ARTICLE

Epixylic vegetation abundance, diversity, and composition varywith coarse woody debris decay class and substrate species inboreal forestPraveen Kumar, Han Y.H. Chen, Sean C. Thomas, and Chander Shahi

Abstract: Although the importance of coarse woody debris (CWD) to understory species diversity has been recognized, thecombined effects of CWD decay and substrate species on abundance and species diversity of epixylic vegetation have receivedlittle attention. We sampled a wide range of CWD substrate species and decay classes, as well as forest floors in fire-origin borealforest stands. Percent cover, species richness, and evenness of epixylic vegetation differed significantly with both CWD decayclass and substrate species. Trends in cover, species richness, and evenness differed significantly between nonvascular andvascular taxa. Cover, species richness, and species evenness of nonvascular species were higher on CWD, whereas those ofvascular plants were higher on the forest floor. Epixylic species composition also varied significantly with stand ages, overstorycompositions, decay classes, substrate species, and their interactions. Our findings highlight strong interactive influences ofdecay class and substrate species on epixylic plant communities and suggest that conservation of epixylic diversity would requireforest managers to maintain a diverse range of CWD decay classes and substrate species. Because stand development andoverstory compositions influence CWD decay classes and substrate species, as well as colonization time and environmentalconditions in the understory, our results indicate that managed boreal landscapes should consist of a mosaic of differentsuccessional stages and a broad suite of overstory types to support diverse understory plant communities.

Key words: boreal forest, coarse woody debris, decay class, epixylic plants, substrate species.

Résumé : Bien que l’importance des débris ligneux grossiers (DLG) pour la diversité des espèces de sous-bois ait été démontrée,les effets combinés de la décomposition des DLG et de l’espèce du substrat sur l’abondance et la diversité des espèces de végétauxépixyliques ont été peu étudiés. Nous avons échantillonné une grande variété d’espèces du substrat de DLG et de classes dedécomposition ainsi que de couvertures mortes dans des peuplements de forêt boréale issus de feu. Le pourcentage de couvert,la richesse en espèces et l’homogénéité de la végétation épixylique étaient significativement différents selon la classe dedécomposition et l’espèce du substrat de DLG. Il y avait des différences significatives entre les taxons vasculaires et nonvasculaires en ce qui a trait aux tendances dans le couvert, la richesse en espèces et l’homogénéité des espèces. Ces attributsprenaient davantage d’ampleur sur les DLG dans le cas des plantes non vasculaires et sur la couverture morte dans le cas desplantes vasculaires. La composition en espèces épixyliques variait également de façon significative selon l’âge du peuplement, lacomposition de l’étage dominant, la classe de décomposition, l’espèce du substrat et leurs interactions. Nos résultats mettent enévidence de fortes influences interactives de la classe de décomposition et de l’espèce du substrat sur les communautés végétalesépixyliques et indiquent que la conservation de la diversité épixylique exigerait que les gestionnaires forestiers maintiennent unéventail diversifié de classes de décomposition des DLG et d’espèces du substrat. Étant donné que le développement dupeuplement et la composition de l’étage dominant influencent les classes de décomposition des DLG et les espèces du substrat,ainsi que le moment de la colonisation et les conditions environnementales dans le sous-bois, nos résultats indiquent que lepaysage boréal aménagé devrait être constitué d’une mosaïque de différents stades successionnels et d’une vaste suite de typesd’étage dominant pour supporter diverses communautés végétales de sous-bois. [Traduit par la Rédaction]

Mots-clés : forêt boréale, débris ligneux grossiers, classe de décomposition, plantes épixyliques, espèces du substrat.

IntroductionCoarse woody debris (CWD) plays a key role in biodiversity and

ecosystem functioning as it provides habitat for many plants andanimals (Bunnell and Houde 2010; Stokland et al. 2012) and con-tributes to carbon and nutrient cycles (Harmon et al. 1986; Laihoand Prescott 1999; Ohtsuka et al. 2014). The importance of CWD tomaintaining understory species diversity has been emphasized bynumerous authors (Andersson and Hytteborn 1991; Crites and Dale1998; Rambo and Muir 1998; Kruys and Jonsson 1999; Humphrey

et al. 2002; Mills and Macdonald 2004; Bunnell et al. 2008; Bottingand DeLong 2009; Caruso and Rudolphi 2009; Dittrich et al. 2014;Checko et al. 2015). It is widely recognized that numerous plant spe-cies are specialized to a greater or lesser extent on CWD substrates,and such species are termed “epixylic”. The focus of most priorstudies of epixylic plants has been on either nonvascular (lichensand bryophytes) or vascular groups occurring on different CWDdecay classes. These studies have generally found that speciesrichness and abundance of nonvascular species (lichens and bryo-phytes) peak in intermediate to later stages of decay, while those

Received 31 July 2017. Accepted 10 November 2017.

P. Kumar, H.Y.H. Chen, and C. Shahi. Faculty of Natural Resources Management, Lakehead University, 955 Oliver Road, Thunder Bay, ON P7B 5E1, Canada.S.C. Thomas. Faculty of Forestry, University of Toronto, 33 Willcocks St., Toronto, ON M5S 3B3, Canada.Corresponding author: Han Y.H. Chen (email: hchen1@lakeheadu.ca).Copyright remains with the author(s) or their institution(s). Permission for reuse (free in most cases) can be obtained from RightsLink.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

1

Can. J. For. Res. 48: 1–13 (2018) dx.doi.org/10.1139/cjfr-2017-0283 Published at www.nrcresearchpress.com/cjfr on 16 January 2018.

Can

. J. F

or. R

es. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y L

akeh

ead

Uni

vers

ity o

n 02

/20/

18Fo

r pe

rson

al u

se o

nly.

of vascular plants continue to increase at advance stages of decay.A few studies have also examined the effect of substrate species ofdeadwood on patterns of epixylic plant abundance and diversity(McAlister 1997; Rambo 2001; Mills and Macdonald 2004; Nowinskaet al. 2009). The simultaneous effects of CWD decay classes andsubstrate species have recently been examined in terms of under-story vegetation species composition (Kumar et al. 2017), but notabundance and species diversity. Species richness and evennessmay respond differently to different ecological processes (Stirlingand Wilsey 2001; Wilsey and Stirling 2007); however, to our knowl-edge, no studies have explicitly assessed patterns of species even-ness in relation to CWD decay. In addition, very few studies havemade a comparison of species abundance and diversity on dead-wood substrates to corresponding forest floor substrates (Dittrichet al. 2014; Checko et al. 2015).

Compared with adjacent forest floor substrates, decaying CWDprovides a variable substrate at different stages of decompositionand is colonized by a different subset of species (Lee and Sturgess2001, 2002; Dittrich et al. 2014). Decaying CWD logs are typicallycolonized by lichen and bryophyte species, which take advantageof lower competition than on the forest floor (Humphrey et al.2002; Dittrich et al. 2014), whereas the forest floor is mostly dom-inated by vascular plant species. The structural and chemical propertiesof CWD change with the decomposition process (Ganjegunte et al.2004; Bütler et al. 2007; Petrillo et al. 2015; Shorohova et al. 2016),until CWD becomes an integral part of the forest floor. These pro-gressive changes in CWD properties over time are likely to be amain driver of the colonization and growth of different nonvas-cular and vascular species. The early stages of CWD decomposi-tion are dominated by nonvascular (mostly lichen) species, someof which normally grow on living trees as epiphytes, whereas thelater stages of decay are characterized by bryophytes and vascularplant species (McCullough 1948; Barkman 1958; Söderström 1988;Jansová and Soldán 2006; Botting and DeLong 2009; Kumar et al.2017). Eventually, the CWD merges with the forest floor in thefinal stages of decay; in boreal systems, the substrate is occupiedby feathermosses and vascular plants that normally grow on theforest floor (Söderström 1988; Andersson and Hytteborn 1991;Kumar et al. 2017). In addition to changes in physical and chemicalproperties of the CWD with decay (Laiho and Prescott 2004; Bütleret al. 2007; Petrillo et al. 2015), the time available for colonizationincreases (Bartels and Chen 2015). Both the decay process and timeelapsed will likely influence abundance and diversity because ofdispersal limitation and successional processes.

Patterns in epixylic plant communities may also differ amongCWD substrate species because of species-specific differences inphysical and chemical traits (pH and texture) of barks (Barkman1958; Weedon et al. 2009; Mezaka et al. 2012; Pereira et al. 2014;Putna and Mezaka 2014; Shorohova and Kapitsa 2014; Shorohovaet al. 2016). For example, the barks of coniferous species are gen-erally drier and more acidic than those of broadleaved tree species(Culberson 1955; Barkman 1958; Hauck and Javkhlan 2009; Jüriadoet al. 2009; Mezaka et al. 2012). Furthermore, broadleaved treespecies, particularly fast-growing species such as Populus spp. andBetula spp. common in boreal forests, generally have higher de-composition rates than coniferous species (Barkman 1958; Weedonet al. 2009; Russell et al. 2014; Shorohova and Kapitsa 2014). As aresult, different epixylic plant communities have been observedon broadleaved and coniferous CWD substratum (Palmer 1986;McAlister 1997; Rambo 2001; Mills and Macdonald 2005). Nonvas-cular species (bryophytes and lichens) can generally tolerate sub-strates with higher acidity (Hauck and Jürgens 2008; Pereira et al.2014), whereas most vascular plants establish on substrates thatare nutrient-rich and less acidic. Although the species specificityof nonvascular species associated with CWD species has been rec-ognized (Culberson 1955; McAlister 1997; Rambo 2001; Spribilleet al. 2008; Caruso and Rudolphi 2009; Kumar et al. 2017), the

combined effects of CWD decomposition and CWD substrate spe-cies on understory vegetation abundance, species richness, andevenness remain unclear.

