ORIGINAL PAPER

Soil microbial response following wildfires in thermic oak-pine forests

Michael S. Huffman1& Michael D. Madritch1

Received: 15 May 2018 /Revised: 30 September 2018 /Accepted: 4 October 2018 /Published online: 12 October 2018# Springer-Verlag GmbH Germany, part of Springer Nature 2018

AbstractThe ecosystem response to wildfire is often linked to fire severity, with potentially large consequences for belowground biogeo-chemistry and microbial processes. While the impacts of wildfire on belowground processes are generally well documented, itremains unclear how fire affects the fine-scale composition of microbial communities. Here, we investigate the composition ofsoil bacterial and fungal communities in burned and unburned forests in an attempt to better understand how these diversecommunities respond to wildfire. We explored the belowground responses to three wildfires in Linville Gorge, NC, USA.Wildfires generally increased soil carbon content while simultaneously reducing soil respiration. We employed amplicon se-quencing to describe soil microbial communities and found that fires decreased both bacterial and fungal diversity. In addition,wildfires resulted in significant shifts in both bacterial and fungal community composition. Bacterial phylum-level distributionsin response to fire were mixed without clear patterns, with members of Acidobacteria being representative of both burned andunburned sites. Fungal communities showed consistent increases in Ascomycota dominance and concurrent decreases inBasidiomycota and Zygomycota dominance in response to burning. Indicator species analysis confirmed shift to Ascomycotain burned sites. These shifts in microbial communities may reflect differences in the quality and quantity of soil organic matterfollowing wildfires.

Keywords Wildfire . Soil microbial community . Amplicon sequencing

Introduction

As wildfire frequency continues to increase, in part, due toanthropogenic climate change (Abatzoglou and Williams2016), it is imperative to understand how fire disturbances willimpact the belowground microbial communities that are re-sponsible for driving essential ecosystem processes. Soil mi-crobial communities directly influence soil respiration, organ-ic matter decomposition, and nutrient cycling (Buchmann2000; Cisneros-Dozal et al. 2006; Dooley and Treseder2012; Hanson et al. 2000). Microbial communities are partic-ularly sensitive to changes in the physical, chemical, and bio-logical properties of soils (Ginzburg and Steinberger 2012;Hernandez et al. 1997; Neary et al. 1999; Van Der Heijdenet al. 2008). Forest soil respiration is of particular importanceto climate change as it constitutes a large terrestrial flux of C tothe atmosphere (Dixon et al. 1994; Raich and Schlesinger1992). Consequently, small changes in soil respiration rates

in response to disturbance events could have large impacts onatmospheric CO2 concentrations (Bond-Lamberty et al. 2004;Raich and Schlesinger 1992). Moreover, concrete links be-tween microbial community composition and diversity withbiogeochemical function remain elusive (Prosser 2012).

Wildfires are an important mechanism of disturbance andalter many belowground processes including soil respiration(Certini 2005; González-Pérez et al., 2004; Knoepp et al.2005; Neary et al. 1999; Wang et al. 2012). The impacts ofwildfires on belowground systems are primarily the result ofthe direct heat transfer to soils, combustion of living plantmaterial, reduction of soil organic matter, increased erosionof topsoil, and loss of fine root biomass (Certini 2005;Gonzales-Perez et al. 2004). The response of an ecosystemto wildfire is influenced by fire severity that is often heteroge-neous across a burn area due to variations in topography, mi-croclimate, fuel moisture levels, and a suite of other environ-mental factors (DeBano et al. 1998). Typically, low to moder-ate severity fires do not transfer large amounts of heat energybelow the surface of the soil, leaving much of the rhizosphereintact, promoting re-sprouting of vegetation and recoloniza-tion of the soil by microflora that survived the burn (Nearyet al. 1999). In contrast, intense fires can be detrimental tobelowground systems, leading to long-lived and sometimes

* Michael D. Madritchmadritchmd@appstate.edu

1 Department of Biology, Appalachian State University,Boone, NC 28608, USA

Biology and Fertility of Soils (2018) 54:985–997https://doi.org/10.1007/s00374-018-1322-5

permanent alterations (Neary et al. 1999). This unpredictabil-ity makes it challenging to generate general conclusions aboutthe effects of wildfire, highlighting the need to examine short-and long-term ecosystem response to wildfires across a rangeof fire severities and spatial scales (Certini 2005; Gonzalez-Perez et al. 2004; Knoepp et al. 2005; Restaino and Peterson2013).

Wildfires were once a frequent disturbance in the southernAppalachian Mountains that shaped the vegetation communi-ty composition, soil environment, and ecological dynamics ofthese forests (Hart et al. 2005; Newell and Peet 1998).Combustion of the organic layer can have a fertilizing effectdirectly following a wildfire, transforming large quantities ofstored C (Gonzalez-Perez et al. 2004) and increasing availableN in the uppermost soil layers (Certini 2005). Several studieshave also demonstrated a pronounced negative effect of fireon belowground microbial biomass (Banning and Murphy2008; Dooley and Treseder 2012; Mikita-Barbato et al.2015). Prieto-Fernandez et al. (1998) found that microbial Cand N decreased in response to burning whereas extractable Cand N contents both increased immediately following a wild-fire. Consequently, while fires may increase C and N in re-cently burned soils, these increases do not always stimulatemicrobial activity. Decreased soil respiration in response tofire may be caused by reduced soil moisture capacity and/ordiminished root and microbial activity (Hernandez et al. 1997;Wang et al. 2012). Variable microbial responses to fires couldbe due to the microbial responses to both fire severity andaboveground plant community (Weber et al. 2014).

While wildfires can stimulate aboveground primary pro-duction and maintain aboveground biodiversity (Swansonet al. 2011), the belowground impacts of wildfires can reducemicrobial activity, biomass, and abundance (Dooley andTreseder 2012). The specific effects of fire on the fine-scalecomposition of belowground communities remains largelyunknown. Here, we examine the biogeochemical and micro-bial responses to wildfires in the Linville Gorge of the south-ern Appalachian Mountains. We measured biogeochemicalresponses including common enzyme activity potentials inan effort to describe microbial function and employedamplicon sequencing to describe the microbial communitycomposition. Our goal was to quantify the potential shifts inmicrobial communities that may accompany the belowgroundbiogeochemical responses to forest fires.

Methods

Study site

Linville Gorge is located in North Carolina, USA, within thePisgah National Forest. Elevations range from 400 m at theriverbed to 1250 m on the upper ridges. Precipitation is

highest in the summer months and averages 1250 to1625 mm annually. Peak temperatures occur during June toAugust (14–17 °C average minimum, 21–27 °C average max-imum) with the coolest temperatures occurring in February (−2 to 0 °C average minimum, 8–12 °C average maximum;Newell and Peet 1998). Upper ridgelines and summits haveTypic or Lithic Dystrochrepts soils in the Ashe, Buladean, andChestnut soil series. All three soil series are strongly to mod-erately acidic and have very high to medium water infiltrationrates and low cation exchange capacities (Knight 2006).

