By: David Ho, P.H.

Watersheds that experience wildfires undergo substantial changes to soil characteristics that influence runoff patterns; such as changes to soil erodibility, soil hydraulic properties, and vegetation cover. The loss of vegetation and the reduction in infiltration rates due to an increase in soil water repellency can exacerbate surface runoff, intensifying the potential for damage to infrastructure and properties. The National Weather Service graphic (Figure 1) shows how watersheds affected by wildfires can potentially increase runoff after a storm. When modeling to account for these conditions, it becomes crucial to adjust the soil loss rates to properly reflect the altered state.

The National Weather Service graphic shows that wildfire burn scars are a flood risk due to surface soil changes

Figure 1. Wildfire burn scars are a flood risk.

A proposal on Post-Wildfire Hydrologic Parameter Adjustment was funded by USACE Headquarters General Investigations program within the Hydrology, Hydraulics and Coastal Community of Practice in Fiscal Year 2024. The primary goal of this proposal is to review existing post-wildfire hydrologic studies and extract relevant pre-burn and post-burn soil loss parameters.  Additionally, the initiative supports a collaborative effort with Dr. Jay Pak to validate soil loss adjustments using the Green and Ampt infiltration model (Green, W. H., & Ampt, G. A., 1911) for pre- and post-wildfire basins.  These efforts will culminate in the creation of a summary table detailing the various soil loss methods identified in these studies and the corresponding parameter adjustments. This table will be incorporated into a larger presentation aimed at guiding Districts in determining suitable ranges for parameter adjustments. Specifically, the aim is to equip Districts with the necessary knowledge to proficiently adjust parameters for post-wildfire modeling within the Hydrologic Modeling System's (HEC-HMS) framework. By consolidating and disseminating this information, the goal of this work is to enhance the capacity of professionals to address the complex challenges posed by post-wildfire hydrologic scenarios.

A common approach within post-wildfire modeling using HEC-HMS has been to use the Curve Number loss method.  The Curve Number (CN) approach is a commonly used method for post-wildfire hydrologic modeling. This approach increases the curve number by a value based on the severity of the burn. However, the CN approach has limitations for post-wildfire modeling: the CN method was originally developed for agricultural and rural settings; the CN method does not work well for long-term simulations; and the CN adjustments have often been subjectively determined.  Prior scholarly research (Ebel et al., 2022; Rengers et al., 2019; McGuire et al., 2018) has explored various alternative hydrologic methodologies, such as the Green and Ampt loss method, for post-wildfire hydrologic modeling. The alternative methodologies can potentially be more applicable and flexible than the CN method for different use cases, such as leveraging measured soil properties in making initial estimates and applying the methods in mountainous watersheds. These investigations typically focus on watersheds that have experienced fire-related disturbances, with models calibrated to accommodate changes brought about by the fire. The calibrated values from pre-fire to post-fire provide valuable insights into parameter adjustments, and guide hydrologic modelers in adapting parameters for other hydrologic methods.

As part of this investigation, several notable studies have been reviewed for their highly useful information. For example, Ebel et al. (2022) performed a soil analysis after the 2017 Thomas Fire in southern California, on moderate to low burn severities. The authors collected soil samples and measured parameters for both burned and unburned conditions to estimate parameters for the Green and Ampt, Smith-Parlange, and Richard Equation infiltration models and recommended an adjustment ratio of 0.37 for hydraulic conductivity, 0.36 for sorptivity, 0.66 for wetting front, and 0.85 for saturated soil content.  Figure 2 demonstrates using the ratio suggested by Ebel et al. (2022) with the Green and Ampt infiltration method during a storm event on the Arroyo Seco watershed in southern California after a wildfire.

Figure 2. Pre- and post-wildfire differences in runoff using Ebel et al., 2022 adjustment ratios.

Another study conducted by McGuire et al. (2018) focused on deriving an area-averaged saturated hydraulic conductivity using the rainfall rate, actual values of saturated hydraulic conductivity within a burnt watershed, and the spatial correlation length of hydraulic conductivity. The study evaluated the area-averaged saturated hydraulic conductivity by using the value in the Green and Ampt soil loss method and compared the capability of predicting debris flow initiation against prediction capabilities obtained through the CN method. Simulations of hydrologic processes in a burned watershed following the 2014 Silverado Fire in southern California showed that the Green and Ampt model using the area-averaged hydraulic conductivities performed better at predicting which storms would lead to debris flows than the CN method.  A more recent analysis by Paredez et al. (2023, unpublished) tested Curve Number adjustment factors from Higginson and Jarnecke (2007) and Livingston et al. (2005), as well as a modified version of Higginson and Jarnecke, to model post-burn conditions. Their testing revealed that careful selection of soil antecedent conditions, in combination with the Higginson and Jarnecke adjustment method, performed best.  A current draft summary of parameter adjustments based on a literature review can be viewed on the HEC-HMS Technical Reference page: Literature Review Post Wildfire Parameter Adjustments

Current research efforts are underway at HEC to use existing models developed by Paredez et al., 2023 (unpublished), to validate reduction factors that were identified in our prior literature review.  The effort centers on shifting from the traditional Curve Numbers method to the Green and Ampt model for calculating losses. We are validating reduction factor estimates for soil hydraulic properties using HEC-HMS software. 

The validation effort will involve applying reduction factors determined during the literature review to Green and Ampt parameters in unburned, calibrated watersheds. Post-burn simulations will be performed using historical rainfall data, with subsequent results compared to actual observed post-burn flows. Moreover, these simulations will be juxtaposed with those generated using the Curve Number method.

This comprehensive process is designed to validate the reduction factors and assist HEC-HMS users in applying these estimates to their specific watershed basins. Furthermore, this process will also serve as a general showcase of prospective parameter and model selection criteria.   

References

  • Ebel, Brian A., and John A. Moody. (2022) “Parameter Estimation for Multiple Post‐Wildfire Hydrologic Models.” Hydrological Processes, vol. 34, no. 21, 4 Aug. 2020, pp. 4049–4066, https://doi.org/10.1002/hyp.13865. Accessed 7 Jan. 2022.

  • Rengers, F. K., McGuire, L. A., Kean, J. W., Staley, D. M., & Youberg, A. (2019). Progress in simplifying hydrologic model parameterization for broad applications to post‐wildfire flooding and debris‐flow hazards. Earth Surface Processes and Landforms, 44(15), 3078–3092.  https://doi.org/10.1002/esp.4697

  • McGuire, L.A., Rengers, F.K., Kean, J.W., Staley, D.M. and Mirus, B.B. (2018). Incorporating spatially heterogeneous infiltration capacity into hydrologic models with applications for simulating post-wildfire debris flow initiation. Hydrological Processes, 32(9), pp.1173–1187. doi:https://doi.org/10.1002/hyp.11458.
  • Paredez, Jose M., Pradhan, Nawa R., Brown, Stephen, Shaloka, Elizabeth, Pak, Jang H., Floyd, Ian. (2023, unpublished) "Evaluation and Improvement of Post-Fire Curve Number Methods in Arid and Semi-Arid Watersheds". U.S. Army Engineer Research and Development Center, Cold Regions Research and Engineering Laboratory; 2023. 
  • Green, W. H., & Ampt, G. A. (1911). Studies on soil physics, 1. The flow of air and water through soils. *The Journal of Agricultural Science, 4*(1), 1-24.

  • National Weather Service. Burn Scar Flash Flood & Debris Flow Risks. [online] www.weather.gov. Available at: https://www.weather.gov/cys/burnscarflashflood.