Impacts of salvage logging on biodiversity: A meta-analysis

Abstract

Logging to "salvage" economic returns from forests affected by natural disturbances has become increasingly prevalent globally. Despite potential negative effects on biodiversity, salvage logging is often conducted, even in areas otherwise excluded from logging and reserved for nature conservation, inter alia because strategic priorities for post-disturbance management are widely lacking. A review of the existing literature revealed that most studies investigating the effects of salvage logging on biodiversity have been conducted less than 5 years following natural disturbances, and focused on non-saproxylic organisms. A meta-analysis across 24 species groups revealed that salvage logging significantly decreases numbers of species of eight taxonomic groups. Richness of dead wood dependent taxa (i.e. saproxylic organisms) decreased more strongly than richness of non-saproxylic taxa. In contrast, taxonomic groups typically associated with open habitats increased in the number of species after salvage logging. By analysing 134 original species abundance matrices, we demonstrate that salvage logging significantly alters community composition in 7 of 17 species groups, particularly affecting saproxylic assemblages. Synthesis and applications. Our results suggest that salvage logging is not consistent with the management objectives of protected areas. Substantial changes, such as the retention of dead wood in naturally disturbed forests, are needed to support biodiversity. Future research should investigate the amount and spatio-temporal distribution of retained dead wood needed to maintain all components of biodiversity.

Full text

J Appl Ecol. 2018;55:279–289. wileyonlinelibrary.com/journal/jpe 
|
279
© 2017 The Authors. Journal of Applied Ecology
© 2017 British Ecological Society
Received:16January2017
|
Accepted:18May2017
DOI: 10.1111/1365-2664.12945
REVIEW
Impacts of salvage logging on biodiversity: A meta- analysis
Simon Thorn1 | Claus Bässler2 | Roland Brandl3 | Philip J. Burton4 | Rebecca
Cahall5 | John L. Campbell5 | Jorge Castro6 | Chang-Yong Choi7 | Tyler Cobb8 |
Daniel C. Donato9 | Ewa Durska10 | Joseph B. Fontaine11 | Sylvie Gauthier12 |
Christian Hebert12 | Torsten Hothorn13 | Richard L. Hutto14 | Eun-Jae Lee15 |
Alexandro B. Leverkus16 | David B. Lindenmayer17 | Martin K. Obrist18 |
Josep Rost19,20 | Sebastian Seibold2,21 | Rupert Seidl22 | Dominik Thom22 |
Kaysandra Waldron23 | Beat Wermelinger24 | Maria-Barbara Winter25 |
Michal Zmihorski26 | Jörg Müller1,2
1FieldStationFabrikschleichach,DepartmentofAnimalEcologyandTropicalBiology(ZoologyIII),Julius-Maximilians-UniversityWürzburg,Rauhenebrach,Germany
2BavarianForestNationalPark,Grafenau,Germany
3DepartmentofEcology,AnimalEcology,FacultyofBiology,Philipps-UniversitätMarburg,Marburg,Germany
4UniversityofNorthernBritishColumbia,Terrace,BC,Canada
5DepartmentofForestEcosystemsandSociety,OregonStateUniversity,Corvallis,OR,USA
6DepartmentofEcology,UniversityofGranada,Granada,Spain
7DepartmentofForestSciences,SeoulNationalUniversity,Seoul,Korea
8RoyalAlbertaMuseum,Edmonton,AB,Canada
9SchoolofEnvironmental&ForestSciences,UniversityofWashington,Seattle,WA,USA
10DepartmentofEcologyandBiodiversity,MuseumandInstituteofZoology,PolishAcademyofSciences,Warsaw,Poland
11SchoolofVeterinaryandLifeSciences,MurdochUniversity,Murdoch,WA,Australia
12NaturalResourcesCanada,CanadianForestService,LaurentianForestryCentre,Quebec,QC,Canada
13DivisionofBiostatistics,UniversityofZürich,Zürich,Switzerland
14DivisionofBiologicalSciences,UniversityofMontana,Missoula,MT,USA
15UrbanPlanningResearchGroup,DaejeonSejongResearchInstitute,Daejeon,Korea
16EcologyUnit,DepartmentofLifeSciences,UniversityofAlcalá,AlcaládeHenares,Madrid,Spain
17FennerSchoolofEnvironmentandSociety,TheAustralianNationalUniversity,Canberra,ACT,Australia
18WSLSwissFederalInstituteforForest,SnowandLandscapeResearch,BiodiversityandConservationBiology,Birmensdorf,Switzerland
19DepartmentofEnvironmentalSciencesandFoodIndustries,UniversityofVic-CentralUniversityofCatalonia,Catalonia,Vic.,Spain
20DepartmentofEnvironmentalSciences,UniversityofGirona,Girona,Spain
21DepartmentofEcologyandEcosystemManagement,ChairforTerrestrialEcology,TechnischeUniversitätMünchen,Freising,Germany
22InstituteofSilviculture,DepartmentofForest-andSoilSciences,UniversityofNaturalResourcesandLifeSciencesVienna,Vienna,Austria
23DepartmentofWoodandForestSciences,LavalUniversity,Québec,QC,Canada
24WSLSwissFederalInstituteforForest,SnowandLandscapeResearch,ForestDynamics–ForestEntomology,Birmensdorf,Switzerland
25ForestResearchInstituteofBaden-Württemberg(FVA),Freiburg,Germany
26DepartmentofEcology,SwedishUniversityofAgriculturalSciences,Uppsala,Sweden

280
|
Journal of Applied Ecology
THORN eT al.
1 | INTRODUCTION
The frequency and extent of stand-replacing natural disturbances,
such as wildfires, windstorms and insect outbreaks, has increased
considerably during recent decades, particularly in the Northern
Hemisphere(Kurz etal.,2008;Seidl, Schelhaas,Rammer,&Verkerk,
2014).Naturaldisturbancescanenhancethestructuralheterogeneity
offorests,createhabitatsforspecies-richassemblagesofhighconser-
vationvalueandincreasethelong-termresilienceofforeststofuture
stressors(Swansonetal.,2011).However,societaldemandfortimber
and/orpestreductioncompelsforestmanagersto“salvage”timberby
logging before it deteriorates,a common practice even in locations
otherwise exempt from conventional green-tree harvesting, such
as national parks orwilderness areas (Figure1) (Chylarecki & Selva,
2016;Thorn etal., 2014).Suchsalvage loggingreducesthe amount
of dead wood, alters successional trajectories, affects biodiversity,
and can influence restoration costs and subsequent fire hazards
(Lindenmayer,Burton, & Franklin, 2008; Waldron,Ruel, & Gauthier,
2013).Consequently,conflictsoftenemergebetweennaturalresource
managers,policy-makersandconservationistsonhowtohandlenatu-
rallydisturbedforests(González&Veblen,2007;Lindenmayer,Thorn,
& Banks, 2017; Lindenmayer etal., 2004; Schmiegelow, Stepnisky,
Stambaugh,& Koivula,2006).This hasresultedinintensepublicde-
bates(Lindenmayeretal.,2017;Nikiforuk,2011;Stokstad,2006).
Differentnaturaldisturbance regimesleavedistincttypesofbio-
logicaland/orstructural legacies(Franklinetal.,2000).Forinstance,
forestskilledbywildfireorinsectoutbreaksarecharacterizedbylarge
numbersofsnags,whilewindstormscreateuprootedtrees(Swanson
etal., 2011). Salvage logging typicallyremoves or alters these lega-
cies.Theresponsesofsaproxylicandnon-saproxylicspeciesgroupsto
salvageloggingthusdependontheirrelationto(deadwood)legacies
affectedbysalvagelogging(Lindenmayeretal.,2008).Consequently,
differenttaxonomicgroupsindifferenttypesofnaturaldisturbances
may respond differently to salvage logging (Zmihorski & Durska,
2011).Numerousstudieshavefocusedontheeffectsofsalvagelog-
gingafternaturaldisturbancesonspeciesrichnessandthecommunity
compositionofvarioustaxasuchasvascularplants(Blair,McBurney,
Blanchard, Banks, & Lindenmayer, 2016; Macdonald, 2007; Stuart,
Correspondence
SimonThorn
Email:simon@thornonline.de
Funding information
GermanEnvironmentalFoundation;Austrian
Science Fund
HandlingEditor:MatthewStruebig
Abstract
1. Loggingto“salvage”economicreturnsfromforestsaffectedbynaturaldisturbances
hasbecomeincreasingly prevalentglobally.Despite potentialnegativeeffects on
biodiversity,salvageloggingisoftenconducted,eveninareasotherwiseexcluded
from logging and reserved for nature conservation, inter alia because
strategicprioritiesforpost-disturbancemanagementarewidelylacking.
