Losing the big ones: human impacts on mammalian body size distribution

Humans have modified the Earth in many ways. We have deforested many areas, some long ago (like Europe) and some more recently (like the Amazon). We hunt many species, some to extinction. We pollute and even change the climate (there is no denying that!). We also do some “good things” to remedy these impacts and even have a scientific discipline dedicate to that purpose: conservation biology. Of course, in order to remedy impacts we also need to understand what has been impacted and how.

A recently published paper in which I collaborated, led by Luca Santini and with Moreno di Marco and Carlo Rondinini as coauthors, asked one of these “what and how” questions exploring the effects of human impacts on mammalian body size distributions (the range of body sizes observed in animals within an area) at a global scale.

Median mammalian body size in 1×1 degree cells around the world.

This question is important from a conservation point of view, and also for another field of study called macroecology, which is interested in large-scale patterns and the processes that cause them. A classic macroecological pattern is the so-called Bergmann’s rule which proposes larger endotherm animals are more often found in temperate, colder regions, with smaller animals present in tropical areas. Originally Carl Bergmann proposed this was caused by heat conservation issues (bigger organisms can deal better with cold weather) but other mechanisms have been later suggested. Many studies have explored this pattern using data from diverse species, but we were curious about whether using today’s data could be leading us astray. Our reasoning was that human impacts, which are not homogeneously distributed around the planet, have affected small and large mammals differently, and thus, could have changed global body size distributions. In particularly, we expected in more humanized areas the larger mammals would likely to gone, so when we use current distribution data to see if different environmental factors can explain observed patterns, we may reach wrong conclusions because today’s body size distributions reflect not only environmental constrains and patterns but also the influence of human activities like habitat modification and hunting.

Our study (published in Diversity and Distributions) supported our expectation. Using regression models (of the fancy type that account for spatial autocorrelation) we first evaluated if current median and maximum body sizes in 1×1 degree cells representing the world’s terrestrial surface were best predicted by environmental factors alone (with variables reflecting like climate, elevation, primary productivity) or by a combination of environmental factors and human impacts (with variables for human population density, agricultural land use, historical human impacts, and accessibility – defined as distance to nearest city). The combined models won by a landslide (or to be more scientific, had stronger support in an Information Theory framework) confirming our expectaction that human impacts have influenced where we can find big and small mammals. We then asked what body size patterns would be if humans have minimal impact everwhere (making predictions from the combined models). The map below shows the difference in observed and those predicted sizes, revealing areas where we seem to have lost many big ones, and showing different areas have been affected differently.

Predicted changes in mammalian body sizes globally. Left panel is change in median and right panel change in maximum values

Despite their magnitude, these changes did not dramatically alter the body size – latitude relationship (Bergmann’s rule), which anyway does not appear to fit the pattern Bergmann suggested very well (you can see that in figure 2 of the published paper, here is a public repository copy in case you cannot access the journal version). However, our study does show human impacts can alter global scale patterns, changing the composition and characteristics of species in different areas, and this cannot be ignored when we seek to understand the hows and whys of how nature is organized.

NOTE: this study by Faurby and Araujo explored a similar question but comparing historical data and current data for some mammalian species. We all reached the same general conclusion.


Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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