In boreal forests, fire is the major natural disturbance respon-sible for creation and elimination of deadwood on the forest floor(Arseneault 2001; Karjalainen and Kuuluvainen 2002; Brassardand Chen 2006); however, the importance of other disturbancessuch as insect outbreak and windthrow in altering stand structurehas been also recognized (Chen and Popadiouk 2002; Brassard andChen 2006; Bergeron et al. 2014). Postdisturbance CWD dynamicsusually results in a U-shaped pattern, where the volume of CWD ofprefire origin decreases logarithmically and the CWD of postfireorigin increases exponentially from increased tree mortality asso-ciated with competition, ageing, and disturbances such as wind-throw and insect outbreaks (Sturtevant et al. 1997; Brassard andChen 2006; Luo and Chen 2011). In addition to stand development,structure and composition of the overstory also affect the volumeand compositional diversity of CWD (Brassard and Chen 2008),because overstory structure and composition influence stand pro-ductivity (Zhang et al. 2012), tree mortality (Chen and Luo 2015),and CWD decomposition rates (Mäkinen et al. 2006). The structureand composition of the overstory varies through forest successionin boreal forests: multiple successional pathways can result inbroadleaf, conifer, or mixedwood-dominated stands (Chen andPopadiouk 2002; Taylor and Chen 2011). CWD in broadleaf standsmay influence the epixylic plant communities differently thanthe CWD in conifer-dominated stands as a result of distinct envi-ronmental conditions in different stand types (Barbier et al. 2008;Chávez and Macdonald 2012; Bartels and Chen 2013; Huo et al.2014). Broadleaf stands have higher resource availability in theunderstory than conifers (Messier et al. 1998; Hart and Chen 2006;Chávez and Macdonald 2010) and are preferred by species thatrequire resource-rich environments as compared with species thatare able to tolerate nutrient-poor conditions in conifer stands.Mixedwood stands, on the other hand, provide heterogeneousconditions and exhibit greater spatial and temporal variability inunderstory resources (Bartemucci et al. 2006; Macdonald andFenniak 2007). Therefore, stand age along with overstory speciescomposition may influence CWD volume and diversity as well asthe extent of postdisturbance recovery of epixylic plant commu-nities on individual CWD substrata.

The purpose of this study was to examine the combined effectof CWD decay classes and substrate species on abundance, speciesdiversity, and composition of understory vegetation in the centralboreal forest of Canada. To cover a wide range of CWD decayclasses and substrate species, we sampled fire-origin boreal foreststands that varied in canopy species composition and stand age.We expected that total vegetation cover would be higher on thelater decay classes of the broadleaved substrate species because ofhigher moisture content at later stages of decay, which wouldparticularly enhance recruitment and growth of feathermossesand vascular plant species at later stages of decay. We hypothe-sized that both species richness and evenness would be higher onintermediate decay classes of coniferous substrates, because theacidic bark and slower rate of fragmentation in conifers may pro-vide favorable opportunities for establishment and colonizationof both nonvascular and vascular species and because some spe-cies would outcompete others at late decay stages. By contrast,CWD of broadleaved species might not show such a pattern due togreater bark sloughing and fragmentation. We also expected thatepixylic plant community composition would differ with standage and overstory composition due to colonization time associ-ated with stand age and environmental conditions, as well asphysical and chemical properties of CWD substrate species asso-ciated with overstory composition. Finally, we predicted thatepixylic plant communities would differ among CWD decayclasses and the forest floor, because of increased time available for

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

2 Can. J. For. Res. Vol. 48, 2018

Published by NRC Research Press

Can

. J. F

or. R

es. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y L

akeh

ead

Uni

vers

ity o

n 02

/20/

18Fo

r pe

rson

al u

se o

nly.

colonization and changes in substrate suitability with CWD decayand forest floor development.

Material and methods

Study areaData analyzed are identical to those of a companion study ex-

amining species composition on coarse woody debris, with a focuson indicator species of decay class and CWD substrate species(Kumar et al. 2017): additional details on study sites and samplingdesign are provided in that paper. Sampling was conducted atboreal mixedwood sites approximately 100 km north of ThunderBay, Ontario, Canada (49°23=N to 49°36=N, 89°31=W to 89°44=W).Mean annual temperature is 2.5 °C and annual precipitation is712 mm at the closest meteorological station (Thunder Bay, Ontario;Environment Canada 2015). Canopy trees include jack pine (Pinusbanksiana Lamb.), trembling aspen (Populus tremuloides Michx.),white birch (Betula papyrifera Marsh.), black spruce (Picea marianaMill. B.S.P.), white spruce (Picea glauca (Moench) Voss), and balsamfir (Abies balsamea L. Mill.). Fire is the main disturbance agent in theregion, with an average fire return interval of approximately100 years (Senici et al. 2010).

Sampling designWe sampled a set of stands representing a replicated chronose-

quence (with stands of ages 34, 98, 146, and 210 years since laststand-replacing fire) and of diverse overstory composition (Brassardand Chen 2008). Fire-origin stands in the region can be dominated byconifers, broadleaved trees, or a mixture (Taylor and Chen 2011): in sofar as possible, we sampled all three overstory stand types for eachage class, with three replicates for each. All combinations of age classand overstory type were sampled, with the exception of a 146-year-old stand that did not have road access, resulting in a total of35 sampled stands. All selected stands were located on mesic sitesat flat or mid-slope positions, with no slope exceeding 5%. Soilparent material consisted of well-drained (sandy or silty loam)glacial moraines, >50 cm in thickness.

Data collectionSampling was conducted in a 400 m2 circular plot randomly

selected in each stand. Logs ≥ 10 cm in diameter at midpointwithin each plot were randomly selected and decay class wasrecorded. We assigned each selected CWD piece (downed woodydebris on forest floor only) to one of five decay class categories,following the field manual of British Columbia Ministry of Forestsand Range and British Columbia Ministry of Environment (2010).Categorization was based on CWD contact with the ground, woodtexture, presence or absence of branches, wood strength, barkintactness, and presence and depth of invading roots. The charac-teristics of decay class 1 were logs elevated from ground, bark orbranches hard and intact, and no invading roots present; those ofdecay class 2 included logs elevated but slightly sagging, presenceof sap rot so that a thumbnail can penetrate, loose bark and softbranches with no invading roots; those of decay class 3 includedlogs sagging or broken, advanced decay (spongy, large pieces) withtrace bark, no branches, and invading roots present in sapwood;those of decay class 4 included logs fully settled on ground, exten-sive decay (crumbly–mushy), bark and branches absent, and in-vading roots present in heartwood; and those of decay class 5included small pieces and soft portions, bark and branches ab-sent, and invading roots present in heartwood. Morphologicalcharacteristics as defined by Brassard and Chen (2008) were usedto identify CWD to species (as detailed in (Kumar et al. 2017)).

Vegetation surveys were conducted in July and August 2014. Wevisually estimated the percent cover of each species following the

method of Mueller-Dombois and Ellenberg (1974) within a 0.10 ×0.50 m (0.05 m2) quadrat, randomly laid lengthwise on top of theeach sampled CWD log. We also sampled forest floor vegetationby establishing an adjacent plot of the same size at a distance of1.0 m in a random direction from the CWD vegetation sample.With a few exceptions, all plants were identified to species; weomitted minute liverwort species; Cladonia coniocraea (Flörke) andCladonia ochrochlora (Flörke) were recorded as Cladonia agg.; andthe moss genera Brachythecium and Mnium were not separated tospecies. The omission of some minor species could lead to poten-tial bias in our results; however, pooled species constituted only11.8% of the total understory vegetation sampled by cover.

Data analysesSpecies abundance was evaluated as the sum of individual spe-

cies’ percent cover in each quadrat, species richness was treated asthe total number of species recorded in each quadrat, and speciesevenness was expressed as how evenly the species in the commu-nity were distributed. Species evenness was calculated followingPielou (1969) as evenness � ���pi logpi�/ln�richness�, where pi isthe proportion of percent cover of species i to total vegetationcover. Total vegetation cover was treated as the sum of species-specific cover values on an individual CWD piece. In addition tototal understory cover, richness, and evenness, we separately an-alyzed species cover, richness, and evenness for nonvascular andvascular groups. The nonvascular species included lichens andbryophytes, whereas vascular plants included herbs, shrubs, trees,ferns, and clubmosses. To test for substrate species and CWD decayclass effects on understory cover, species richness, and evenness, weused the following general linear mixed-effects model:

(1) Yijkl � � � Di � Sj � D × Sij � �k � � l(ijk)

where Yijkl is understory plant species cover, richness, or evenness(separately analyzed by total, nonvascular, and vascular groups),� is the overall mean, Di (i = 1, 2, …, 5) is decay class, Sj (j = 1, 2, …, 5)is substrate species, �k is plot random effect that accounts forspatial autocorrelation among logs within each sample plot aswell as the variation in stand age and overstory type related toplot, and �l(ijk) (l = 1, 2, …, n) is random sampling error within decayclass, substrate species, and plot. We conducted the mixed-effectsanalysis using restricted maximum likelihood estimation usingthe lme4 package in R (Bates et al. 2016; R Core Team 2017). How-ever, vascular plant data did not conform to the assumption ofnormality based on the Shapiro–Wilk test and were skewed to theleft. To mitigate the violation to the normality assumption and toimprove coefficient estimates, we bootstrapped the fitted coeffi-cients of linear models by using “ggplot2” (Wickham 2009). Webootstrapped the 95% confidence intervals and considered esti-mates to be significantly different if their confidence intervals didnot overlap others’ means. For species richness, we specified theresidual distribution as Poisson. The significance of the predictorswas tested using F tests by means of analysis of variance (ANOVA)or deviance, and the variance or deviance explained by each vari-able in the model was calculated as a percentage of variance ordeviance explained by the variable to that of null model. To ex-plicitly test the effect of stand age and overstory composition onunderstory cover, species richness, or evenness, we added thesevariables and their two-way interactions (removing nonsignifi-cant higher order interactions) in a revised general linear mixed-effects model. The results of the revised model are presented inSupplementary Table S11.