2. Areviewoftheexistingliteraturerevealedthatmoststudiesinvestigatingtheef-
fects of salvage logging on biodiversity have been conducted less than 5years
followingnaturaldisturbances,andfocusedonnon-saproxylicorganisms.
3. Ameta-analysisacross24speciesgroupsrevealedthatsalvageloggingsignificantly
decreasesnumbers ofspecies ofeighttaxonomicgroups.Richnessofdeadwood
dependenttaxa(i.e.saproxylicorganisms)decreasedmorestronglythanrichnessof
non-saproxylictaxa.Incontrast, taxonomicgroupstypically associatedwithopen
habitatsincreasedinthenumberofspeciesaftersalvagelogging.
4. Byanalysing134originalspeciesabundancematrices,wedemonstratethatsalvage
logging significantly alters community composition in 7 of 17 species groups,
particularlyaffectingsaproxylicassemblages.
5. Synthesis and applications.Ourresultssuggestthatsalvageloggingisnotconsistent
withthe managementobjectivesofprotectedareas. Substantialchanges,suchas
theretentionofdeadwoodin naturallydisturbedforests, areneededto support
biodiversity.Futureresearchshouldinvestigatetheamountandspatio-temporal
distribution of retained dead wood needed to maintain all components of
biodiversity.
KEYWORDS
barkbeetle,climatechange,deadwood,disturbedforest,fire,naturaldisturbance,post-
disturbancelogging,salvagelogging,saproxylictaxa,windstorm

|
281
Journal of Applied Ecology
THORN eT al.
Grifantini,Fox,&Fox,1993),carabids(Cobb,Langor,&Spence,2007;
Koivula & Spence, 2006; Phillips, Cobb,Spence, & Brigham, 2006),
birds(Castro,Moreno-Rueda,&Hódar,2010;Choi,Lee,Nam,Lee,&
Lim,2014;Nappi&Drapeau, 2009; Saab, Russell,&Dudley, 2009;
Thorn,Werner,etal.,2016; Zmihorski,2010),and saproxylicorgan-
isms(i.e.thosedependingon dead wood during somepartoftheir
life cycles; Cobb etal., 2011; Norvez, Hébert, Bélanger, Hebert, &
Belanger,2013).
Twomain effects of salvage logging on biodiversityarise recur-
rentlyfrom the existing bodyofliterature. First,salvageloggingre-
duces the richness of taxonomicgroups or abundance of particular
speciesthatdependondeadwood.Forinstance,salvageloggingde-
creasednestingdensity ofcavity-nesting-birdsthat usuallybreedin
fire-killedtrees(Hutto&Gallo,2006).Similarly,post-stormloggingde-
creasedthetotalnumberofsaproxylicbeetlespeciesandthenumber
ofthreatenedspecies(Thornetal.,2014).Second,studiesthatinves-
tigateasetofdifferenttaxonomicgroupshavedemonstratedthatsal-
vageloggingcanalterthecommunitycompositionofbothsaproxylic
andnon-saproxylicorganisms,whiletheeffectsontheoverallnumber
ofspeciescanbesmall(Thorn,Bässler,Bernhardt-Römermann,etal.,
2016).Forinstance, post-stormsalvagelogginginMinnesota greatly
diminishedbirdcommunities,whilefewerdifferencesinthetreecover
weredetected(Lain,Haney,Burris,&Burton,2008).However,previ-
ousattemptstosummarizeknowledgeontheeffectsofsalvagelog-
gingonbiodiversityhavefocusedmainlyonsalvageloggingofburned
forests(Lindenmayer&Noss,2006;Lindenmayeretal.,2008;McIver
&Starr,2000;Thorn,Bässler,Svoboda,&Müller,2016),andaquan-
titativeassessment ofsalvagelogging impactsonbiodiversityisstill
lacking,particularlyacrossdifferenttaxonomicgroupsandinresponse
todifferenttypesofdisturbances(Figure1).
FIGURE1 Salvagelogging(SL)iscommonlyappliedafterwildfires,windstormsorinsectoutbreaks,andleadstochangesinhabitatsand
communitycompositionsinvariousforestecosystemsaroundtheworld(ashighlightedbythestudiesillustratedinpanels(a–l).Studylocations
(colouredcircles)representstudysitesthatcontributeddatatoourmeta-analysis.Photographsbyauthors.[Colourfigurecanbeviewedat
wileyonlinelibrary.com]

282
|
Journal of Applied Ecology
THORN eT al.
Here,wereviewedthescientificliteratureandcompiledexisting
data to quantify the effectsof salvage logging after wildfire,wind-
storms and insect outbreaks on (1) species numbers via a meta-
analysisof238individualcomparisonsofsalvaged/unsalvagedareas;
and (2) community composition, based on a subset of134 original
speciesabundancematrices.Wealsotestedthehypothesisthat the
impactsofsalvagelogging aremorepronouncedfor saproxylicspe-
ciesgroups thanfor non-saproxylicgroupsregardingthenumber of
speciesandcommunitycompositionwithindifferenttypesofnatural
disturbances.
2 | MATERIALS AND METHODS
2.1 | Literature search
We followed guidelines for systematic literature reviews (Pullin &
Stewart,2006) tocompile comparisonsofspeciesrichnessbetween
salvagedandunsalvagedfire-, wind- or insect-affectedforests.We
screenedtheelectronicdatabasesWebofScience,ScopusandGoogle
Scholaron15 February 2016 byusingthe simplified search strings
[salvageloggingORpost$disturbance*ORsalvaging]and[forest$OR
vegetationORdisturbanceORecosystem].Fromthis bodyoflitera-
ture(>2,000articles),weretainedonlyfield-basedstudiesafterhav-
ingscreenedthetitleandabstract.Modellingstudieswereexcluded.
Wealsoaddedrelevantpapersfromreferencelistsinpublishedstud-
ies. We restricted studies to those providing comparisons between
completely salvage logged plots and completely unsalvaged control
plotsaccordingtotheinformationgivenintherespectivestudies.This
meansthatonsalvageloggedplots,morethan75%ofthetreeswere
affectedbynatural disturbanceandthen completelysalvage logged
withoutfurthertreatment suchastree plantingorlegacy retention.
Lowerintensitiesofnaturaldisturbanceshavebeenrarelytargetedby
scientificstudies.Salvageloggingoperationsthusresembledconven-
tionalclear-cutting.Unsalvaged control plots hadtobeaffected by
thesamenaturaldisturbanceeventbutwithoutanyhumaninterven-
tion.Salvageloggedplotshadtobeofsimilarsize,surveyedwiththe
samefieldmethodsduringthesamestudyperiodandwiththesame
samplingeffortasunsalvagedcontrolplots.
Toexaminewhetherpseudo-replication(i.e.allplotsnestedwithin
onearea)mightbiasthe resultsofour meta-analysis(Ramage etal.,
2013), we carefullyselected the studies according to their designs,
andweusedstatisticsthataccountforpseudo-replication(seebelow).
Thespatialarrangementofplotsinallstudieswascheckedbasedon
methoddescriptionsand/ororiginalgeographiccoordinates.Wecon-
tactedauthorstoprovidedataortoclarifytheirstudydesignswhere
necessary(seeDatasourcessection).Studieswithouttrue replicates
(e.g.allsalvagedplotsnestedandseparatedfromunsalvagedcontrol
plots) were excluded from the analysis to ensure valid effect sizes
(Halme etal., 2010). Studies using the same set offield plots and/
orthesame studyarea(e.g.Samcheok Forest,Korea)wereidentified
andnestedinallsubsequentstatisticalanalysestocontrolforpseudo-
replicationwithinstudyareas.Wealsoexcludedstudiesthatsampled
forests undergoing multiple types of disturbances. Salvage logging
hadtobe conducted immediately(<12months)afternaturaldistur-
bancetookplace.Meannumberofspeciesandstandarddeviationval-
uespersamplingunitwereextractedfrompublishedtextandtables,
or from figures using PLOT DIGITIZER 2.6.2. (www.plotdigitizer.
sourceforge.net). Last, we compiled data on covariates by extract-
ing information on the disturbance type and the time since distur-
bance, and the time since subsequent salvage logging. In addition,
we compiled original species abundance matrices that underpinned
the published papers, which allowed us to explore the effects of
salvageloggingoncommunitycomposition.
