Synergistic feedbacks among drought, nitrogen eutrophication, and plant functional traits determine grassland plant-insect interactions

Background/Question/Methods – Anthropogenically driven increases in drought and nitrogen deposition should transform plant-insect dynamics given how they reconfigure the availability of limiting resources and disrupt species interactions. However, the direction of these changes is unclear because of varying possible influences of drought and eutrophication on plant tissue quality for insect herbivores.

Because insects are nitrogen-limited, herbivores generally find plants higher in nitrogen more appealing. Drought and eutrophication can increase foliar nitrogen, the former doing so via physiological-based changes (e.g. protein degradation). Alternatively, drought may decrease plant palatability by reducing ion transport and turgor. Or, tissue may be the most palatable following re-occurrence of rainfall post drought by restoring turgor and uptake of surplus nitrogen, as per the “pulsed-stress hypothesis”.

Here, we tested these alternatives with a factorial field experiment quantifying the interactions among drought (100% rainfall reduction), drought followed by a 30mm water pulse, and nitrogen addition (10 g/m²) on five response variables: plant aboveground biomass, cover, foliar quality (C:N ratios), herbivory (sap-sucking, chewing), and insect abundance and diversity.

We conducted this work in a C4-dominated tallgrass prairie and a C3-dominated old field. We predicted drought to constrain biomass and tissue quality in the old field, reducing insect herbivory and diversity.

Results/Conclusions – For plant-based responses, we observed anticipated increases in biomass and cover with nitrogen addition in both communities. Drought and drought-nitrogen responses, in contrast, unfolded unexpectedly. Community-level drought effects on plants were relatively minor, explained by compensatory responses by some species in both communities. Drought negated the benefits of nitrogen addition on aboveground biomass.  As well, rainout shelters only significantly reduced soil moisture to a depth of 20 cm, whereas many plants in these systems have deep rooting systems. The pulsed treatment led to significant increases in biomass and cover after one week compared to the drought treatments. Nitrogen addition increased percent nitrogen of foliage. Contrary to our predictions, however, neither drought nor pulsed treatments influenced carbon to nitrogen ratios.

For insect-based responses, most differences were explained by community type. The old field had significantly more chewing damage and herbivore abundance and diversity. The pulsed treatment increased sap-sucking damage in the old field but not in the tallgrass prairie.

In summary, our work shows strong system resiliency to drought and eutrophication interactions in two distinct grasslands. Responses were best predicted by plant species composition, both for compensation to drought but also in regulating insect diversity and abundance.