Impacts of wildfire on benthic insects in burned streams: community and population responses at multiple scales

Abstract

This research combines several approaches, spanning a range of spatial and temporal scales and biological organization, to document the impacts of wildfire on stream insects of the Pajarito Plateau, near Los Alamos, New Mexico. I present results of an eight-year monitoring study to describe physicochemical and insect population and community responses to the 1996 Dome fire in a burned stream (Chapter 1). To better understand mechanisms behind community resistance and resilience in this long-term study, I explored how insect population attributes (i.e. life-history, morphological, and behavioral traits) were associated with recovery time and environmental conditions after the Dome fire (Chapter 2). Fortuitously, I was able to conduct these studies with a before-after-impact-control design, where benthic insect responses and trait-environment associations were compared between a burned stream and an unburned stream for two years before and six years after the wildfire. I also conducted a short-term in situ experiment at the microhabitat level, where I compared community colonization patterns on cobbles in six watersheds across a gradient of wildfire history (Chapter 3). Specifically, I measured community resilience in two streams burned in the 1996 Dome fire, in two streams burned in the 1977 La Mesa fire, and in two unburned streams. I also compared representation of population attributes shown to convey community resilience after the Dome fire (Chapter 2) across these six streams. Finally, after documenting low resilience of shredder populations to both wildfire and experimental disturbance, I conducted experiments to investigate colonization and growth of stonefly shredders, Pteronarcella badia Hagen, on two types of burned detritus (deciduous and coniferous) (Appendix A). The 1996 Dome fire had moderate to severe impacts on physical, chemical, and hydrological stream characteristics in the burned stream, which resulted in strong negative impacts on stream insect communities (Chapter 1). Insect communities showed low resistance to a 100-year flash flood following the fire in 1996, where insect densities and taxa richness were reduced to near zero. Total insect density showed high resilience and recovered within a year of the fire. This was largely due to the colonization of disturbance-tolerant and strong-dispersing taxa such as Simulium sp. blackflies, Baetis tricaudatus Dodds mayflies, and Chironomidae. Taxa richness was less resilient and recovered after 3 years, while taxa composition remained markedly different from pre-fire composition in the burned stream. Several stonefly taxa and elmid beetles common in pre-fire years were greatly reduced or absent at the end of the study. In contrast, dramatic changes in total insect density, taxa richness, and community composition were not observed in a nearby and geomorphically similar unburned stream. Relationships between insect population attributes and environmental conditions explained patterns of community resistance and resilience in the burned stream (Chapter 2). Specifically, taxa with traits conveying high resilience (i.e. strong larval and adult dispersal, multi-voltinism, resource generalists) colonized earlier and were positively associated with higher peakflows after the Dome fire. Traits conveying high resistance (i.e. ability to avoid or tolerate high flows through swimming, attachment to substrate, or burrowing; low drag forces) were also strongly represented in early post-fire years with floods in the burned stream. These relationships made sense in light of strong colonization barriers presented to both larval and adult insects in streams of the Pajarito Plateau. Specifically, these streams lack tributaries, have upper and lower ephemeral reaches, and are embedded in a harsh canyon-mesa landscape. Despite common landscape factors, the unburned stream did not show the same relationships between resistant and resilient traits and environmental conditions as the unburned stream. Patterns of resilience to a small-scale experimental disturbance corresponded to a gradient of wildfire history, where insect communities in recently burned streams were more resilient than communities in unburned streams (Chapter 3). Taxa with attributes that convey resilience (multi-voltinism, strong larval dispersal, generalist-feeding strategies) dominated recently burned streams. For instance, Chironomidae and Baetidae recolonized quickly via drift. In contrast, taxa dominating unburned streams had poorly resilient attributes, such as weak larval dispersal, specialist feeding strategies, and semivoltine life-cycles (i.e. grazing heptageniids, shredder stoneflies). This study demonstrated that wildfire and post-fire flooding can have lasting effects on community composition, and suggests that post-disturbance communities are more resilient to local scale disturbances. In a separate set of experiments, I found that P. badia were able colonize and grow on burned detritus, regardless of lower microbial conditioning and nutrition (Appendix A). These results suggest that low resistance and resilience of shredders in burned streams, as well as on experimentally-disturbed cobbles, may have been due to reductions in quantity rather than quality of detritus. A multiple-level approach, including monitoring studies and experiments across different spatial and temporal scales, and endpoints spanning levels of biological organization, was necessary to understand impacts of wildfire on stream insect communities. Further, I demonstrated the utility of adopting a trait-based perspective to reveal mechanisms behind stream insect community responses to disturbance. Linking species traits with environmental conditions in burned streams, and interpreting relationships in light of colonization barriers of the landscape, revealed distinct post-fire succession patterns. In general, trait-based approaches are both ecologically and evolutionarily relevant since real-time trait-environment interactions determine species persistence, while traits are a result of natural selection. Traits also reflect functional significance (i.e. foodweb position and productivity) of insect species. Finally, trait representation in communities can be compared across regions with different species pools, allowing for broader tests of ecological theory.