lable at ScienceDirect

Quaternary Science Reviews 180 (2018) 63e74

Contents lists avai

Quaternary Science Reviews

journal homepage: www.elsevier .com/locate/quascirev

Late Holocene influence of societies on the fire regime in southernQu�ebec temperate forests

Olivier Blarquez a, *, Julie Talbot a, Jordan Paillard a, b, Lyna Lapointe-Elmrabti a,Nicolas Pelletier a, c, Christian Gates St-Pierre d

a D�epartement de G�eographie, Universit�e de Montr�eal, Pavillon 520, Ch. de la Cote-Sainte-Catherine C. P. 6128, Succursale Centre-ville, Montr�eal, Qu�ebec,H3C 3J7, Canadab Chaire en Am�enagement Forestier Durable, Universit�e du Qu�ebec en Abitibi-T�emiscamingue (UQAT), Rouyn-Noranda, Qu�ebec, Canadac Geography and Environmental Studies, Carleton University, Ottawa, Ontario, Canadad D�epartement d'anthropologie, Universit�e de Montr�eal, Montr�eal, Qu�ebec, Canada

a r t i c l e i n f o

Article history:Received 31 August 2017Received in revised form15 November 2017Accepted 17 November 2017

Keywords:CharcoalPaleoecologyMultiproxyArcheologyLand useHoloceneCanada

* Corresponding author.E-mail address: blarquez@gmail.com (O. Blarquez)

https://doi.org/10.1016/j.quascirev.2017.11.0220277-3791/© 2017 Elsevier Ltd. All rights reserved.

a b s t r a c t

Climatic change that occurred during the Holocene is often recognized as the main factor for explainingfire dynamics, while the influence of human societies is less apparent. In eastern North America, humaninfluence on fire regime before European settlement has been debated, mainly because of a paucity ofsites and paleoecological techniques that can distinguish human influences unequivocally from climate.We applied a multiproxy analysis to a 12 000-year-old paleoecological sequence from a site in the vi-cinity of known settlement areas that were occupied over more than 7000 years. From this analysis, wewere able detect the human influence on the fire regime before and after European colonization. Fireoccurrence and fire return intervals (FRI) were based on analysis of sedimentary charcoals at a hightemporal and spatial resolution. Fire occurrence was then compared to vegetation that was reconstructedfrom pollen analysis, from population densities deduced from archeological site dating, from de-mographic and technological models, and from climate reconstructed using general circulation modelsand ice-core isotopes. Holocene mean FRI was short (164 ± 134 years) and associated with small charcoalpeaks that were likely indicative of surface fires affecting small areas. After 1500 BP, large vegetationchanges and human demographic growth that was demonstrated through increased settlement evidencelikely caused the observed FRI lengthening (301 ± 201 years), which occurred without significantchanges in climate. Permanent settlement by Europeans in the area around 1800 AD was followed by asubstantial demographic increase, leading to the establishment of Gatineau, Hull and Ottawa. This trendwas accompanied by a shift in the charcoal record toward anthropogenic particles that were reflective offossil fuel burning and an apparent absence of wood charcoal that would be indicative of complete firesuppression. An anthropogenic fire regime that was characterized by severe and large fires and long fire-return intervals occurred more than 1000 years ago, concomitant with the spread of native agriculture,which intensified with European colonization over the past two centuries.

© 2017 Elsevier Ltd. All rights reserved.

1. Introduction

Fire is a natural component of most ecosystems worldwide(Bowman et al., 2009) and has had a long-lasting and widespreadinfluence on ecosystem dynamics (Marlon et al., 2016; Power et al.,2008), biodiversity (Blarquez et al., 2010; Colombaroli et al., 2013)and biogeochemical cycles (Carcaillet et al., 2002; Levine, 1990;

.

Mack et al., 2011). Certain ecosystems, such as tropical rainforestsor temperate forests, are rarely subjected to fires, except duringyears that are characterized by extreme climatic conditions(Bernier et al., 2016; Millar and Stephenson, 2015). Many temperateforests in Europe and North America burn relatively rarely due tolow ecosystem flammability and fuel spatial arrangements that areunfavorable to fires (H�ely et al., 2000; Pausas et al., 2017). Thesetemperate forests now face an increase in climate-driven droughtfrequency, which is associated with an increase in extreme fireevents (Flannigan et al., 2009; Millar and Stephenson, 2015). Severe

O. Blarquez et al. / Quaternary Science Reviews 180 (2018) 63e7464

and unusual fire events tend to occur more frequently. Thesedirectly threaten temperate forest resilience through increased treemortality and susceptibility to disturbances that interact with fire,such as insect outbreaks or biological invasions (Fleming et al.,2002; Millar and Stephenson, 2015).

Most temperate ecosystems in Europe and North America havebeen shaped largely by humans, sometimes over the course ofmillennia (Vanni�ere et al., 2015). Yet it is unclear whether fires are anatural component of these ecosystems where the increase in largeand severe fire events is a result of climate change alone, or they arefavored by ecosystemmanagement and other human practices. Forexample, fire suppression during the 20th century in the westernUSA has long been recognized as one of the main factors explainingthe rise of extreme wildfire events during the 1980s (e.g., Turnerand Romme, 1994). Landscape modification through livestockproduction, agriculture and, later, urbanization has profoundlyshaped the long-term fire regime (Marlon et al., 2012). In NorthAmerica, a large portion of fire outbreaks are of human origin: 85%of fires in the USA account for 44% of burned areas and slightlymore than 50% in Canada, which account for only ~20% of burnedareas; lightning fires, most notably in northern regions of Canada,represent 80% of burned areas (Balch et al., 2017; Stocks et al., 2002;Wotton et al., 2003).

Human-ignited fires are an important disturbance in temperateforests of eastern Canada, but few studies have analyzed past hu-man contributions to the fire regime in temperate forests (Clark andRoyall 1996). Most studies have focused on boreal forests. In theboreal forests of Eastern Canada, the fire regime is generally char-acterized by short fire-return intervals and long-term dependenceupon climatic conditions (Carcaillet et al., 2001; Ali et al. 2009a),especially spring temperatures that are suspected of driving largefire years (Ali et al., 2012). In boreal mixedwood forests, vegetationcomposition at the landscape scale, particularly the balance be-tween conifers and broadleaved trees, can explain Holocene firedynamics and has been shown to override the direct influence ofclimate (Blarquez et al., 2015; Girardin et al., 2013). In temperateforests that are dominated by hardwood species, recent changes inbiomass burning have been attributed to humans. During the pasttwo centuries, European settlements were accompanied by anincreased use of fire for land clearance and agriculture (Clark andRoyall 1996; Clark et al., 1996). However, the dependence of Ho-locene fire regime dynamics upon climate, vegetation and humanactivities in Canadian temperate forests is not fully resolved (Clarkand Royall 1996; Blarquez et al., 2015).