Most of Linville Gorge has not experience logging due toits rugged topography. Thermic oak-pine communities domi-nate the upper ridges and slopes where this study was con-ducted. Historically, these communities experienced frequent,uncontrolled wildfires, with 7–12 year mean fire intervals(Newell and Peet 1998). However, fire suppression practicesled to the exclusion of large, uncontrolled wildfires prior to2000. Since 2000, large wildfires have occurred in the LinvilleGorge including the Brushy Ridge Fire in 2000 (BR-2000),the Pinnacle Fire in 2007 (PNCL-2007), and more recently theTable Rock Fire in 2013 (TR-2013).

Sampling methods

The study region consisted of three areas that were exposed tomoderate to severe wildfires. The definition of Bfire severity^varies widely among publications (Lentile et al. 2006;Perrakis and Zell 2008). Here, severe wildfires are those thatcause mortality of large, overstory trees. Moderate wildfiresreplace understory species and leave visual scars on overstorytrees but are not stand-replacing fires. Both leave blackenedsoils and apparent ash deposition. Each burned sampling areawas paired with an adjacent unburned control area that did notexperience a wildfire (6 sites). We identified sampled areasusing burn perimeter maps provided by the US ForestService supported by visual indicators of a wildfire occur-rence. Visual indicators included standing and fallen trees thathad visible burn scars, blackened soils, and ash deposition.Each sampling site was approximately 50 × 50 m, with ten2-m diameter circular plots located randomly within the pe-rimeter of each of the six sites. Burned and control sites werelocated within the perimeters of the Brushy Ridge (BR-2000),Pinnacle (PNCL-2007), and Table Rock (TR-2013) fires.Control sites that lacked any visual indicators of severe fireswere located < 100 m from burned sites. We do not have pre-burn information on any sites used in this study. However,using visual indicators to identify burned and unburned con-trol sites is common in fire studies (e.g., Holden and Treseder2013; Holden et al. 2016). In addition, we employed a pairedsampling scheme such that each burned site was paired with anearby unburned site in an attempt to account for spatial var-iation in soil properties and aboveground plant communities.

986 Biol Fertil Soils (2018) 54:985–997

Soil biogeochemistry

All sampling took place during the last week of July and firstweek of August 2014, during peak soil respiration and micro-bial activity (Raich and Schlesinger 1992). Paired sites weresampled over subsequent days without rainfall to minimizetemporal variation in temperature and moisture. For each plot,soil respiration and temperature measurements were recordedat the center using a Li-COR 8100, yielding 60 measurementsof soil respiration and temperature. In situ soil moisture wasestimated with a Hydrosense soil moisture meter (CampellScientific, USA).

Directly following respiration measurements, five soil sam-ples were collected from each plot using a 2 × 15-cm soil corerto a depth of 15 cm. Between each sample, the corer wascleaned with bleach and ethanol to reduce the possibility ofmicrobial cross contamination. Cores were taken at the centerof each subplot and at four equally spaced locations on thediameter of the 2-m plot. The five cores from each plot werepooled into a single sample, yielding 60 soil samples. Soilswere stored in sterile bags, transported on ice, and stored at −20 °C until further analysis. Soils were thoroughly mixedwithin the sterile bags, but not sieved prior to DNA extractions(below) in order to reduce the possibility of cross contamina-tion. After subsampling for DNA extraction, soils were sievedto 2-mm subsampled again for enzyme activity potentials(below) and then freeze-dried to determine pH, total carbonC, and N. Soil pH was determined by adding 10 g dried soil to20mLH2O. Total C andN content of soils was determined viacombustion analyses on a FlashEA 1112 NC Analyzer(Thermo Fisher Scientific, Waltham, MA).

Enzyme activity

In an effort to characterize the functionalmicrobial response towildfires, we measured the potential activity of six soil en-zymes in burned and unburned soils: cellobiohydrolase andβ-glucosidase (involved with the degradation of cellulose),leucine aminopeptidase (involved with the degradation of pro-teins), phenol oxidase and peroxidase (involved with the deg-radation of aromatic compounds), and urease (hydrolyzesurea). Enzyme assays were based on protocols by Carreiroet al. (2000) and Saiya-Cork et al. (2002) and are describedin detail by Madritch et al. (2007). Composite soil sampleswere sieved at 2 mm, mixed, and duplicate subsamples of 2 gof equivalent dry mass were blended in 15 ml 50 mM acetatebuffer at room temperature using steel balls and a modifiedpaint shaker to create a soil suspension. Aliquots of soil sus-pensions were then immediately aliquoted with wide-mouthedpipettes into duplicate 2-ml microcentrifuge tubes for each ofthe six enzyme assays as well as a set for sample blanks.Cellobiohydrolase, β-glucosidase, and leucine aminopepti-dase assays were conducted with the p-nitrophenol (pNP)-

labeled substrates 4-pNP-β-D-cellobioside, pNP-β-glucopyranoside, and leucine p-nitroanilide, respectively.Soil suspensions and substrates were allowed to react for 2 hat 29C, then centrifuged at 1800×g for 10 min to separate soilparticles. Sample supernatant was removed and aliquoted intriplicate into 96-well microplates, amended with NaOH, andthen read at 410 nm with a spectrophotometer. A separate p-nitrophenol standard curve was used for each incubationbatch. Phenol oxidase and peroxides activities of each extractwere estimated in triplicate by measuring L-DOPA metabo-lism with and without H2O2. Soil extracts and substrates wereallowed to react for 3 h at 29 °C before centrifuging as above.We used a purified horseradish peroxidase standard curve as areference and read absorbance at 460 nm. Urease was mea-sured in triplicate by analyzing soil extract ammonium con-centration before and after a 2-h incubation using urea as asubstrate. Ammonium concentrations were determined color-imetrically using assays based on the indophenol blue methodmodified by Mulvaney (1996) and using sodiumdichloroisocyanurate as a hypochlorite source. The enzymeassays employed here have limitations in that they measurepotential enzyme activity instead of actual in situ enzymeactivities (Nannipieri et al. 2018).

DNA extraction and amplification

Soil DNAwas extracted from ~ 0.5 g of pooled soil samplesusing MPBIO FastDNA™ SPIN Kit for Soils (MPBiomedical, Solon, OH) according to the manufacturer’s in-structions. Partial bacterial 16S rRNA genes were amplifiedfrom composite DNA samples by polymerase chain reaction(PCR) using primers 515f and 806r (Caporaso et al. 2011).Fungal ITS1 regions were amplified with ITS1F and ITS2primers (Smith and Pea, 2014). To yield 25 μL reaction mix-tures, 1 μL of ~ 20 ng/μL DNA template, 1 μL of each 5 μMprimer, 12 μL of Nuclease-Free water (Quiagen, Hilden,Germany), and 10 μL Q5 High-Fidelity 2X Master Mix(New England BioLabs, Ipswich, MA) were added to eachreaction mixture. Bacterial PCR cycling parameters were94 °C for 180 s, followed by 35 cycles of 94 °C for 45 s,55 °C for 60 s, and 72 °C for 90 s, followed by 600-s extensionat 72 °C. Fungal PCR cycling parameters were 95 °C for10 min, followed by 30 cycles of 95 °C for 30 s, 55 °C for20 s, and 72 °C for 00 s, followed by 8-min extension at 72 °C.Positive and negative controls were included with each PCRreaction set. All PCR reactions were run in duplicate thencombined before quantification.

