When it comes to invertebrate body size, it’s hard to go big toward the poles

Ecologists have long been fascinated with body size as the “one functional trait that rules them all”. An organism’s body size is just so good at helping us understand how it fits into the rest of the community. Thus an exhaustive dataset on how communities of organisms (e.g., all the birds that occupy a woodlot, all the spiders collected from a single tropical tree) vary as you move from place to place—we’re talking real Geographic Ecology here—has always been a grail for ecologists. Or certainly, at least for me and my colleagues.

We sampled 99 invertebrate communities from the NEON network, with up to 4 land cover types per site. This standardized sampling captures the ecosystem variation across the continent, and allows us to effectively analyze the drivers of body size.

Enter NEON, the National Ecological Observatory Network. Constructed to monitor North American ecosystems for 30 years, NEON generates some of the highest quality community data out there—all in pretty much the same way—and has allowed our team (including coPIs Michael Weiser, Katie Marshall, and Cam Siler, and a host of students) to begin putting together a picture of how insect communities are built and how they vary from the subtropics of Puerto Rico to the Arctic Circle. Already, we’ve generated the first such studies of insect activity, diversity, and invasive species, with many more to come. Today, let’s look at the insect size results, shall we?

The invertebrates captured from one two week pitfall trap in North Dakota, from the North American Great Plains.

So this is what a NEON sample—3-4 pitfall traps sampling a given ecosystem for 2 weeks—looks like. Using computer imaging, we can turn these images into a metric of body size (in this case, the area (mm2). From there we generate community distributions like the 10 below, with the five communities on top representing warm ecosystems with little or no winter, and those at the bottom with some of the longest winters (and shortest growing seasons).

Ten body size distributions of ground invertebrates from the NEON North American monitoring network. Southern distributions (in red) tend to have a smaller mode, but larger maximum size. Colder ecosystems tend to have slightly larger modes, but do not have the season lengths that allow the generation of the largest body sizes.

A close inspection reveals that the sites with the longer, warmer growing seasons supported insect communities where the modal bug size was smaller. But at the same time, those same communities had more individuals in the largest size classes than the colder sites with longer winters.

How the mean size and size diversity varies with latitude across 99 invertebrate communities (a and c), and 8 widely distributed taxa making up >90% of the individuals (b and d). Data come from early, middle, and late season pitfall samples from across U.S. North America. Significant LS Mean regression in c (+ 95% CI). Regression fits for sub taxa are on log10 axis to fit most intercepts).

When we do the statistics, we find little evidence for a geography of mean body size, in part because the bigger taxa (like the Orthoptera get smaller as you move North, while the smallest, like the mites, get bigger). But the range of sizes, the size diversity of this functional trait becomes more and more constrained as you approach the arctic. This pattern is consistent across the taxa.

Using AICc informed regression to test 4 drivers of size mean and diversity across 99 invertebrate communities in the US North America. Lines are best fit linear regressions (+ 95% CI), partial adjusted r2’s associated with each driver (i.e., model r2 = 0.11 for mean size and 0.49 for size diversity). The leftmost scatterplot accounts for the greatest variation in the variable; rightward plots account for residual variation (i.e., Mean size increases with GPP; the remaining variation increases with Number of Winter Days.

Why is this so? The top two graphs above capture what little variation (11%!) we can account for body size, but this number is a little misleading. This is because the plot of size vs an ecosystem’s plant productivity (GPP) suggests that communities of tiny bugs trend toward depauperate ones, but that large average size is promoted, but not guaranteed, by productive ones.

But we appear to have a much stronger handle on the diversity of sizes–if you want a community with large and small bugs, you want a long growing season/short winter. We think that is because it takes time to grow, and if you have 12 months to do so, your community will be more likely to produce a behemoth than if you only have two.

A tiny grasshopper of the genus Aulocara.
A much, much larger grasshopper of the genus Romalea, found in the same Nebraska prairie.

The upshot? We now have a much, much better picture of another feature of insects at a continental scale. And as the Earth warms, we can begin to make intelligent predictions as to how insects—described by EO Wilson famously as “the little things that run the world”—will respond. For example, if growing seasons are prolonged, and temperatures increased, we may begin to see larger bugs in our gardens and agricultural fields. If so, the NEON monitoring network may be in a position to reveal this change. Check out the manuscript in Ecosphere.

Discover more from The Kaspari Lab

Subscribe now to keep reading and get access to the full archive.

Continue reading