2.2 | Meta- analysis
Allanalyseswereconductedinr3.3.1(www.r-project.org).Priortosta-
tisticalanalysis,specieswereassignedtooneofthefollowingtaxonomic
groupsandtoassociationwithdeadwood(i.e.saproxylic/non-saproxylic)
basedonthedescriptioninthearticles.Thesewhere:amphibians,ants,
bats, bees and wasps, birds, carabids, epigeal lichens, epigeal mosses,
epigealspiders,epixyliclichens,epixylicmosses,harvestmen,hoverflies,
landsnails,nocturnalmoths,non-saproxylicbeetles(excludingcarabids),
reptiles,rodents,saproxylic beetles,scuttleflies, springtails, truebugs,
vascular plants and wood-inhabiting fungi. For the analysis compar-
ingresponses ofsaproxylicand non-saproxylicspecies groups,wede-
finedsaproxylicbeetles,wood-inhabitingfungi,andepixyliclichensand
mossesassaproxylicandallotherspeciesgroupsasnon-saproxylic.
Forcomparing numbersofspecies betweensalvagedand unsal-
vagednaturallydisturbedplotsdescribedin thepublished literature,
weusedHedges’d,whichaccountsfordifferencesinsamplingeffort
across studies and for small sample sizes (Hedges & Olkin, 1985).
PositivevaluesofHedges’dindicatehighernumbersofspeciesinsal-
vageloggedplots,whereasnegativevaluesindicatealossinnumbers
ofspeciesattributedtosalvagelogging(i.e.highernumbersofspecies
inunsalvagednaturallydisturbedplots).Meanabsoluteeffectsizesof
d=0.2indicateasmalleffect,d=0.5amoderateeffect,andd = 0.8 a
largeeffect(Koricheva,Gurevitch,&Mengersen,2013).
We used multi-level linear mixed-effectsmodels, provided by
the r function “rma.mv” in the “metafor” package (Viechtbauer,
2010),totest theeffectoftaxonomicgroup asacategorical pre-
dictor and year since disturbance as a numerical covariate on
Hedges’dastheresponsevariable.Hedges’dvalueswereweighted
bythecorrespondingsamplingvariancewithinthestatisticalmodel.
Furthermore,thestudysitewasincludedasarandomeffectinthe
model (i.e. moderator term) to control for unmeasured site spec-
ificities and repeated measurements (pseudo-replication) within
onestudysite.Thismeansthatmultipledatapointsperstudywere
possibleifstudiesexaminedmultipletaxonomicgroupsorifstudies
lastedformorethan1year.Wesubtracted theinterceptfromthe
effectsizes (byincluding “−1”in themodelformula)to evaluateif
observedHedges’ddifferedsignificantlyfrom zero(fordetails and
modelformulaseeTableS1).
Toevaluate theeffectsof salvagelogging on saproxylicvs.non-
saproxylic groups, we fitted a second model with Hedges’d as re-
sponsevariable.Weagainincludedtheyearafternaturaldisturbance

|
283
Journal of Applied Ecology
THORN eT al.
andsubsequentlogging as a numericalpredictorvariable andstudy
siteas wellastaxonomicgroupasrandomfactors. Furthermore,we
addedtheinteractionofdeadwooddependence(i.e.saproxylic/non-
saproxylic)withnaturaldisturbancetypeaspredictorstotestwhether
theeffectofsalvageloggingonthe numberofspecies insaproxylic
andnon-saproxylicgroups differedwithin different typesofnatural
disturbances. Weimplemented a simultaneous inference procedure
tocomparesaproxylicandnon-saproxylicspeciesgroupswithineach
disturbancetype(Hothorn, Bretz,& Westfall,2008). Thisprocedure
allowedus totest ifresponsesofsaproxylicandnon-saproxylictaxa
varyamong fire-,wind-and insect-disturbedforests(fordetails and
model formula see TableS2). Last, we conducted funnel plots by
meansofthefunction“funnel”fromthe“metafor”packagetoassess
publicationbias(Korichevaetal.,2013;FigureS1).
2.3 | Analysis of community composition
Based on the reviewed literature, we compiled original species
abundance matrices to quantify changes in community composi-
tioninducedby salvage logging.Quantifyingchangesin community
compositionamong largeheterogeneousdatasetsischallenging and
requires statistical methods able to deal with issues such as unbal-
ancedsamplingeffortandwhichgenerateastandardizedeffect size
thatis comparableamong differentspecies groupsand surveytech-
niques.Thus,weusedpermutationalmultivariateanalysisofvariance
usingdistance matrices(Legendre&Anderson,1999),performedby
meansofthefunction“adonis”inthepackage“vegan”(Oksanenetal.,
2016).ThisanalysisprovidesapseudoF-value,basedon999permu-
tations,thatquantifies thedeviancefrom thenull-hypothesis,while
simultaneouslyaccounting forimbalanced studydesigns (McArdle&
Anderson,2001).Consequently,largevaluesofFcorrespondtolarge
changesin communitycompositioninducedbysalvagelogging.This
F-value represents the standardized difference between communi-
tiesinsalvageloggedandunsalvagednaturallydisturbedplotswithin
onespeciesabundancematrix(e.g. differences in bird communities
6yearsafterwildfire andsalvagelogging inOregon).We rigorously
restrictedthisanalysistothoseabundancematricesthatyieldedvalid
pseudoF-values overthe courseof permutations;that is,those ma-
triceswhichgeneratedlessthan99realpermutationswereexcluded.
Theserestrictions resultedin atotal numberof 134matrices, which
suppliedF-valuesfortheanalysisoutlinedbelow.
Totestifsalvageloggingchangedcommunitycompositionindif-
ferenttaxonomicgroups,wemodelledpseudoF-valuesinlinearmixed
modelsprovidedbythefunction“lmer”inthe“lme4”packageassum-
ingaGaussianerrordistribution(Bolkeretal.,2009).Weincludedthe
taxonomicgroupasacategoricalpredictorand theyearsincedistur-
banceas a numericalcovariate. Furthermore,weincludedthe study
siteasarandomeffecttocontrolforpossibledifferencesamongstudy
sitesand repeatedmeasurementswithin onestudysite.Weomitted
theinterceptfromthemodelformulatodetermineifF-valuesdiffered
significantlyfromzero.Thus,significant changesin communitycom-
positionofataxonomicgroupduetosalvageloggingwereindicated
by F-valuessignificantlylargerthanzero(fordetailsandmodelformula
seeTableS3).
AsfortheanalysisofHedges’d,asecondmodelwasfittedtotest
whether the effects of salvage logging on community composition
differed between saproxylic and non-saproxylic species groups in
differenttypesofdisturbances.Therefore,weincludedtheyearafter
disturbance and the interaction of saproxylic/non-saproxylic with
disturbancetypeaspredictors.Taxonomicgroupandstudysitewere
includedasrandomfactorsinthismodel.Weimplementedasimulta-
neousinferenceproceduretocomparesaproxylicandnon-saproxylic
species groups within each disturbance type (for details and model
formulaseeTableS4).
3 | RESULTS
Our meta-analysis showed that the effects of salvage logging have
been studied primarily for birds, vascular plants and carabids, par-
ticularly in burned forests. Studies were conducted primarily in
NorthAmerica andEurope, butlackingintropicalregions(Figure1).
Furthermore,therewasaclearlackofstudiesinvestigatingsaproxylic
taxa.Ofthe238compileddatapoints,170coveredaperiodof5years
or less after disturbance, with studies addressing the long-term ef-
fectsofsalvageloggingbeingrare(Figure2).Onlyonestudy(Hutto&
Gallo,2006)wasavailablethatprovideddataontheeffectsofsalvage
loggingformorethan20yearsafterdisturbances(Figure2).
Halfofthe individual comparisonsproducedvalues ofHedges’d
lowerthanzero,indicatinghighernumbers ofspeciesinnon-salvage
loggedareasthan salvageloggedareas (Figure3).Wefoundsignifi-
cantly lowerspecies numbers of epigeal and epixylic mosses, birds,
wood-inhabiting fungi, saproxylic beetles, springtails and epixylic
aswellas epigeallichensin salvage loggedareas comparedtonon-
salvagelogged areas(Figure3a).Incontrast,thenumbers ofspecies
of land snails, epigeal spiders and carabids were higher in salvage
logged areas than in unsalvaged areas (Figure3a). Thirteen of the
24 taxonomic groups, includingvascular plants, exhibited no signif-
icantresponsein numbers ofspeciestosalvage logging (Figure3a).
The numbers of species of saproxylic taxa significantly decreased
compared to non-saproxylic taxa in storm-affected and burned for-
ests(Figure4a).Thenegativeeffectofsalvageloggingonnumberof
speciesincreasedwithtimeelapsedsincedisturbanceandsubsequent
salvagelogging,althoughlong-termdataonsalvageloggingarescarce.