Aboriginal use of fire prior to European colonization is not fullyunderstood and has been frequently debated (Campbell andMcAndrews, 1995; Clark, 1995; Clark and Royall, 1995; Dey andGuyette, 2000; Guyette et al., 2006; Parshall and Foster, 2002).On one hand, low population densities do seem to have precludedany influence on the fire regime and associated ecosystem modi-fications (Campbell and Campbell,1994). On the other hand, the useof fire in land clearance for firewood supplies or hunting, and slash-and-burn agriculture was widespread before European coloniza-tion in the northeastern United States, where even low-densityaboriginal populations were suspected of burning large areas ofthe landscape through escaped fires (Pyne 1997). Indeed, lightningfires are rare in Eastern North American temperate forests andsouthern Canada in particular (Stocks et al., 2002), but wildfires didoccur during the Holocene (e.g., Power et al., 2013; Blarquez et al.,2015).

Ecosystems became inhabited by humans as soon as theybecame biologically viable (Pedersen et al., 2016), which wouldhave occurred after ca. 12 500 years BP in southern Canada. Duringthis period, mega-herbivores and vegetation started to recolonizedeglaciated land (Gill et al., 2009), a period after which human

influence upon the long-term fire regime cannot be ruled out. Themagnitude of human influence on the fire regime is difficult tomeasure because archeological archives tend to distort our vision ofprocesses occurring at the landscape scale; indeed, numeroustaphonomic biases are rarely taken into account (which are relatedto charcoal production, and conservation or representativeness inarcheological assemblages; see Th�ery-Parisot et al., 2010; Carcaillet,2017). Fire data from natural archives, such as lake or peat bogsediments, are also difficult to relate to human practices, which ispartly due to the difficulty of reconstructing past population den-sities and a lack of methods for correlating sedimentary archiveswith human history. Moreover, sedimentary archives are oftenlocated far from known archeological sites and cultural centers(Bowman et al., 2011; Clark and Royall, 1995).

In this study, we used a multiproxy approach to understand theinteraction between fire and humans. We studied a naturalpaleoecological site that was located within the vicinity of an areathat was occupied for at least 7000 years, and which had experi-enced a gradual increase in human density and land-use that leadto the modern city complex of Hull-Gatineau-Ottawa (Lalibert�e,2000, 2002). From that natural archive, we analyzed fire occur-rence from a high-resolution charcoal record and vegetation frompollen analysis. Holocene fire intervals were compared to humanpopulation densities that were inferred from the analysis of a largedataset of archeological radiocarbon dating in the area (Chaput andGajewski, 2016), and demographic and technological models (KleinGoldewijk et al., 2010). Fire dynamics for the past 200 years werecompared with the known post-settlement history of the region toestimate the influence of early European farming and later industryon the fire regime.

We hypothesize that humans influenced the Holocene fireregime and that this influence was modulated by population den-sity. We hypothesize that humans did light fires well before Euro-pean colonization and that human landscape modificationsinfluenced the long-term fire regime (eventually beyond its rangeof natural variability, sensuWillis and Birks, 2006). Ultimately, highpopulation densities should eliminate fire from the ecosystem,primarily through fuel management and secondarily through fireexclusion. Consistent with Pyne (1997), we hypothesized that Eu-ropean colonization took place within open to semi-open land-scapes and forests that were partly cleared by aboriginal fires. Thepost-settlement increase in fire use and later suppression wouldthus be the continuation of a preexisting history of human fire usein southeastern Canada.

2. Material and methods

2.1. Site description

The Folly peat bog is situated in Gatineau Park near the cities ofHull and Gatineau, Quebec, which are north of the Ottawa River andthe urban area of Ottawa, Ontario (Fig. 1). Folly peat bog(45.4552�N, �75.7813�W) is situated near the park entrance at anelevation of 133 m asl. The peatland area is 14 ha within a smallwatershed of about 28 ha in area. Although it is commonly referredto as a bog, the main part of the peatland is a nutrient-rich, forestedpeat swamp surrounding a small treeless bog that is located near itscenter. The bog portion of the peatland was cored in September2014 with a Russian corer and a box corer for the uppermost part ofthe core (i.e., sphagnum peat). Fibrous peat sediments dominatefrom 0 to 250 cm depth. From 220 to 250 cm, the sedimentsgradually change from peat to gyttja and, from 250 to 850 cm, shiftto gyttja that is indicative of lake sediments. Below 850 cm, claysediments that originated from marine transgression of theChamplain Sea (13 000e10 500 years BP) are found (Fig. 2). The

Fig. 1. Location of Folly peat bog (a), and the sediment age-depth model (b). The map insert shows the sites in the CARD 2.1 database (Martindale et al., 2016) that are located lessthan 200 km from Folly peat bog. The best age-depth model from the bootstrap procedure is represented using a black line (b). The gray area represents 95% confidence intervals forthe model. For each dating, calibrated year ranges are indicated using blue shading. (For interpretation of the references to colour in this figure legend, the reader is referred to theweb version of this article.)

Fig. 2. Pollen percentage diagram for Folly peat bog. Each division on the X-axis represents 5%; gray areas are exaggerated 10X. Total influx (grain.cm�2.yr�1), rarefied richness intaxon number, organic matter %, and Charcoal Accumulation Rate (CHAR, mm2.cm�2.yr�1) are plotted, together with pollen percentages of the main taxa (pollen sums > 5% for theentire record). Dashed lines represent main pollen zone divisions based upon CONISS analysis and the broken-stick model (i.e., 4 significant zones).

O. Blarquez et al. / Quaternary Science Reviews 180 (2018) 63e74 65

O. Blarquez et al. / Quaternary Science Reviews 180 (2018) 63e7466

Folly peat bog lies on bedrock composed of Precambrian meta-morphic rocks that are typical of the Canadian Shield. Climate in thearea is humid continental with an annual mean temperature of5.9 �C and marked seasonal differences, with �11.0 and 20.6 for thecoldest and hottest months (January and July), respectively. Pre-cipitation is evenly distributed throughout the year, totalling980.6 mm, with 793.6 mm falling as rain and 187.1 cm as snow(Canadian Climate Normals, 1981e2010 at Chelsea, QC, 113 m asl;http://climate.weather.gc.ca).

Almost 50 species of trees and 1100 vascular plants are repre-sented within Gatineau Park, including 90 species that areconsidered endangered in Canada or Qu�ebec (Cahill et al., 2008).Vegetation in the vicinity of the bog is typical of temperate decid-uous forest from the maple-linden bioclimatic domain with Acersaccharum Marshall as the dominant tree. Tree diversity is impor-tant and A. saccharum and A. rubrum L. are accompanied bynumerous species, such as Quercus rubra L. Fraxinus americana L.,Populus grandidentata Michaux, P. tremuloides Michaux, Betulapapyrifera Marshall, Fagus grandifolia Ehrhart, Betula alleghaniensisBritton, Tilia americana L., Prunus serotina Ehrhart. Within the areaof the peatland, treed portions are dominated by A. rubrum, Frax-inus nigraMarshall, Larix laricina (Du Roi) K.Koch, Thuja occidentalisL. and Picea mariana (Miller) BSP. The understory is dominated bythe invasive shrub Frangula alnus Miller and the fern Matteucciastruthiopteris (L.) Todaro. The bog portion of the peatland where thecore was extracted contained small saplings of L. laricina andP. mariana, while the understory is dominated by ericaceous shrubs(mostly Chamaedaphne calyculata [L.] Moench) and a continuouscover of Sphagnum mosses.

