The Climate in Emergency

A weekly blog on science, news, and ideas related to climate change

Predicting Extinction

Leave a comment

This is the third in a four-part series on the relationship between global climate change and mass extinction.

In 2004, in an article published in the journal, Nature, Chris B. Thomas, Alison Cameron, and over a dozen of their colleagues asserted that some 37% of species, both animal and plant, could be “committed to extinction” by the year 2050. Over the past nine years, further research has added detail and greater clarity to the picture, but the prediction is still quite grim. Now, avoiding catastrophe usually requires knowing a bit about the threat, so, to this end, let us look more carefully at how Thomas et al. arrived at their prediction, what it means, and what it does not mean.

First, the phrase “committed to extinction” does not mean “extinct,” nor even “hopeless.” Instead, the phrase is an acknowledgement of the fact that extinction is seldom an instantaneous event. For example, say that a population is shrinking by 2% every year; eventually, this population will die out, unless something changes, but it could take many years. To be committed to extinction does not mean to be beyond hope, but it does mean that the process of extinction has begun, that the conditions the animal or plant needs to continue no longer exist.

Second, let us look at the methods these researchers used. At the heart of their work is the species/area relationship, a mathematically expressible law that states that the number of species in a distinct place (say, an island or a forest) is proportionate to its size. The law says nothing about which species will be present, only how many. Some species will die out locally and new ones will migrate in or even evolve over the years, but the total number of species at any given time will still hover right around whatever the size of the place dictates. So, if the place shrinks, if part of the forest is logged, say, a predictable number of species will die out. Usually, the lost species will come back as soon as the forest regrows, but if one of these species is endemic, that is, found nowhere else, it will be gone for good.

There are very large parks, supposedly protected in perpetuity for our grandchildren to enjoy, that are nevertheless slowly losing species. Once, these parks were just imaginary lines around land that was about the same inside as outside. They were part of islands of habitat that, for some species, covered most of North America or even beyond. Being very big, these areas had a lot of species, so to save the grandeur of that biodiversity, we made parks to protect them. But development or hunting pressure around those parks has cut them off, turned them into islands in an inhospitable sea. Even Yellowstone is not big enough to hold as many species as wild North America once did.

The species area relationship itself says nothing about which species will be lost, although educated guesses can be made based on other ecological principles. What this neat equation does instead is allow ecologists to be sure of themselves, within certain parameters. With this equation, Thomas et al. can be as sure of disaster as could a passenger on the Titanic who had counted the lifeboats.

The reason that the species area relationship matters for anticipating the impact of global warming is that one of the things climate change does is to shrink some habitats, polar ice caps being the most obvious, but hardly the only, example. By looking only at areas where species distributions and their climate parameters are fairly well known, and looking only at endemics, these researchers were able to calculate probable extinction rates for several different possible warming scenarios.

The scientists did not look at distinctly bounded places, however. Instead, they used anticipated changes in the distribution of particular species, either individually or as a group, in order to create a larger picture of extinction risk, using different permutations of the species area relationship equation. For example, if a forest were to shrink by half, than each species in the forest would lose an average of half its range. Therefore, if all the species under consideration lose an average of 50% of their range, then that is equivalent of the forest losing half its area and the number of species lost should be the same. They used three different methods to make the calculation so that minor potential flaws in each method would cancel each other out.

In all three methods, they looked at the climate that currently exists in the ranges of each of the species they studied and then looked at climate projections to see what the area of that particular climate type would be in the future. If the area is predicted to shrink, they assumed that the range of that species would shrink accordingly.

Each method was used to make several different predictions, based on different assumptions about variables that are impossible to actually measure. We do not know how much warming will happen and how fast because we do not know when, or if, society will really get serious about cutting carbon emissions. Therefore, the scientists considered three different possible warming scenarios and made a prediction for each one. Also, sometimes the climate conditions for a species expand in one area while shrinking in another, and species vary a lot in their ability to move to new areas. Since actually including each species’ ability to move in the calculation would make the whole study ridiculously complex, the scientists simply ran each calculation once under the assumption that none of the species could migrate and once under the assumption that all of them could. So for each of the three predictive methods, there are six predictions, one for each possible combinations of initial assumptions.

Additionally, the scientists considered a fourth, less precise method by applying expert judgment to make an educated guess. Then, they added in additional predicted extinctions due to other factors, such as habitat destruction by logging, and looked at specific extinction risk in particular ecological regions.

Naturally, the results for each region considered were different, and the results of each of the eighteen different simulations (three methods, three warming scenarios, and two dispersal scenarios) were also different. At first such a wide variety of answers to a single question might not sound helpful, but this is why scientists record and publish their methods as well as their results. With the methods in mind, a few interesting points can be drawn out.

First and foremost, our position is very scary. Some of the simulations predict upwards of 50% of species lost. To get an idea of what this really means, make a long list of animals and plants (the list can be random, since this prediction did not identify which species might be lost), think carefully and fondly about each entry, and then cross off every other one.

Perhaps more useful than the pure impact of horror, though, is the difference between the three warming scenarios. For minimum expected warming, the range is anywhere from 9% to 31% of species lost. Mid-range warming would cause between 15% and 37% losses. The maximum expected warming would drive extinction up to between 21% and 52%. Although there is some overlap, there are also clear differences between these scenarios—a clear difference between what could happen if we do stop pumping greenhouse gases into the atmosphere (the first or, quite possibly second warming scenario) and what could happen if we don’t stop. This is not merely a warning; is a practical tip. It is clear evidence that what we do still matters.

It is true that this prediction has a lot of wiggle-room. It is dependent on a lot of assumptions and it is focused on very particular species and regions that may or may not be representative of the whole planet. Again, this is why scientists record their methods. All scientists make assumptions in the course of setting up their studies, because it is impossible to study everything in the entire world at the same time. The system works because they keep track of their assumptions and clearly identify what they know, what they are assuming for the sake of the study, and what they did and how. So under the circumstances described in the paper, the results reported in the paper are reliable. The species-area relationship is reliable, the warming scenarios are all possible, and so we know that these extreme extinction rates are likely for the systems described. This isn’t an alarmist supposition or a vague guess. It is a small, keyhole view of the workings of oncoming disaster, and while it is small, it is accurate. With this view in mind, we can confidently take action.

Now, is it true that one of those carefully documented assumptions could be so wrong that the entire prediction is wrong? Yes, it is possible—but it’s unlikely, and so far, when scientists have been wrong about global warming, it’s because they have underestimated the scope of the problem. So while it is in the realm of possibility that Thomas et al. are wrong, how about we not wait around to find out?


Author: Caroline Ailanthus

I am a creative science writer. That is, most of my writing is creative rather than technical, but my topic is usually science. I enjoy explaining things and exploring ideas. I have one published novel and another on the way. I have a master's degree in Conservation Biology and I work full-time as a writer.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s