Salvageloggingwas associatedwithsignificant changes incom-
munitycompositionin 7 of 17taxonomicgroups(Figure3b).These
seven groups were epigeal spiders, carabids, vascular plants, birds,
wood-inhabiting fungi, saproxylic beetles and epixylic lichens
(Figure3b). Time elapsed since disturbance had no effect on the
strength of logging-induced changes to community composition
(TableS3).Furthermore,logging-inducedchangesincommunitycom-
positionwerestrongerforsaproxylictaxathanfornon-saproxylictaxa
in storm-disturbed forests. However, data availabilitywas scarce in
insect-affectedforestandlackinginburnedforests(Figure4b).

284
|
Journal of Applied Ecology
THORN eT al.
4 | DISCUSSION
Our study revealed that salvage logging can result in significant
changes in species numbers and/or in altered community composi-
tion.Negativeeffects were particularlystrongfor taxa thatdepend
ondeadwood. Incontrast,thenumbers ofspeciesof taxa thatare
commonly characterized by species-rich communities in open habi-
tats, such as carabids and epigeal spiders, responded positively to
salvage logging. Despite positive effects of salvage logging on taxa
associated with open habitats, strong negative effects on saprox-
ylicgroupscall forsubstantialchanges inhowdisturbed forestsare
routinelymanaged.
Naturallydisturbedforestsarecharacterizedbylargevolumesof
deadwoodwithhighstructuraldiversity(Swansonetal.,2011).Incon-
trast,salvageloggingtypicallyreducestheamountandheterogeneity
of deadwood by removing tree trunks (Keyser, Smith, & Shepperd,
2009;Priewasser,Brang,Bachofen,Bugmann,&Wohlgemuth,2013).
Notsurprisingly, salvageloggingreduced thenumbersof speciesof
saproxylicgroups (Figures3and4). However,not onlyadecreasing
deadwoodamountbutlikewisealogging-inducedshiftindeadwood
qualitymay haveadditionalimpacts onsaproxylictaxa. Salvagelog-
gingnotonlyreducestheamountoflargetreetrunksbutalsoalters
characteristicconditions,such as decaystages or diameterdistribu-
tions,oftheremainingdeadwood(Waldronetal.,2013).Forinstance,
branchescutduringpost-stormloggingremainonthegroundbutare
overgrownbygroundvegetation.Theresultingshiftin microclimatic
conditionsthenmodifiesresourcequality,leadingtoalossofsaprox-
ylicbeetles thatdependon sun-exposed, drybranches (Thornetal.,
2014).
It is important to note that losses of saproxylic species can be
presentalso within taxonomicgroups thatdisplayedno responsein
theiroverallspeciesnumbers(Figure3a).Forinstance,birds(themost
studiedvertebrategroup)wereslightlynegativelyaffectedbysalvage
logging(Figure3a),despitefewspeciesbeing directlydependent on
dead wood. Nevertheless, several forest-dwelling bird species de-
pend on snags, cavities or natural regenerationin post-disturbance
forest stands. The removal of such legacies bysalvage logging can
causealossofassociated bird speciesandconsequentlyan overall
lowernumber ofbirdspecies in loggedareas(Hutto &Gallo,2006;
Werner,Müller,Heurich,&Thorn,2015).Althoughtheoverallnumber
ofbirdspeciesdecreasedlessstronglythan,forinstance,thenumber
ofsaproxylic beetlespecies(Figure3a),birdspecies thatdependon
post-disturbancehabitatcharacteristicsareoftenofhighconservation
interest.Forinstance,salvageloggingafterhighseveritywildfirescan
leadtolowersite occupanciesofNorthernSpottedOwls(Strix occi-
dentalis caurina)onlogged than onunloggedsites in Oregon(Clark,
Anthony,&Andrews,2013).
Ourstudyrevealedthatsalvageloggingcausedsignificantchanges
incommunitycomposition forsevenspeciesgroups(Figure3b),with
saproxylic species groups being affected most strongly (Figure4b).
Suchalterationsin community compositionmightreflect the estab-
lishmentofopen-habitatspeciesand/orasimultaneouslossofforest
specialists.Forinstance, salvagelogging canincreasetheabundance
ofopen-habitatcarabidbeetles(Koivula&Spence,2006)orpromote
the establishment of non-forest vegetation (Stuartetal., 1993; Van
Nieuwstadt,Sheil, &Kartawinata,2001).Hence,speciesgroupsthat
are commonly characterized by species-rich communities in open
habitats,suchascarabidsorepigealspiders,candisplayanoverallin-
creaseinnumbersofspeciesinresponsetosalvagelogging(Figure3a).
Likewise, salvagelogging can cause an increase in herb- and grass-
feedingmothspeciesbutadecreaseinsaproxylicanddetritus-feeding
mothspecies (Thornetal., 2015).Such contrastingresponseswithin
andbetweenspeciesgroupscan masktheoverallimpactofsalvage
logging on biodiversity in coarse-scale analyses (i.e. Thom & Seidl,
2016).Numerousspecies ofhighconservationinterest, suchasthe
Red-cockadedWoodpecker(Leuconotopicus borealis),dependondead
woodin burnedforests(Conner,Rudolph,& Walters,2001).The re-
sultsofourstudythereforeindicatethat thebiodiversityofsaprox-
ylictaxacouldbeenhanced bya modifiedmanagementofnaturally
disturbedforests.Incontrast,populationsofspeciesassociatedwith
open habitats, such as the Sharp-tailed Grouse (Tympanuchus pha-
sianellus)inNorthAmerica,maypersistorevenincreaseinthelarger
remaining area subject to unmodified management, that is, salvage
logging(Radeloff,Mladenoff,&Boyce,2000).
Thetwomajor incentivesfor salvageloggingare toreduceeco-
nomic losses caused by a naturaldisturbance and to omit mass re-
productionand spreadofinsectpests thatdevelop intrees killedor
weakenedbya preceding naturaldisturbance.Forinstance,salvage
logging of storm-felled Norway spruce (Picea abies) decreased new
infestationsofnearbytreesbytheEuropeansprucebarkbeetle (Ips
FIGURE2 Distributionofstudies
investigatingtheeffectsofsalvagelogging
onbiodiversityafterwildfire,windstorms
andinsectoutbreaksaccordingtotheyears
afterdisturbance.[Colourfigurecanbe
viewedatwileyonlinelibrary.com]

|
285
Journal of Applied Ecology
THORN eT al.
typographus) at a landscape scale (Stadelmann, Bugmann, Meier,
Wermelinger,&Bigler,2013).Salvageloggingisthereforethepredom-
inantresponsetonaturaldisturbancesinwoodproductionforests,but
pestcontrolis regularlyusedtojustify salvage logginginprotected
areas.Forinstance,theBiałowieżaForestNationalParkontheborder
betweenPolandandBelarus,whichis thelast primevallowland for-
estinEurope,iscurrentlyobligedtosalvageloggingofareasaffected
by I. typographusonattempttoavoidfurtherinfestations(Chylarecki
&Selva,2016). Suchan approachtodisturbed forestsneglects that
regionalfactors,suchassummerdrought,canpromoteoutbreaksof
I. typographus more strongly than local standvariables (Seidl etal.,
2015).Furthermore,salvageloggedtimberis usuallyofsubstantially
lowereconomicvaluethannormallyharvestedtimberduetoarapid
colonization by wood-inhabiting fungi and to the fact that distur-
bancesaffect forestsofanyage,sothat generalizedsalvagelogging
operationsnecessarilyincludeyoungerstands thatotherwisewould
notbeharvested(Leverkus,Puerta-Pinero,Guzmán-Álvarez,Navarro,
& Castro, 2012). Our results demonstrate that salvage logging has
strongandnegative effects on manytaxonomic groups, particularly
thoseassociatedwith deadwood,and thatitisthus notconsistent
with biodiversity conservation goals.Along with questionable eco-
nomicoutputs andpestreducingeffects,weargue thatsalvagelog-
gingshouldbeexcludedfromprotectedareassuchasnationalparks.
The incidence of stand-replacing natural disturbances remains
spatially and temporally unpredictable (Berry etal., 2015), creating
inherentuncertaintyaboutappropriatemanagementofnaturallydis-
turbedforests.Hence,managementplansneedtobejointlydeveloped
with(andconfirmedby)stakeholders,scientistsandnaturalresource
managers before the nextdisturban ceoccurs (Lindenmayer, Likens,
& Franklin, 2010). Such management plans could,for instance, en-
compassan aprioriidentificationofsalvageloggingexclusionzones
basedonecologicaldata(e.g.Nappietal.,2011).Forestmanagersalso
may targetthe preservation of structural key attributes in naturally
disturbed forests, including snags or tipped uproot plates of wind-
throwntrees(Hutto,2006).Retentionoftreesduringgreen-treehar-
vestshas becomeanincreasingly commontool aroundtheglobe to
helpconserveforestbiodiversity(Fedrowitzetal.,2014;Gustafsson
etal.,2012;Mori&Kitagawa,2014).Toobtainsomeeconomicreturn
whileretainingdeadwood-dependenttaxa,werecommendasimple
expansionofthe green-treeretentionapproach toincludenaturally
disturbedforests.Retentionapproachesinnaturallydisturbedforests
couldbeexpectedtobelesscostlythaningreen-treeharvestdueto
theloweropportunitycostofnotharvestingdisturbance-killedtrees.