Amplified products were gel purified with a 1% agarose geland quantified using the Molecular Imager Gel Doc XR sys-tem (Bio-Rad Laboratories, Hercules, CA) and again withPicoGreen (Invtirogen, Paisley, UK). PCR products werecombined in equimolar concentrations and sent to theGenomics Core Facility in Manchester, WV, for sequence

Biol Fertil Soils (2018) 54:985–997 987

analysis with the Illumina MiSeq platform (Illumina Inc., SanDiego, CA).

Sequence processing

Bacterial IlluminaMiSeq pair-end reads were processed usingMOTHUR version 1.35.1 using the MOTHUR MiSeq stan-dard operating procedure (Schloss et al. 2011; Kozich et al.2013). We screened joined sequences and remove any se-quence that was longer than 275 bp, contained at least oneambiguous base, and/or had > 8 nt homopolymers.Sequences were further screened for singletons. Aligned andscreened sequences were clustered at 97% similarity and OTUclassification of unique sequences was completed using theRibosomal Database Project (RDP) taxonomic database(release 9; Cole et al. 2009). To account for variation in thenumber of sequence reads per sample, the number of sequencereads in each sample was rarefied to 10,000, which removedtwo samples from further bacterial analysis due to poor sam-pling depth. Fastq files are stored with NCBI (SRA accession:SRP078568).

Fungal Illumina MiSeq forward and reverse reads wereassembled using MOTHUR (version 1.35.1, Schloss et al.2009). Primers were removed with cutadapt (Martin, 2011).For quality filtering, we discarded reads with 9 or more ho-mopolymers and any reads less than 125 bp (Schloss et al.,2009). Further analyses were done in QIIME (Caporaso et al.2010). Reads were filtered for singletons and chimeras usingde novo chimera function in VSEARCH (Rognes et al. 2016).Sequences were clustered at 95% similarity. Taxonomies wereassigned using BLAST (Altschul et al. 1990) againstthe UNITE database (v7, Koljalg et al. 2013). To account forvariation in the number of sequence reads per sample, thenumber of sequence reads in each sample was rarefied to1000, which removed five samples from further fungal anal-ysis due to poor sampling depth. Fastq files are stored withNCBI Sequence Read Archive (SRP078568).

Statistical analyses

Shannon diversity indices for both bacterial and fungal com-munities we calculated using the vegan package (Oksanenet al. 2013) in R (R Development Core Team 2013). As yearof burn is confounded with site, we concentrated our analyseson determining the belowground effects of fires but do notepatterns associated with time since burn. We employed nestedanalysis of variance (ANOVA) with fire treatment (burned orunburned) nested within area (TR-2013, PNCL-2007, or BR-2000) to determine the effect of fire and site on soil biogeo-chemistry and microbial diversity metrics.

The composition of bacterial and fungal communities wereanalyzed in parallel. To determine the effect of both site andwildfire on microbial communities, we performed

PERMANOVAs with burn treatment nested within site usingvegan (Oksanen et al. 2013). We generated NMS scores usingvegan (Oksanen et al. 2013) to visualize community differ-ences as a complimentary analysis to the PERMANOVAs.Microbial community composition was summarized withstacked barplots for both bacterial and fungal communitiesvia Phyloseq (McMurdie and Holmes 2013). Lastly, we per-formed nested indicator species analysis (PC-ORD v.6 MJMSoftware) to determine which bacterial and fungal OTUs wereassociated with burn and unburned treatments.

Results

Biogeochemistry

Both site and exposure to wildfire influenced soil respiration,soil C, and soil N (Fig. 1a–c). Wildfires reduced soil respira-tion across all sites with the largest reduction occurring at themost recently burned site TR-2013(Fig. 1a). Wildfires consis-tently increased soil C (Fig. 1b), and the increase in soil C waslargest in the older fire sites. Soil N at BR-2000 and PNCL-2007 roughly mirrored trends in soil C, with burned sitescontaining higher levels of soil N than did the unburned sites.However, the more recently burned TR-2013 plots had lowersoil N than did TR-2013 unburned plots (Fig. 1c).

Soil temperatures in recently burned sites were higher thanin unburned sites (Fig. 1d). Soil moisture varied widely acrosssites and less so between burn treatments (Fig. 1e). Soil pHalso varied by both site and burn treatment, with burned plotshaving higher pH values than did unburned plots (Fig. 1f).

Of the six enzyme activity potentials analyzed, only ureaseand phenol oxidase activity showed a response to either site orexposure to wildfire. Urease activity decreased in response toburning. BR-2000 exhibited the highest urease activity levels,followed by TR-2013, and PNCL-2007 (Fig. 1g). Variation inphenol oxidase activity across sites and increased slightly inresponse to exposure to wildfire (Fig. 1h).

Microbial community analysis

Approximately 2 million bacterial and 1.2 million fungal se-quences passed quality filtering and were used for furtheranalysis. A total of 52,564 bacterial operational taxonomicunits (BOTUs) were observed among all samples with anaverage of 1644 BOTUs observed per sample. A total of35,585 fungal operational taxonomic units (FOTUs) were ob-served among all samples with an average of 791 FOTUsobserved per sample. The effect of site and exposure to wild-fire on bacterial diversity was variable (Fig. 2a). Bacterialdiversity decreased in two of the three sites in response toburning, while fungal diversity decreased in all sites in re-sponse to burning (Fig. 2B).

988 Biol Fertil Soils (2018) 54:985–997

Nested PERMANOVA results indicate that both bacterialand fungal community composition was influenced by siteand exposure to wildfire (Table 1). Nonmetric multidimen-sional scaling (NMDS) analyses describe community com-position between burn treatments across all sites. NMDSplots are complementary to PERMANOVA results and showmoderate clustering of burn treatments, but with largeamounts of overlap indicating that both burn and unburnedsoils contained many of the same bacterial and fungal OTUs(Fig. 3).

Phylum-level distributions show differences in microbialcommunities between burn and unburned sites (Fig. 4).Bacterial communities at all locations were dominated bymembers of the Acidobacteria and Proteobacteria phyla, withActinobacteria, Bacteroidetes, Planctomycetes, andVerrucomicrobia comprising minor fractions. Roughly 14%of the OTUs were unclassified. Bacterial responses to bothsite and exposure to wildfire were mixed and showed littleconsistent changes across sites or burn treatment. Fungal com-munities were dominated by member of the Ascomycota and

Fig. 1 a–h Biogeochemical responses to wildfire in Linville Gorgevaried among sites and between burn treatments. Bars represent the

mean of 10 subplots and error bars represent standard error. Statisticalresults are from nested ANOVA of burn treatment within site

Biol Fertil Soils (2018) 54:985–997 989

Zygomycota, and to a lesser extent by Basidiomycota. Fungalcommunities increased in Ascomycota dominance and de-creased in both Basiodiomycota and Zygomycota dominancein response to exposure to wildfire at all sites.

