Wildfire is a natural disturbance, though elemental losses and changes that occur during combustion and post-fire erosion can have long-term impacts on soil properties, ecosystem productivity, and watershed condition. Here we evaluate the potential of forest residue-based materials to rehabilitate burned soils. We compare soil nutrient and water availability, and plant recovery after application of 37 t ha−1 of wood mulch, 20 t ha−1 of biochar, and the combination of the two amendments with untreated, burned soils. We also conducted a greenhouse trial to examine how biochar influenced soil nutrient and water content under two wetting regimes. The effects of wood mulch on plant-available soil N and water content were significant and seasonally consistent during the three-year field study. Biochar applied alone had few effects under field conditions, but significantly increased soil pH, Ca, P, and water in the greenhouse. The mulched biochar treatment had the greatest effects on soil N and water availability and increased cover of the most abundant native plant. We found that rehabilitation treatments consisting of forest residue-based products have potential to enhance soil N and water dynamics and plant recovery following severe wildfire and may be justified where erosion risk or water supply protection are crucial.
High-severity wildfires can cause significant, lasting impacts on forest soils and watersheds [
Managers are working on US Forest Service land use Burned Area Emergency Response (BAER) treatments to counter immediate post-fire soil erosion losses [
Ecosystem restoration is enhanced by greater understanding of how disturbance and subsequent rehabilitation treatments alter soil processes and properties [
Experimental trials that evaluate how both established and novel rehabilitation treatments influence post-fire soil nutrient and water availability and plant establishment are required in order to develop guidelines for rehabilitating severely burned landscapes. Here we investigate the potential of wood mulch and biochar created from lodgepole pine
Research was conducted on the Arapaho-Roosevelt National Forest near Fraser, Colorado, in forests burned by the October 2010 Church’s Park fire (39°56′25′′N; 105°57′00′′W). The fire burned a 200-hectare area with 17%, 30%, and 53% classified as high, moderate, and low fire severity, respectively (E. Schroder, USDA USFS BAER report). The prefire forest was dominated by mountain pine bark beetle-killed lodgepole pine with patches of quaking aspen
Post-fire rehabilitation treatments were compared in areas that burned at high severity and had visual evidence of post-fire erosion and sparse plant recovery. In June 2014, six replicate blocks of 5 × 5 m treatment plots were established on randomly selected sites that had relatively similar prefire tree species composition (>75% lodgepole pine), slope (5–15%), and aspect (south-facing). Rehabilitation treatment comparisons included (1) wood mulch, (2) biochar, (3) biochar with wood mulch, and (4) untreated burned conditions. Treatment plots were randomly assigned within each block and arranged parallel to the slope contour.
Surface amendments were designed to evaluate the potential of forest residue amendments to alter post-fire soil nutrient availability and water relations. Both woodchip and biochar treatments were created from small diameter, beetle-killed lodgepole pine. Biochar was added at an application rate of 20 t ha−1 and hand raked into the upper 2-3 cm of mineral soil. Woodchip mulch was applied to create a 2 cm deep surface layer, equivalent to 37 t ha−1. Mulch was applied above the biochar in the combined treatment. A small hand-dug trench was created upslope of the plots to reduce surface runoff reaching the experimental plots.
The study biochar was created from oven dry, lodgepole pine chips (8–10% water content) using a two-step pyrolysis process that combined an O2-limited step (700–750°C, <1 minute) followed by an O2-free step (400–550°C, 10–15 minutes). Pyrolysis was conducted by Biochar Engineering Corporation (BEC), formerly of Golden CO. Biochar consisted of 87.2% carbon (C), 1.4% 0, 0.4% N, 9.4% ash, and 1.1% water. The biochar had a pH of 9.4, surface area of 176 m2 g−1, and total pore volume of 0.11 cm3 g−1 [
We compared the effects of the rehabilitation treatments on soil nutrients, water content, and plant recovery over the course of 3 years following treatment establishment. We assessed the effects of rehabilitation treatments on plant-available soil nitrogen and potential nitrate leaching using ion exchange resin (IER) bags [
In August 2016, two 10 cm deep soil cores were collected and composited from each treatment plot. We measured net N mineralization and nitrification at the end of the third growing season using aerobic laboratory incubations to estimate how the treatments influenced the production of inorganic soil N [
We also characterized water-soluble nutrients and C released from soil sampled at the end of the field study in a 5 g of soil (<2 mm size) and 100 mL of deionized water mixture. Samples were agitated for one hour, settled for 24 hours, and agitated for a second one-hour period. Samples were then filtered through 0.45
We measured the volumetric soil water content (0–10 cm depth) twice monthly during the 2014, 2015, and 2016 growing seasons (June–August) using a hand-held, time domain reflectometry probe (CD 620, HydroSense Campbell Scientific, Logan, UT). For each sample date, five mineral soil water values were recorded per plot beneath surface mulch or organic soil layers.