We used permutation multivariate ANOVA (perMANOVA) toexamine the effect of stand age, overstory composition, decay

1Supplementary data are available with the article through the journal Web site at http://nrcresearchpress.com/doi/suppl/10.1139/cjfr-2017-0283.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

Kumar et al. 3

Published by NRC Research Press

Can

. J. F

or. R

es. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y L

akeh

ead

Uni

vers

ity o

n 02

/20/

18Fo

r pe

rson

al u

se o

nly.

class, substrate species, and their interactions on understory spe-cies composition. PerMANOVA is a nonparametric, multivariateanalysis that uses permutation techniques to test for composi-tional differences between more than one factor (Anderson 2005).We used perMANOVA with Bray–Curtis dissimilarity and 1000 per-mutations for the composition data. To examine trends in thecomposition data, we used nonmetric multidimensional scaling(NMDS) (Kruskal 1964), a robust ordination technique for commu-nity data that are non-normal or evaluated on discontinuous orordinal scales (McCune and Grace 2002). The analysis was per-formed (i) to examine differences in the species composition onCWD between stand ages and overstory compositional types and(ii) to examine the differences in species composition betweenCWD decay classes, CWD substrate species, and the forest floor.All statistical analyses were conducted in R (ver. 3.4.2; R CoreTeam 2017).

ResultsWe recorded a total of 68 understory species on CWD and forest

floor, including 33 nonvascular species and 35 vascular species.Total epixylic cover differed with decay class (P < 0.001; Table 1),and the effect of substrate species was dependent on decay class,indicated by a significant interaction of decay class and substratespecies (Table 1). Decay class was the main contributing factor,explaining 47.2% of the variance, while the interaction explained4.2% of variance. Within decay classes 1 and 2, epixylic vegetationcover did not differ among the five substrate species, while atdecay class 3, Pinus banksiana and Populus substrate species had ahigher percent cover than other species (Fig. 1). At decay classes 4and 5, percent cover was higher on Populus spp., Picea spp., andPinus banksiana than on Betula papyrifera and Abies balsamea. Com-pared with forest floor, Populus spp., Picea spp., and Pinus banksianaat decay classes 4 and 5 had higher percent cover, while all sub-strate species had lower percent cover on decay classes 1 and 2.

Across all sampling units, percent cover was 33.2% for nonvas-cular species and 4.6% for vascular species, on average. The effectsof decay class and substrate species on nonvascular percent coverwere similar to those for total percent cover (Table 1; Fig. 1). Thepercent cover of nonvascular species increased with decay classes,peaked on decay class 4, and decreased thereafter on Populus spp.,Pinus banksiana, and Abies balsamea substrates, whereas the percentcover on Betula papyrifera and Picea spp. increased throughout thedecay classes. On average, the percent cover was highest on Populusspp. The percent cover of vascular plant species varied signifi-

cantly with decay class but not with substrate species (Table 1).The percent cover of nonvascular species on the forest floor wassimilar to that on substrates of decay class 2. Vascular plant spe-cies cover was much higher on the forest floor than on CWD of alldecay classes. No vascular plant species were found on decayclasses 1 and 2, with the percent cover of vascular plant speciesincreasing from decay class 3 to decay class 5 on all substratespecies.

The linear mixed-effects model indicated that total species rich-ness differed strongly with decay class and substrate species with-out a significant interaction (Table 1), but the model violated thenormality assumption (P = 0.005). Bootstrap analysis showed thattotal species richness increased continuously with decay class onbroadleaved substrates, but peaked at decay class 4 on coniferoussubstrates (Fig. 2). Nonvascular species richness peaked on decayclass 3 or 4 for all substrate species. Vascular plant species rich-ness varied significantly only with decay class (P < 0.001; Table 1)and increased from decay class 3 to decay class 5 (Fig. 2). Comparedwith the forest floor, total species richness on decay classes 1 and 2was lower, and species richness of nonvascular species was higheron all decay classes except class 1, while species richness of vascu-lar plants was lower on all decay classes. Total species richness onconiferous substrates was higher than that on broadleaved sub-strates at decay classes 3 and 4, but there was little differencebetween two substrate groups in other decay classes (Fig. 2). Thispattern was similar for nonvascular species richness (Fig. 2).

Total species evenness differed significantly with decay classand substrate species, with a significant interaction between thetwo factors (Table 1). Decay class was the main contributing factor,explaining 10.2% of the variance, while the interaction explained13.6% of the variance in species evenness. Due to sparseness ofspecies occurrence, species evenness could not be evaluated ondecay class 1. Pinus banksiana CWD substrate species at decayclasses 2 and 4, Picea spp. at decay class 3, and Abies balsamea CWDsubstrate species at decay class 5 had higher total species evennessthan that on other CWD substrate species (Fig. 3). The effect ofdecay class and substrate species on nonvascular species evennesswas similar to that for total species evenness (Table 1; Fig. 3).Vascular plant species evenness increased significantly from de-cay class 4 to decay class 5. Compared with forest floor, decayclasses 2 and 3 had, on average, similar total species evenness, butdecay classes 4 and 5 had lower total species evenness; CWD hadlower vascular species evenness at decay class 4 and similar vas-cular species evenness at class 5.

Table 1. Results of general or generalized linear mixed effects model showing the effects ofdecay class (D) and substrate species (S) on understory cover, species richness, and speciesevenness, separately analyzed by total, nonvascular, and vascular vegetation.

Source df

Total Nonvascular Vascular

Devianceor varianceexplained (%) P

Devianceor varianceexplained (%) P

Devianceor varianceexplained (%) P

CoverD 4 47.2 <0.001 34.6 <0.001 48.2 <0.001S 4 0.5 0.339 0.5 0.524 0.3 0.918D × S 16 4.2 0.003 4.3 0.031 4.5 0.654

RichnessD 4 42.8 <0.001 29.0 <0.001 68.2 <0.001S 4 12.6 <0.001 13.0 <0.001 0.5 0.577D × S 16 7.1 0.21 7.5 0.137 0.9 0.995

EvennessD 4 10.2 <0.001 11.4 <0.001 65.3 <0.001S 4 3.9 0.009 4.1 0.011 1.0 0.447D × S 16 13.6 <0.001 15.8 <0.001 5.1 0.274

Note: Bold font indicates statistical significance (� = 0.05). The columns provide the degrees offreedom (df), variance or deviance explained, and P values.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

4 Can. J. For. Res. Vol. 48, 2018

Published by NRC Research Press

Can

. J. F

or. R

es. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y L

akeh

ead

Uni

vers

ity o

n 02

/20/

18Fo

r pe

rson

al u

se o

nly.

The species composition of epixylic plant communities differedsignificantly among stand ages, overstory composition, decayclasses, and substrate species, with significant interactions amongthese predictors (Table 2). While main effects of these predictorsaccounted for the majority of the variations in the species com-position of epixylic plant communities (as indicated by partial R2

values), significant interaction terms showed that the composi-tional responses to one predictor were significantly dependent onthe levels of other predictor(s). When patterns of epixylic speciescomposition were visualized using NMDS (Fig. 4a), epixylic com-

munities on CWD in stands of different ages separated well inordination space, with young stands (34 years old) showing a fairseparation from older stands (210 years old), while intermediate-aged stands (98 and 146 years old) were positioned between youngand old. Epixylic species composition on CWD differed betweenbroadleaf and conifer stands, but there was overlap between co-nifer and mixedwood stands (Fig. 4a). The NMDS ordination alsoshowed clear differentiation of the species composition amongCWD decay classes, substrate species, and the forest floor (Fig. 4b).Most notably, there is a distinction in species composition accord-

Fig. 1. Total, nonvascular, and vascular understory vegetation cover (means and bootstrapped 95% confident intervals) in relation to decayclass and substrate species. Substrates: Po, Populus spp.; Bw, Betula papyrifera; Pj, Pinus banksiana; Sx, Picea spp.; Bf, Abies balsamea; FF, forestfloor. [Colour online.]

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

Kumar et al. 5

Published by NRC Research Press

Can

. J. F

or. R

es. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y L

akeh

ead

Uni

vers

ity o

n 02

/20/

18Fo

r pe

rson

al u

se o

nly.

ing to decay classes: decay classes 1 and 2 are grouped apart fromlater decay classes and there are two distinct groupings of sub-strates: conifer species (Pinus banksiana and Picea spp.) and broad-leaf species (Betula papyrifera and Populus spp.), but with Abiesbalsamea showing species composition common to conifers andbroadleaf species. On the other hand, species composition on theadjacent forest floor substrates differed strongly from all CWDsubstrates, particularly those not well decayed (Fig. 4b).

