UCLA

Wildfires result in changes in soil water retention in post-wildfire hillslopes. Even though there have been previous studies regarding the post-wildfire water retention of soil, the temporal changes in soil water retention and corresponding slope stability are not well known. Firstly, this thesis analyzed long-term changes in hillslope stability after the 2019 Williams Flats, WA Fire. Our investigations were performed both near a tree that was burned by the fire (i.e., dead tree), and near a tree that was not affected by the fire (i.e., live tree) within the fire perimeter that is 300 m away from each other. The results showed that the lowest factor of safety (FS) against translational slope failure near the burned tree was in June 2020, which was attributed to the loss of vegetation and preferential flow through macropores. The change in vegetation was analyzed through normalized difference vegetation index (NDVI), which is a remote sensing method that quantifies the amount and health of the vegetation. The FS near the burned tree increased over time as a result of vegetation regrowth, which was also observed through the NDVI of the site. The temperature data indicated that the macropore clogging through wildfire ash also may be a contributing factor for the increase in water retention at the dead tree over time.The spatial properties of the hillslope and spatiotemporal variation of the soil water retention after the 2023 Happy Camp, CA Fire were analyzed. This was part of a larger project that aims to quantify the changes in soil mechanical and hydrological behavior and forest system dynamics (vegetation loss/regrowth, root decay, and ash redistribution) and develop a probabilistic wildfire-specific slope stability model that is a function of time after fire and degree of saturation. As with the Williams Flats, WA Fire study, we considered two sites, one in burned hillslope and the other in unburned hillslope. The site that was considered for burned hillslope is 100 m long  60 m wide and has a relatively uniform slope of 30° with a northwest-facing orientation. The burned hillslope was used to investigate the soil properties and soil water retention changes after the fire. The unburned hillslope with a similar slope angle and with a northeast-facing orientation that is around 2000 m away from the burned hillslope was selected to investigate the saturated hydraulic conductivity and the soil water retention changes as a baseline for comparison. The results indicate that the hillslope soil properties are variable within the selected site between the upper, middle, and lower portions of the burned hillslope. The variation of saturated hydraulic conductivity on the burned hillslope was significantly higher than on the unburned hillslope. The short-term water retention of the burned hillslope had spatiotemporal variation between the upper, middle, and lower portions of the hillslope, and aligned well with the large rain events (>15 mm). The comparison between burned and unburned hillslope showed that the unburned hillslope generally has higher suction and lower saturation due to evapotranspiration through vegetation. In addition, the layered ash formation and its water retention were examined after the 2022 Bolt Creek, WA Fire. This was part of a larger study that aims to understand the effect of ash erodibility and mechanical behavior on sediment movement and debris flow initiation that will lead to the development of a mechanistic debris flow model. Samples of the ash were collected one month and one year after the fire. Field and laboratory experiments showed that surface ash, subsurface ash, water repellent soil, and soil layers have different characteristics and water retention behavior. The short-term volumetric water content measurements and laboratory experiments showed that the surface ash has higher water retention than the subsurface ash. The long-term volumetric water content analysis showed that the water content of the ash layer changed significantly through precipitation events meanwhile the soil layer was not affected by the precipitation events over a year after the wildfire.