Indicator species analyses identified that bacterial and fun-gal OTUs that only, or predominantly, occurred in eitherburned or unburned sites. The indicator value is an indexranging from 0 (no indication) to 100 (perfect indication;Dufrêne and Legedre 1997). As bacteria phyla are extremelydiverse, different OTUs within the same major bacterial phylawere indicators of both burn and unburned sites (Table 2).Members of Acidobacter ia , Act inobacter ia , andProteobacteria were moderate indicators of burned sites,whereas Acidobacteria and Proteobacteria were moderate in-dicators of unburned sites. Our fungal analyses indicated thatall sampled soils contained less fungal diversity than they didbacterial diversity. Consequently, indicator species analysisidentified fewer fungal OTUs compared to bacterial analyses.Members of the Ascomycota were weak indicators of both

Fig. 3 Nonmetric multidimensional scaling of bacterial (a) and fungal (b)OTUs grouped by burn treatments. Ellipses represent 95% confidenceintervals for groups

Fig. 2 Bacterial (a) and fungal (b) diversity generally declined inresponse to burning. Bars represent the mean of 10 subplots and errorbars represent standard error. Statistical results are from nested ANOVAof burn treatment within site

Table 1 PERMANOVA results of bacterial and fungal communitieswith burn treatment nested within site

d.f. F R2 P

Bacterial

Site 2 3.9 0.14 0.001

Site (burn) 3 3.9 0.21 0.001

Fungal

Site 2 1.9 0.08 0.001

Site (burn) 3 1.7 0.10 0.001

990 Biol Fertil Soils (2018) 54:985–997

burned and unburned sites, whereas members of theZygomycota were weak indicators of the unburned sites(Table 3). These results corroborated a shift from shift fromZygomycota in unburned to Ascomycota in burned sites thatwas also shown by stacked barplots (Fig. 4).

Discussion

Wildfires can influence both aboveground communities andprocesses (Hart et al. 2005; Reilly et al. 2006) andbelowground biogeochemical processes (Certini 2005;DeBano et al. 1998; Dooley and Treseder 2012; Neary et al.1999). Here we demonstrate that wildfires can also drive var-iation in the composition of soil microbial communities. Ourresults indicate that wildfires in Linville Gorge reduced mi-crobial diversity, and caused shifts in microbial communities

that likely coincide with changes in soil organic matter andrhizosphere following fires.

Biogeochemical responses to fires

Wildfires in Linville Gorge have led to persistent reductionsin soil respiration, in agreement with previous investigations(Hamman et al. 2007; Holden and Treseder 2013;Wang et al.2012). Reductions in soil respiration following wildfires arelikely the result of several interacting factors in addition tothe direct, heat-induced mortality of soil microbes (Certini2005). Heterotrophic microbes rely on the aboveground litterinputs and belowground root exudates. Wildfires can directlyreduce the aboveground biomass and reduce soil organicmatter (SOM), leading to altered soil processes (González-Pérez et al., 2004). The reduction in the quantity and qualityof litter inputs to below ground systems after fire often limitslabile C sources that are easily mineralized leading to

Fig. 4 Phylum-level distributionof bacteria (a) and fungal (b)communities with a 0.1% cutoff

Biol Fertil Soils (2018) 54:985–997 991

reduced soil respiration (Certini 2005). While site locationand time since burn are confounded in our study, patternsamong sites may reflect patterns due to time since burn,and soil respiration was reduced the most in the most

recently burned TR-2013 sites. Decreased root respirationand reduced substrate availability for surviving microbeslikely decreased soil respiration in response to recentwildfires.

Table 2 Indicator species analysis of bacterial OTUs associated with either burn or unburn treatments. Several genera were represented by multipleOTUs, as indicated by the #OTUs column

Burn/unburn

#OTUs Indicatorvalue

P Phyla Class Order Family Genus

Burn 8 58.2–73.9 < 0.0356 Acidobacteria Acidobacteria Incertae sedis Incertae sedis Gp1

Burn 62.1 0.0116 Acidobacteria Acidobacteria Incertae sedis Incertae sedis Gp13

Burn 7 55.5–63.3 < 0.0394 Acidobacteria Acidobacteria Incertae sedis Incertae sedis Gp2

Burn 57.7 0.0412 Actinobacteria Actinobacteria Actinomycetales Thermomonosporaceae Actinoallomurus

Burn 2 65–68.1 < 0.0026 Actinobacteria Actinobacteria Actinomycetales Unclassified Unclassified

Burn 59.1 0.0386 Bacteroidetes Sphingobacteria Sphingobacteriales Sphingobacteriaceae Mucilaginibacter

Burn 62.5 0.0096 Chloroflexi Unclassified Unclassified Unclassified Unclassified

Burn 57.2 0.0246 Proteobacteria Alphaproteobacteria Caulobacterales Caulobacteraceae Phenylobacterium

Burn 61.4 0.0272 Proteobacteria Alphaproteobacteria Rhodospirillales Acetobacteraceae Unclassified

Burn 63 0.0086 Proteobacteria Alphaproteobacteria Rhodospirillales Unclassified(99) Unclassified

Burn 63.7 0.0046 Proteobacteria Betaproteobacteria Unclassified Unclassified Unclassified

Burn 59.8 0.0462 Proteobacteria Gammaproteobacteria Unclassified Unclassified Unclassified

Burn 66.7 0.0028 Verrucomicrobia Opitutae Opitutales Opitutaceae Opitutus

Burn 2 62.5–64 < 0.25 Verrucomicrobia Subdivision3 Incertae sedis Incertae sedis Incertae sedis

Unburn 61.2 0.0008 Acidobacteria Acidobacteria_Gp1 Incertae sedis Incertae sedis Gp1

Unburn 2 55.6–58.1 < 0.0398 Acidobacteria Acidobacteria_Gp3 Incertae sedis Incertae sedis Gp3

Unburn 61.2 0.0362 Bacteroidetes Sphingobacteria Sphingobacteriales Chitinophagaceae Unclassified

Unburn 78.7 0.0002 Chlamydiae Chlamydiae Chlamydiales Unclassified Unclassified

Unburn 62.6 0.0036 Planctomycetes Planctomycetacia Planctomycetales Planctomycetaceae Planctomyces

Unburn 67.1 0.0002 Planctomycetes Planctomycetacia Planctomycetales Planctomycetaceae Unclassified

Unburn 63.2 0.0008 Proteobacteria Alphaproteobacteria Rhizobiales Unclassified Unclassified

Unburn 2 64.4–64.8 < 0.0022 Proteobacteria Alphaproteobacteria Unclassified Unclassified Unclassified

Unburn 65.6 0.0192 Proteobacteria Betaproteobacteria Burkholderiales Comamonadaceae Variovorax

Unburn 65.4 0.0056 Proteobacteria Betaproteobacteria Burkholderiales Oxalobacteraceae(94) Unclassified

Unburn 78.8 0.0012 Proteobacteria Gammaproteobacteria Pseudomonadales Pseudomonadaceae Pseudomonas

Unburn 4 59.6–73 < 0.0136 Verrucomicrobia Spartobacteria Incertae sedis Incertae sedis Incertae sedis

Table 3 Indicator species analysis of fungal OTUs associated with either burn or unburn treatments

Burn/unburn

Indicatorvalue

P Phylum Class Order Family Genus

Burn 45.2 0.0032 Ascomycota Archaeorhizomycetes Archaeorhizomycetales Archaeorhizomycetaceae Archaeorhizomyces

Burn 23.1 0.0476 Ascomycota Archaeorhizomycetes Archaeorhizomycetales Archaeorhizomycetaceae Archaeorhizomyces

Burn 26.9 0.0278 Ascomycota Eurotiomycetes Eurotiales Trichocomaceae Aspergillus

Unburn 43.2 0.005 Ascomycota Archaeorhizomycetes Archaeorhizomycetales Archaeorhizomycetaceae Archaeorhizomyces