DiscussionWe found that abundance, species richness, and species evenness

of epixylic vegetation differed with both decay class and CWD sub-strate species. Epixylic vegetation abundance, measured as percentcover, increased with decay class from 1 to 5. This finding is consis-tent with previous studies (Andersson and Hytteborn 1991; Mills andMacdonald 2004; Botting and DeLong 2009; Checko et al. 2015).Likewise, species richness generally increased through the decay

Fig. 2. Total, nonvascular, and vascular understory species richness (means and bootstrapped 95% confident intervals) per 0.05 m2 quadratin relation to decay class and substrate species. Substrates: Po, Populus spp.; Bw, Betula papyrifera; Pj, Pinus banksiana; Sx, Picea spp.; Bf, Abies balsamea;FF, forest floor. [Colour online.]

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

6 Can. J. For. Res. Vol. 48, 2018

Published by NRC Research Press

Can

. J. F

or. R

es. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y L

akeh

ead

Uni

vers

ity o

n 02

/20/

18Fo

r pe

rson

al u

se o

nly.

process, though a decrease in richness of nonvascular plant spe-cies was seen in the later stages of decay. In addition, we foundthat substrate species strongly influenced understory vegetationpatterns in terms of abundance, species richness, and speciesevenness of total, nonvascular, and vascular species. Moreover,there were significant decay class and substrate species interac-tions in the analyses, consistent with the conclusion that succes-sional dynamics of epixylic plant communities varies withsubstrate species. Finally, patterns of epixylic species compositionwere also influenced by stand age and overstory composition, as

well as their interaction, highlighting the importance of coloni-zation and understory environments associated with stand ageand overstory composition, respectively.

Nonvascular and vascular plants responded differently to decayclass and substrate species. In general, all CWD substrate speciesshow a trend of increased vegetation abundance as CWD decayadvances; in particular, feathermosses and vascular plants in-crease in abundance with the process of CWD decay. The timeavailable for colonization increases (Bartels and Chen 2015), alongwith changing structure, chemistry, and moisture content of the

Fig. 3. Total, nonvascular, and vascular understory species evenness (means and bootstrapped 95% confident intervals) per 0.05 m2 quadratin relation to decay class and substrate species. Substrates: Po, Populus spp.; Bw, Betula papyrifera; Pj, Pinus banksiana; Sx, Picea spp.; Bf, Abies balsamea;FF, forest floor. [Colour online.]

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

Kumar et al. 7

Published by NRC Research Press

Can

. J. F

or. R

es. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y L

akeh

ead

Uni

vers

ity o

n 02

/20/

18Fo

r pe

rson

al u

se o

nly.

CWD substrate along the decay continuum (Alban and Pastor1993; Laiho and Prescott 2004; Bütler et al. 2007; Petrillo et al.2015). Thus, the combination of substrate receptivity and an in-creasing probability of recruitment contribute to increasingabundance of bryophytes, particularly feathermosses and vascu-lar plants in later stages of decay. Although data do not permit anevaluation of the relative importance of these factors, the impor-tance of increasing CWD moisture content and pH is also sug-gested by the high abundance of understory vegetation found onlater decay classes of Populus spp., which tends to have higherwood moisture content and pH than other taxa (Culberson 1955;Alban and Pastor 1993; Putna and Mezaka 2014).

Our findings highlight the importance of both decay class andsubstrate species as determinants of epixylic plant communityrichness and evenness. The observed increase in species richnessof understory vegetation in higher decay classes is in agreementwith the results from previous studies, which have also notedhigher species richness in advanced stages of decay (Crites andDale 1998; Rambo and Muir 1998; Laiho and Prescott 2004; Millsand Macdonald 2004; Bunnell et al. 2008; Botting and DeLong2009; Caruso and Rudolphi 2009; Dittrich et al. 2014; Checko et al.2015). However, most of these studies analyzed species richness ofindividual groups (lichens or bryophytes or vascular plants),whereas our study includes overall species diversity for all groups(nonvascular and vascular plants). The increase in species richnesswith decay reflects continuous colonization of different species;in addition, heterogeneity in the decay process may also contrib-ute to accumulation of species, as some “holdover” species fromprior decay classes persist later in the decay process. High abun-dance of a few species early in the decay process was associatedwith low species richness. Early epixylic communities start todominate with bark sloughing and further decomposition, andfinally, they are replaced by late epixylic species with humusdevelopment in the later stages of decay (McCullough 1948;Barkman 1958; Söderström 1988; Laaka 1995; Rambo and Muir1998; Rambo 2001; Jansová and Soldán 2006; Botting and DeLong2009; Kumar et al. 2017). The final stage of succession was charac-terized by epigeic species, mostly feathermosses and vascularplants that characteristically grow on the forest floor (Söderström1988; Andersson and Hytteborn 1991; Kumar et al. 2017). The ob-served decrease in the species evenness at late CWD decay stagesis attributable to increasing dominance of specific nonvascularspecies that varied among substrate species. For example, in acompanion study that utilized indicator species analysis on thesame data set, we found that decay class 5 logs of Betula papyrifera

were dominated by Brachytheciaceae and Mniaceae mosses andMitella nuda, logs of Pinus banksiana were characterized by Pleuroziumschreberi and Cornus canadensis, logs of Picea spp. were character-ized by Dicranum polysetum, Ptilium crista-castrensis, and Gaultheriahispidula, and logs of Abies balsamea were dominated by Hylocomiumsplendens (Kumar et al. 2017).

In addition to effects of decay class, CWD substrate species alsoaffected understory species richness and evenness. Previous stud-ies have noted higher species diversity on broadleaved substratespecies than on conifers (McAlister 1997; Rambo 2001; Mills andMacdonald 2004; Nowinska et al. 2009; Putna and Mezaka 2014;Checko et al. 2015). In contrast, we found, on average, higherepixylic species richness and evenness on coniferous substratesthan on broadleaved substrates. More importantly, we found thatthe effects of substrate species on epixylic species diversity aredependent on decay class and evaluated understory species groups.Species-specific changes in CWD physical and chemical propertiesmay contribute to these differences (Jüriado et al. 2009; Fritz andHeilmann-Clausen 2010; Pereira et al. 2014; Putna and Mezaka2014; Shorohova and Kapitsa 2014). The coniferous log generallydecays slowly and remains stable for many years (Harmon et al.1986; Harmon 1989; Mäkinen et al. 2006; Shorohova and Kapitsa2014), which may provide better opportunities and time for colo-nization of lichens and bryophytes, thereby increasing their spe-cies diversity. In addition, coniferous bark is generally more acidic(Hauck and Javkhlan 2009; Hauck 2011; Putna and Mezaka 2014)compared with that of broadleaved species (Culberson 1955;Kuusinen 1996; McAlister 1997; Mezaka et al. 2008). Among sub-strate species in the present study, the highest species diversity ofepixylic plants was found on Pinus banksiana; this may be attribut-able to its low bark pH and thicker bark (Hauck and Javkhlan2009). Differences in trends of species richness and evenness be-tween conifers and broadleaf substrates also include a decline inspecies richness of nonvascular plants from decay class 4 to decayclass 5 that is associated with a large increase in the cover ofvascular plants (correlation coefficient = −0.55, P < 0.001) and adecrease in evenness (correlation coefficient = 0.23, P = 0.002),particularly of nonvascular plants. We suggest that this pattern isattributable to the growth of feathermosses and the competitiveeffects of vascular plants at later stages of decay (Kumar et al.2017).

Sampling in the present study utilized a replicated chronose-quence that allowed for the representation of multiple canopysuccessional pathways in boreal forests (Chen and Popadiouk2002; Taylor and Chen 2011), allowing analyses that demonstratedistinct epixylic plant communities on CWD decay class and sub-strate species that depend on stand age and overstory composi-tion type. Variation in epixylic plant communities with stand ageand overstory composition suggest different successional trajec-tories for different stands (Fig. 4a). For instance, many lichen spe-cies such as Xanthoria fallax, Evernia mesomorpha, and Cladonia spp.and mosses such as Dicranum spp. and Pleurozium schreberi werepredominant in young stands, whereas the lichen Usnea subfloridanaand the moss species Hylocomium splendens and Rhytidiadelphustriquetrus were found exclusively in older stands. Furthermore, thelichen species Xanthoria fallax and Peltigera canina and the mossesBrachythecium spp., Mnium spp., and Callicladium haldanianum werefound exclusively in broadleaf stands, while Hypogymnia physodesand Cladonia lichens and Dicranum spp. and Pleurozium schreberimosses were found predominantly in conifer stands. Moreover,epixylic species composition differed with CWD decay class andsubstrate species, as well as between CWD and forest floor sub-strates, as found in a companion study (Kumar et al. 2017). Previ-ous studies have also revealed the differences in lichens andbryophytes species composition along the CWD decompositiongradient (Muhle and LeBlanc 1975; Mills and Macdonald 2005;Nascimbene et al. 2008; Botting and DeLong 2009). Differences inunderstory plant species composition between conifer and broa-

Table 2. Results of permutation multivariate analysis of variance (per-MANOVA) testing the effects of stand ages (A), overstory compositions (C),decay classes (D), substrate species (S), and their interactions on epixylicspecies composition.

