tipping point that caused the Antarctic ozone hole — a regional, episodic event. A five-per-cent decrease in ozone might be appropriate as a planetary boundary, but that's only true if the concept is expanded to include limits that are well within the linear regime for that Earth-system process.

Comment on this article at the Climate Feedback blog