Unburn 30 0.0022 Ascomycota Archaeorhizomycetes Archaeorhizomycetales Archaeorhizomycetaceae Archaeorhizomyces

Unburn 20 0.0286 Basidiomycota Tremellomycetes Tremellales Incertae sedis Cryptococcus

Unburn 62.7 0.0002 Zygomycota Incertae sedis Mortierellales Mortierellaceae Mortierella

Unburn 60.6 0.0098 Zygomycota Incertae sedis Mortierellales Mortierellaceae Mortierella

Unburn 49.3 0.0002 Unidentified Unidentified Unidentified Unidentified Unidentified

Unburn 43.5 0.0006 Zygomycota Incertae sedis Mortierellales Mortierellaceae Mortierella

Unburn 39.2 0.0026 Zygomycota Incertae sedis Mortierellales Mortierellaceae Mortierella

992 Biol Fertil Soils (2018) 54:985–997

While exposure to wildfire decreased soil respiration, soilC content increased in response to fire. Others have also re-ported that fire increases soil C (e.g., Boerner et al. 2005), butthe effects of fire on soil C are not consistent across studiesand systems. For instance, fire can also decrease soil C(Pourreza et al. 2014; Neff et al. 2005), or may have little tono effect on soil C (Hamman et al. 2007; Knoepp et al. 2004;Mikita-Barbato et al. 2015). Inconsistencies in soil C re-sponses to fire can be attributed to variation in fire severity,soil type, vegetation, erosion, and sampling methods amongstudies. Increased soil C in this study was most pronounced inolder burn sites, suggesting that soil C may be accumulatingand remaining in the soil for several years after a fire. Dumaset al. (2007) reported both decreased respiration and decreaseddecomposition rates in burned plots as compared to unburnedplots following the BR-2000 fires in the same area.Consequently, soil C accumulation in burned plots could bethe result of decreased decomposition driven by reduced mi-crobial activity.

Severe wildfires can reduce soil N via volatilisation dur-ing combustion, and by causing increased leaching in burnedsoils (Certini 2005). Conversely, soil N can also be increasedby fires, but those increases may be limited to NO3-N ratherthan NH4-N (Choromanska and DeLuca, 2001). We ob-served contrasting trends with decreased soil N in the recent-ly burned sites (TR-2013) and increased soil N in the olderburned sites (PNCL-2007andBR-2000).Wildfiresmayneg-atively impact functionally important soil microbes causinga reduction in total soil N as N-cycling microbes are particu-larly sensitive to soil heating (Choromanska and DeLuca,2002; Dunn et al. 1985). Related to soil N-cycling, fire de-creased urease activity potential at all sites. Heat-inducedreductions of N-cycling bacteria could be responsible forthe observed decreases in urease activity potential. The re-duction in urease enzyme activity potential could also be aresult of increase nitrogen following fires as at the two olderburn sites. Allison and Vitousek (2005) argue that an abun-dance of labile soil N could lead to decreased urease activity,as a consequence of microbial economics whereby enzymeproduction is dependent on microbial demand and availabil-ity of N for enzyme synthesis.

Phenol oxidase production is associated with the degra-dation of lignin and other recalcitrant forms of soil organicmatter and can be an indicator of fungal activity (Baldrianand Valášková, 2008). Wildfires can transform soil organic Cinto highly recalcitrant forms including those associated withlignin degradation (González-Pérez et al., 2004). The in-creased phenolic activity potential seen at all burned sitesin our study may be associated with the enzymatic degrada-tions of lignin-based C pools. Boerner et al. (2005) also not-ed elevated phenol oxidase activity in burned soils, possiblyresulting from increased soil organic carbon content follow-ing fire.

A limitation of enzymatic assays employed here is that theydescribe enzyme activity potential instead of in situ enzymeactivity, and they are highly susceptible to variation in labora-tory protocols (Nannipieri et al. 2018). For instance, the mi-croplate pNP-based protocols we used provide higher esti-mates and have lower detection limits compared to otherbench-top pNP or microplate MUF-based methods (Denget al. 2013). Nonetheless, microplate pNP estimates are highlycorrelated with MUF methods (Deng et al. 2013) and bothmethods measure the same pool of enzymes.

Microbial community response to fire

Belowground microbial responses to wildfire can depend up-on the frequency and intensity of fire (Weber et al. 2014;Williams et al. 2012). Some have found reductions in below-ground diversity in both bacterial (Weber et al. 2014) andfungal (Dove and Hart 2017; Lumley et al. 2001) communi-ties in response to fire. Reductions in bacterial diversity oc-curred in two of the three sites studied here, whereas reduc-tions in fungal diversity were fairly uniform across all threesites. Although soil respiration declined in response to burningat all sites, it is unclear if the observed reductions in diversityper se influence biogeochemical cycles. The high OTU counts(mean 1644 bacteria and 791 fungal per plot) supports theconcept that forest soils harbor hyper-diverse microbial com-munities that likely contain many functionally redundant spe-cies (Prosser 2012).

While wildfires influenced both bacterial and fungal com-munity composition, burned and unburned communitiescontained many of the samemicrobial members. The bacterialphylum level composition found here was similar to thosereported in other studies (Cleveland et al. 2007; Janssen2006; Weber et al. 2014), with Acidobacteria andProteobacteria representing the dominant phyla across allgroups. The dominance of Acidobacteria is expected as thesoils in Linville Gorge are moderately to highly acidic (Knight2006) and the low pH is likely a major factor shaping theoverall composition of these communities. Actinobacteriaand Bacteriodetes comprised minor portions of the communi-ty, and both have been reported as common phyla in forestsoils (Jansen 2006). While Planctomycetes are more typicallyassociated with brackish aquatic communities, they havemorerecently been associated with widespread soil communities(Axelrood et al. 2002; Buckley et al. 2006). Likewise, al-though Chloroflexi and Verrucomicribia are not as well de-scribed as other phyla, they commonly are found in soils(Janssen 2006).

Indicator species analysis identified specific bacterial taxathat were associated primarily with either burned or unburnedplots. Members of Actinobacteria were the most frequent indi-cators of burned sites. Actinobacteria have been found to in-crease after fires in other forest systems potentially acting as

Biol Fertil Soils (2018) 54:985–997 993

plant growth promoters and able to survive high temperaturesvia spores (Fernández-González et al., 2017). Changes in spe-cific taxa may reflect the chemical composition of soil organicmatter in burned sites. Phenylobacterium are commonly foundin forest soils (Nacke et al. 2011) and are capable of aromaticcompound degradation (Lingens et al. 1985). Conversely,Variovorax is a chitinolytic genus has been associated withdegrading fungal mycelium (Bers et al. 2011) and was an in-dicator of unburned sites (Table 2) where fungal decomposersof lignin may have been abundant. Some indicators species didnot follow patterns found in previous work, as Pseudomonashave been reported elsewhere to increase after fire treatments(Pandey et al., 2011) but were indicative of unburned sites inour study. The amplicon sequencing techniques used here areable to describe microbial community composition, but to datehave little power to describe functional responses (Vestergaardet al. 2017). In addition, OTU detection via amplicon sequenc-ing does not necessarily mean that these OTUs are active in thesoil microbial community (Carini et al. 2016). This disconnectbetween OTUs as measured via amplicon sequencing and theactive members of the microbial community in situ makes itdifficult to draw clear cause and effect patterns.

