Climate change leads to species extinctions and exponentially so: the loss of biodiversity is set to accelerate under continuation of global average temperature rise.

This graph illustrates two important things: (1) the pattern of exponential decline of biodiversity for a linear increase in global temperatures (which means that for the preservation of Earth’s biodiversity a business as usual emissions scenario is much worse than ambitious global climate policy) – and (2) that there is large variation among individual published biodiversity projections (illustrating high complexity/uncertainty).

We learn this from a very interesting metastudy performed by ecologist Mark Urban of the University of Connecticut that was published in Science in 2015. In this study Urban assessed 131 existing studies in scientific literature investigating extinction risk under anthropogenic climate change.

Models are simplifications, real world ecological responses are very complex – and for instance differ geographically…

All assessed studies contain extinction predictions and focus on multiple species – and most of the climate-biodiversity models used define extinction as species habitat falling below a critical limit, yet, as the author notes, often ignore further complexities as species interactions (both ecosystem interdependance and competition), dispersal differences and evolution.

Main conclusions from the meta-analysis are that extinction risk varies geographically and that overall biodiversity loss increases exponentially with climatic warming.

The latter may be unsurprising if you give it a thought (for instance due to the cascading nature of extinctions), but it’s also a very important realisation. It means that if we want to prevent biodiversity loss – we do, of course – then raising the global climate policy ambition can be very rewarding:

For instance the average values from Urban’s assessment show that the RCP8.5 scenario that (under conservative climate sensitivity assumption) is linked to a global average warming of slightly more than 4 degrees would lead to more than three times as much biodiversity loss as a 2 degrees warming scenario. Meanwhile the difference in biodiversity decline comparing the 2 degrees and the more ambitious 1.5 degrees scenario could also be significant.

What can we learn from variation in climate-biodiversity studies?

Another thing Urban’s meta-analysis clearly shows, is that there is a large variation in studies assessing extinction risk under anthropogenic climate change, ‘depending on the specific assumptions and geographic and taxonomic focus of each study’. However when he synthesised these studies it showed that extinction risk did not vary by taxonomic group – in other words, the damaging effect of climate change on biodiversity is universal. Variation was however strongly correlated with geography, with significant differences per continent:

Extinction risk under climate change per continent, reflecting higher vulnerability for hot regions (tropics) and smaller land masses/islands. [Speculation from our side: used biodiversity-climate models may ignore full extent of regional climate warming feedbacks, that lead to dramatic increase of relative warming at high latitudes, aggravating local ecological effects – thereby possibly evening out above geographical pattern.]

This shows two overlapping phenomena: the immediate biodiversity decline in hot, tropical regions (where species do not migrate to) plus the relatively high sensitivity of relatively small and geographically isolated habitats, most notably islands – which goes for Australia and New Zealand, and to some extent also South America – as the Panama Isthmus is still an ecological bottleneck for northward climate migration. (If you want to understand Earth’s biodiversity, first understand plate tectonics – they are very closely correlated. Rule of thumb: having many islands and complicated coastlines is good for biodiversity.)

Within the larger pool of biodiversity-climate research, the more pessimistic publications seem to deserve special attention, as Urban writes that those studies that used ‘realistic assumptions about extinction debt and dispersal capacity’ showed ‘substantially increased extinction risks’.

Hmm… that does not sound like good news. We’ll try to find out more.

© Rolf Schuttenhelm | www.bitsofscience.org