On January 5th of this year, Alaska Governor and Vice Presidential candidate Sarah Palin wrote an op-ed in the New York Times opposing listing polar bears on the endangered species list. Her argument was well reasoned and thoughtful, although in the end, unsatisfactory. In her piece Governor Palin noted her support for policies that helped preserve polar bears:
"We have a ban on most hunting – only Alaska Native subsistence families can hunt polar bears – and measures to protect denning areas and prevent harassment of the bears. We are also participating in international efforts aimed at preserving polar bear populations worldwide."
In that op-ed Palin observed that:
"…polar bears are magnificent animals… They are worthy of our utmost efforts to protect them and their Arctic habitat. But adding polar bears to the nation’s list of endangered species, as some are now proposing, should not be part of those efforts."
Her argument against listing the polar bears is that the threat to them is not based on evidence of proven threat, but a projection of threat based on models of the impact of climate change on habitats. The governor makes clear that she does not oppose the Endangered Species Act:
"We’re not against protecting plants and animals under the Endangered Species Act. Alaska has supported listings of other species, like the Aleutian Canada goose. The law worked as it should – under its protection the population of the geese rebounded so much that they were taken off the list of endangered and threatened species in 2001.
Listing the goose – then taking it off – was based on science. The possible listing of a healthy species like the polar bear would be based on uncertain modeling of possible effects. This is simply not justified."
(Emphasis is mine)
For me, the most distressing part of the op-ed is its attempt to contrast "science" to "modeling". There are a number of methods used by scientists to add to our understanding of our world and our environment. Sometimes we study samples and extrapolate to an entire population. Sometimes we can study a discrete phenomenon in a laboratory and observe it with equipment that allows us to view details and relationships invisible to the naked eye. Sometimes we study a complicated set of relationships by developing mathematical models that allow us to simulate probable future effects of facts we can now observe. These models are just as "scientific" as the other methods used to understand our world. To make policy on a scientific phenomenon like species extinction we must by definition rely on some form of modeling. If we don’t project extinction, then by the time we move to protect a species it is already gone.
The question for policy makers is what type of risk we are willing to allow and what type of information convinces us that extinction is a real risk. Governor Palin believes that accepting projections of species impact based on climate change opens up the possibility that the Endangered Species Act would be used to regulate carbon dioxide emissions. To her, this would be an overly broad interpretation of the Endangered Species Act.
Fair enough, she makes a good point. A better policy approach would be to leave that type of regulation to a new law specifically designed to regulate carbon dioxide. However, we all need to pay close attention to politicians who make a distinction between "science" and "modeling". As Palin correctly notes, scientists should present their analyses to elected leaders who conduct an open discussion about the policy approach needed to address the issues raised by their studies. However, if elected leaders are going to make policy based on scientific information, they need to develop at least a modest level of scientific literacy themselves.
Excellent science requires an understanding of probability and the reasoned use of extrapolation. The polls that politicians read are based on these same techniques. Just like some elections are too close to call based on the polls, some models generate results that are too uncertain to be used for policy making. However, just like the exit polls that show a landslide–sometimes the model is so predictive that it can and should inform policymaking. Effective policy in our complex world requires modeling projected impacts. Modeling is an essential method of scientific inquiry. It is science.