The timing and magnitude of changes to Hortonian overland flow at the watershed scale during the post‐fire recovery process

Abstract

Extreme hydrologic responses following wildfires can lead to floods and debris flows with costly economic and societal impacts. Process-based hydrologic and geomorphic models used to predict the downstream impacts of wildfire must account for temporal changes in hydrologic parameters related to the generation and subsequent routing of infiltration-excess overland flow across the landscape.However, we lack quantitative relationships showing how parameters change with time-since-burning, particularly at the watershed scale. To assess variations in bestfit hydrologic parameters with time, we used the KINEROS2 hydrological model to explore temporal changes in hillslope saturated hydraulic conductivity (K sh ) and channel hydraulic roughness (n c ) following a wildfire in the upper Arroyo Seco watershed (41.5 km 2 ), which burned during the 2009 Station fire in the San Gabriel Mountains, California, USA. This study explored runoff-producing storms between 2008 and 2014 to infer watershed hydraulic properties by calibrating the model to observations at the watershed outlet. Modelling indicates K sh is lowest in the first year following the fire and then increases at an average rate of approximately 4.2 mm/h/year during the first 5 years of recovery. The estimated values for K sh in the first year following the fire are similar to those obtained in previous studies on smaller watersheds (<1.5 km 2 ) following the Station fire, suggesting hydrologic changes detected here can be applied to lower-order watersheds. Hydraulic roughness, n c , was lowest in the first year following the fire, but increased by a factor of 2 after 1 year of recovery. Post-fire observations suggest changes in n c are due to changes in grain roughness and vegetation in channels. These results provide quantitative constraints on the magnitude of fire-induced hydrologic changes following severe wildfires in chaparral-dominated ecosystems as well as the timing of hydrologic recovery.

Temporal changes in rainfall intensity-duration thresholds for post-wildfire flash floods and sensitivity to spatiotemporal distributions of rainfall

“…Rengers et al (2019) calibrated a hydrologic model using data from small watersheds (0.01-2 km 2 ) burned by the Station Fire and found relatively low values for saturated hydraulic conductivity (K s ), generally between 2-10 mm/h. These results are consistent with values for saturated hydraulic conductivity inferred by Liu et al (2021) via model calibration in the upper Arroyo Seco watershed.…”

“…Due to wildfire-induced changes in surface conditions, including canopy cover and soil-hydraulic properties, runoff generation in the first year following the fire was likely dominated by infiltration excess overland flow (Schmidt et al, 2011, Liu et al, 2021. Enhanced soil water repellency (SWR), which helps promote low infiltration capacity, and extensive dry ravel, which loads channels with fine-grained hillslope sediment, are both commonly observed after fires in the San Gabriel Mountains (e.g., Watson and Letey, 1970;Hubbert and Oriol, 2005;Lamb et al, 2011;Hubbert et al, 2012).…”

“…The impact of dry ravel, which reduces grain roughness in the channel network, and reduced vegetation density led to estimates of Manning's n in the channels of the upper Arroyo Seco of approximately 0.09 s m -1/3 in the first year following fire (Liu et al, 2021). These hydrologic changes led to widespread flooding and debris flows during multiple rainstorms in the first winter after the fire Oakley et al, 2017).…”

“…These hydrologic changes led to widespread flooding and debris flows during multiple rainstorms in the first winter after the fire Oakley et al, 2017). As hydrologic recovery began over the next several years, the watershedscale K s and Manning's n generally increased and likely started to mitigate the flash flood risk (Liu et al, 2021).…”

“…Both saturated hydraulic conductivity on hillslopes (K sh ) and hydraulic roughness in channels (n c ) primarily determine runoff generation and the shape of hydrograph, including total runoff volume, peak discharge rate, time to peak (Canfield et al, 2005;Yatheendradas et al, 2008;Menberu et al, 2019). Other parameters, such as hydraulic roughness (n h ) and capillary drive (G h ) on hillslopes, had a relatively minor impact on modelled runoff after the Station Fire in the upper Arroyo Seco watershed (Liu et al, 2021). (2021) were unable to calibrate the necessary K2 parameters in post-fire year 4 so we do not perform any simulations to constrain flash flood thresholds in that year.…”