Approximately70%ofthestudieswecompiledspannedlessthan
5years; studies addressing the long-term effects of salvage logging
arerare(Figure2). However,deadwood, and particularlysnags, are
long-lastingkeybiologicallegacies,andtheirlosscanhavelong-lasting
effectson biodiversity(Hutto,2006). Hence,future researchshould
target the long-term effects of salvage logging after natural distur-
bances.Therearealsotaxonomicbiasesinexistingstudiesinvestigating
FIGURE3 (a)Estimatedresponse
ofHedges’dbasedon238individual
comparisonsofspeciesnumbersin
salvageloggedandunsalvagedforests
affectedbynaturaldisturbances.Higher
speciesnumbersinsalvageloggedareas
correspondtopositiveHedges’d,whereas
negativevaluesindicatelowerspecies
numbersinsalvageloggedareas.(b)Pseudo
F-valuesofpermutationalmultivariate
analysisofvariancebasedon134individual
speciesabundancematrices.Largerpseudo
F-valuescorrespondtolargerchangesin
communitycompositioninducedbysalvage
logging.Asterisksindicatesignificant
responses(seeTablesS1andS2for
statisticaldetails).Forillustrativepurposes,
greydots(andthegreylinejoiningthemfor
emphasis)representthemeaneffectsizein
eachtaxonomicgroup.[Colourfigurecan
beviewedatwileyonlinelibrary.com]

286
|
Journal of Applied Ecology
THORN eT al.
theeffectsofsalvageloggingafternaturaldisturbances.Inparticular,
saproxylicgroupssuchaswood-inhabitingfungihavebeenunderrep-
resentedin empirical studiesdespitetheir high diversityand impor-
tance for ecosystem functioning. Future research should therefore
targetparticularlysaproxylicspeciesgroups.Incontrast,othergroups
havebeenrelativelywellstudiedinonedisturbancetype(e.g.birdsin
burnedforests),but lessinothers,and studies wereconducted pri-
marilyin NorthAmerica, Europeand Asia,butlacking intropicalre-
gions(Figure1).However,differenttypesofnatural disturbancesin
differentpartsoftheworldcanactatverydifferentspatialscalesand
mayrequiredifferentretentionapproaches(Kulakowskietal.,2016).
Furthermore,coniferousforestsoftheNorthernHemisphere—incon-
trasttotropicalforests—arenaturallypronetolarge-scalenaturaldis-
turbances(Lindenmayeretal.,2008),whereasdisturbancesintropical
forestsmostlyhave anthropogeniccausesassociatedwithlong-term
land-use change (e.g. fire to open space for livestock grazing and
agriculture; Peres,Barlow, & Laurance, 2006). Nevertheless, natural
disturbancessuchaswindstormsaffecttropicalforestsaswellastem-
perateforests,andsalvageloggingeffectson tropicalforestsshould
betargetedinfutureresearch(e.g.Lawton&Putz,1988).
In conclusion, these data from a wide range of studies demon-
strate that salvage logging has a range of effects on species num-
bers and community composition ofvarious taxonomic groups,with
important negative consequences for several groups, especially
saproxylic ones. Whilecurrent policies for enhancing biodiversity and
ecosystemservices,suchasgreen-treeretention(e.g.,Gustafssonetal.,
2012), focus mainly on forestssubjected to traditional logging opera-
tions,suchpolicies are largelyabsent fromnaturallydisturbed forests.
Wethereforecallforanexpansionofthegreen-treeretentionapproach
toincludenaturallydisturbedforestsbyleavingsubstantialamountsof
deadwoodonsitetoreducetheimpactofsalvageloggingonbiodiversity.
ACKNOWLEDGEMENTS
Wethanknumerouscontributorsforclarifyingtheirstudiesandthree
anonymousreviewersfortheircommentsonanearlierversionofthis
manuscript.S.T.andS.S.werefundedbytheGermanEnvironmental
Foundation. R.S. and D.T. acknowledge support from the Austrian
ScienceFund(FWF,STARTgrantY895-B25).J.C.acknowledgessup-
portfromgrantP12-RNM-2705andA.L.fromSpanishMINECO(FJCI-
2015-23687).D.B.L.wassupportedbyanARCLaureateFellowship.
AUTHORS' CONTRIBUTIONS
S.T.andJ.M.initiatedthestudy.S.T.analysedandinterpretedthedata
andwrotethefirstdraftofthepaper.TheauthorsnamedfromS.T.to
J.M.arelistedalphabetically,astheycontributedequallyingathering
fielddata,providingcorrectionstosubsequentmanuscriptdraftsand
discussingideas.
DATA ACCESSIBILITY
All data are from previously published articles, see “Data sources”.
Datafromthesearticlescan bemadeavailable uponreasonable re-
questtooriginaldataowners.Alistofdatasourcesusedinthestudy
isprovidedintheDataSourcessection.
REFERENCES
Berry,L.E.,Driscoll,D.A.,Stein,J.A.,Blanchard,W.,Banks,S.C.,Bradstock,
R.A.,&Lindenmayer,D.B.(2015).Identifyingthelocationoffireref-
ugesinwetforestecosystems.Ecological Applications,25,2337–2348.
Blair,D.P.,McBurney,L.M.,Blanchard,W.,Banks,S.C.,&Lindenmayer,D.
B.(2016).Disturbancegradientshowsloggingaffectsplantfunctional
groupsmorethanfire.Ecological Applications,26,2280–2301.
FIGURE4 (a)Estimatedresponseandcorrespondingstandard
errorofsaproxylicandnon-saproxylictaxatosalvageloggingbased
on238individualcomparisons(basedonHedges’d)ofnumbers
ofspeciesinburned,storm-andinsect-affectedforests.Negative
estimatescorrespondtoadecreaseinnumbersofspeciesby
salvagelogging(TableS3).(b)Estimatedresponseandcorresponding
standarderrorofcommunitycompositionofsaproxylicand
non-saproxylictaxabasedonpseudoF-valuesofpermutational
multivariateanalysisofvarianceobtainedfrom134individual
speciesabundancematrices.IncreasingF-valuescorrespondto
largerchangesincommunitycompositioninducedbysalvagelogging
(TableS4).Note,insufficientdata(DD)wereavailableforsaproxylic
taxainburnedforests.Asterisksabovedotsindicatesignificant
differencesintheresponsesbetweensaproxylicandnon-saproxylic
taxawithineachdisturbancetype.Numberofunderlyingdatapoints
isindicatedbythesizeofthecircles,with10and100sizeshownfor
reference.[Colourfigurecanbeviewedatwileyonlinelibrary.com]

|
287
Journal of Applied Ecology
THORN eT al.
Bolker, B. M., Brooks, M. E., Clark, C. J., Geange, S. W., Poulsen, J. R.,
Stevens,M. H. H.,&White,J.-S.S. (2009). Generalizedlinear mixed
models:Apracticalguideforecologyandevolution.Trends in Ecology &
Evolution,24,127–135.
Castro, J., Moreno-Rueda, G., & Hódar, J. (2010). Experimental test of
postfire management in pine forests:Impact of salvage logging ver-
suspartialcutting and noninterventiononbird-species assemblages.
Conservation Biology,24,810–819.
Choi,C.-Y.,Lee,E.-J.,Nam,H.-Y.,Lee,W.-S.,&Lim,J.-H.(2014).Temporal
changes in the breeding bird community caused by post-fire treat-
mentsaftertheSamcheokforestfireinKorea.Landscape and Ecological
Engineering,10,203–214.
Chylarecki,P.,&Selva, N.(2016).Ancient forest:Spareit fromclearance.
Nature,530,419.
Clark,D.A.,Anthony,R.G.,&Andrews,L.S.(2013).Relationshipbetween
wildfire,salvagelogging,andoccupancyofnestingterritoriesbynorth-
ernspottedowls.The Journal of Wildlife Management,77,672–688.
Cobb,T.P.,Langor,D. W.,&Spence,J.R. (2007).Biodiversityandmulti-
pledisturbances: Borealforestgroundbeetle(Coleoptera:Carabidae)
responses to wildfire, harvesting,and herbicide. Canadian Journal of
Forest Research,37,1310–1323.