While there were no consistent phyla-level changes in bac-terial communities across treatments, several general trendswere apparent in the fungal communities. All sites were dom-inated largely byAscomycota, often a dominant component oflitter decomposers (Osono and Takeda, 2006; Schneider et al.,2012) that are associated with cellulose decomposition(Osono 2007). Burned sites experienced a large shift towardsdominance of Ascomycota at the expense of Basidiomycotaand Zygomycota. Basidiomycota can be particularly sensitiveto fires (Holden et al. 2016). The shift from lignin-degradingBasidiomycetes in unburned sites to cellulose-degradingAscomycota in burned sites could reflect a change from ligninto cellulose as the major constituents of soil organic matter inunburned and burned sites, respectively. Here, the reduction inBasidiomycota was largest in the most recently burned site(TR-2013), perhaps because the other sites had several yearsof litter inputs that replenished belowground cellulose SOMstocks. Wildfire severity in Linville Gorge for the 2000 fireswas highly variable (Wimberly and Reilly 2007). In addition,the effects of fire on SOM are also highly variable (González-Pérez et al., 2004), and differences in fire severity, fuel load, aswell as pre- and post-fire plant community drove variation inpost-fire SOM stocks.

Oliver et al. (2015) report that recurring burns had no ob-servable effect on richness or diversity of the fungal communi-ty, but that burned sites did support a fungal community thatwas distinct from unburned soils. Our study indicated both areduction in fungal diversity and a shift in community compo-sition. Indicator species analysis confirmed the shifts in com-munities shown by phylum-level distributions (Fig. 4).Important changes included a shift from Basidiomycota and

Zygomycota in unburned sites to Ascomycota in burned sites.Different members of Archaeorhizomyces were indicators ofboth burned and unburned sites. Archaeorhizomyces are wide-spread soil fungi commonly associated with plant roots(Rosling et al. 2011) and contain an estimated 500 species(Menkis et al. 2014). Shifts within the Archaeorhizomyces taxabetween burned and unburned sites could reflect the responseof specific fungal species to changes in plant communities and/or fine root dynamics following wildfires. The large portion ofZygomycota in all of our sites is somewhat unexpected but issupported by other studies of forest soils (e.g., Oliver et al.2015). Lastly, is not clear why Mortierella were indicators ofunburned sites as they are known associates with litter decom-position and likely focus on labile carbon sources(Asemaninejad et al. 2016) which were present at both burnedand unburned sites. Associating specific functional roles withOTUs identified by indicator species analysis is difficult be-cause the majority of OTUs lack functional meaning(Vestergaard et al., 2017). In addition, both amplicon sequenc-ing and the indicator species analyses used here consider onlywhich OTUs were present in each treatment, and not necessar-ily who was playing an active role in the microbial community.

Conclusions

Combining soil biogeochemistry with soil microbial commu-nity data can provide insight into the functional roles of soilbacterial communities. The observed reductions in soil respi-ration and increase in soil C could indicate that in this studywildfire decreased overall microbial activity and negativelyimpacted functionally important soil microbial communitiesthat would have otherwise catabolized soil C stocks. Increasedsoil C could have also been facilitated by heat-induced trans-formations that made soil C more recalcitrant. These twomechanisms are not mutually exclusive. Diversity declinedin both bacterial and fungal communities in response to wild-fire, but it is unknown if the reductions in diversity per se werelarge enough to account for variation in biogeochemical pro-cess. Changes in biogeochemistry are likely also driven bychanges in microbial community composition. While expo-sure to wildfire influenced the microbial communities in forestsoils, the composition burned soil communities overlappedwith those of unburned sites. The observed changes in below-ground microbial communities likely reflect differences inboth SOM and fine root dynamics pre- and post-fire soils,highlighting above-and belowground links. Future researchshould include methods that focus on active gene expressionin order to establish a stronger link between microbial com-munity composition and specific function. In addition, quali-tative assessment of SOM changes post fire may provide bet-ter insight towards understanding changes in the belowgroundmicrobial community.

994 Biol Fertil Soils (2018) 54:985–997

Acknowledgements This work was supported by the Cratis D. WilliamsGraduate School, the Appalachian State University Biology Department,and the Grandfather Ranger district of the U.S.D.A Forest Service.

References

Abatzoglou JT, Williams PA (2016) Impact of anthropogenic climatechange on wildfire across western US forests. PNAS 113:11770–11775

Allison SD, Vitousek PM (2005) Responses of extracellular enzymes tosimple and complex nutrient inputs. Soil Biol Biochem 37:937–944

Altschul SF, GishW,MillerW,Myers EW, Lipman DJ (1990) Basic localalignment search tool. J Mol Biol 215:403–410

Asemaninejad A, Thorn RG, Lindo Z (2016) Experimental climatechange modifies degradative succession in boreal peatland fungalcommunities. Microb Ecol 73:521–531

Axelrood PE, ChowML, Radomski CC,McDermott JM, Davies J (2002)Molecular characterization of bacterial diversity from BritishColumbia forest soils subjected to disturbance. Can J Microbiol48:655–674

Baldrian P, Valášková V (2008) Degradation of cellulose by basidiomy-cetous fungi. FEMS Microbiol Rev 32:501–552

Banning NC, Murphy DV (2008) Effect of heat-induced disturbance onmicrobial biomass and activity in forest soil and the relationshipbetween disturbance effects and microbial community structure.Appl Soil Ecol 40:109–119

Bers K, Sniegowski K, Albers P, Breugelmans P, Hendrickx L, DeMot R,Springael D (2011) A molecular toolbox to estimate the number anddiversity of Variovorax in the environment: application in soils treat-ed with the phenylurea herbicide linuron. FEMSMicrobiol Ecol 76:14–25

Boerner REJ, Brinkman JA, Smith A (2005) Seasonal variations in en-zyme activity and organic carbon in soil of a burned and unburnedhardwood forest. Soil Biol Biochem 37:1419–1426

Bond-Lamberty B, Wang C, Gower ST (2004) A global relationshipbetween the heterotrophic and autotrophic components of soil res-piration? Glob Chang Biol 10(10):1756–1766

Buchmann N (2000) Biotic and abiotic factors controlling soil respirationrates in Picea abies stands. Soil Biol Biochem 32:1625–1635

Buckley DH, Huangyutitham V, Nelson TA, Rumberger A, Thies JE(2006) Diversity of planctomycetes in soil in relation to soil historyand environmental heterogeneity. App Environ Microbiol 72:4522–4531

Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD,Costello EK, Fierer N, Pena AG, Goodrich JK, Gordon JI, HuttleyGA, Kelley ST, Knights D, Koenig JE, Ley RE, Lozupone CA,McDonald D, Muegge BD, Pirrung M, Reeder J, Sevinsky JR,Turnbaugh PJ, Walters WA, Widmann J, Yatsunenko T, ZaneveldJ, Knight R (2010) QIIME allows analysis of high-throughput com-munity sequencing data. Nat Methods 7:335–336

Caporaso JG, Lauber CL, Walters WA, Berg-Lyons D, Lozupone CA,Turnbaugh PJ, Fierer N, Knight R (2011) Global patterns of 16SrRNA diversity at a depth of millions of sequences per sample.PNAS 108:4516–4522