2.2. Dating and age-depth modeling

Five 210Pb measurements were performed on sediments fromthe top core to date the most recent part of the sequence. 210Pbactivity was inferred by measuring the activity of the daughterisotope 210Po by alpha spectrometry after acid digestion and sub-sequent deposition on an Ag disk. 210Pb activity was then used tocalculate sediment ages using the Constant Rate of Supply model(CRS; Appleby and Oldfield, 1983). Errors from the 210Pb datingwere assessed using Monte Carlo Bayesian uncertainties calcula-tions, which have been described and implemented within EXCELby Sanchez-Cabeza et al. (2014). The software is freely available athttp://www.sciencedirect.com/science/article/pii/S1871101414000569.

Six AMS 14C measurements were obtained from bulk sediments(Table 1). 14C was calibrated to calendar years using the IntCal13calibration curve (Reimer et al., 2013) and expressed in years BeforePresent (thereafter, BP; by convention, the present is 1950 AD). Anage-depth model was reconstructed using the procedure that was

Table 1Radiocarbon and 210Pb dating information. 14C ages are uncalibrated, 210Pb ages arecalibrated using the CRS model and given in years BP (Before Present, i.e., 1950 AD).

Laboratory code Age 14C Age BP Error Depth (cm) Thickness (cm)

Pb-0001 e �64 0.1 0 1Pb-0002 e �60.83 0.27 0.5 1Pb-0003 e 3.84 2.17 10.5 1Pb-0004 e 44 3.18 20.5 1Pb-0005 e 61.25 3.95 30.5 1D-AMS 019847 476 e 25 123.5 3C1410 975 e 35 215 5D-AMS 019848 3562 e 40 375 4D-AMS 019849 4766 e 64 531.5 5D-AMS 019850 6847 e 81 692.5 5Beta392480 10310 e 30 852.5 5

described by Blaauw (2010) with the clam 2.2 package that isavailable at http://www.chrono.qub.ac.uk/blaauw/clam.html andwritten in R (R Core Team, 2017). For each calibrated date, an agewas randomly picked according to its probability density function.An age-depth model then was reconstructed using a smoothingspline curve with a smoothing parameter of l ¼ 0.67, which rep-resented the most parsimonious model. The procedure wasrepeated 1000 times to estimate a 95% confidence interval aroundthe ‘best’ median age-depth model. Iterations resulting in age in-versions were automatically discarded (Blaauw, 2010). Subsamples(1 cm3) were taken in increments of 10 cm and analyzed by loss-on-ignition (LOI; Dean, 1974) at 600 �C for 30 min for dry-mass organicmatter content (g and %) and C content (g).

2.3. Vegetation reconstruction

Pollen was extracted from 16 sediment samples (1 cm3)distributed along the sequence using the heavy-liquid separationtechnique. Bulk sediments were digested using hydrochloric acid(HCl) and organic matter was deflocculated using potassium hy-droxide (KOH). Sediments were sieved (15 and 250 mm), density-separated using sodium polytungstate and acetolyzed. Lycopo-dium capsules were added to quantify pollen concentrations.Samples were mounted in glycerin and analyzed under light mi-croscope at 400x magnification. A minimum of 500 pollen grainswere counted, except one sample (the deepest one, from the claylayer), where 300 pollen grains were counted. Pollen zonationrelied upon (i) an optimal number of zones that were assessedusing a broken-stick model (Birks and Gordon, 1985) and (ii)temporally constrained zonation based upon the CONISS procedure(Grimm, 1987). Pollen diagrams and zonation were performed us-ing the rioja 0.9e15 R package (Juggins, 2015). Pollen data could beretrieved from the Neotoma database at https://www.neotomadb.org.

2.4. Fire occurrence reconstruction

Contiguous subsamples of 1 cm3 were analyzed for their char-coal content. Subsamples were soaked ~24 h in aqueous 8% sodiumhypochlorite (NaClO) to deflocculate and bleach the sediment.Charcoal particles were sieved with a 150 mm mesh, then countedand measured using a stereoscope coupled to an image analysissystem (WinSeedle, Regent Instruments, Quebec, QC). Charcoalareal accumulation rates (CHAR) were assessed using the age-depth model and a proportion matrix approach for calculatingCHAR in mm2.cm�2.yr�1 for samples of equivalent duration, whichin turn were equivalent to the median resolution of the record, i.e.,13 years (for details, see Higuera et al., 2009 or Blarquez et al.,2013). Here, we used charcoal area because charcoal abundancecould be subject to fragmentation and because the two estimatesconverged for our record (Ali et al. 2009b; Leys et al., 2013).

Fire occurrence reconstruction is based upon the ensemblemember procedure that was proposed by Blarquez et al. (2013) andwhich is freely available at http://paleoecologie.umontreal.ca/category/code/. Briefly, fire-history reconstructions were per-formed with the CharAnalysis 1.1 (Higuera et al., 2009, available athttps://sites.google.com/site/charanalysis/). To obtain residualhigh-frequency series (CHARpeak), the CHAR series were interpo-lated for removing the low-frequency background signals (CHAR-back), which corresponded to variation in charcoal production, thesedimentation process, mixing and sampling. CHARpeak series wereseparated into two sub-populations, which are referred to asCHARnoise and CHARfire, using a Gaussianmixturemodel that used alocally defined threshold. Each CHARpeak that exceeded the 99thpercentile threshold could be considered representative of one or

O. Blarquez et al. / Quaternary Science Reviews 180 (2018) 63e74 67

more local fire events. The interpolation step can use severalfiltering methods and different smoothing window widths. Toobtain a robust fire reconstruction and to minimize potential biasthat was related to the choice of a given filter, we used an ensemblemember approach that iterated the CharAnalysis runs, by changingthe filtering method (moving mode, moving median, movingaverage, lowess, robust lowess) and the smoothing-window width(t ¼ 50, 100, 150, …, 1500 years), which resulted in 150 re-constructions (Blarquez et al., 2013). The signal-to-noise index(SNI; Kelly et al., 2011), which is used to evaluate the effectivenessof the discrimination between CHARfire and CHARnoise and, there-fore, peak detection accuracy, was used to select the most accuratereconstructions. Reconstructions with an SNI 5th percentile >3were retained. From that ensemble member, fire date occurrencesthat were detected in more than 50% of the reconstructions wereselected as robust fire dates. Temporal trend in fire occurrence wasevaluated using a kernel density estimation procedure, which isavailable within the paleofire 1.2.0 package under R (Blarquez et al.,2014). Charcoal data are freely available within the Global CharcoalDatabase at http://paleofire.org.