Cobb,T.P.,Morissette,J. L.,Jacobs,J.M.,Koivula,M.J.,Spence,J.R., &
Langor,D.W.(2011).Effectsofpostfiresalvageloggingondeadwood-
associatedbeetles.Conservation Biology,25,94–104.
Conner, R., Rudolph, D. C., & Walters, J. R. (2001). The red-cockated
woodpecker surviving in a fire-maintained ecosystem. Austin, TX: The
UniversityofTexasPress.
Fedrowitz, K., Koricheva, J., Baker, S. C., Lindenmayer, D. B., Palik, B.,
Rosenvald, R., … Gustafsson, L. (2014). Can retention forestry help
conservebiodiversity?Ameta-analysis.Journal of Applied Ecology,51,
1669–1679.
Franklin,J.F.,Lindenmayer,D.,Macmahon,J.A.,Mckee,A.,Perry,D.A.,
Waide,R.,&Foster,D. (2000).Threadsofcontinuity:Ecosystemdis-
turbances, biological legacies and ecosystem recovery. Conservation
Biology in Practice,1,8–16.
González, M. E., & Veblen, T. T. (2007). Wildfire in Araucaria araucana
forests and ecological considerationsabout salvage logging in areas
recentlyburned.Revista Chilena de Historia Natural,80,243–253.
Gustafsson,L., Baker,S. C.,Bauhus, J.,Beese, W.J.,Brodie,A.,Kouki,J.,
…Franklin,J.F.(2012).Retentionforestrytomaintain multifunctional
forests:Aworldperspective.BioScience,62,633–645.
Halme, P., Toivanen,T., Honkanen,M., Kotiaho, J. S., Mönkkönen, M., &
Timonen, J. (2010). Flawed meta-analysis of biodiversity effects of
forestmanagement.Conservation Biology,24,1154–1156.
Hedges,L.V.,&Olkin, I.(1985).Statistical methods for meta-analysis. San
Diego,CA:AcademicPress.
Hothorn,T.,Bretz,F.,&Westfall,P.(2008).Simultaneousinferenceingen-
eralparametricmodels.Biometrical Journal,50,346–363.
Hutto,R.L. (2006). Towardmeaningfulsnag-management guidelines for
postfiresalvagelogginginNorthAmericanconiferforests.Conservation
Biology,20,984–993.
Hutto,R.L.,&Gallo,S.M.(2006).Theeffectsofpostfiresalvageloggingon
cavity-nestingbirds.Condor,108,817–831.
Inbar,M.,Wittenberg,L.,&Tamir,M.(1997).Soilerosionandforestryman-
agementafterwildfireinamediterraneanwoodland,Mt.Carmel,Israel.
International Journal of Wildland Fire,7,285–294.
Keyser,T.L.,Smith,F.W.,&Shepperd,W.D.(2009).Short-termimpactof
post-firesalvageloggingonregeneration,hazardousfuelaccumulation,
andunderstoreydevelopmentinponderosa pineforestsoftheBlack
Hills,SD,USA.International Journal of Wildland Fire,18,451–458.
Koivula, M., & Spence,J. R. (2006). Effects of post-fire salvage logging
on boreal mixed-wood ground beetle assemblages (Coleoptera,
Carabidae).Forest Ecology and Management,236,102–112.
Koricheva, J., Gurevitch, J., & Mengersen, K. (2013). Handbook of
meta-analysis in ecology and evolution(edsJ.Koricheva,J.Gurevitch&
K.Mengersen).Princeton,NJ:PrincetonUniversityPress.
Kulakowski,D.,Seidl,R.,Holeksa,J.,Kuuluvainen,T.,Nagel,T.A.,Panayotov,
M.,… Bebi,P.(2016).Awalkon thewild side:Disturbancedynamics
andtheconservationandmanagement ofEuropean mountainforest
ecosystems.Forest Ecology and Management,388,120–131.
Kurz,W.,Dymond,C.C.,Stinson,G.,Rampley,G.J.,Neilson,E.T.,Carroll,
A.L., …Safranyik,L.(2008).Mountainpinebeetle andforest carbon
feedbacktoclimatechange.Nature,452,987–990.
Lain,E.J.,Haney,A.,Burris,J. M.,&Burton,J.(2008).Responseofvege-
tationandbirds tosevere winddisturbance and salvagelogging ina
southernborealforest.Forest Ecology and Management,256,863–871.
Lawton,R.,&Putz,F.E.(1988).Naturaldisturbanceandgap-phaseregen-
erationinawind-exposedtropicalcloudforest.Ecology,69,764–777.
Legendre,P.,&Anderson,M.J.(1999).Distance-basedredundancyanaly-
sis:Testingmultispeciesresponsesinmultifactorialecologicalexperi-
ments.Ecological Monographs,69,1–24.
Leverkus,A. B., Puerta-Pinero, C., Guzmán-Álvarez, J. R., Navarro,J., &
Castro,J.(2012). Post-fire salvageloggingincreasesrestorationcosts
inaMediterraneanmountainecosystem.New Forests,43,601–613.
Lindenmayer,D.,Burton,P.J.,&Franklin,J.F.(2008).Salvage logging and its
ecological consequences.Washington:IslandPress.
Lindenmayer,D.B.,Foster,D.R., Franklin,J.F.,Hunter,M.L.,Noss,R. F.,
Schmiegelow,F.A.,&Perry,D.(2004).Salvageharvestingpoliciesafter
naturaldisturbance.Science,303,1303.
Lindenmayer,D.B.,Likens,G.E.,& Franklin,J.F.(2010).Rapidresponses
tofacilitateecologicaldiscoveriesfrommajordisturbances.Frontiers in
Ecology and the Environment,8,527–532.
Lindenmayer, D. B., & Noss, R. F. (2006). Salvage logging, ecosystem
processes, and biodiversity conservation. Conservation Biology, 20,
946–948.
Lindenmayer,D.B.,&Ough,K.(2007).Salvagelogginginthemontaneash
eucalyptforestsoftheCentralHighlands ofVictoriaand itspotential
impactsonbiodiversity.Conservation Biology,20,1005–1015.
Lindenmayer, D.,Thorn, S., & Banks, S. (2017). Please do not disturb
ecosystemsfurther.Nature Ecology and Evolution,1,1–8.
Macdonald,S.E.(2007).Effectsofpartialpost-firesalvageharvestingon
vegetation communities in the boreal mixedwood forest region of
northeastern Alberta, Canada. Forest Ecology and Management, 239,
21–31.
McArdle,B. H., &Anderson,M. J.(2001).Fitting multivariate modelsto
communitydata:Acommenton distance-basedredundancyanalysis.
Ecology,82,290–297.
McIver, J. D., & Starr, L. (2000). Environmental effects of postfire log-
ging: Literature review and annotated bibliography. Portland, OR: U.S.
DepartmentofAgriculture,ForestService,PacificNorthwestResearch
Station.
Mori,A.S.,&Kitagawa,R.(2014).Retentionforestryasamajorparadigm
forsafeguardingforestbiodiversityinproductivelandscapes:Aglobal
meta-analysis.Biological Conservation,175,65–73.
Morimoto,J.,Morimoto,M.,&Nakamura,F.(2011).Initialvegetationre-
covery following a blowdown of a conifer plantation in monsoonal
EastAsia:Impactsoflegacyretention,salvaging,sitepreparation,and
weeding.Forest Ecology and Management,261,1353–1361.
Nappi,A.,&Drapeau,P.(2009).Reproductivesuccessoftheblack-backed
woodpecker (Picoides arcticus) in burned boreal forests: Are burns
sourcehabitats?Biological Conservation,142,1381–1391.
Nappi,A.,Stéphane,D.,Bujold,F.,Chabot,M.,Dumont,M.-C.,Duval,J.,…
Bergeron,I.(2011).Harvestinginburnedforests–Issuesandorienta-
tionsforecosystem-basedmanagement.MinistèredesRessourcesna-
turellesetdelaFaune,Directiondel’environnementetdelaprotection
desforêts,Québec.

288
|
Journal of Applied Ecology
THORN eT al.
Nikiforuk,A.(2011).Empire of the beetle: How human folly and a tiny bug
are killing North America’s great forests,FirstEdit.Vancouver,Canada:
DavidSuzukiFoundationSeries.
Norvez, O., Hébert, C., Bélanger, L., Hebert, C., & Belanger, L. (2013).
Impact of salvagelogging on stand structure and beetle diversity in
borealbalsamfir forest,20years afteraspruce budworm outbreak.
Forest Ecology and Management,302,122–132.
Oksanen,J.,Blanchet,F.G.,Kindt,R.,Legendre,P.,Minchin,P.R.,O’Hara,R.