We evaluated the effect of rehabilitation treatments on plant, mineral soil, litter, and rock cover in August 2016 with a gridded point-intercept method in 1 m2 sample quadrats. Plant and surface cover sampling was conducted and plant nativity was classified according to the USDA NRCS Plants Database [
We also isolated the effects of biochar on water content, nutrients, and chemistry of soils from the Church’s Park burn under controlled greenhouse conditions. We compared two biochar levels (0 and 20 t biochar ha−1) under two wetting regimes (average and dry) during a six-month trial. The biochar treatment was equal to the field application rate. The dry and average wetting treatments received 4 and 8 cm of water per month, respectively. The wetting amounts were based on local long-term summer precipitation records [
We compared the cumulative effects of the four field-scale rehabilitation treatments in August 2016 using one-way analysis of variance (SPSS V. 22, IBM CO, Chicago, IL). To compare treatment effects on multiple measurements of IER-N and volumetric soil water content we used a mixed model, repeated measures analysis of variance. In the greenhouse trial, biochar and wetting treatments were fixed effects in a two-way analysis of variance. Levene’s statistic was used to test homogeneity of variance and data were log-transformed prior to conducting analysis of variance with that corrected normality or unequal variance. Statistical significance was assigned for
Mulch reduced the pool of total plant-available N measured with ion exchange resins and the proportion comprised of nitrate during the first and second snowmelt periods after treatment establishment (Figure
Ion exchange resin soil nitrogen comparing post-wildfire rehabilitation treatments at the Church’s Park burn, Colorado. Ion exchange resin bags were installed in mineral soil (5–10 cm depth) in September and removed in early June to sample nutrients percolating in spring snowmelt. Bars show means and standard errors of six replicate treatment blocks. Letters denote significant differences among means at
The rehabilitation treatment effects on KCl-extractable soil N measured once after three growing seasons (Table
Soil properties and net N incubations 3 years after establishment of rehabilitation treatments at the Church’s Park fire, Colorado. Treatments include lodgepole pine-derived biochar and wood chip mulch, their combination, and untreated, severely burned soils. Data are means and standard error in parentheses (
Control | Biochar | Mulch | Mulch + biochar | | | |
---|---|---|---|---|---|---|
| 5.7 | 5.8 | 5.7 | 6.4 | 4.6 | 0.014 |
| 5.3 (0.16) | 5.1 (0.10) | 5.1 (0.17) | 5.7 (0.20) | 2.7 | 0.073 |
NH4-N (mg N/kg) | 1.6 | 1.2 | 0.6 | 0.2 | 3.5 | 0.032 |
NO3-N (mg N/kg) | 0.4 | 0.5 | 0.1 | 0.1 | 4.3 | 0.015 |
Total N (g N/kg) | 0.8 | 0.7 | 0.8 | 1.0 | 6.6 | 0.002 |
Total C (g N/kg) | 17.4 | 19.3 | 20.2 | 24.5 | 3.5 | 0.031 |
C : N | 21.4 | 29.6 | 24.3 | 23.8 | 6.3 | 0.003 |
Net mineralization (mg N/kg/28 d) | −0.7 | −0.4 | 0.4 | 0.6 | 3.5 | 0.032 |
Net nitrification (mg N/kg/28 d) | −0.2 | −0.3 | 0.1 | 0.2 | 5.1 | 0.007 |
Letters denote significant differences among treatment means at
Nitrogen analyzed in water-soluble extracts of mineral soils collected at the end of the study further confirmed the treatment effects measured by IER-N bags (Figure
Individual and combined rehabilitation treatments influenced growing season volumetric water content (Figure
Volumetric soil water content (0–10 cm depth) under post-wildfire rehabilitation treatment at the Church’s Park burn, Colorado. Soil water was measured at 6 replicate blocks of treatments, 6 times per year during 2014, 2015, and 2016. Bars show means and standard errors of six replicate treatment blocks. Letters denote significant differences among means at
In summer of 2016, five growing seasons after the Church Park fire, total plant cover averaged 38% on untreated plots and consisted primarily of forbs (29%) with lesser amounts of graminoid (5%) and shrub (4%) species. The rehabilitation treatments had no general effects on the cover of plant functional groups (i.e., forbs, graminoids, and shrubs), but had specific effects on the most common species. Fireweed
Total forb and fireweed