Charcoal from the uppermost sediment, which corresponded tothe 20th century, was visually identified as originating from fossilfuels and anthropogenic activities, rather than fromwood burning.The former were distinguished by their spherical shape and strongresemblance to Spheroidal Carbonaceous Particles (SCP; e.g., Roseet al., 1995). In order to confirm this finding, we calculated thesphericity of each charcoal particle with the following formula:

S ¼ 4pA.P2

where A and P were charcoal area and perimeter, respectively. S isbounded by the interval [0 1], where 1 is a perfect circle. Wecalculated the width-to-length (W/L) ratio, following Aleman et al.(2013). W/L is also bounded by [0 1]. For samples with the highest Svalues, we selected several particles for scanning electron micro-scopy (SEM), followed by characterization of submicron particlestructure using a Quanta 200 3D SEM (FEI Company).

2.5. Population density modeling and statistical analyses

We extracted archeological 14C dates from the CARD databasefor sites that were located within a 200 km radius of Folly peat bog(Martindale et al., 2016). We completed those dates with an addi-tional set of 7 dates from an archeological site in Gatineau, whichhad yet to be included in the database (Arch�eotec, 2014). We cali-brated each date in the database using the IntCal13 calibrationcurve (Reimer et al., 2013). For reconstructing the temporal densityof archeological dating that should relate to population density(Chaput and Gajewski, 2016), we used an approach inspired by‘clam’ package age-depth modeling (Blaauw, 2010). For eacharcheological dating, we randomly picked one age according to itscalibrated probability density (i.e., ages with a higher probabilityare more likely to be selected). We then used a kernel densityfunction to assess the temporal density of archeological dating. Werepeated this procedure 1000 times to assess the 95% confidenceinterval (CI) around the median trend in archeological sites datingduring the Holocene. The procedure is freely available at http://paleoecologie.umontreal.ca/category/code/.

The density of archeological site datings (Martindale et al.,2016), together with known regional occupation history (Pilon,1999), was used to delineate four different periods based uponexpected population densities showing no population (>7000 BP),low population density (7000e4600 BP), medium populationdensity (4600e1500 BP) and high population density (<1500 BP).

The most recent period with very high population (past 200 years),which corresponds to the industrial/urban period, was includedwithin the preceding period in subsequent analyses due to its shortduration. For each period, fire return intervals (FRI) were calculatedand median FRI was estimated. For each period, FRI distributionswere fitted using a Weibull model; the scale (Wb) and shape (Wc)parameters were estimated, together with their standard errors bymaximum likelihood using the ‘fitdistrplus’ R package (Delignette-Muller and Dutang, 2015).

2.6. Climate data

In order to detect rapid responses of fire frequency and vege-tation to climate change, we used continuous simulations fromCCSM3 that were resolved in 30-year time steps using Paleoviewv1.1 freeware, which was available at https://github.com/GlobalEcologyLab/PaleoView/releases (Fordham et al., 2017a,2017b). Paleoview climate data came from the TRaCE21ka experi-ment that spanned the period 22 000 BP to the present (1989 AD),and which was based on the CCSM3 GCM (Liu et al., 2009). Meansummer temperature anomalies (June, July, August in �C) andannual precipitation anomalies (mm.day�1) were obtained at a 30-year time resolution by subtracting simulated paleoclimate valuesto the simulated preindustrial period (ca. 1750 AD). Compared tothe original spatial resolution of CCSM3 (i.e., 2.5� latitude by 3.75�

longitude), Paleoview spatial resolution was downscaled to a2.5 � 2.5� grid using bilinear interpollation (Fordham et al., 2017a).We thus extracted climate data for pixels that were located be-tween 42.5 and 47.5�N and between �77.5 and �72.5�W, whichbounded our study area. Medians and 90% confidence intervals forprecipitation and temperature were based upon distributions ofanomalies for the different pixels at each 30-year time step.

We used d18O of the NGRIP ice-core record as an additionalproxy for regional temperatures (Andersen et al., 2004), and icecore CO2 values from the Epica Dome C coring as a proxy of past CO2atmospheric concentrations (Monnin et al., 2004), which weresupplemented with modern (1958e2016) data from the Mauna LoaObservatory that are available from NOAA at https://www.esrl.noaa.gov/gmd/ccgg/trends/.

3. Results and discussion

3.1. Sedimentation and vegetation history

The age-depth model included six 14C AMS and five 210PB dates(Table 1 and Fig. 1). The transition from clay and silts to a sedimentthat was composed from gyttja occurs at the 850 cm depth corre-sponding to an age of 11 961 BP [11 756e12 146 BP] (95% CI, cali-brated calendar ages). Our basal date at 850e855 cm depth(10 310 ± 30 14C BP) closely agrees with the basal dates at Pink(10 600 ± 150 14C BP) and Ramsay Lake (10 800± 180 14C BP), whichwere estimated by Mott and Farley-Gill (1981) for the claydgyttjaboundaries. We did not find mollusk shells or marine microfossilsin the basal clay sediment that would indicate a marine origin.Microfossils were also absent from Pink and Ramsay Lake sedi-ments (Mott and Farley-Gill, 1981), indicating that the clay at thosetwo sites could be of lacustrine origin. Folly peat bog is at a lowerelevation (133 m asl) compared to Pink and Ramsay Lakes (170 and200 m a.s.l., respectively). Reconstructions place the site beneaththe surface of the Champlain Sea (Occhietti, 2007; Occhietti andRichard, 2003). At 790 cm depth, the sediment is gyttja with 76%organic content (Fig. 2), which is indicative of freshwater lacustrineorigin; this level is dated at 10 221 BP [10 078e10 426 BP]. Since10 000 BP, Folly peat bog was definitively isolated from theChamplain Sea and functioned as a freshwater lake, which agrees

O. Blarquez et al. / Quaternary Science Reviews 180 (2018) 63e7468

with the known history of marine phases in the region (Occhiettiand Richard, 2003). From 10 000 BP to 930 BP, the sediment is anorganic-rich (>90% organic matter content) lacustrine gyttja(Figs. 1e2). From 215 cm and upward to the surface, we observed atransition from gyttja to peat. The topmost peat was fibric withmacrofossil occurrences of Sphagnum magellanicum Bridel-Brideri,with an organic matter content above 0.05 g cm�3 and a C con-tent greater than 0.02 g cm�3 (Fig. S2). The gyttjadpeat transitionthat indicates the end of the lacustrine phase is dated to 1009 BP[909e1225 BP]. Sediment accumulation rate showed a low degreeof variation for the observed period; the mean (±SD) time resolu-tion for the Holocene was 14.2 ± 7.6 year cm�1.