Source df SS F P Partial R2

A 3 8.4 22.4 0.001 0.08C 2 11.8 47.6 0.001 0.11D 4 25.4 51.1 0.001 0.23S 4 5.5 10.9 0.001 0.05A × D 12 5.8 3.9 0.001 0.05C × D 8 5.2 5.3 0.001 0.05A × S 9 2.1 1.9 0.001 0.02C × S 7 1.3 1.5 0.013 0.01D × S 14 4.9 2.8 0.001 0.04A × C × D 18 3.4 1.5 0.002 0.03A × D × S 11 2.6 1.9 0.001 0.02C × D × S 6 0.6 0.8 0.855 0.01Plot 29 10.1 2.8 0.001 0.09Residual 186 23.1

Note: Bold font indicates statistical significance (� = 0.05). The columns pro-vide the degrees of freedom (df), sums of squares (SS), partial R2, and F andP values.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

8 Can. J. For. Res. Vol. 48, 2018

Published by NRC Research Press

Can

. J. F

or. R

es. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y L

akeh

ead

Uni

vers

ity o

n 02

/20/

18Fo

r pe

rson

al u

se o

nly.

dleaf substrates have likewise been documented (Barkman 1958;Palmer 1986; McAlister 1997). Differences in moisture, texture,and pH levels of the bark of substrate species have been suggestedas likely determinants (Culberson 1955; Hale 1955; McAlister 1997;Mills and Macdonald 2005; Mezaka et al. 2012; Pereira et al. 2014).The beneficial effect of CWD substrates on lichens and bryophytesis in part attributable to reduced competition on deadwood sur-face relative to the forest floor (Humphrey et al. 2002; Dittrichet al. 2014).

Our results indicate that CWD decay class and substrate specieshave strong influences on epixylic vegetation abundance, diver-sity, and composition. Furthermore, stands of different ages andoverstory composition types support different epixylic speciescomposition, indicating that shifts in forest age structure andcomposition can strongly influence the successional dynamics ofepixylic plant communities. Management activities strongly af-fect both forest age structure and composition (Chen et al. 2017),CWD decay class, and substrate species (Brassard and Chen 2008)

Fig. 4. Two-dimensional nonmetric multidimensional scaling (NMDS) ordination. (a) The differences in the epixylic species composition onCWD between stand ages and overstory compositional types. Overstory types are broadleaf (green in online version), conifer (blue in onlineversion), and mixedwood (black) and symbol shapes on legends differentiate stand ages. (b) The differences in epixylic species compositionbetween CWD decay classes, CWD substrate species, and adjacent forest floor substrates. Substrates: Po, Populus spp. (purple in online version);Bw, Betula papyrifera (blue in online version); Pj, Pinus banksiana (green in online version); Sx, Picea spp. (yellow in online version); Bf, Abies balsamea(red in online version); FF, forest floor (black). Symbol shapes on legends differentiate decay classes. Points nearest each other in ordinationspace have similar floristic assemblages, whereas those located farther apart are less similar. Vectors (arrows) indicate the significant(P < 0.05) joint axis correlation with decay class and substrate species, and length of the vector represents the strength of the correlation.[Please refer to the online version for colour.]

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

Kumar et al. 9

Published by NRC Research Press

Can

. J. F

or. R

es. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y L

akeh

ead

Uni

vers

ity o

n 02

/20/

18Fo

r pe

rson

al u

se o

nly.

and, in turn, epixylic vegetation abundance, diversity, and com-position. For example, short-rotation harvesting regimes couldresult in the loss of epixylic species with affinity for suitable hab-itat conditions in older stands. On the other hand, epixylic speciesassociated with specific CWD substrate species would becomeextinct without sufficient presence of the tree species in the over-story to generate species-specific CWD; therefore, to minimizeimpacts to sensitive lichens and bryophytes associated with CWD,it is important to maintain a diverse range of CWD decay classesand substrate species by a diverse range of forest age structure andoverstory composition. These results concur with the previoussuggestions that managed boreal landscapes should consist of amosaic of different successional stages and a broad suite of dom-inant overstory species (Boudreault et al. 2002; Brassard and Chen2008; Bartels and Chen 2015), thereby supplying appropriateamounts of CWD across the full range of decay classes and sub-strate species to maintain bryophyte and lichen species habitat inforests.

ConclusionOur results show that abundance and species richness and even-

ness of understory vegetation differed with both CWD decayclasses and substrate species and between CWD and the forestfloor. The trends in abundance, richness, and evenness also dif-fered for nonvascular and vascular plant groups. Abundance ofunderstory vegetation on CWD increased along the decomposi-tional gradient in all CWD substrate species. These findings indi-cate that understory species require suitable substrates and anextended period of time to colonize CWD. Epixylic species com-position also differed with stand age, overstory composition, de-cay class, and substrate species, as well as with their interactions.These results highlight that conservation of epixylic diversity andrestoration of their ecological functions would require forestmanagers to maintain a diverse range of CWD decay classes andsubstrate species in boreal forests. Moreover, that distinct epixylicplant communities were found in young and older stands andamong overstory composition types further suggests that man-aged boreal landscapes should consist of a mosaic of differentstand ages and a broad suite of dominant overstory species tosupport diverse understory plant communities. Future researchin this area should focus on further elucidation of species-specificrelations between CWD substrate and understory vegetation com-munities and on better understanding of the underlying mecha-nisms that determine community patterns.

AcknowledgementsWe thank Alexandra Hume, Allan Chen, and Magali Nehemy

for assisting with fieldwork and Erika North for help with theidentification of bryophytes and macrolichens. Financial supportfrom the Natural Sciences and Engineering Research Council ofCanada Strategic Grant Project (RGPIN-2014-0418 and STPGP428641)is gratefully acknowledged.

ReferencesAlban, D.H., and Pastor, J. 1993. Decomposition of aspen, spruce, and pine boles

on two sites in Minnesota. Can. J. For. Res. 23(9): 1744–1749. doi:10.1139/x93-220.

Anderson, M.J. 2005. Permutational multivariate analysis of variance. Depart-ment of Statistics, University of Auckland, Auckland, New Zealand.

Andersson, L.I., and Hytteborn, H. 1991. Bryophytes and decaying wood: a com-parison between managed and natural forest. Holarctic Ecol. 14(2): 121–130.

Arseneault, D. 2001. Impact of fire behavior on postfire forest development in ahomogeneous boreal landscape. Can. J. For. Res. 31(8): 1367–1374. doi:10.1139/x01-065.

Barbier, S., Gosselin, F., and Balandier, P. 2008. Influence of tree species onunderstory vegetation diversity and mechanisms involved — a critical reviewfor temperate and boreal forests. For. Ecol. Manage. 254(1): 1–15. doi:10.1016/j.foreco.2007.09.038.

Barkman, J.J. 1958. Phytosociology and ecology of cryptogamic epiphytes, in-cluding a taxonomic survey and description of their vegetation units inEurope. Van Gorcum & Comp. N.V., Assen, Netherlands.

Bartels, S.F., and Chen, H.Y.H. 2013. Interactions between overstorey and under-storey vegetation along an overstorey compositional gradient. J. Veg. Sci.24(3): 543–552. doi:10.1111/j.1654-1103.2012.01479.x.

Bartels, S.F., and Chen, H.Y.H. 2015. Dynamics of epiphytic macrolichen abun-dance, diversity and composition in boreal forest. J. Appl. Ecol. 52(1): 181–189.doi:10.1111/1365-2664.12360.

Bartemucci, P., Messier, C., and Canham, C.D. 2006. Overstory influences onlight attenuation patterns and understory plant community diversity andcomposition in southern boreal forests of Quebec. Can. J. For. Res. 36(9):2065–2079. doi:10.1139/x06-088.

Bates, D., Bolker, B., Walker, S., Christensen, R.H.B., Singmann, H., Dai, B., andGrothendieck, G. 2016. lme4: linear mixed-effects models using Eigen and S4.R package version 1.1-10. R Foundation for Statistical Computing, Vienna,Austria.

Bergeron, Y., Chen, H.Y.H., Kenkel, N.C., Leduc, A.L., and Macdonald, S.E. 2014.Boreal mixedwood stand dynamics: ecological processes underlying multiplepathways. For. Chron. 90(02): 202–213. doi:10.5558/tfc2014-039.

Botting, R.S., and DeLong, C. 2009. Macrolichen and bryophyte responses tocoarse woody debris characteristics in sub-boreal spruce forest. For. Ecol.Manage. 258: S85–S94. doi:10.1016/j.foreco.2009.08.036.

Boudreault, C., Bergeron, Y., Gauthier, S., and Drapeau, P. 2002. Bryophyte andlichen communities in mature to old-growth stands in eastern boreal forestsof Canada. Can. J. For. Res. 32(6): 1080–1093. doi:10.1139/x02-027.

Brassard, B.W., and Chen, H.Y.H. 2006. Stand structural dynamics of NorthAmerican boreal forests. Crit. Rev. Plant Sci. 25(2): 115–137. doi:10.1080/07352680500348857.

Brassard, B.W., and Chen, H.Y.H. 2008. Effects of forest type and disturbance ondiversity of coarse woody debris in boreal forest. Ecosystems, 11(7): 1078–1090. doi:10.1007/s10021-008-9180-x.

British Columbia Ministry of Forests and Range and British Columbia Ministryof Environment. 2010. Field manual for describing terrestrial ecosystems.2nd ed. Research Branch, British Columbia Ministry of Forests and Range,Victoria, B.C.

Bunnell, F.L., and Houde, I. 2010. Down wood and biodiversity — implications toforest practices. Environ. Rev. 18: 397–421. doi:10.1139/A10-019.

Bunnell, F.L., Spribille, T., Houde, I., Goward, T., and Björk, C. 2008. Lichens ondown wood in logged and unlogged forest stands. Can. J. For. Res. 38(5):1033–1041. doi:10.1139/X07-206.