B.,…Wagner,H.(2016).vegan:CommunityEcologyPackage.Retrieved
fromhttps://cran.r-project.org/package=vegan
Peres,C.A.,Barlow,J.,&Laurance,W.F.(2006).Detectinganthropogenic
disturbance in tropical forests. Trends in Ecology and Evolution, 21,
227–229.
Phillips,I. D.,Cobb,T.P.,Spence,J. R.,&Brigham, R.M. (2006).Salvage
logging,edgeeffects, andcarabidbeetles: Connectionstoconserva-
tionandsustainableforestmanagement.Environmental Entomology,35,
950–957.
Priewasser, K., Brang, P., Bachofen, H., Bugmann, H., & Wohlgemuth, T.
(2013).Impactsofsalvage-loggingonthestatusofdeadwoodafterwind-
throwinSwissforests.European Journal of Forest Research,132,231–240.
Pullin,A. S.,&Stewart, G.B.(2006). Guidelinesforsystematicreviewin
conservation and environmental management. Conservation Biology,
20,1647–1656.
Radeloff,V.C.,Mladenoff,D.J.,&Boyce,M.S.(2000).Effectsofinteracting
disturbancesonlandscapepatterns:Budwormdefoliationandsalvage
logging.Ecological Applications,10,233–247.
Ramage,B.S., Sheil, D.,Salim,H. M. W.,Fletcher,C.,Mustafa, N. Z.A.,
Luruthusamay,J.C.,… Potts,M.D.(2013).Pseudoreplicationintrop-
ical forests and the resulting effects on biodiversity conservation.
Conservation Biology,27,364–372.
Saab,V.,Russell,R.E.,&Dudley,J.G.(2009).Nest-siteselectionbycavity-
nestingbirdsinrelationtopostfiresalvagelogging.Forest Ecology and
Management,257,151–159.
Schmiegelow, F. K.A., Stepnisky, D. P.,Stambaugh, C. A., & Koivula,
M. (2006). Reconciling salvage logging of boreal forests with a
natural-disturbance managementmodel.Conservation Biology, 20,
971–983.
Seidl,R.,Müller,J.,Hothorn,T.,Bässler,C.,Heurich,M.,&Kautz,M.(2015).
Smallbeetle, large-scaledrivers:Howregionaland landscapefactors
affectoutbreaksoftheEuropeansprucebarkbeetle.Journal of Applied
Ecology,53,530–540.
Seidl,R., Schelhaas,M.-J.,Rammer,W.,&Verkerk,P.J.(2014). Increasing
forest disturbances in Europe and their impact on carbon storage.
Nature Climate Change,4,806–810.
Stadelmann, G., Bugmann, H., Meier, F., Wermelinger, B., & Bigler, C.
(2013).Effectsofsalvageloggingandsanitationfellingonbarkbeetle
(Ips typographusL.) infestations.Forest Ecology and Management,305,
273–281.
Stokstad,E.(2006).Salvageloggingresearchcontinuestogeneratesparks.
Science,311,761.
Stuart,J. D., Grifantini,M.C., Fox,L.III., & Fox,L. (1993). Early succes-
sionalpathwaysfollowingwildfireand subsequentsilvicultural treat-
mentinDouglas-fir/hardwoodforests, NWCalifornia. Forest Science,
39,561–572.
Swanson,M.E.,Franklin,J.F.,Beschta,R.L.,Crisafulli,C.M.,DellaSala,D.
A.,Hutto,R. L.,…Swanson,F.J.(2011).Theforgottenstageofforest
succession:Early-successionalecosystemsonforestsites.Frontiers in
Ecology and the Environment,9,117–125.
Thom, D., & Seidl,R. (2016). Natural disturbance impacts on ecosystem
services and biodiversity in temperate and boreal forests. Biological
Reviews,91,760–781.
Thorn, S., Bässler,C., Bernhardt-Römermann, M., Cadotte, M., Heibl, C.,
Schäfer, H., … Müller, J. (2016). Changes in the dominant assembly
mechanismdrivesspecieslosscausedbydecliningresources.Ecology
Letters,19,163–170.
Thorn, S., Bässler, C., Gottschalk, T., Hothorn, T., Bussler, H., Raffa, K.,
& Müller, J. (2014). New insights into the consequences of post-
windthrowsalvageloggingrevealedbyfunctionalstructureofsaprox-
ylicbeetlesassemblages.PLoS ONE,9,e101757.
Thorn,S.,Bässler,C.,Svoboda,M.,&Müller,J.(2016).Effectsofnatural
disturbancesand salvageloggingonbiodiversity–Lessonsfromthe
BohemianForest.Forest Ecology and Management,388,113–119.
Thorn,S.,Hacker,H.H.,Seibold,S.,Jehl,H.,Bässler,C.,&Müller,J.(2015).
Guild-specificresponsesofforestLepidopterahighlightconservation-
oriented forest management – Implications from conifer-dominated
forests.Forest Ecology and Management,337,41–47.
Thorn,S.,Werner,S.A.B.,Wohlfahrt,J.,Bässler,C.,Seibold,S.,Quillfeldt,
P.,&Müller,J.(2016).Responseofbirdassemblagestowindstormand
salvagelogging—Insightsfromanalysesoffunctionalguildandindica-
torspecies.Ecological Indicators,65,142–148.
VanNieuwstadt,M.G.L.,Sheil,D.,&Kartawinata,K.(2001).Theecologi-
calconsequencesofloggingintheburnedforestsofEastKalimantan,
Indonesia.Conservation Biology,15,1183–1186.
Viechtbauer,W.(2010).Conductingmeta-analysesinRwith themetafor
package.Journal of Statistical Software,36,1–48.
Waldron,K.,Ruel,J.,&Gauthier,S.(2013).Foreststructuralattributesafter
windthrow and consequences ofsalvage logging. Forest Ecology and
Management,289,28–37.
Werner,S.A.B.,Müller,J.,Heurich,M.,&Thorn,S.(2015).Naturalregen-
erationdetermineswinteringbirdpresenceinwind-damagedconifer-
ousforeststandsindependentofpost-disturbance logging.Canadian
Journal of Forest Research,45,1232–1237.
Zmihorski, M. (2010). The effect of windthrow and its managementon
breeding bird communities in a managed forest. Biodiversity and
Conservation,19,1871–1882.
Zmihorski,M.,&Durska,E.(2011).Theeffectofcontrastingmanagement
typesontwodistincttaxonomic groupsin alarge-scaledwindthrow.
European Journal of Forest Research,130,589–600.
DATA SOURCES
Barnouin,T.(2005).ÉvaluationdeL‘ImportanceDesForêtsAffectéesParLaTordeuse
DesBourgeons deLâ€TMÉpinette,ChoristoneuraFumiferana(Clem.),Dans Le
MaintiendeLaDiversitéDesColéoptèresSaproxyliques.UniversitéLavalQuebec.
Blair,D.P.,McBurney,L.M.,Blanchard,W.,Banks,S.C.,&Lindenmayer,D.B.(2016).
Disturbance gradient shows loggingaffects plant functional groups more than
fire.Ecological Applications,26,2280–2301.
Boucher,D.,Gauthier,S.,Noël,J.,Greene,D.F.,&Bergeron,Y.(2014).Salvagelogging
affectsearlypost-firetreecomposition inCanadianboreal forest.Forest Ecology
and Management,325,118–127.
Bradbury,S.M.(2006).Responseofthepost-firebryophytecommunitytosalvagelog-
ginginborealmixedwoodforestsofnortheasternAlberta,Canada.Forest Ecology
and Management,234,313–322.
Cahall,R.E.,&Hayes,J.P.(2009).Influencesofpostfiresalvageloggingonforestbirdsinthe
EasternCascades,Oregon,USA.Forest Ecology and Management,257,1119–1128.
Castro,J.,Moreno-Rueda,G.,&Hódar,J.(2010).Experimentaltestofpostfiremanage-
mentinpineforests:Impactofsalvageloggingversuspartialcuttingandnoninter-
ventiononbird-speciesassemblages.Conservation Biology,24,810–819.
Choi,C.-Y.(2007).Effectsofpostfireloggingonbirdpopulationsandcommunitiesin
burnedforests.Journal of the Korean Forest Society,96,115–123.
Choi,C.-Y.,Lee,E.-J.,Nam,H.-Y.,Lee,W.-S.,&Lim,J.-H.(2014).Temporalchangesin
thebreedingbirdcommunitycausedbypost-firetreatmentsaftertheSamcheok
forestfireinKorea.Landscape and Ecological Engineering,10,203–214.
