POST-FIRE TREATMENT EVIDENCE · Reviewed 9 October 2026
Does post-fire mulching reduce erosion? Research supports reductions under studied conditions, particularly where dry mulch supplies ground cover to burned hillslopes. The result depends on the treatment, observation period and measured outcome. A claim about reduced sediment is not automatically a claim about reduced runoff or protection from every downstream hazard. [1] [3]
The first question is what protects the exposed ground
A Forest Service synthesis identifies exposed mineral soil, or its inverse ground cover, as a central treatment factor in post-fire hillslope erosion. It reviews dry mulch, hydromulch and erosion barriers, distinguishing materials that cover soil from barriers that provide little surface cover. [1]
Tecnobosque reading principle: identify the intervention that actually occurred. “Straw was applied” and “the slope retained protective cover during the storm” describe different observations. Save the material, application date, measured cover and subsequent condition before connecting a product name to a performance claim.
These studies concern emergency rehabilitation of burned hillslopes and catchments. Begin with soil burn severity and BAER assessment to understand the ground conditions behind a treatment decision.
Three study designs, three different comparisons
Hillslope plots across four US wildfires
Robichaud and colleagues measured sediment from natural rainfall on hillslope plots following four western US wildfires. Straw, wood strands and hydromulch were among the treatments, with untreated controls followed for four to seven years. Untreated sediment yields declined over time, while mulch persistence differed: wood strands remained detectable longer than straw, and hydromulch was not detected after the first post-fire year at the two fires where it was tested. [2]
That establishes the importance of documenting both recovery and treatment longevity. It does not establish that one commercial material will have the same lifetime on every slope.
Small catchments after Hayman and Cedar
A companion study monitored matched catchments for five to seven years after the 2002 Hayman and 2003 Cedar fires. Hayman responses involved short, intense summer storms; Cedar responses involved longer winter rainfall. At Hayman, wheat-straw mulch significantly reduced peak flow and sediment yield, but the statistical model did not find a significant treatment effect on runoff reduction. [3]
Outcome distinction: peak flow describes the highest discharge, runoff describes water leaving the catchment, and sediment yield describes exported material. A statistically significant result for one must not be copied into the other two. Read the Hayman case study for the watershed context.
Three experimental methods after Terrace Mountain, Canada
Following the 2009 Terrace Mountain Fire in British Columbia, researchers compared straw and wood shreds using rainfall simulation, concentrated-flow experiments and natural-rainfall plots. Both mulches reduced sediment in the first study year. Performance was sufficiently similar for the treatments to be combined in the analysis. Vegetation subsequently recovered similarly across treated and untreated plots. [4]
This comparison supports examining material characteristics and site needs rather than assuming that a different material name proves a different erosion outcome. It also demonstrates why simulation results and natural-storm measurements should retain their original labels.
What a ten-year study changes about the success claim
A later Hayman study compared treatments using simulated rill erosion immediately after fire and again ten years later, with additional observations of untreated recovery between those endpoints. The paper reports the following early sediment flux rates. [5]
| Experimental condition | Reported sediment flux |
|---|---|
| Burned control | 1.9 g/s |
| Contour scarification | 2.8 g/s |
| Agricultural straw mulch | 0.9 g/s |
| Wood mulch | 1.1 g/s |
These are experimental flux rates: grams of sediment per second. They are not annual catchment sediment yields. The study found lower early flux for the mulch treatments, but no statistically significant differences in flux among treatments at year ten. [5]
Tecnobosque interpretation: a later convergence does not erase a measured early benefit. Conversely, early differences cannot establish permanent superiority. Nor does the difference between 0.9 and 1.1 alone establish that straw outperformed wood statistically.
Worked example: normalise sediment by the measured area
Invented numbers—not measurements from the cited studies. Suppose collectors record dry sediment from two bounded plots during the same observation interval:
| Plot | Plot area | Dry sediment mass | Yield |
|---|---|---|---|
| Untreated | 20 m² | 12 kg | 12 ÷ 20 = 0.60 kg/m² = 6 Mg/ha |
| Treated | 30 m² | 9 kg | 9 ÷ 30 = 0.30 kg/m² = 3 Mg/ha |
The raw mass is 25% lower, but the yield per unit area is 50% lower. A hectare contains 10,000 m²; one Mg is 1,000 kg, so 1 kg/m² corresponds to 10 Mg/ha. Uppercase Mg means megagram, not milligram.
Area normalisation fixes this accounting difference. It does not make the plots equivalent in slope, rainfall exposure, soil or starting condition, and it does not turn a single observation into a causal treatment estimate. Keep the observation interval visible; an event yield is not an annual yield.
Worked example: a before-and-after decline includes recovery
A separate hypothetical annual-yield example: untreated land produces 10 Mg/ha in year one and 4 Mg/ha in year two, a 60% decline without treatment. Another treated area produces 2 Mg/ha in year two.
Comparing 2 with the previous year’s 10 gives an 80% decline. Comparing the two areas during year two gives 50% less sediment on the treated area: (4 − 2) ÷ 4 = 50%. Those are different comparisons.
Neither percentage alone establishes the causal effect. The between-area comparison still requires a defensible design and uncertainty assessment; the before-and-after comparison mixes treatment with changes in weather and terrain condition. Replication, comparable controls and analysis of storm characteristics determine what the observations can support.
Six checks before quoting a mulch-effectiveness percentage
Tecnobosque editorial checklist: attach these details to the reported result.
- Material and condition: name the mulch and record how much protective cover remained when the response was measured.
- Comparison: identify untreated controls, their starting conditions and the number of independent experimental units.
- Scale: distinguish rainfall-simulation plots, natural-rainfall plots, hillslopes and catchments. Preserve the measurement boundary.
- Storm and period: record the rainfall measure, observation interval and time since fire. Do not silently combine event and annual totals.
- Outcome and units: preserve runoff, peak flow, sediment mass, yield or flux as actually measured. Include area normalisation where applicable.
- Uncertainty and scope: retain statistical findings and study limitations. A non-significant result is not proof of exactly zero effect.
A bounded conclusion: “Under [study conditions], [treatment] changed [measured outcome and units] relative to [comparison] during [period], with [reported uncertainty].” Fill in the measured fields before adding an explanation about why it worked.
For downstream processes, continue with the debris-flow evidence guide. Hillslope erosion reduction is one part of a wider assessment; an individual treatment percentage does not define a downstream hazard boundary.
Primary studies and a Forest Service synthesis
- Robichaud, Ashmun and Sims (2010): Post-fire treatment effectiveness for hillslope stabilization. Forest Service synthesis, RMRS-GTR-240. DOI: 10.2737/RMRS-GTR-240.
- Robichaud and colleagues (2013): Post-fire mulching, Part I—effectiveness at reducing hillslope erosion rates. Catena 105, 75–92. DOI: 10.1016/j.catena.2012.11.015.
- Robichaud and colleagues (2013): Post-fire mulching, Part II—runoff and sediment yields from small catchments. Catena 105, 93–111. DOI: 10.1016/j.catena.2012.11.016.
- Robichaud and colleagues (2013): Wood shred and agricultural straw mulches after wildfire in southern British Columbia. Geomorphology 197, 21–33. DOI: 10.1016/j.geomorph.2013.04.024.
- Robichaud and colleagues (2020): Quantifying long-term post-fire sediment delivery and erosion mitigation effectiveness. Earth Surface Processes and Landforms 45, 771–782. DOI: 10.1002/esp.4755.