Insect community responses to climate and weather across elevation gradients in the Sagebrush Steppe, eastern Oregon

Abstract

In this study, the U.S. Geological Survey investigated the use of insects as bioindicators of climate change in sagebrush steppe shrublands and grasslands in the Upper Columbia Basin. The research was conducted in the Stinkingwater and Pueblo mountain ranges in eastern Oregon on lands administered by the Bureau of Land Management. We used a "space-for-time" sampling design that related insect communities to climate and weather along elevation gradients. We analyzed our insect dataset at three levels of organization: (1) whole-community, (2) feeding guilds (detritivores, herbivores, nectarivores, parasites, and predators), and (3) orders within nectarivores (i.e., pollinators). We captured 59,517 insects from 176 families and 10 orders at the Pueblo Mountains study area and 112,305 insects from 185 families and 11 orders at the Stinkingwater Mountains study area in 2012 and 2013. Of all the individuals captured at the Stinkingwater Mountains study area, 77,688 were from the family Cecidomyiidae (Diptera, gall gnats). We found that the composition of insect communities was associated with variability in longterm (30-yr) temperature and interannual fluctuations in temperature. We found that captures of certain fly, bee, moth, and butterfly pollinators were more strongly associated with some climate and vegetation variables than others. We found that timing of emergence, as measured by first detection of families, was associated with elevation. When analyzed by feeding guilds, we found that all guilds emerged later at high elevations except for detritivores, which emerged earlier at high elevations. The abundance of most taxa varied through time, mostly in response to temperature and precipitation. Of the pollinators, bees (particularly, Halictidae and Megachilidae) peaked in abundance in late June and early July, whereas butterflies and moths peaked in August. Flies peaked in abundance in July. Overall, our interpretation of these patterns is that insect communities respond positively and negatively to weather and local vegetation more than to long-term climate. Given increasing variability in weather and high probability of extreme weather events, insect communities in sagebrush steppe also may experience considerable fluctuations in composition and abundance, as well as phenology. These findings have implications for many ecosystem services, including pollination, decomposition, and food resources for predatory birds and other vertebrates.

Differences in functional trait responses to elevation among feeding guilds of Aculeata community

“…Our findings for predator SD are in agreement with many studies on elevation (e.g. Corcos et al, 2018 ), and the same is true for our finding that predator richness increases with sites with warmer minimum winter temperatures (Pilliod & Rohde, 2016 ). The increase in the elevation range with increasing elevation indicates that highland predator species have broader environmental tolerance.…”

“…Humidity increases the probability of parasite and pathogen spread (Fornoff et al, 2021 ; Hranitz et al, 2009 ). In addition, colder winter temperatures may indicate greater snow insulation for individuals buried just below the soil surface in pollinator communities (Pilliod & Rohde, 2016 ). The exact amount of snowfall at our sites was not measured, but the Tokachi region, where our site is located, is known to have low snowfall and deep soil freezing (Hirota et al, 2006 ).…”

“…Feeding guilds grouped by food resource and foraging type have different responses to the environment (Berg et al, 2010 ; Fontana et al, 2020 ; Pilliod & Rohde, 2016 ). In a multi‐trophic study, species richness (SR) patterns along an elevational gradient were different in each guild (Fontana et al, 2020 ).…”

“…Particularly in the alpine zone, the seasonal duration time of pollinators is linked to the limited flowering period (Kudo, 2014 ). Pollinators are predicted to respond to abiotic and phenological factors (Hoiss et al, 2012 ; Pilliod & Rohde, 2016 ).…”