Cobb,T. P.,Langor, D. W.,& Spence, J. R. (2007). Biodiversity and multiple distur-
bances:Borealforestgroundbeetle(Coleoptera:Carabidae)responsestowildfire,
harvesting,andherbicide.Canadian Journal of Forest Research,37,1310–1323.
Cobb,T.P.,Morissette,J.L.,Jacobs,J.M.,Koivula,M.J., Spence,J.R., &Langor, D.
W.(2011). Effectsof postfiresalvageloggingon deadwood-associatedbeetles.
Conservation Biology,25,94–104.
Donato, D. C., Fontaine,J. B., Campbell, J. L., Robinson, W.D., Kauffman, J. B., &
Law,B.E.(2009).Coniferregenerationinstand-replacementportionsofalarge

|
289
Journal of Applied Ecology
THORN eT al.
mixed-severitywildfire in the Klamath-SiskiyouMountains. Canadian Journal of
Forest Research,39,823–838.
Donato,D.C.,Fontaine,J.B.,Kauffman,J.B.,Robinson,D., &Law,B.E.(2013).Fuel
massandforeststructurefollowingstand-replacementfireandpost-firelogging
inamixed-evergreenforest.International Journal of Wildland Fire,22,652–666.
Duelli,P.,Obrist,M.K.,&Wermelinger,B.(2002).Windthrow-induced changesinfau-
nisticbiodiversityinalpinespruceforests.Forest Snow and Landscape Research,131,
117–131.
Durska,E. (2013).Effectsofdisturbanceson scuttleflies (Diptera:Phoridae)in Pine
Forests.Biodiversity and Conservation,22,1991–2021.
Fontaine,J. B.(2007). Influences of high severity fire and postfire logging on avian and
small mammal communities of the Siskiyou Mountains.Corvallis,OR:OregonState
University.
Fontaine,J.B., Donato,D.C.,Robinson,W.D.,Law,B. E.,& Kauffman,J.B. (2009).
Birdcommunitiesfollowinghigh-severityfire:Responsetosingleandrepeatfires
inamixed-evergreenforest,Oregon,USA.Forest Ecology and Management,257,
1496–1504.
Hutto,R. L.,& Gallo,S.M.(2006).The effectsofpostfire salvageloggingoncavity-
nestingbirds.Condor,108,817–831.
Jeffrey,O.(2013).EffetsDesCoupesdeRécupérationSurLesSuccessionsNaturelles
deColéoptèresSaproxyliquesLeLongD’une Chronoséquencede15 AnsAprès
FeuEnForêtBoréaleCommerciale.Univer.
Jonasova,M.,&Prach,K. (2008).Theinfluenceofbark beetlesoutbreakvs. salvage
loggingongroundlayervegetationinCentralEuropeanmountainspruceforests.
Biological Conservation,141,1525–1535.
Koivula, M., & Spence, J.R. (2006). Effects of post-fire salvage logging on boreal
mixed-wood groundbeetle assemblages(Coleoptera,Carabidae).Forest Ecology
and Management,236,102–112.
Kurulok,S. E.,&Macdonald,S.E. (2007).Impacts ofpostfiresalvage loggingonun-
derstoryplantcommunitiesoftheborealmixedwoodforest2and34yearsafter
disturbance.Canadian Journal of Forest Research,37,2637–2651.
Lee,E.L.W.,Hun,S., Rhim,S.-J.,Lee,W.-S.,& Son,S.H.(2011).Differencesinbird
communitiesinpostfiresilviculturalpracticesstandswithinpineforestofSouth
Korea.Landscape and Ecological Engineering,7,137–143.
Lee,E., Lee,W., &Rhim, S.-J.(2008). Characteristics ofsmall rodentpopulations in
post-fire silvicultural management stands within pine forest.Forest Ecology and
Management,255,1418–1422.
Lee,E.-J.,Rhim,S.,Son, S.,&Lee,W.(2012).Differencesinsmall-mammalandstand
structuresbetween unburned andburned pine standssubjectedto twodiffer-
entpost-firesilvicultural managementpractices.Annales Zoologici Fennici,2450,
129–138.
Leverkus,A. B.,Lorite,J., Navarro,F.B., Sánchez-Cañete, E.P.,&Castro, J.(2014).
Post-fire salvage logging alters species composition and reduces cover, rich-
ness,anddiversityin Mediterraneanplantcommunities.Journal of Environmental
Management,133,323–331.
Liepold,K.(2004).VergleichendeUntersuchungenZurFaunistischenUndGenetischen
DiversitätvonKäferzönoseninGenutzten UndUngenutzten Bergmischwäldern
DesBayerischenWaldes.TUMMünchen.
Mehr,M.,Brandl,R.,Kneib,T.,&Müller,J.(2012).Theeffectofbarkbeetleinfestation
andsalvageloggingonbatactivityinanationalpark.Biodiversity and Conservation,
21,2775–2786.
Norvez,O.,Hébert,C., Bélanger,L.,Hebert,C.,& Belanger,L.(2013).Impactofsal-
vageloggingon standstructureand beetlediversityinboreal balsamfirforest,
20yearsafterasprucebudwormoutbreak.Forest Ecology and Management,302,
122–132.
Park,C. (2012).Differences in characteristics of amphibian and reptile communities due
to different forest environment at forest fired area in Samcheok, Gangwon Province.
Seoul,Korea:SeoulNationalUniversity.
Peterson,C.J.,&Leach,A.D.(2008).Salvageloggingafterwindthrowaltersmicrosite
diversity,abundanceandenvironment,butnotvegetation.Forestry,81,361–376.
Rost,J.,Clavero,M.,Brotons,L.,& Pons,P.(2012).Theeffectofpostfiresalvagelog-
gingonbirdcommunitiesinMediterraneanpineforests:Thebenefitsfordeclining
species.Journal of Applied Ecology,49,644–651.
Rumbaitis,C.M., &RumbaitisdelRio, C.M.(2006).Changesinunderstorycomposi-
tionfollowing catastrophicwindthrowand salvagelogging ina subalpineforest
ecosystem.Canadian Journal of Forest Research – Revue Canadienne De Recherche
Forestiere,36,2943–2954.
Thorn,S., Bässler,C.,Bernhardt-Römermann,M.,Cadotte, M.,Heibl, C.,Schäfer,H.,
…Müller,J.(2016).Changesinthedominantassemblymechanismdrivesspecies
losscausedbydecliningresources.Ecology Letters,19,163–170.
Thorn, S., Bässler,C., Gottschalk, T., Hothorn,T., Bussler,H., Raffa, K., & Müller, J.
(2014).New insightsinto theconsequencesofpost-windthrow salvagelogging
revealedbyfunctionalstructureofsaproxylicbeetlesassemblages.PLoS ONE,9,
e101757.
Thorn,S.,Hacker,H.H., Seibold,S.,Jehl, H.,Bässler, C., &Müller, J.(2015). Guild-
specific responses of forestLepidoptera highlight conservation-oriented forest
management – Implications fromconifer-dominated forests. Forest Ecology and
Management,337,41–47.
Thorn, S., Werner, S. A. B., Wohlfahrt,J., Bässler, C., Seibold, S., Quillfeldt, P., &
Müller,J. (2016).Responseof birdassemblages towindstorm and salvagelog-
ging—Insightsfromanalyses offunctionalguildandindicatorspecies.Ecological
Indicators,65,142–148.
Waldron,K.,Ruel,J.-C.,Gauthier,S.,DeGrandpré,L.,&Peterson,C.J.(2014).Effects
ofpost-windthrowsalvageloggingonmicrosites,plantcompositionandregener-
ationed.J. Ewald. Applied Vegetation Science,17,323–337.
Wermelinger,B.,Moretti,M.,Duelli,P.,Lachat,T.,Pezzatti,G.B.,&Obrist,M.K.(2017).
Impactofwindthrowandsalvage-loggingontaxonomicandfunctionaldiversityof
forestarthropods.Forest Ecology and Management,391,9–18.
Winter,M.-B.,Ammer,C.,Baier,R.,Donato,D.C.,Seibold,S., &Müller,J.(2015).Multi-
taxonalphadiversityfollowingbarkbeetledisturbance:Evaluatingmulti-decadeper-
sistenceofadiverseearly-seralphase.Forest Ecology and Management,338,32–45.
Zmihorski,M.(2010).Theeffectofwindthrowanditsmanagementonbreeding bird
communitiesinamanagedforest.Biodiversity and Conservation,19,1871–1882.
SUPPORTING INFORMATION
Additional Supporting Information may be found online in the
supportinginformationtabforthisarticle.
How to cite this article:ThornS,BässlerC,BrandlR,etal.
Impactsofsalvageloggingonbiodiversity:Ameta-analysis.
J Appl Ecol. 2018;55:279–289. https://doi.org/10.1111/1365-
2664.12945