Bütler, R., Patty, L., Le Bayon, R.-C., Guenat, C., and Schlaepfer, R. 2007. Log decayof Picea abies in the Swiss Jura Mountains of central Europe. For. Ecol. Manage.242(2–3): 791–799. doi:10.1016/j.foreco.2007.02.017.

Caruso, A., and Rudolphi, J. 2009. Influence of substrate age and quality onspecies diversity of lichens and bryophytes on stumps. Bryologist, 112(3):520–531. doi:10.1639/0007-2745-112.3.520.

Chávez, V., and Macdonald, S.E. 2010. The influence of canopy patch mosaics onunderstory plant community composition in boreal mixedwood forest. For.Ecol. Manage. 259(6): 1067–1075. doi:10.1016/j.foreco.2009.12.013.

Chávez, V., and Macdonald, S.E. 2012. Partitioning vascular understory diversityin mixedwood boreal forests: the importance of mixed canopies for diversityconservation. For. Ecol. Manage. 271: 19–26. doi:10.1016/j.foreco.2011.12.038.

Checko, E., Jaroszewicz, B., Olejniczak, K., and Kwiatkowska-Falinska, A.J. 2015.The importance of coarse woody debris for vascular plants in temperatemixed deciduous forests. Can. J. For. Res. 45(9): 1154–1163. doi:10.1139/cjfr-2014-0473.

Chen, H.Y.H., and Luo, Y. 2015. Net aboveground biomass declines of four majorforest types with forest ageing and climate change in western Canada’s bo-real forests. Global Change Biol. 21(10): 3675–3684. doi:10.1111/gcb.12994.

Chen, H.Y.H., and Popadiouk, R.V. 2002. Dynamics of North American borealmixedwoods. Environ. Rev. 10(3): 137–166. doi:10.1139/a02-007.

Chen, S., Shahi, C., Chen, H.Y.H., and McLaren, B. 2017. Economic analysis offorest management alternatives: compositional objectives, rotation ages,and harvest methods in boreal forests. For. Policy Econ. 85: 124–134. doi:10.1016/j.forpol.2017.09.006.

Crites, S., and Dale, M.R.T. 1998. Diversity and abundance of bryophytes, lichens,and fungi in relation to woody substrate and successional stage in aspenmixedwood boreal forests. Can. J. Bot. 76(4): 641–651. doi:10.1139/b98-030.

Culberson, W.L. 1955. The corticolous communities of lichens and bryophytes inthe upland forests of northern Wisconsin. Ecol. Monogr. 25(2): 215–231. doi:10.2307/1943551.

Dittrich, S., Jacob, M., Bade, C., Leuschner, C., and Hauck, M. 2014. The signifi-cance of deadwood for total bryophyte, lichen, and vascular plant diversity inan old-growth spruce forest. Plant Ecol. 215(10): 1123–1137. doi:10.1007/s11258-014-0371-6.

Environment Canada. 2015. Canadian climate normals 1981–2010 [online]. Avail-able from http://climate.weather.gc.ca/climate_normals/index_e.html [accessed25 August 2015].

Fritz, Ö., and Heilmann-Clausen, J. 2010. Rot holes create key microhabitats forepiphytic lichens and bryophytes on beech (Fagus sylvatica). Biol. Conserv.143(4): 1008–1016. doi:10.1016/j.biocon.2010.01.016.

Ganjegunte, G.K., Condron, L.M., Clinton, P.W., Davis, M.R., and Mahieu, N.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

10 Can. J. For. Res. Vol. 48, 2018

Published by NRC Research Press

Can

. J. F

or. R

es. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y L

akeh

ead

Uni

vers

ity o

n 02

/20/

18Fo

r pe

rson

al u

se o

nly.

2004. Decomposition and nutrient release from radiata pine (Pinus radiata)coarse woody debris. For. Ecol. Manage. 187(2–3): 197–211. doi:10.1016/S0378-1127(03)00332-3.

Hale, M.E., Jr. 1955. Phytosociology of corticolous crytogams in the upland for-ests of southern Wisconsin. Ecology, 36(1): 45–63. doi:10.2307/1931430.

Harmon, M.E. 1989. Effects of bark fragmentation on plant succession on coniferlogs in the Picea–Tsuga forests of Olympic National Park, Washington. Am.Midl. Nat. 121(1): 112–124. doi:10.2307/2425662.

Harmon, M.E., Franklin, J.F., Swanson, F.J., Sollins, P., Gregory, S.V., Lattin, J.D.,Anderson, N.H., Cline, S.P., Aumen, N.G., Sedell, J.R., Lienkaemper, G.W.,Cromack, K., and Cummins, K.W. 1986. Ecology of coarse woody debris intemperate ecosystems. Adv. Ecol. Res. 15: 133–302. doi:10.1016/S0065-2504(08)60121-X.

Hart, S.A., and Chen, H.Y.H. 2006. Understory vegetation dynamics of NorthAmerican boreal forests. Crit. Rev. Plant Sci. 25(4): 381–397. doi:10.1080/07352680600819286.

Hauck, M. 2011. Site factors controlling epiphytic lichen abundance in northernconiferous forests. Flora, 206(2): 81–90. doi:10.1016/j.flora.2010.02.001.

Hauck, M., and Javkhlan, S. 2009. Epiphytic lichen diversity and its dependenceon bark chemistry in the northern Mongolian dark taiga. Flora, 204(4): 278–288. doi:10.1016/j.flora.2008.03.001.

Hauck, M., and Jürgens, S.R. 2008. Usnic acid controls the acidity tolerance oflichens. Environ. Pollut. 156(1): 115–122. doi:10.1016/j.envpol.2007.12.033.PMID:18262699.

Humphrey, J.W., Davey, S., Peace, A.J., Ferris, R., and Harding, K. 2002. Lichensand bryophyte communities of planted and semi-natural forests in Britain:the influence of site type, stand structure and deadwood. Biol. Conserv.107(2): 165–180. doi:10.1016/S0006-3207(02)00057-5.

Huo, H., Feng, Q., and Su, Y.-h. 2014. The influences of canopy species andtopographic variables on understory species diversity and composition inconiferous forests. Sci. World J. 2014: 252489. doi:10.1155/2014/252489.

Jansová, I., and Soldán, Z. 2006. The habitat factors that affect the compositionof bryophyte and lichen communities on fallen logs. Preslia, 78(1): 67–86.

Jüriado, I., Liira, J., Paal, J., and Suija, A. 2009. Tree and stand level variablesinfluencing diversity of lichens on temperate broad-leaved trees in boreo-nemoral floodplain forests. Biodivers. Conserv. 18(1): 105–125. doi:10.1007/s10531-008-9460-y.

Karjalainen, L., and Kuuluvainen, T. 2002. Amount and diversity of coarse woodydebris within a boreal forest landscape dominated by Pinus sylvestris in Vien-ansalo wilderness, eastern Fennoscandia. Silva Fenn. 36(1): 147–167. doi:10.14214/sf.555.

Kruskal, J.B. 1964. Multidimensional scaling by optimizing goodness of fit to anonmetric hypothesis. Psychometrika, 29(1): 1–27. doi:10.1007/BF02289565.

Kruys, N., and Jonsson, B.G. 1999. Fine woody debris is important for speciesrichness on logs in managed boreal spruce forests of northern Sweden. Can.J. For. Res. 29(8): 1295–1299. doi:10.1139/x99-106.

Kumar, P., Chen, H.Y.H., Thomas, S.C., and Shahi, C. 2017. Effects of coarsewoody debris on plant and lichen species composition in boreal forests.J. Veg. Sci. 28(2): 389–400. doi:10.1111/jvs.12485.

Kuusinen, M. 1996. Epiphyte flora and diversity on basal trunks of six old-growthforest tree species in southern and middle boreal Finland. Lichenologist,28(5): 443–463. doi:10.1006/lich.1996.0043.

Laaka, S. 1995. Epixylic lichens on conifer logs in four natural forests in Finland.Graphis Scripta, 7(1): 25–31.

Laiho, R., and Prescott, C.E. 1999. The contribution of coarse woody debris tocarbon, nitrogen, and phosphorus cycles in three Rocky Mountain conifer-ous forests. Can. J. For. Res. 29(10): 1592–1603. doi:10.1139/x99-132.

Laiho, R., and Prescott, C.E. 2004. Decay and nutrient dynamics of coarse woodydebris in northern coniferous forests: a synthesis. Can. J. For. Res. 34(4):763–777. doi:10.1139/x03-241.

Lee, P., and Sturgess, K. 2001. The effects of logs, stumps, and root throws onunderstory communities within 28-year-old aspen-dominated boreal forests.Can. J. Bot. 79(8): 905–916. doi:10.1139/b01-072.

Lee, P., and Sturgess, K. 2002. Assemblages of vascular plants on logs and stumpswithin 28-year-old aspen dominated boreal forests. In Proceedings of theSymposium on the Ecology and Management of Dead Wood in WesternForests, Reno, Nev., 2–4 November 1999. Edited by W.F. Laudenslayer, Jr.,P.J. Shea, B.E. Valentine, C.P. Weatherspoon, and T.E. Lisle. USDA ForestService, Pacific Southwest Research Station, Albany, Calif., Gen. Tech. Rep.PSW-GTR-181. pp. 369–380. doi:10.2737/PSW-GTR-181.

Luo, Y., and Chen, H.Y.H. 2011. Competition, species interaction and ageingcontrol tree mortality in boreal forests. J. Ecol. 99(6): 1470–1480. doi:10.1111/j.1365-2745.2011.01882.x.