Carini R, Marsden PJ, Leff JW, Morgan EE, Strickland MS, Fierer N.(2016) Relic DNA is abundant in soil and obscures estimates of soilmicrobial diversity. Nature Microbiology https://doi.org/10.1038/nmicrobiol.2016.242

Carreiro MM, Sinsabaugh RL, Repert DA, Parkhurst DF (2000)Microbial enzyme shifts explain litter decay responses to simulatednitrogen deposition. Ecology 81:2359–2365

Certini G (2005) Effects of fire on properties of forest soils: a review.Oecologia 143:1–10

Choromanska U, DeLuca TH (2001) Prescribed fire alters the impact ofwildfire on biochemical properties in a ponderosa pine forest.SSSAJ 65:232–238

Choromanska U, DeLuca TH (2002) Microbial activity and nitrogenmineralization in forest mineral soils following heating: evaluationof post-fire effects. Soil Biol Biochem 34:263–271

Cleveland CC, Nemergut DR, Schmidt SK, Townsend AR(2007).Increases in soil respiration following labile carbon additionslinked to rapid shifts in soil microbial community composition.Biogeochemistry 82:229-240.

Cisneros-Dozal LM, Trumbore S, Hanson P (2006) Partitioning sourcesof soil-respired CO2 and their seasonal variation using a uniqueradiocarbon tracer. Glob Chang Biol 12:194–204

Cole JR, Wang Q, Cardenas E, Fish J, Chai B, Farris RJ, Kulam-Syed-Mohideen AS, McGarrell DM, Marsh T, Garrity GM, Tiedje JM(2009) The ribosomal database project: improved alignments andnew tools for rRNA analysis. Nucleic Acids Res 37:D141–D145

DeBano LF, Neary DG, Ffolliott PF (1998) Fire’s effects on ecosystems.New York: Wiley; 1998

Deng S, Popova IE, Dick L, Dick R (2013) Bench scale and microplateformat assay of soil enzyme activities using spectroscopic and fluo-rometric approaches. Appl Soil Ecol 64:84–90

Dixon RK, Brown S, Houghton RA, Solomon AM, Trexler MC,Wisniewski J (1994) Carbon pools and flux of global forest ecosys-tems. Science 263:185–190

Dooley SR, Treseder KK (2012) The effect of fire on microbial biomass:a meta-analysis of field studies. Biogeochemistry 109:49–61

Dove N, Hart S (2017) Fire reduces fungal species richness and in situmycorrhizal colonization: a meta-analysis. Fire Ecol 13:37–65

DufrêneM, Legedre P (1997) Species assemblages and indicator species:the need for a flexible asymmetrical approach. Ecol Monogr 67:345–366

Dumas S, Neufeld HS, Melany CF (2007) Fire in a thermic oak-pineforest in Linville Gorge Wilderness Area, North Carolina: impor-tance of the shrub layer to ecosystem response. Castanea 72:92–104

Dunn PH, Barro SC, Poth M (1985) Soil moisture affects survival ofmicroorganisms in heated chaparral soil. Soil Biol Biochem 17:143–148

Fernández-González AJ, Martínez-Hidalgo P, Cobo-Díaz JF, Villadas PJ,Martínez-Molina E, Toro N, Tringe SG, Fernández-López M (2017)The rhizosphere microbiome of burned holm-oak: potential role ofthe genus Arthrobacter in the recovery of burned soils. Sci Rep 7:6008

Ginzburg O, Steinberger Y (2012) Effects of forest wildfire on soil mi-crobial community activity and chemical components on atemporal-seasonal scale. Plant Soil 360:243–257

González-Pérez JA, Gonzalez-Vila GA, Gonzalo A, Knicker H (2004)The effect of fire on soil organic matter—a review. Environ Int 30:855–870

Hamman ST, Burke IC, Stromberger ME (2007) Relationships betweenmicrobial community structure and soil environmental conditions ina recently burned system. Soil Biol Biochem 39:1703–1711

Hanson PJ, Edwards NT, Garten CT, Andrews JA (2000) Separating rootand soil microbial contributions to soil respiration: a review ofmethods and observations. Biogeochemistry 48:115–146

Hart SC, DeLuca TH, Newman SG, MacKenzie MD, Boyle SI (2005)Post-fire vegetative dynamics as drivers of microbial communitystructure and function in forest soils. For Ecol Manag 220:166–184

Hernandez T, Garcia C, Reinhardt I (1997) Short-term effect of wildfireon the chemical, biochemical andmicrobiological wildfire effects onboreal soil Fungi 45 properties of Mediterranean pine forest soils.Biol Fertil Soils 25:109–116

Holden SR, Treseder KK (2013) A meta-analysis of soil microbial bio-mass responses to forest disturbances. Front Microbiol 4:163

Biol Fertil Soils (2018) 54:985–997 995

Holden SR, Rogers BM, Treseder KK, Randerson JT (2016) Fire severityinfluences the response of soil microbes to a boreal forest fire.Environ Res Lett 11:035004

Janssen PH (2006) Identifying the dominant soil bacterial taxa in librariesof 16S rRNA and 16S rRNA genes. Appl Environ Microbiol 72:1719–1728

Knight D (2006) Soil survey of Burke County, North Carolina. UnitedStates Department of Agriculture, Natural Resources ConservationService

Knoepp JD, Vose JM, SwankWT (2004) Long-term soil responses to sitepreparation burning in the southern Appalachians. For Sci 50:540–550

Knoepp JD, DeBano LF, Neary DG (2005) Soil chemistry. In: Neary DG,Ryan KC, DeBano LF (eds)Wildland fire in ecosystem; effect of fireon soils and water. General Technical Report RMRS-GTR 42-4,U.S. Department of Agriculture, Forest Service, Rocky MountainResearch Station, Ogden, UT, pp 53–71

Kozich JJ, Westcott SL, Baxter NT, Highlander SK, Schloss PD (2013)Development of a dual-index sequencing strategy and curation pipe-line for analyzing amplicon sequence data on the MiSeq Illuminasequencing platform. Appl Environ Microbiol 79:5112–5120

Lentile LB, Holden ZA, Smith AMS, Falkowski MJ, Hudak AT, MorganP, Lewis SA, Gessler PE, Benson NC (2006) Remote sensing tech-niques to assess active fire characteristics and post-fire effects. Int JWildland Fire 15:319–345

Lingens F, Blecher R, Blecher H, Blobel F, Eberspächer J, Fröhner C et al(1985) Phenylobacterium immobile gen. Nov., sp. Nov., a gram-negative bacterium that degrades the herbicide chloridazon. Int JSyst Bacteriol 35:26–39

Lumley RTC, Gignac D, Currah RS (2001) Microfungus communities ofwhite spruce and trembling aspen logs at different stages of decay indisturbed and undisturbed sites in the boreal mixedwood region ofAlberta. Can J Bot 79(1):76–92

Madritch MD, Donaldson JR, Lindroth RL (2007) Canopy herbivory canmediate the influence of plant genotype on soil processes throughfrass deposition. Soil Biol Biochem 39:1192–1201

Martin M (2011). Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet.journal 17. https://doi.org/10.14806/ej.17.1.200

