Predicting the effects of range shifts in high-elevation pond caddisflies on ecosystem processes

Background/Question/Methods

Shifts in species distributions along local and regional environmental gradients are perhaps the most frequently reported impact of climate change, yet we often know little about how the resulting community rearrangements affect ecosystem function. Over the past several decades, we have observed range shifts among species of limnephilid caddisflies in high-elevation ponds near the Rocky Mountain Biological Laboratory where changes in climatic variables (snow pack, melt dates, summer temperature and precipitation) are affecting whether and when ponds dry. Caddisfly larvae are the dominant group of detritivores in these ponds and we know from previous research that their abundance affects multiple ecosystem processes including the transfer of detrital energy to detritivore secondary production, release of detritus-bound nutrients for algal uptake, and bottom-up effects of algae on herbivores. Here we report on experiments designed to determine the degree to which different resident and invading species in this detritivore guild are functional replacements with respect to these processes. We conducted experiments in replicate microcosms containing standardized starting quantities of detritus, inocula of heterotrophic and autotrophic microbes from source ponds, and periphyton tiles for measuring the biomass of benthic algae.  We measured detritus decay rates, caddisfly growth, ammonium nitrogen, soluble reactive phosphorus, and chlorophyll a on tile substrates.

Results/Conclusions

The species of limnephilid caddisflies that dominate at different elevations and along permanence gradients at a given elevation at our study sites differ in per-unit-biomass detritus processing rates, detritivore growth, and mobilization of detrital nutrients. We use those data to parameterize linear models that predict how substitutive and additive changes in species composition in this detritivore guild will affect the total amount of detritus processed and converted to detritivore secondary production. Experimentally derived regression coefficients that link algal productivity to nutrient levels, and in turn to bottom-up effects on herbivores, provide additional predictions for how changes in species composition in this guild will affect the algal trophic path in pond foodwebs. Our results indicate that whether shifts in species distributions will change the overall rates of ecosystem processes will depend on the particular combinations of species that are assembled. Model predictions provide the basis for the design of future experiments in which we experimentally manipulate combinations of species in relatively natural habitats to determine the net ecosystem-level effects of newly assembled species combinations; and the extent to which species interactions modify expected outcomes from additive models parameterized with short-term single-species data.