Macdonald, S.E., and Fenniak, T.E. 2007. Understory plant communities of bo-real mixedwood forests in western Canada: natural patterns and response tovariable-retention harvesting. For. Ecol. Manage. 242(1): 34–48. doi:10.1016/j.foreco.2007.01.029.

Mäkinen, H., Hynynen, J., Siitonen, J., and Sievänen, R. 2006. Predicting the

decomposition of Scots pine, Norway spruce, and birch stems in Finland.Ecol. Appl. 16(5): 1865–1879. doi:10.1890/1051-0761(2006)016[1865:PTDOSP]2.0.CO;2. PMID:17069378.

McAlister, S. 1997. Cryptogam communities on fallen logs in the Duke forest,North Carolina. J. Veg. Sci. 8(1): 115–124. doi:10.2307/3237249.

McCullough, H.A. 1948. Plant succession on fallen logs in a virgin spruce–firforest. Ecology, 29(4): 508–513. doi:10.2307/1932645.

McCune, B., and Grace, J.B. 2002. Analysis of ecological communities. MjM Soft-ware Design, Gleneden Beach, Ore.

Messier, C., Parent, S., and Bergeron, Y. 1998. Effects of overstory and understoryvegetation on the understory light environment in mixed boreal forests.J. Veg. Sci. 9(4): 511–520. doi:10.2307/3237266.

Mezaka, A., Brumelis, G., and Piterans, A. 2008. The distribution of epiphyticbryophyte and lichen species in relation to phorophyte characters in Latviannatural old-growth broad leaved forests. Folia Cryptog. Est. 44: 89–99.

Mezaka, A., Brumelis, G., and Piterans, A. 2012. Tree and stand-scale factorsaffecting richness and composition of epiphytic bryophytes and lichens indeciduous woodland key habitats. Biodivers. Conserv. 21(12): 3221–3241. doi:10.1007/s10531-012-0361-8.

Mills, S.E., and Macdonald, S.E. 2004. Predictors of moss and liverwort speciesdiversity of microsites in conifer-dominated boreal forest. J. Veg. Sci. 15(2):189–198. doi:10.1111/j.1654-1103.2004.tb02254.x.

Mills, S.E., and Macdonald, S.E. 2005. Factors influencing bryophyte assemblageat different scales in the western Canadian boreal forest. Bryologist, 108(1):86–100. doi:10.1639/0007-2745(2005)108[86:FIBAAD]2.0.CO;2.

Mueller-Dombois, D., and Ellenberg, H. 1974. Aims and methods of vegetationecology. The Blackburn Press, Caldwell, N.J.

Muhle, H., and LeBlanc, F. 1975. Bryophyte and lichen succession on decayinglogs. 1. Analysis along an evaporational gradient in eastern Canada. J. HattoriBot. Lab. 39: 1–33.

Nascimbene, J., Marini, L., Motta, R., and Nimis, P.L. 2008. Lichen diversity ofcoarse woody habitats in a Pinus–Larix stand in the Italian Alps. Lichenologist,40(2): 153–163. doi:10.1017/S0024282908007585.

Nowinska, R., Urbanski, P., and Szewczyk, W. 2009. Species diversity of plantsand fungi on logs of fallen trees of different species in oak–hornbeam forests.Rocz. Akad. Roln. Poznaniu, Bot.-Steciana, 13: 109–124.

Ohtsuka, T., Shizu, Y., Hirota, M., Yashiro, Y., Shugang, J., Iimura, Y., andKoizumi, H. 2014. Role of coarse woody debris in the carbon cycle of Ta-kayama forest, central Japan. Ecol. Res. 29(1): 91–101. doi:10.1007/s11284-013-1102-5.

Palmer, M.W. 1986. Pattern in corticolous bryophyte communities of the NorthCarolina Piedmont: do mosses see the forest or the trees? Bryologist, 89(1):59–65. doi:10.2307/3243078.

Pereira, I., Mueller, F., and Moya, M. 2014. Influence of Nothofagus bark pH on thelichen and bryophytes richness, Central Chile. Gayana Bot. 71(1): 120–130.doi:10.4067/S0717-66432014000100012.

Petrillo, M., Cherubini, P., Sartori, G., Abiven, S., Ascher, J., Bertoldi, D.,Camin, F., Barbero, A., Larcher, R., and Egli, M. 2015. Decomposition of Nor-way spruce and European larch coarse woody debris (CWD) in relation todifferent elevation and exposure in an Alpine setting. iForest, 9: 154–164.doi:10.3832/ifor1591-008.

Pielou, E.C. 1969. An introduction to mathematical ecology. Wiley Interscience,New York.

Putna, S., and Mezaka, A. 2014. Preferences of epiphytic bryophytes for foreststand and substrate in north-east Latvia. Folia Cryptog. Est. 51: 75–83. doi:10.12697/fce.2014.51.08.

R Core Team. 2017. R: a language and environment for statistical computing.Version 3.4.0. R Foundation for Statistical Computing, Vienna, Austria.

Rambo, T.R. 2001. Decaying logs and habitat heterogeneity: implications forbryophyte diversity in western Oregon forests. Northwest Sci. 75(3): 270–279.

Rambo, T.R., and Muir, P.S. 1998. Bryophyte species associations with coarsewoody debris and stand ages in Oregon. Bryologist, 101(3): 366–376. doi:10.2307/3244175.

Russell, M.B., Woodall, C.W., Fraver, S., D’Amato, A.W., Domke, G.M., andSkog, K.E. 2014. Residence times and decay rates of downed woody debrisbiomass/carbon in eastern US forests. Ecosystems, 17(5): 765–777. doi:10.1007/s10021-014-9757-5.

Senici, D., Chen, H.Y.H., Bergeron, Y., and Cyr, D. 2010. Spatiotemporal varia-tions of fire frequency in central boreal forest. Ecosystems, 13(8): 1227–1238.doi:10.1007/s10021-010-9383-9.

Shorohova, E., and Kapitsa, E. 2014. Influence of the substrate and ecosystemattributes on the decomposition rates of coarse woody debris in Europeanboreal forests. For. Ecol. Manage. 315: 173–184. doi:10.1016/j.foreco.2013.12.025.

Shorohova, E., Kapitsa, E., Kazartsev, I., Romashkin, I., Polevoi, A., andKushnevskaya, H. 2016. Tree species traits are the predominant control onthe decomposition rate of tree log bark in a mesic old-growth boreal forest.For. Ecol. Manage. 377: 36–45. doi:10.1016/j.foreco.2016.06.036.

Söderström, L. 1988. Sequence of bryophytes and lichens in relation to substrate

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

Kumar et al. 11

Published by NRC Research Press

Can

. J. F

or. R

es. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y L

akeh

ead

Uni

vers

ity o

n 02

/20/

18Fo

r pe

rson

al u

se o

nly.

variables of decaying coniferous wood in Northern Sweden. Nord. J. Bot. 8(1):89–97. doi:10.1111/j.1756-1051.1988.tb01709.x.

Spribille, T., Thor, G., Bunnell, F.L., Goward, T., and Björk, C.R. 2008. Lichens ondead wood: species–substrate relationships in the epiphytic lichen floras ofthe Pacific Northwest and Fennoscandia. Ecography, 31(6): 741–750. doi:10.1111/j.1600-0587.2008.05503.x.

Stirling, G., and Wilsey, B. 2001. Empirical relationships between species rich-ness, evenness, and proportional diversity. Am. Nat. 158(3): 286–299. doi:10.1086/321317. PMID:18707325.

Stokland, J.N., Siitonen, J., and Jonsson, B.G. 2012. Biodiversity in dead wood.Cambridge University Press, New York.

Sturtevant, B.R., Bissonette, J.A., Long, J.N., and Roberts, D.W. 1997. Coarsewoody debris as a function of age, stand structure, and disturbance in borealNewfoundland. Ecol. Appl. 7(2): 702–712. doi:10.1890/1051-0761(1997)007[0702:CWDAAF]2.0.CO;2.

Taylor, A.R., and Chen, H.Y.H. 2011. Multiple successional pathways of borealforest stands in central Canada. Ecography, 34(2): 208–219. doi:10.1111/j.1600-0587.2010.06455.x.

Weedon, J.T., Cornwell, W.K., Cornelissen, J.H., Zanne, A.E., Wirth, C., andCoomes, D.A. 2009. Global meta-analysis of wood decomposition rates: a rolefor trait variation among tree species? Ecol. Lett. 12(1): 45–56. doi:10.1111/j.1461-0248.2008.01259.x. PMID:19016827.

Wickham, H. 2009. ggplot2: elegant graphics for data analysis [online]. Springer-Verlag, New York. Available from http://ggplot2.org/.

Wilsey, B., and Stirling, G. 2007. Species richness and evenness respond in adifferent manner to propagule density in developing prairie microcosm com-munities. Plant Ecol. 190(2): 259–273. doi:10.1007/s11258-006-9206-4.

Zhang, Y., Chen, H.Y.H., and Reich, P.B. 2012. Forest productivity increases withevenness, species richness and trait variation: a global meta-analysis. J. Ecol.100(3): 742–749. doi:10.1111/j.1365-2745.2011.01944.x.

Appendix A

Table A1. Average percent cover values of nonvascular and vascular understory vegetation found on each stand age, overstory composition, CWDsubstrate species, and decay class in the central boreal forest of Canada.

