Date of Award


Document Type

Open Access Master's Thesis

Degree Name

Master of Science in Applied Ecology (MS)

Administrative Home Department

College of Forest Resources and Environmental Science

Advisor 1

Joseph Wagenbrenner

Committee Member 1

Peter Robichaud

Committee Member 2

Casey Huckins


Wildfires can increase soil erosion by orders of magnitude over rates in unburned forests and negatively impact aquatic resources. Rill erosion is a dominant erosion and sediment transport mechanism in burned forests, and hydrologically connected rills can form networks on burned hillslopes. At the swale scale (< 10,000 m2), little is known about how rill networks develop under different burn severities over time, their relationship with sediment yields, and the effect of post-fire salvage logging on rill networks and sediment yields.

The first study assessed rill networks and sediment yields in three burn severities in the inland Coast Range of California, USA, after the 2015 Valley Fire. The results indicated the rill networks in high burn severity areas reached nearly the entire extent of the burned hillslopes. Rill densities in high burn severity areas were significantly higher (19-23 cm m-2) than low and moderate severity areas (0.5-2.1 cm m-2). Sediment yields from high burn severity areas (13-15 Mg ha-1) were significantly higher than the low to moderate burn severity areas (0.1-3.4 Mg ha-1), and highly correlated with rill density (r2 = 0.97). Results indicate that the extensive rill networks in high burn severity areas can greatly increase connectivity, resulting in increased sediment delivery downslope.

The second study assessed the effects of post-fire salvage logging on soil bulk density, field saturated hydraulic conductivity, ground cover, rill networks and sediment yields in the central California Sierra Nevada following the 2013 Rim Fire. Post-fire logging resulted in 9-56% percent soil disturbance, which was dominated by high traffic skid trails (9-29%). Feller buncher tracks averaged 2% of swale areas, while mixed traffic areas averaged 7% and were only found in five of nine logged swales. Within high traffic logging disturbance soil bulk density was increased, field saturated hydraulic conductivity reduced, and bare soil increased. When scaled up to the swale scale, logging had no significant effect on ground cover, but did result in significantly higher wood cover relative to unlogged swales. High traffic skid trails typically initiated extensive rill networks, with up to 20 cm m-2 sourced from skid trails. Rills in skid trails were typically concentrated at waterbars and directed downslope, where they often connected to the outlet. Control swales had 4-18 rills from untrafficked areas, and logged swales had 0-15 rills from untrafficked areas and 1-12 rills from disturbed areas. As with the ground cover, rill densities in logged and unlogged swales were not significantly different. Mean annual sediment yields in logged and unlogged swales were not significantly different, and the high input of wood cover from logging did not have a significant effect on rill density or sediment yield. In unlogged and logged swales, both rill density and sediment yield appeared to be reduced by large areas of undisturbed ground with high surface roughness, which disconnected rills from the stream network. To reduce rill connectivity and sediment delivery from logging disturbance, land managers should optimize skid trail layouts to reduce the spatial footprint of skid trails, increase ground cover on skid trails and below waterbar outlets, and retain buffers with high ground cover between logging disturbance and ephemeral channels.