On the challenge of interpreting Activity Density from NEON’s pitfall arrays

As fossils fuels burn—with all the attendant effects—we are becoming increasingly concerned with how Earth’s insects—the little things that run the world—may be declining. Follow along, and let met tell you about a wee complication toward understanding what’s happening.

In a new paper in Ecology we use NEON‘s network to explore how changes in climate affect insect communities monitored via Activity Density PDF: https://bit.ly/3IThjfu.

Frequently insects are monitored with traps—like these malaise traps that capture flying insects, and NEON’s pitfall traps that capture bugs running on the ground. In each case traps catch more when 1) bugs are more abundant and/or 2) bugs move around more.

Malaise traps at the Niobrara River Prairie Preserve

This gets tricky when we use the traps to say something about insect populations. Why? Because ecologists know that bugs (as ectotherms) can move more when its warm, and can reach higher numbers in productive, rich environments. Hence Activity (movement) Density (abundance).

The rate that bugs fall in a pitfall trap increases with temperature (because they move faster) and productivity (because more bugs are supported with more productivity).

Consider, when

Kirsten DeBeurs and I reanalyzed a global dataset from tundra to rainforest, the number of ants running across branches, sure enough, increased predictably as the plant production grew richer. PDF: https://bit.ly/3wOOfDC

An analysis of a geographic study of ant predation on clay caterpillars revealed that 40% of the global variation was accounted for by the habitat’s productivity.

But here’s the complication. A lot of other things change as you go from deserts to forest including the ability to move, unimpeded, thru the environment. Our dear ant would be slowed by a lot more stuff on the way to a trap in forest litter compared to desert pavement.

And the pitfall traps in the NEON network sample from *all* major habitat types, often more than 1 at a site. High in the Rocky Mountains, for example, the NIWO site samples grasslands and evergreen forests. Could these habitats yield different responses to changing climate?

Niwot, one of NEON’s administrative sites in the Rocky Mountains, monitors both grasslands and evergreen forests.

When we—Cam Siler, Michael Weiser, Kirsten de Beurs and Katie Marshall—examined the effects of a site’s mean temperature and productivity on its bug community’s Activity Density, we got very different results for each.In desert scrub, as temp increased so did activity density. In grasslands this increase plateaus. Now look at forests: AD increases to a peak, then declines as temps warm. A global monitoring network, this suggests, promises all sorts of responses to an increase in temperature.

Four different habitat types generated four different patterns of activity density with temperature and productivity.

Warming in the desert scrub may continue to generate higher AD, with or without changes in abundance, simply because bugs move faster along desert pavement. Forests? 1°C of warming may increase—or decrease—AD, with or without changes in the numbers of bugs.

Why is this important? If we use traps to monitor Earth’s bugs populations—particularly if we want to see how forest bug populations are changing relative to desert bugs—we need to be *very careful* to consider their habitat and how bugs move through it. More on this soon.