McMurdie PJ, Holmes S (2013) phyloseq: an R package for reproducibleinteractive analysis and graphics of micro-biome census data. PLoSOne 8(4):e61217

Menkis A, Urbina H, James TY, Rosling A (2014) Archaeorhizomycesborealis sp. nov. and a sequence- based classification of related soilfungal species. Fungal Biol 118:943–955

Mikita-Barbato RA, Kelly JJ, Tate RL (2015) Wildfire effects on theproperties and microbial community structure of organic horizonsoils in the New Jersey Pinelands. Soil Biol Biochem 86:67–76

Mulvaney R (1996) Nitrogen - inorganic forms. In: Bartels J (ed)Methods of soil analysis part 3 chemical methods. Soil Sci SocAm, Madison, WI, pp 1123–1184

Nacke H, Thürmer A, Wollherr A, Will C, Hodac L, Herold N, SchoningI, Schrumpf M, Daniel R (2011) Pyrosequencing-based assessmentof bacterial community structure along different management typesin German forest and grassland soils. PLoS One 6:e17000

Nannipieri P, Trasar-Cepeda C, Dick RP (2018) Soil enzyme activity: abrief history and biochemistry as a basis for appropriate interpreta-tions and meta-analysis. Biol Fertil Soils 54:11–19

Neary DG, Klopatek CC, DeBano LF, Elliott PF (1999) Fire effects onbelowground sustainability: a review and synthesis. For EcolManag122:51–71

Neff JC, Harden JW, Gleixner G (2005) Fire effects on soil organic mattercontent, composition, and nutrients in boreal interior Alaska. Can JFor Res 35:2178–2187

Newell CL, Peet RK (1998) Vegetation of Linville Gorge Wilderness,North Carolina. Castanea 63:275–322

Oksanen J, Blanchet FG, Kindt R, Legendre P, O’Hara RB, Simpson GL,Solymos P, Stevens MHH, Wagner H (2013) vegan: communityecology package version 2.4–4. https://cran.r-project.org/web/packages/vegan/index. html

Oliver AK, Callaham MA, Jr, Jumpponen A (2015) Soil fungal commu-nities respond compositionally to recurring frequent prescribedburning in a managed southeastern US forest system. For EcolManag 345:1–9

Osono T (2007) Ecology of ligninolytic fungi associated with leaf litterdecomposition. Ecol Res 22:955–974

Osono T, Takeda H (2006) Fungal decomposition of Abies needle andBetula leaf litter. Mycologia 98:172–179

Pandey A, Chaudhry S, Sharma A, Choudhary VS, Malviya MK,Chamoli S, Rinu K, Trivedi P, Palni LMS (2011) Recovery ofBacillus and Pseudomonas spp. from the ‘Fired Plots’ under shiftingcultivation in Northeast India. Curr Microbiol 62:273–280

Perrakis D, Zell D (2008).Remote assessment of burn severity: a pilotstudy in landscape monitoring. Parks Canada Agency: Western andNorthern Service Centre and National Fire Centre

Pourreza M, Hosseini SM, Sinegani AAS, Matinizadeh M, Dick W(2014) Soil microbial activity in response to fire severity in Zagrosoak (Quercus brantii Lindl.) forests, Iran, after one year. Geoderma213:95–102

Prieto-Fernandez A, Acea MJ, Carballas T (1998) Soil microbial andextractable C and N after wildfire. Biol Fertil Soils 27:132–142

Prosser JI (2012) Ecosystem processes and interactions in a morass ofdiversity. FEMS Microbiol Ecol 81:507–519

Raich JW, Schlesinger WH (1992) The global carbon dioxide flux in soilrespiration and its relationship to vegetation and climate. Tellus 44B:81–99

Reilly MJ, Wimberly MC, Newell CL (2006) Wildfire effects on beta-diversity and species turnover in a forested landscape. J Veg Sci 17:447–454

Restaino JC, Peterson DL (2013) Wildfire and fuel treatment effects onforest carbon dynamics in the western United States. For EcolManage 303:46–60

Rognes T, Flouri T, Nichols B, Quince C, Mahé F (2016) VSEARCH: aversatile open source tool for metagenomics. PeerJ 4:e2584

Rosling A, Cruz-Martinez K, Ihrmark K, Grelet GA, Lindahl B, MenkisA, James T (2011) Archaeorhizomycetes: unearthing an ancientclass of ubiquitous soil fungi. Science 33:876–879

Saiya-Cork K, Sinsabaugh RL, Zak DR (2002) The effects of long termnitrogen deposition on extracellular enzyme activity in an Acersaccharum forest soil. Soil Biol Biochem 34:1309–1315

Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M, Hollister EB,Lesniewski RA, Oakley BB, ParksDH, RobinsonCJ, Sahl JW, StresB,Thallinger GG, Van Horn DJ, Weber CF (2009) Introducingmothur: opensource, platform-independent, community-supportedsoftware for describing and comparing microbial communities.Appl Environ Microbiol 75: 7537–7541.

Schloss PD, Gevers D, Westcott SL (2011) Reducing the Effects of PCRAmplification and Sequencing Artifacts on 16S rRNA-BasedStudies. PLoS ONE 6(12):e27310

Schneider T, Keiblinger KM, Schmid E, Sterflinger-Gleixner K,Ellersdorfer G, Roschitzki B, Richter A, Eberl L, Zechmeister-Boltenstern S, Riedel K (2012) Who is who in litter decomposition?Metaproteomics reveals major microbial players and their biogeo-chemical functions. ISME J 6:1749–1762

Smith DP, PeaKG (2014) Sequence depth, not PCR replication, improvesecological inference from next generation DNA sequencing. PloSOne 9(2):e90234

Swanson ME, Franklin JF, Beschta RL, Crisafulli CM, DellaSala DA,Hutto RL, Lindenmayer DB, Swanson FJ (2011) The forgotten stageof forest succession: early-successional ecosystems on forest sites.Front Ecol Environ 9:117–125

996 Biol Fertil Soils (2018) 54:985–997

Van der Heijden MGA, Bardgett RD, Van Straalen NM (2008) The un-seen majority: soil microbes as drivers of plant diversity and pro-ductivity in terrestrial ecosystems. Ecol Lett 11:296–310

Vestergaard G, Schulze S, Scholer A, Schloter A (2017) Making big datasmart – how to use metagenomics to understand soil quality. BiolFertil Soils 53:479–484

Wang Q, Zhong M, Wang S (2012) A meta-analysis on the response ofmicrobial biomass, dissolved organic matter, respiration, and Nmin-eralization in mineral soil to fire in forest ecosystems. For EcolManage 271:91–97

Weber CF, Lockhart J, Charaska E, Aho K, Lohse KA (2014) Bacterialcomposition of soils in ponderosa pine and mixed conifer forestsexposed to different wildfire burn severity. Soil Biol Biochem 69:242–250

Williams RJ, Hallgren SW,Wilson GWT (2012) Frequency of prescribedburning in an upland oak forest determines soil and litter propertiesand alters the soil microbial community. For Ecol Manag 265:241-247

Wimberly MC, Reilly MJ (2007) Assessment of fire severity and speciesdiversity in the southern Appalachians using Landsat TM andETMþ imagery. Remote Sens Environ 108:189–197

Biol Fertil Soils (2018) 54:985–997 997

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具