The broken-stick model indicated that the pollen diagram couldbe divided into four different zones where zone boundaries weredetermined using CONISS. The first zone (12 000e10 000 BP) cor-responds to vegetation colonization after Champlain Sea regres-sion. Vegetation was dominated by Picea sp., Pinus banksianaLambert and Populus, while the presence of pollen grains of Betulacf. glandulosa seems to indicate open vegetation (Fig. 1). WhilePopulus was abundant before 10 000 BP, it never reached levelscomparable to what was observed for the Pink Lake, which islocated only 2 km northwest of Folly peat bog (Mott and Farley-Gill,1981). Pink Lake is a meromictic lake of 9 ha surrounded by abruptcliffs. Postglacial vegetation history of the lake was studied by Mottand Farley-Gill (1981), together with another lake in the GatineauPark, Ramsay Lake, which is situated 29 km northwest of Folly peatbog (Fig. 1). Like Folly peat bog, these two lakes showed sustainedlevels of Populus pollen (i.e., >5%) until ca. 9000 and 10000 BP forPink and Ramsay Lakes, respectively (Fig. S1). Maximum abun-dance (>30%) dated back to between 10 500 and 10 000 BP, a periodfor which we lack data at Folly peat bog (Fig. 2). Higher Populuspollen percentages, which were likely associated with higher thanpresent tree biomass (Blarquez and Aleman, 2016), are presumed tohave occurred at Folly peat bog, but were probably not detectedwith our low-resolution record.

The second zone (10 000 to 8000 BP) corresponds to affores-tation of the area, with an increase in Pinus strobus L. and decreasesin taxa characteristic of tundra. Across this zone, charcoal wasfound continuously and fires occurred regularly (Fig. 3). Identifi-cation of soil charcoal contained within cave deposits from theEardley Escarpment (Fig. 1) that dated from 9200 ± 110 BP indi-cated local presence and abundance of P. banksiana, Betula andRosaceae (Lauriol et al., 2003). These taxa are associatedwithminorconstituents, such as Quercus, Thuja/Juniperus, A. saccharum andP. strobus, which became dominant after 8000 BP.

The third zone (8000e600 BP) corresponds to the developmentof temperate forest with the continuous occurrence ofA. saccharum, F. grandifolia, Tsuga canadensis (L.) Carri�ere, Betulaalleghaniensis Britton, Quercus sp. and Ulmus sp. The establishmentof temperate forest is characterized by an increase in hardwoodtaxa and the likely dominance of maple forests. Acer saccharum is alow pollen producer, and a pollen abundance of 2% likely indicatesthe dominance of the taxa within the forest canopy. The impor-tance of A. saccharum in the forests is confirmed by its dominancewithin the Eardley Escarpment charcoal record after 5700 BP(Lauriol et al., 2003). Over this zone, T. canadensis declines around5200 BP in agreement with other regional pollen reconstructionsdating Tsuga decline between 5500 and 5100 BP (Fuller,1998). Plantrichness oscillates between 25 and 30 taxa, with Holocenemaximum richness being attained at 4100 BP (30.4 taxa, Fig. 2).

The last zone (the past 600 years) is characterized by declines inseveral taxa such as Tsuga, Ulmus and F. grandifolia, and an increasein anthropogenic markers, such as Ambrosia, Poacaeae and light-demanding taxa, such as Alnus incana subsp. rugosa (Du Roi)R.T.Clausen. The zone is also characterized by a rise in P. strobus

after an 8000-year decline that was experienced during the pre-vious period where a minimum was recorded (1400 BP).

3.2. Fire occurrence

Fire history reconstructions from sieved charcoal records, wheresignal decomposition can provide information on past fire occur-rences (Blarquez et al., 2013; Higuera et al., 2005), are rare in thearea (Clark et al., 1996). From our charcoal record, we reconstructedlocal past fire occurrence, where the term “local” is used for fireevents occurring at distances generally <10 km (Higuera et al.,2007) and less frequently, up to ~30 km from the accumulationsite (Oris et al., 2014).

For the Holocene, the mean fire return interval at Folly peat bogis 164 ± 134 years, Wb is 181 ± 17, and Wc is 1.37 ± 0.12, whichindicate that the probability of fire occurrence is dependent uponthe duration since the previous fire (Johnson and Gutsell, 1994).Sixty-five fire episodes that corresponded to one or more firesoccurring locally were detected for the entire period (Table 2).These results should not be confoundedwith the expected fire cyclein the area, which is over ca. 1200 years (Bridge, 2001). Twoconsecutive fires that were inferred from the charcoal record didnot necessarily occur in the same area, implying that FRI is notequivalent to the fire cycle. Also, charcoal peaks retrieved from theFolly peatland aremuch lower compared to peaks from boreal lakesthat experience stand-replacing fires (Ali et al., 2012; Carcailletet al., 2001; Higuera et al., 2009). Those low charcoal peaksappear to indicate the occurrence of either surface fires or moredistant fires. Maximum fire occurrence occurred around 4700 BP,reaching 12 fires per millennium. Low fire occurrence is recordedsince 1500 BP, where four fires occurred over 1500 years (~2.6 fireper millennium). For all periods that were determined using pop-ulation density analysis, the shape parameter Wc is consistently >1(Table 2 and Fig. 4), indicating that fire occurrence was alwaysdependent upon the duration since the last fire and, therefore,intuitively on fuel build-up (Johnson and Gutsell, 1994).

Themost recent fire event that was detected using the ensemblemember procedure at 0 BP was discarded because charcoal origi-nated from either fossil fuel emissions or coal burning. The peakwas largely dominated by SCP (Fig. 5), where both particle sphe-ricity and W/L ratio were always greater than 0.5, which indicatesno elongated particles. The ratio was briefly above 0.75, indicatingvery round to circular particles. SEM microscopy helped confirmthe nature of SCP, as no wood structure was visible on the vitrifiedsurface of the particles (Fig. 5). Moreover, particle structure did notfit into any known burned plant material categories (Mustaphi andPisaric, 2014). We did not use specific sediment treatments for SCPextractions (Rose, 1994), but our W/L and Sphericity indices agreedwith known regional and global dynamics of SCP and, thus, are agood proxy for SCP (Charles et al. 1990; Rose, 2015). Indeed, SCPincreased globally around 1950 (Rose, 2015) and has since beenproposed as a stratotype for defining the Anthropocene (Rose,2015; Swindles et al., 2015). Although no SCP record is availablefor Qu�ebec or Ontario, records from the nearby AdirondackMountains in New York State (~200 km distance) indicates a rise incoal soot particles around 1900, attaining a maximum in 1960(Charles et al. 1990). Charles et al. (1990) differentiated betweencoal and oil soot classes. The latter class was not retrieved by ourprotocol andwe can assume that our SCP record is indicative of coaland solid fuel burning (i.e., production of particles > 150 mm indiameter).

Several anthropogenic activities that took place around 1900,and which continued until the mid 20th century, could be theorigin of SCP. These particles are in the diameter class of sievedcharcoal particles that originate from local sources of emissions

Fig. 3. Long-term fire regime versus anthropogenic proxies and climate. CHAR gray areas represents 10X exaggeration. Detected fire events are represented using red plus signs.Gray bands around fire occurrence and the number of archeological datings represents 95% CI, while 90% CI were used for temperature and precipitations anomalies (see Methods).d18O from NGRIP and CO2 are represented using a black line. Dotted lines are zone demarcations based upon population density deduced from temporal densities of archeologicaldates. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Table 2Fire return intervals (FRI, years) for different periods based upon human population density (see Fig. 3, and Material and Methods for details). Standard deviations for meanFRIs and standard errors for Weibull b and c parameters are indicated in parentheses.

