Persistent Effects of Wildfire and Debris Flows on the Invertebrate Prey Base of Rainbow Trout in Idaho Streams

Wildfire and debris flows are important physical and ecological drivers in headwater streams of western North America. Past research has primarily examined short-term effects of these disturbances; less is known about longer-term impacts. We investigated wildfire effects on the invertebrate prey base for drift-feeding rainbow trout (Oncorhynchus mykiss, Walbaum) in Idaho headwater streams a decade after wildfire. Three stream types with different disturbance histories were examined: 1) unburned, 2) burned, and 3) burned followed by debris flows that reset channel morphology and riparian vegetation. The quantity of macroinvertebrate drift (biomass density) was more variable within than among disturbance categories. Average body weight and taxonomic richness of drift were significantly related to water temperature and influenced by disturbance history. During the autumn sampling period, the amount of terrestrial insects in rainbow trout diets varied with disturbance history and the amount of overhead canopy along the stream banks. Results indicate that there are detectable changes to macroinvertebrate drift and trout diet a decade after wildfire, and that these responses are better correlated with specific characteristics of the stream (water temperature, canopy cover) than with broad disturbance classes. 1 Author to whom correspondence should be addressed: Email: [email protected] 2 Current Address: Fisheries Division, University of Alaska Fairbanks School of Fisheries and Ocean Sciences, Fairbanks, Alaska 99775 3 US Forest Service, Rocky Mountain Research Station Boise Aquatic Sciences Laboratory, Boise, Idaho 83702 4 USGS Alaska Cooperative Fish and Wildlife Research Unit, Institute of Arctic Biology, University of Alaska Fairbanks, Fairbanks, Alaska 99775 Introduction Natural disturbances such as wildfire can play a key role in aquatic ecosystems, altering physical conditions that drive habitat availability and species productivity (Reeves et al. 1995, Gresswell 1999). In the short-term, wildfire can alter basin hydrology and hill slope erosional thresholds, increasing the probability of severe erosional events (postfire floods and debris flows; e.g., Benda et al. 2003) that result in extensive loss of streamside vegetation, channel reorganization (change in channel form and habitat availability), and local extirpation of aquatic species (e.g., Pilliod et al. 2003). Most studies of wildfire and aquatic ecosystems focus on the short-term impacts of wildfire on stream ecosystems (e.g., Minshall et al. 1989, Minshall 2003); less is known about longer-term effects (Bisson et al. 2003). A decade after wildfire, the condition of streamside vegetation and its effects on the prey of drift-feeding rainbow trout may vary depending on the severity of the disturbance. Deciduous streamside vegetation immediately adjacent to the stream can recover rapidly (5 yr; e.g., willows [Salix spp.] and alders [Alnus spp.]); whereas forest trees (e.g., Douglas fir [Pseudotsuga menziesii, Mirbel]) recover over decades (Taylor and Skinner 1998, Dwire and Kauffman 2003). Further, postfire debris flows can reset successional dynamics of streamside vegetation and slow riparian recovery (Dwire and Kauffman 2003). The quantity of terrestrial macroinvertebrates falling into streams––an important energy source for trout (Kawaguchi et al. 2003)––could decrease with changes in the abundance and Northwest Science, Vol. 85, No. 1, 2011 56 composition of streamside vegetation (Wipfli 1997, Kawaguchi and Nakano 2001, Koetsier et al. 2007, McCarthy et al. 2009). The effects of altered streamside vegetation on drifting aquatic macroinvertebrates are less straightforward, but changes in riparian cover can dramatically change drift composition and density (Piccolo and Wipfli 2002). Loss of riparian vegetation reduces the amount of allochthonous energy sources for the stream (e.g., terrestrial plant material; Cummins et al. 1989), while the increase in sunlight boosts available autochthonous energy sources (e.g., algae; Hawkins et al. 1982, Hill et al. 1995, Fuller et al. 2004, Mihuc and Minshall 2005). An increase in either of these energy sources could translate into increased productivity of invertebrate prey for rainbow trout. Further, aquatic macroinvertebrates that enter the drift may benefit from the combined input of autochthonous and allochthonous energy sources in streams that have burned in the past, but have retained their streamside deciduous vegetation (Mihuc and Minshall 1995, 2005 but see Perry et al. 2003). Increased solar radiation and warmer stream temperatures after wildfire may also affect the aquatic component of the macroinvertebrate drift (Minshall et al. 1989, Dunham et al. 2007). Some taxa may disappear entirely, either through loss of intolerant species (Vannote and Sweeney 1980) or earlier onset of adult insect emergence (Frutiger and Imhof 1997). Taxa that remain may be smaller in size if their metabolism has increased with temperature at a rate disproportionate to assimilation (Vannote and Sweeney 1980). These effects on the macroinvertebrate community could translate to lower quality and reliability of the aquatic component of the drifting invertebrate prey base for rainbow trout. Based on the above considerations, we compared the following features of aquatic drifting prey: 1) density; 2) taxonomic richness; and 3) body weight among headwater streams contrasting in disturbance histories and habitat characteristics (canopy cover, stream temperature) in selected Idaho headwater streams. We also compared the diet of rainbow trout to investigate patterns in the predominance of terrestrial prey and the average size and taxonomic richness of aquatic macroinvertebrates. We investigated both summer and autumn characteristics of the trout prey base, with the autumn feeding season particularly important for overwintering survival of rainbow trout. Autumnal shifts in both streamside vegetation (deciduous leaf drop) and aquatic macroinvertebrate composition and density (e.g., through larval insect emergence) could exacerbate or eliminate differences in the prey base among streams with different disturbance histories.