Stream Water Quality and Soil Health After Compound Disturbance: A History of Forest Harvesting, Severe Wildfire, and Post-Fire Management in a Western Oregon Watershed
The impact of large, high severity wildfires on soil health and stream water quality remains poorly understood, in part, due to the lack of tightly controlled, landscape scale experiments that integrate the effects from fire and land use activities. In 2020, the Archie Creek fire burned ~53,230 ha of forested land in Oregon, with ~77% at moderate to high severity. The burn area included the Hinkle Creek watershed, an intensively managed Douglas-fir (Pseudotsuga menziesii) plantation, which was previously studied from 2002–2011 to investigate the effects of forest harvesting on streamflow and water quality. We are leveraging the existing study sites to compare pre-harvest, post-harvest, and post-fire stream water nitrate concentrations (NO3-). We are also quantifying the effect of fire on saturated hydraulic conductivity (Ksat), soil nitrogen (NO3-, NH4+, potentially mineralizable N), and active soil carbon to understand the mechanisms driving in-stream responses. We applied a chronosequence study design across a matrix of forest harvesting to disentangle the potential variability in fire effects. In 2021, we began collecting stream water samples at six of the historic locations. Within the riparian zones, we collected soil cores (0–5 cm) for soil hydraulic properties and soil samples (0–15 cm) for nutrient analysis at ~60 sites. Historically, stream NO3- did not vary substantially in harvested catchments; the average concentrations were 0.15 ± 0.03 (SE) mg L-1 during pre-harvest and 0.18 mg L-1 ± 0.02 during post-harvest. However, preliminary results from our post-fire study indicate a substantial variation among streams, with post-fire concentrations ranging from 0.16 ± 0.04 mg L-1 to 1.46 ± 0.15 mg L-1. We have observed a positive linear relationship between stream NO3- concentrations and the extent of catchment burned at low and moderate severities (r2 = ~0.55, p <0.001). Contrary to findings from other studies (Rhoades et al., 2011), we have observed a negative relationship with high severity burn and stream NO3- (r2 = -0.76, p<0.001), signaling the potential influence of other localized factors. Extractable soil N in year one appears to be elevated by fire. However, by year two, we observed a ~54% reduction in NO3-, ~59% reduction in NH4+, and ~69% increase in potentially mineralizable N. Critical questions remain about the timing and legacy of nutrient pulses after wildfire. Watershed-scale studies, such as this, allow us to evaluate the links between soil properties and water quality following fire disturbance in western forest ecosystems.
Graduate Student Forest Ecosystems & Society, College of Forestry, Oregon State University