Last Modified: 2025-08-18 10:36:05.461

This tutorial walks you through configuring and calibrating an HEC-HMS model for long-term post-wildfire hydrology simulations.

Software Version

HEC-HMS version 4.13 was used to develop this tutorial. 

Incorporation of the Pak & Lee Dynamic Surface Method

In Step 2, we identified the need to represent post-fire recovery processes—such as time-varying soil loss and canopy dynamics—to capture long-term landscape evolution. In Step 3, our first option is to incorporate the Pak & Lee Dynamic Surface Method (Applying the Pak and Lee Dynamic Surface Method for Post-wildfire Hydrologic Modeling). Wildfires often bake the soil surface, creating a water-repellent crust (hydrophobic layer) that severely limits infiltration and surface storage, thereby increasing runoff. As rainfall gradually erodes this layer, infiltration rates recover over years—behavior that constant-rate methods cannot reproduce. The Pak & Lee method overcomes this limitation by dynamically updating key surface parameters, allowing the model to simulate both the immediate post-fire runoff surge and the slower, multi-year rebound in infiltration capacity.

We implemented and recalibrated this approach in HEC-HMS for the Arroyo Seco watershed, fitting the January 18–19, 2010 event and then extending the simulation through February 14, 2019.

In Step 3, we will

• Compare the two calibration runs for the January 18–19, 2010 event—each using a different approach—and select the optimal setup for the long-term simulation.

• quantify the improvement gained by the Pak & Lee Dynamic Surface, and 

• identify opportunities to further refine the model for long-term post-wildfire hydrology.

Compare two calibrated results with Pak & Lee Infiltration Limit Dynamic Surface method for Post-Wildfire hydrology Model

  1. Select the Basin Models folder to expand the Watershed Explorer. 
  2. Select LGA_CUH_LR_Dys and  LGA_CUH_LR_Dys_R Basin Models and review all its methods and input parameters.
  3. Run the LGA_CUH_LR_DyS_Event and  LGA_CUH_LR_DyS_R_Event simulations, calibrated to the January 18–19, 2010 event using the Layered G&A infiltration loss method and Pak & Lee Infiltration Limit Dynamic Surface method. 

  4. Examine the simulation results, then answer the questions below.


    Question 1: 

    What are the results of the calibration?

    • Computed peak discharge (value, date, and time)
    • Total runoff volume
    • Nash–Sutcliffe Efficiency

    Which model—LGA_CUH_LR_DyS_Event or LGA_CUH_LR_DyS_R_Event—shows better calibration according to these metrics, and is it robust enough to run continuously for nine years?

    compared to USGS Arroyo Seco gage observations, both the layered and standard Green-Ampt models exhibit a strong match.

    • Computed peak discharge for LGA_CUH_LR_DyS_Event: 4,902 cfs on 18 Jan 2010 at 22:30
      Computed peak discharge for LGA_CUH_LR_DyS_R_Event: 4,229 cfs on 18 Jan 2010 at 22:45
      Observed peak discharge: 4,230 cfs on 18 Jan 2010 at 22:30
    • Total runoff volume: 0.82 in (LGA_CUH_LR_DyS_Event) vs. 1.04 in (LGA_CUH_LR_DyS_R_Event) vs. 1.04 in (observed)
    • Nash–Sutcliffe efficiency: 0.912 (LGA_CUH_LR_DyS_Event) vs. 0.863 (LGA_CUH_LR_DyS_R_Event)

    Both models deliver similarly high‐quality event‐based calibration. Although LGA_CUH_LR_DyS_Event posts a slightly higher Nash–Sutcliffe efficiency (0.912 vs. 0.863), LGA_CUH_LR_DyS_R_Event more accurately captures peak discharges and total runoff volume. Therefore, LGA_CUH_LR_DyS_R_Event was chosen to define the initial conditions for the nine‐year continuous simulation.

    Calibration 1 results with the Pak & Lee Infiltration Limit Dynamic Surface method (LGA_CUH_LR_DyS_Event)

     

    Calibration 2 results with the Pak & Lee Infiltration Limit Dynamic Surface method (LGA_CUH_LR_DyS_R_Event)

  5. Run the LGA_CUH_LR_DyS_R simulation (17Jan2020 - 14Feb1019) using the Pak & Lee Infiltration Limit Dynamic Surface method.
  6. Examine the simulation results, then answer the questions below.


    Question 2: 

    What are the results for 9-year simulation?

    • Computed peak discharge (value, date, and time)
    • Total runoff volume
    • Nash–Sutcliffe Efficiency

    Based on these metrics, does the Layered G&A Method with the Pak & Lee Infiltration Limit Dynamic Surface method (LGA_CUH_LR_DyS_R) outperform the Layered G&A Method without Pak & Lee Infiltration Limit Dynamic Surface method (LGA_CUH_LR), and is it robust enough to support a nine-year continuous simulation?

    Both layered G&A models—with and without the Pak & Lee Infiltration Limit Dynamic Surface method—were evaluated against USGS Arroyo Seco gage data. The version with the Pak & Lee method delivers substantial improvement, particularly in the later years of the simulation, and closely matches the observed flows.

    Computed peak discharge for with Pak & Lee Infiltration Limit Dynamic Surface: 4,229 cfs on 18 Jan 2010 at 22:45
    Computed peak discharge for without Pak & Lee Infiltration Limit Dynamic Surface: 4,293 cfs on 28 Feb 2014 at 17:00
    Observed peak discharge: 4,620 cfs on 06 Feb 2010 at 13:45
    Total runoff volume: 31.36 in (with Pak & Lee Infiltration Limit Dynamic Surface) vs. 101.09 in (without Pak & Lee Infiltration Limit Dynamic Surface) vs. 42.45 in (observed)
    Nash–Sutcliffe efficiency: 0.552 (with Pak & Lee Infiltration Limit Dynamic Surface) vs. -0.966 (without Pak & Lee Infiltration Limit Dynamic Surface)

    These metrics demonstrate that the layered G&A model incorporating the Pak & Lee infiltration-limit dynamic surface method delivers high-quality, consistent performance over a nine-year continuous simulation, with computed results matching observed data particularly well in the later years.

    New Results for the Pak & Lee Infiltration Limit Dynamic Surface method

     Previous Results for the Layered G&A Method without Pak & Lee Infiltration Limit Dynamic Surface method

     

    Question 3: 

    What strategies can we employ to improve the model and ensure reliable long-term post-wildfire hydrology estimates?

    Although integrating the Pak & Lee infiltration‐limit dynamic surface method into the layered Green–Ampt framework markedly improves the nine-year continuous simulation, it remains insufficient for post-wildfire hydrology, as it models only soil recovery dynamics and omits canopy recovery after fire.

    Next, incorporate post‐fire recovery processes—such as dynamic canopy modules—to capture long‐term landscape changes. Based on this review, the following are recommended as minimum requirements for the long‐term post‐wildfire hydrology model:

    • Incorporate the Dynamic Canopy Method to adjust canopy interception and evapotranspiration in long-term continuous simulations. This approach accommodates time-varying crop coefficients—as seen during post-fire vegetation recovery—ensuring a more realistic representation of changing canopy conditions.

    Results from the layered Green–Ampt model with the Pak & Lee infiltration-limit dynamic surface method highlight the need for dynamic canopy recovery features in nine-year continuous simulations to support post-wildfire evacuation and maintenance planning. Further refinements are required to ensure reliable long-term hydrologic estimates.