At the end of our six-month greenhouse trial, Church’s Park soil mixed with 20 t ha−1 of biochar had higher concentrations of most exchangeable nutrients and higher soil pH and gravimetric water (Table
Soil properties after a six-month greenhouse trial with lodgepole pine-derived biochar and soil from areas affected by the Church’s Park fire, Colorado. Data are means and standard error in parentheses (
Biochar | Watering regime | | |||||
---|---|---|---|---|---|---|---|
t/ha | Dry | Wet | Biochar | Water | Char × water | ||
Gravimetric water content | (%) | 0 | 12.1 (0.7) | 20.5 (0.6) | <0.001 | <0.001 | <0.001 |
20 | 21.7 (1.0) | 23.0 (0.9) | |||||
| — | 0 | 5.6 (0.05) | 5.6 (0.05) | <0.001 | 0.570 | 0.731 |
20 | 5.9 (0.06) | 6.0 (0.08) | |||||
| — | 0 | 4.5 (0.07) | 4.6 (0.03) | <0.001 | 0.128 | 0.401 |
20 | 4.9 (0.04) | 5.0 (0.06) | |||||
NO3-N | (mg/L) | 0 | 1.8 (0.3) | 0.8 (0.2) | <0.001 | 0.002 | 0.297 |
20 | 7.9 (0.7) | 5.8 (0.5) | |||||
NH4-N | (mg/L) | 0 | 0.3 (0.1) | 0.5 (0.1) | 0.524 | 0.231 | 0.441 |
20 | 0.3 (0.1) | 0.3 (0.1) | |||||
P | (mg/L) | 0 | 36.0 (1.3) | 38.2 (1.0) | <0.001 | 0.520 | 0.026 |
20 | 49.5 (0.6) | 45.7 (1.9) | |||||
K | (mg/L) | 0 | 23.6 (1.2) | 22.5 (0.9) | <0.001 | 0.009 | 0.115 |
20 | 30.5 (0.9) | 26.2 (0.9) | |||||
Mg | (mg/L) | 0 | 16.6 (1.6) | 16.6 (1.7) | 0.253 | 0.883 | 0.888 |
20 | 18.7 (1.6) | 18.2 (1.5) | |||||
Ca | (mg/L) | 0 | 99.5 (4.0) | 95.7 (3.7) | 0.001 | 0.166 | 0.785 |
20 | 112.2 (2.7) | 106.6 (2.8) |
This project compared rehabilitation treatments designed to improve soil conditions and speed native plant recovery after high-severity wildfire. The treatments, in particular those including mulch, had significant effects on soil water and nutrient relations (Figures
Biochar has been widely promoted as a soil amendment to improve plant nutrient and water availability [
Our greenhouse trial demonstrated substantial effects of biochar on soil nutrient and water availability in a controlled environment. The 20 t ha−1 biochar treatment increased soil water, pH, nitrate, P, Ca, and K (Table
Like numerous other studies [
Where applied for restoration of burned soils, mulch can both stimulate and suppress native plants [
For most the factors we measured, biochar combined with a wood mulch layer had the greatest effects, consistently exceeding those of biochar alone. This agrees with findings that biochar treatments are enhanced by addition of fertilizer, microbial inocula, or organic amendments [
The increase in understory cover that occurred during this study indicates that unassisted post-fire recovery is well underway. Nevertheless, bare soil remains high and shrub cover remains negligible (<4%) in the control plots, and tree regeneration was absent in the study treatments and surrounding areas. The timing of the Church’s Park fire, 5–8 years after bark beetles killed most of the overstory lodgepole pine, has delayed forest regeneration compared to the profuse and rapid establishment and growth of tree seedlings in nearby unburned and salvage-logged stands [
Immediate post-fire treatments aimed at avoiding soil nutrient and organic matter losses have the combined benefit of supporting ecosystem productivity while limiting nutrient export and potential water quality degradation. Our work began 2 years after the Church’s Park fire; the treatment effects may have been greater had rehabilitation been conducted within months of the fire, as is typical of the BAER program [
This paper was written and prepared by US Government employees on official time, and therefore it is in the public domain and not subject to copyright.
The authors declare that there is no conflict of interests of any sort regarding the publication of this paper. Specifically, there is no conflict of interests with the sources of financial support.
Research was supported by the US Forest Service, Region 2–BAER Program, Rocky Mountain Research Station, Colorado State University, Department of Soil and Crops and Colorado Agricultural Experiment Station. Thanks are due to Jonah Levine (Biochar Solutions, Inc.), for manufacturing the study biochar. The authors gratefully acknowledge field and laboratory assistance from Corey Hanson, Roger Tyler, Sam Block, Ryan Davies, Clayton Bliss, and Aisha Jama. Comments from Susan Miller greatly improved the manuscript.