Period Age BP # events Mean FRI Weibull b Weibull c

High population >1500 5 301 (201) 342 (90) 1.81 (0.61)Medium population 4600e1500 23 135 (132) 145 (27) 1.21 (0.18)Low population 7000e4600 16 141 (112) 157 (29) 1.44 (0.26)No population <7000 21 181 (125) 205 (29) 1.66 (0.25)Holocene 12000e0 65 164 (135) 181 (17) 1.37 (0.12)

Fig. 4. FRI distribution for the Holocene (a) and for different temporal periods experiencing different population densities, based upon the analysis of the temporal density ofarcheological datings in the study area (see Fig. 3). The red line is the fitted Weibull distribution for each FRI distribution (Table 2). (For interpretation of the references to colour inthis figure legend, the reader is referred to the web version of this article.)

O. Blarquez et al. / Quaternary Science Reviews 180 (2018) 63e74 69

Fig. 5. Scanning electron micrographs of spheroidal carbonaceous particles (SCP) occurring within the 25-0 BP samples (aed). Shape analysis results of the charcoal particlesoccurring during the period 100 to �50 BP (e). Sample mean sphericity and W/L ratio are the closed circles and their sample standard deviations are the vertical bars. Total CHARthat is analyzed using SEM is indicated within the gray area.

O. Blarquez et al. / Quaternary Science Reviews 180 (2018) 63e7470

(<10 km; Higuera et al., 2007). Among possible sources, we can citethe steam locomotive train station that was located 2 km from theFolly peatland (Hull-Chelsea-Wakefield Railway; Fig. 1) or Hull in-dustrial complexes, notably sawmills, the activity of which peakedduring the previous century. Several brief events may also havecontributed to the SCP record, such as the 1900 Hull-Ottawa firethat burnedmost of the city of Hull and parts of Ottawa. House firescan reach temperatures that are high enough to produce vitrifiedparticles, such as those found between 1900 and 1950 (Fig. 5ced).Although the hypothesis linking fire temperature and vitrificationhas been debated (McParland et al., 2010), these particles likelyoriginated from anthropogenic activities that experienced adownturn in the 1950s through a reduction in coal use (for trans-portation, industry, heating or cooking) and the increased effi-ciency of protection measures against catastrophic urban fires.

3.3. Fire, human and climate

Bowman et al. (2011, P2224) defined some key concepts fordistinguishing human from natural burning in paleorecords, suchas

(1) temporal or spatial changes in fire activity and vegetationapparent from palaeoecological proxies, (2) a demonstration thatthese changes are not predicted by climateefuelsefire relation-ships and palaeoclimate reconstructions for the period of fireregime change, and (3) a demonstration that fire regime changescoincide in space and time with changes in human history (e.g.,technological, economic, political, or demographic changes,including colonization of new lands) known from archaeology,anthropology and historical sources.

We would add to these demonstrations the use of paleoeco-logical sites that are in close proximity to historical occupation sites(Clark and Royall, 1995) or directly within urban areas (Lavoie andRichard, 2016). In the region, very few archeological sites have beendated from before 7000 BP, indicating very low population density

or near absence of human occupation (Fig. 3). This finding concurswith the HYDE 3.1 human population density dataset, indicatingclose to zero population density. While the HYDE dataset indicatesnear zero population density up to the last 1000 years, it should benoted that the model is highly affected by technological use andagriculture, such that it likely underestimates the populationdensity from hunter-gatherers onwards. From 7000 to 4600 BP,humans are present around the Folly peatland (Arch�eotec, 2014;Lalibert�e, 2002, 2000), when archeological sites start being recor-ded regularly (Fig. 3). After 4600 BP and up to 1500 BP, archeo-logical findings are widespread (Pilon, 1999) and this period ofmedium population density is associated with prevalent fires withthe lowest FRI for the entire period (135 years, Table 2).

The past 1500 years have been characterized by a large increasein population density that accelerated during the past 200 yearsafter the European settlement of the region and the founding ofGatineau after 1800 AD (Gatineau Park and the Folly peatlandsurroundings were settled after 1826 by the Pink family). Duringthat period, FRI changed greatly from the previous period toward alengthening of the FRI (301 ± 201 years, Table 2, Fig. 4). Althoughimportant, this change is not significant at the 5% level, but could beconsidered at the 10% level (KS test: P ¼ 0.06). All other periodsshowed no significant difference in their FRI distributions at the10% level (results not shown). This change in FRI is associatedwith asignificant change in vegetation at least for the last 500 years;indeed, CONISS analysis detected a vegetation shift. Although thelast pollen zone corresponding to the past 500 years can beattributed to the appearance of anthropogenic taxa (e.g., Ambrosiaand Chenopodiaceae [now Amaranthaceae], among others), vege-tation changes occurred earlier. Since 1500 BP, we recorded an in-crease in P. strobus, and a decrease in T. canadensis, F. grandifolia andBetula spp. (Fig. 2). It appears that temporal changes in vegetationand fire were apparent during the last 1500 years (Bowman et al.,2011).

Concurrent with FRI change, the highest charcoal peaks thatwere recorded during the Holocene are indicative of larger or moresevere fire events (i.e., high biomass burning, but low fire

O. Blarquez et al. / Quaternary Science Reviews 180 (2018) 63e74 71

frequency; Ali et al., 2012). Similar large charcoal peaks wererecorded by Clark and Royall (1995) at Crawford Lake in southernOntario, which they attributed to early aboriginal burning. Togetherwith this site, several other paleoecological records in southernOntario exhibited vegetation changes that were likely influencedby aboriginal burning (Clark and Royall 1996; Munoz and Gajewski,2010). Munoz and Gajewski (2010) found that the sites displayingsignificant indigeous influences were all characterized by ~5%anthropogenic markers in the pollen record (i.e., the sum of Am-brosia, Artemisia, Rumex, Ranunculus, Pteridium, Brassicaceae,Plantago, Urtica, Caryophyllaceae and Polygonum), together with alarge decrease in Fagus and an increase in Pinus. For our record,anthropogenic indicators attained the 5% level as early as 1300 BP(Fig. S3), along with a fire that was associated with a very largecharcoal peak (Fig. 3), and during a period characterized byincreasing archeological radiocarbon site datings (Fig. 3).