Stand age (years) Overstory Substrate species Decay class

34 98 146 210 Bro Mix Con Po Bw Pj Sx Bf 1 2 3 4 5 FFTotalabundance

Number of logs 74 89 73 78 92 109 113 97 21 73 32 91 9 50 78 82 95

Nonvascular speciesHypogymnia physodes 0.80 1.20 1.21 1.82 0.90 1.30 1.52 0.57 2.98 1.18 1.75 1.48 1.06 5.36 1.19 0.23 0.05 395Usnea subfloridana 0.20 0.34 0.45 0.86 0.27 0.56 0.62 0.10 1.21 0.47 0.38 0.81 0.83 2.62 0.22 156Xanthoria fallax 0.32 0.09 0.05 0.03 0.27 0.05 0.04 0.31 0.11 0.22 0.14 0.13 0.11 0.11 38Parmelia sulcata 0.14 0.35 0.21 0.15 0.11 0.16 0.22 0.26 0.06 0.02 0.3 0.37 44Peltigera canina 0.07 0.27 0.19 0.10 0.28 0.04 0.12 0.32 0.07 0.08 0.14 0.22 0.21 0.02 43Phaeophyscia pusilloides 0.19 0.82 0.48 0.14 0.60 0.12 0.70 0.12 0.15 0.36 0.78 0.12 0.02 91Evernia mesomorpha 0.43 0.03 0.02 0.11 0.09 0.13 0.10 0.14 0.30 0.33 0.16 0.19 0.11 35Vulpicida pinastri 0.15 0.11 0.06 0.11 0.08 0.10 0.15 0.34 0.05 21Leptogium milligranum 0.11 0.06 0.03 0.05 0.03 0.04 0.08 0.07 0.06 0.06 0.03 13Cladonia coniocraea 1.09 0.55 0.46 0.08 0.33 0.46 0.73 0.31 0.96 0.81 0.42 0.04 1.13 0.74 0.14 164Cladonia cenotea 0.03 0.06 0.03 0.04 0.03 0.10 0.06 0.04 0.06 0.01 10Cladonia ochrochlora 0.09 0.06 0.10 0.03 0.06 7Cladonia agg. 2.62 1.33 1.62 1.06 1.07 1.67 2.30 0.91 0.10 2.86 1.06 2.27 0.62 3.71 2.26 0.37 540Cladonia chlorophaea 0.07 0.10 0.10 0.04 0.05 0.11 0.04 0.03 0.19 0.10 0.13 0.13 21Cladonia cervicornis 0.95 0.04 0.03 0.04 0.18 0.19 0.35 0.13 0.81 0.09 0.03 0.50 0.43 0.04 0.01 80Cladonia rangiferina 0.70 0.09 0.05 0.13 0.11 0.32 0.13 0.55 0.16 0.03 0.22 0.23 0.26 0.04 74Brachythecium spp. 5.16 9.21 8.02 8.88 18.8 3.94 1.88 17.9 12.2 0.41 0.78 3.40 0.22 2.98 9.82 9.39 7.06 0.02 2365Mnium spp. 0.12 4.62 3.84 2.49 6.14 1.17 1.18 6.19 1.95 0.07 0.88 1.63 0.10 1.54 4.21 3.71 0.01 825Dicranum ontariense 4.03 1.02 1.47 0.49 0.60 1.25 3.4 0.10 5.64 1.00 1.33 2.54 4.2 0.35 587Dicranum fuscescens 0.73 0.07 0.04 0.59 0.05 0.49 0.08 0.07 0.10 0.54 0.08 0.04 60Dicranum polysetum 2.69 1.88 1.70 0.96 0.89 1.42 2.77 0.65 0.10 2.34 4.03 2.00 0.02 0.31 2.60 3.25 0.38 666Dicranum flagellare 0.11 0.02 0.01 0.07 0.02 0.08 0.03 0.12 10Polytrichum commune 0.20 0.07 0.04 0.19 0.21 0.07 0.08 0.16 0.11 54Polytrichum juniperinum 0.02 0.01 0.05 0.02 0.02 0.05 7Callicladium haldanianum 0.17 0.29 0.28 0.19 0.09 0.05 27Orthotrichum speciosum 0.06 0.06 0.11 0.10 0.20 10Hylocomium splendens 0.26 0.38 1.74 0.45 0.81 0.61 0.19 0.03 1.96 0.54 1.62 0.48 350Ptilium crista-castrensis 3.15 3.08 2.86 1.27 1.35 1.83 4.17 1.76 0.86 2.78 6.44 1.97 1.06 2.65 5.02 3.31 1815Pleurozium schreberi 14.0 9.15 8.26 2.62 3.51 7.83 12.4 4.57 0.43 13.5 17.0 6.27 0.06 2.08 9.99 16.5 2.43 3317Rhytidiadelphus triquetrus 0.17 0.20 1.06 0.22 0.59 0.27 0.18 0.19 1.01 0.08 0.50 0.20 0.57 1.32 526Sphagnum wulfianum 0.05 0.03 0.02 0.06 0.05 0.03 0.07 0.05 22Ptilidium pulcherrimum 4.26 1.73 1.53 0.19 1.79 1.91 1.81 1.79 0.62 3.15 0.94 1.42 0.6 3.10 3.22 0.42 0.01 579Jamesoniella autumnalis 0.07 0.19 0.25 0.05 0.10 0.35 0.16 0.38 0.29 0.14 0.22 0.27 0.04 50

Vascular speciesMaianthemum canadense 1.07 0.48 0.59 0.63 1.01 0.34 0.82 0.80 0.14 0.81 0.66 0.64 0.3 2.04 3.00 1162Coptis trifolia 0.08 0.14 0.04 0.01 0.09 0.13 0.02 0.16 0.03 0.12 0.02 0.25 0.31 123Cornus canadensis 1.88 0.34 0.72 0.76 0.96 0.97 1.07 1.06 0.24 1.45 0.75 0.78 0.56 2.77 5.05 1894Clintonia borealis 0.20 0.22 0.28 0.08 0.26 0.14 0.10 0.28 4.00 1291Aralia nudicaulis 0.06 0.05 0.05 0.06 2.72 859Chamerion angustifolium 0.08 26Viola renifolia 0.04 0.03 0.03 0.04 0.32 103Streptopus roseus 0.07 0.05 0.16 0.11 0.04 0.16 1.41 458Trientalis borealis 0.26 0.55 0.39 0.59 0.87 0.30 0.13 0.66 0.52 0.14 0.16 0.37 0.51 0.86 0.61 316Mitella nuda 0.35 0.27 0.72 0.55 0.18 0.26 0.43 0.33 0.56 0.37 0.74 0.94 395Viola blanda 0.05 0.07 0.07 0.12 0.04 0.11 0.05 0.16 0.68 227Galium triflorum 0.07 21Aster macrophyllus 1.15 360Equisetum sylvaticum 0.18 57

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

12 Can. J. For. Res. Vol. 48, 2018

Published by NRC Research Press

Can

. J. F

or. R

es. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y L

akeh

ead

Uni

vers

ity o

n 02

/20/

18Fo

r pe

rson

al u

se o

nly.

Table A1 (concluded).

Stand age (years) Overstory Substrate species Decay class

34 98 146 210 Bro Mix Con Po Bw Pj Sx Bf 1 2 3 4 5 FFTotalabundance

Lycopodium annotinum 0.04 0.10 0.14 0.26 0.13 0.27 0.09 0.10 0.13 0.27 0.05 0.43 0.75 282Lycopodium obscurum 0.07 0.14 0.17 0.05 0.12 0.17 0.89 295Lycopodium lucidulum 0.06 0.12 0.15 0.04 0.14 0.16 0.07 0.14 0.22 89Lycopodium complanatum 0.07 22Linnea borealis 1.18 0.40 0.38 0.21 0.13 0.48 0.89 0.26 0.19 1.14 0.47 0.46 0.01 0.73 1.14 0.57 348Gaultheria hispidula 0.51 1.62 1.23 0.37 0.18 1.62 0.32 1.07 3.69 0.14 0.08 0.57 1.97 0.26 323Acer spicatum 0.05 0.09 0.08 0.08 1.32 422Calamagrostis canadensis 0.04 12Diervilla lonicera 1.03 324Lonicera canadensis 1.40 441Vaccinium angustifolium 0.94 295Ledum groenlandicum 0.58 182Rosa acicularis 0.48 150Vaccinium myrtilloides 0.05 0.05 0.05 0.62 199Rubus pubescens 0.07 0.38 0.21 0.14 0.29 0.21 0.29 0.20 0.48 2.12 712Ribes triste 0.24 0.20 0.19 0.22 0.06 36Petasites palmatus 0.07 21Salix spp. 0.11 36Picea spp. 0.08 26Abies balsamea 0.15 0.02 0.09 0.13 0.07 0.06 0.19 72Danthonia spicata 0.03 0.02 0.03 0.04 0.08 29

Note: Overstory compositions: Bro, broadleaf; Mix, mixedwood; Con, conifer. Substrate species: Po, Populus spp.; Bw, Betula papyrifera; Pj, Pinus banksiana; Sx, Piceaspp.; Bf, Abies balsamea. FF, forest floor. Total abundance, total percent cover values of each species on the present sites.

Pagination not final (cite DOI) / Pagination provisoire (citer le DOI)

Kumar et al. 13

Published by NRC Research Press

Can

. J. F

or. R

es. D

ownl

oade

d fr

om w

ww

.nrc

rese

arch

pres

s.co

m b

y L

akeh

ead

Uni

vers

ity o

n 02

/20/

18Fo

r pe

rson

al u

se o

nly.