10/20/2011 02:39 pm ET Updated Dec 20, 2011

Underestimation Is Never a Good Thing When It Comes to Risks

A recent blog by Climate Progress brought to mind some musings about human nature that I have had for a while. Human nature is an interesting beast. There are issues for which we will fight to death, and there are ones on which we tread very carefully. Among the latter is a puzzling tendency for unwarranted underestimation, especially when it comes to science. Scientists have long been very careful with what they say, and are famous for using expressions such as "results suggest," "data indicate," and "the analyses seem to show" in their peer-reviewed papers, as opposed to "results show" or "data confirm." And in some instances, this tendency can lead to public confusion and poor policy decisions.

Take climate change. It has been known and shown and proven that (1) there is more CO2 in the atmosphere than there was before; (2) CO2 in the atmosphere makes it warmer; (3) that the earth is, indeed, getting warmer; and (4) that there is no other plausible explanation for the recent warming trend, be it natural of otherwise. However, when it comes to predictions and forecasts related to climate change itself and its various consequences, as mentioned in the Climate Progress blog, there is a striking bias for scientists to keep them on the "safe" side -- that is, estimates where they have highest confidence and are safe from being wrong.

When you are talking about issues with large risks, like climate change, this is not a smart strategy, to say the least. As communicator Naomi Oreskes very wisely puts it, it is always better to be safe than sorry, but in the case of climate change, scientific underestimations might put us in a situation where we will be sorry and not at all safe. Recently, at the Stephen H. Schneider Symposium on Climate Change, Oreskes addressed "the question of how scientific communication may have contributed to public confusion, particularly the scientific tendency to downplay dramatic results." She calls this tendency "ESLD - Erring on the Side of Least Drama," a tendency probably "aris[ing] from the scientific virtues of skepticism, dispassion, and restraint, but which has perhaps inadvertently led scientists to under-estimate the tempo, mode, and severity of climate change, and given the public the impression that the scientific findings are less secure, and their implications less alarming, than they actually are". I thought that was brilliant, and probably one of the most important reasons behind the large amount of misconceptions and skepticism we see. She gave various examples of climate change-related predictions and estimates that have been (sometimes grossly) underestimated. Here are a few examples modified from Think Progress:
• A Norwegian expert in 2005 stated that the Arctic sea ice retreat at the time was larger than predicted by any IPCC models. The retreat has accelerated in the past two years.
• Penn State climatologist Richard Alley in March 2006 said that the ice sheets appeared to be shrinking 100 years ahead of schedule.
• Greenland and Antarctica have already lost significant mass, when the IPCC predicted no significant loss by 2100.
• The temperature rise of 0.33°C from 1990 to 2005 was close to the top range of IPCC climate model predictions, in other words, the range the IPCC estimated lest likely to occur.
• Average sea-level rise from 1993 and 2006 was3.3 mm/yr, higher than predicted by IPCC models.
• Expansion of the subtropics is happening faster than projections from models.
• Since 2000, carbon dioxide emissions have grown much faster than predicted, with 2010 having the highest emissions increase ever.

The take-home message here is simple: climate change is real, it is happening now, and downplaying it is not going to get us any allies. If we keep underestimating its consequences, we aren't conveying the real risks to the public and policy makers, and we will continue to see inaction to address climate change.