The main change in climate that can be inferred from the datathat are presented here corresponds to the termination of the lastglacial period and the transition from the Pleistocene to HoloceneEpoch, which is accompanied by an increase in temperature. A 7 �Cgain in temperature is observed from 12 000 to 10 000 BP, ac-cording to the CCSM3model simulation and this trend matches thed18O record from the NGRIP ice core (Liu et al., 2009, Fig. 3). CCSM3simulated precipitation followed the temperature trend but withgreater variability, as shown by larger confidence intervals.Maximum temperatures are attained around 7500 BP during aperiod that corresponds to the Holocene Thermal Maximum (ca.9000e5000 BP). The Neoglacial period that corresponds to a colderand drier period starting around 4000 BP is not apparent from thedata that are presented here. Main trends in CO2 were expressed bya gradual increase in concentration from 12 000 to 11 000 BP andafter 8000 BP, and a rise after 1800 AD that matched the HYDEpopulation density dataset (Fig. 3). Since 10 700 BP, fires occurredaround the mean FRI, but several periods of no fire occurrence (i.e.,long FRI), which lasted several hundred years, occurred at ca. 8000,6000, 4200, 2600 and 1000 BP (Fig. 3). The no-fire period at ca.8000 BP (8300-7900 BP) largely coincides with the North Atlanticclimate cooling known as the “8200 BP event,” which is related tothe discharge of the proglacial Lake Ojibway into the Atlantic Ocean(Alley et al., 1997; Dansgaard et al., 1989). This cooling is apparent inboth temperature anomalies from the CCSM3 simulation and d18OfromNGRIP (Fig. 3, Monnin et al., 2004). Subsequent no-fire periodsat 6200-6000, 4400-3900, and 2800-2400 BP all match knownclimate cooling periods that were evidenced by North Atlantic ice-rafting events (Bond, 2001), times of polar cell expansion fromGISP2 ice core (O'Brien et al., 1995), glacial advances in Scandinavia(Denton and Karl�en, 1973) or pine pollen declines in eastern NorthAmerica (Willard et al., 2005). Low temperatures during thoseperiods may have prevented the occurrence of fires, due to lowerecosystem productivity and fuel availability (Carcaillet andBlarquez, 2017), while precipitation effects are less clear orapparent in our data (Fig. 3). The most recent no-fire periodoccurred at 1300-700 BP and does not correspond to knownclimate cooling; rather, it is centered on the Medieval WarmClimate anomaly (Mann et al., 2009). Counter-intuitively, low firenumber during this period is related to warmer conditions and lowprecipitation, which further argues in favor of human effects on thefire regime (Fig. 3).

4. Conclusion

Vegetation changes that occurred during the last millenniumwere first attributed to climate, particularly to the Little Ice Agecooling (Campbell and McAndrews, 1993). It has since beendemonstrated that large vegetation changes, such as the abrupt

Fagus decline that were observed at several southern Ontario sites,were related to a combination of factors that must include humanimpacts and fire (Campbell and McAndrews, 1995; Clark, 1995;Munoz and Gajewski, 2010). Algonquian populations that hadinhabited the area around Folly peat bog probably did not practiceagriculture during prehistoric times. They certainly used fire forother purposes, for example, to drive game, to clear underbrush orto favor the growth of specific wild plant species (Day, 1953;Patterson III and Sassaman, 1988), although the frequency, in-tensity and duration of such practices have been the subjects ofdebate (see Barrett et al. 2005; Lewis 1982; Myers and Peroni 1983;Russell 1983; Williams 2000). Nevertheless, it is impossible toexclude human influence through fire use and fire regime modifi-cation given the substantial changes in vegetation and fire that weobserved. The fire regime of the past 200 years, which is un-doubtedly associated with anthropogenic activities (SCPs) andcharacterized with few and large charcoal peaks, appears as acontinuation of human-fire history rather than a complete shift.Population density and associated human fire uses are thus drivers,and modulated, the long-term fire history. Human-fire history andrelationships that can be deduced from our paleorecord seem tocomply with concepts advanced by Bowman et al. (2011). They arealso peculiar compared to the Holocene trends. These peculiaritiesshould be considered for future ecosystem conservation andmanagement.

Author contributions

OB and JT designed the study; LL-E analyzed the pollen; JPanalyzed the charcoal record; CGSP provided archeological data;NP performed sediment lead dating; NP and JP performed LOI an-alyses; OB performed the statistical analyses; OB wrote themanuscript that was commented on, improved, and approved by allco-authors.

Acknowledgments

We thank Pierre J.H. Richard for comments on the manuscriptand Sabrina Demers-Thibault and Jules Regard for their assistancein the field. We thank two anonymous reviewers for constructivecomments on the manuscript. We thank WFJ Parsons for Englishediting. This work was funded by the Natural Sciences and Engi-neering Research Council of Canada (OB, JT), the Fonds de rechercheNature et technologies du Qu�ebec (OB), and by a research grantfrom the National Capital Commission, Gatineau Park and theFriends of Gatineau Park (JT).

Appendix A. Supplementary data

Supplementary data related to this article can be found athttps://doi.org/10.1016/j.quascirev.2017.11.022.

References

Aleman, J.C., Blarquez, O., Bentaleb, I., Bont�e, P., Brossier, B., Carcaillet, C., Gond, V.,Gourlet-Fleury, S., Kpolita, A., Lef�evre, I., Oslisly, R., Power, M.J., Yongo, O.,Bremond, L., Favier, C., 2013. Tracking land-cover changes with sedimentarycharcoal in the Afrotropics. Holocene 23, 1853e1862. https://doi.org/10.1177/0959683613508159.

Ali, A.A., Blarquez, O., Girardin, M.P., Hely, C., Tinquaut, F., El Guellab, A.,Valsecchi, V., Terrier, A., Bremond, L., Genries, A., Gauthier, S., Bergeron, Y., 2012.Control of the multimillennial wildfire size in boreal North America by springclimatic conditions. Proc. Natl. Acad. Sci. 109, 20966e20970. https://doi.org/10.1073/pnas.1203467109.

Ali, A.A., Carcaillet, C., Bergeron, Y., 2009a. Long-term fire frequency variability inthe eastern Canadian boreal forest : the influence of climate vs local factors.Glob. Chang. Biol. 15, 1230e1241.

Ali, A.A., Higuera, P.E., Bergeron, Y., Carcaillet, C., 2009b. Comparing fire-history

O. Blarquez et al. / Quaternary Science Reviews 180 (2018) 63e7472

interpretations based on area, number and estimated volume of macroscopiccharcoal in lake sediments. Quat. Res. 72, 462e468. https://doi.org/10.1016/j.yqres.2009.07.002.

Alley, R.B., Mayewski, P.A., Sowers, T., Stuiver, M., Taylor, K.C., Clark, P.U., 1997.Holocene climatic instability: a prominent, widespread event 8200 yr ago.Geology 25, 483e486. https://doi.org/10.1130/0091-7613(1997)025<0483:HCIAPW>2.3.CO;2.

Andersen, K.K., Azuma, N., Barnola, J.-M., Bigler, M., Biscaye, P., Caillon, N.,Chappellaz, J., Clausen, H.B., Dahl-Jensen, D., Fischer, H., Flückiger, J.,Fritzsche, D., Fujii, Y., Goto-Azuma, K., Grønvold, K., Gundestrup, N.S.,Hansson, M., Huber, C., Hvidberg, C.S., Johnsen, S.J., Jonsell, U., Jouzel, J.,Kipfstuhl, S., Landais, A., Leuenberger, M., Lorrain, R., Masson-Delmotte, V.,Miller, H., Motoyama, H., Narita, H., Popp, T., Rasmussen, S.O., Raynaud, D.,Rothlisberger, R., Ruth, U., Samyn, D., Schwander, J., Shoji, H., Siggard-Andersen, M.-L., Steffensen, J.P., Stocker, T., Sveinbj€ornsd�ottir, A.E., Svensson, A.,Takata, M., Tison, J.-L., Thorsteinsson, T., Watanabe, O., Wilhelms, F.,White, J.W.C., 2004. High-resolution record of Northern Hemisphere climateextending into the last interglacial period. Nature 431, 147e151. https://doi.org/10.1038/nature02805.

Appleby, P.G., Oldfieldz, F., 1983. The assessment of 210Pb data from sites withvarying sediment accumulation rates. Hydrobiologia 103, 29e35. https://doi.org/10.1007/BF00028424.

Arch�eotec, 2014. Travaux de r�eam�enagement de la rue Jacques-Cartier, ville deGatineau Site BiFw-172.

Balch, J.K., Bradley, B.A., Abatzoglou, J.T., Nagy, R.C., Fusco, E.J., Mahood, A.L., 2017.Human-started wildfires expand the fire niche across the United States. Proc.Natl. Acad. Sci. 114, 2946e2951. https://doi.org/10.1073/pnas.1617394114.

Barrett, S.W., Swetnam, T.W., Baker, W.L., 2005. indian fire use: deflating the legend.Fire Manag. Today 65, 31e33.

Bernier, P.Y., Gauthier, S., Jean, P.O., Manka, F., Boulanger, Y., Beaudoin, A.,Guindon, L., 2016. Mapping local effects of forest properties on fire risk acrossCanada. Forests 7, 157. https://doi.org/10.3390/f7080157.

Birks, H.J.B., Gordon, D., 1985. Numerical methods in quaternary pollen analysis. J.Quat. Sci. https://doi.org/10.1016/0305-4403(86)90048-8. Academic Press.Harcourt Brace Jovanovich, Orlando.

Blaauw, M., 2010. Methods and code for “classical”age-modelling of radiocarbonsequences. Quat. Geochronol. 5, 512e518.

Blarquez, O., Aleman, J.C., 2016. Tree biomass reconstruction shows no lag inpostglacial afforestation of eastern Canada. Can. J. For. Res. 46, 485e498.https://doi.org/10.1139/cjfr-2015-0201.

Blarquez, O., Ali, A.A., Girardin, M.P., Grondin, P., Fr�echette, B., Bergeron, Y., H�ely, C.,2015. Regional paleofire regimes affected by non-uniform climate, vegetationand human drivers. Sci. Rep. 5, 13356. https://doi.org/10.1038/srep13356.

Blarquez, O., Bremond, L., Carcaillet, C., 2010. Holocene fires and a herb-dominatedunderstorey track wetter climates in subalpine forests. J. Ecol. 98, 1358e1368.https://doi.org/10.1111/j.1365-2745.2010.01721.x.

Blarquez, O., Girardin, M.P., Leys, B., Ali, A.A., Aleman, J.C., Bergeron, Y., Carcaillet, C.,2013. Paleofire reconstruction based on an ensemble-member strategy appliedto sedimentary charcoal. Geophys. Res. Lett. 40, 2667e2672. https://doi.org/10.1002/grl.50504.

Blarquez, O., Vanni�ere, B., Marlon, J.R., Daniau, A.L., Power, M.J., Brewer, S.,Bartlein, P.J., 2014. Paleofire: an R package to analyse sedimentary charcoalrecords from the Global Charcoal Database to reconstruct past biomass burning.Comput. Geosci. 72, 255e261. https://doi.org/10.1016/j.cageo.2014.07.020.

Bond, G., 2001. Persistent solar influence on North Atlantic climate during theHolocene. Sci. 294, 2130e2136. https://doi.org/10.1126/science.1065680.

Bowman, D.M.J.S., Balch, J., Artaxo, P., Bond, W.J., Cochrane, M.A., D’Antonio, C.M.,Defries, R., Johnston, F.H., Keeley, J.E., Krawchuk, M.A., Kull, C.A., Mack, M.,Moritz, M.A., Pyne, S., Roos, C.I., Scott, A.C., Sodhi, N.S., Swetnam, T.W., 2011. Thehuman dimension of fire regimes on Earth. J. Biogeogr. 38, 2223e2236. https://doi.org/10.1111/j.1365-2699.2011.02595.x.

Bowman, D.M.J.S., Balch, J.K., Artaxo, P., Bond, W.J., Carlson, J.M., Cochrane, M.A.,D'Antonio, C.M., DeFries, R.S., Doyle, J.C., Harrison, S.P., Johnston, F.H.,Keeley, J.E., Krawchuk, M.A., Kull, C.A., Marston, J.B., Moritz, M.A., Prentice, I.C.,Roos, C.I., Scott, A.C., Swetnam, T.W., van der Werf, G.R., Pyne, S.J., 2009. Fire inthe earth system. Sci. 324, 481e484. https://doi.org/10.1126/science.1163886.

Bridge, S.R.J., 2001. Spatial and Temporal Variations in the Fire Cycle Across Ontario.OMNR, Northeast Science & Technology. NEST Technical Report TN-043 41p.

Cahill, C., Glendenning, J., Buchan, K., Seguin, J., Smythe, D., Murray, J., Knowles, S.,2008. Le parc de la Gatineau Un tr�esor menac�e. Societ�e pour la nature et lesparcs du Canada, section vall�ee de l’Outaouais, 28 p.

Campbell, I.D., Campbell, C., 1994. The impact of late woodland land use on theforest landscape of southern Ontario. Gt. Lakes Geogr. 1, 21e29.

Campbell, I.D., McAndrews, J.H., 1995. Charcoal evidence for indian-set fires: acomment on Clark and Royall. Holocene 5, 369e370. https://doi.org/10.1177/095968369500500314.

Campbell, I.D., McAndrews, J.H., 1993. Forest disequilibrium caused by rapid LittleIce Age cooling. Nature 366, 336e338. https://doi.org/10.1038/366336a0.

Carcaillet, C., 2017. Unlimited fuel wood during the middle Mesolithic (9650e8300cal. yr BP) in northern Sweden: fuel typology and pine-dominated vegetationinferred from charcoal identification and tree-ring morphology. Holocene 27,1370e1378. https://doi.org/10.1177/0959683617693894.

Carcaillet, C., Almquist, H., Asnong, H., Bradshaw, R.H.W., Carri�on, J.S., Gaillard, M.-J.,Gajewski, K., Haas, J.N., Haberle, S.G., Hadorn, P., Müller, S.D., Richard, P.J.H.,Richoz, I., R€osch, M., S�anchez Go~ni, M.F., von Stedingk, H., Stevenson, A.C.,

Talon, B., Tardy, C., Tinner, W., Tryterud, E., Wick, L., Willis, K.J., 2002. Holocenebiomass burning and global dynamics of the carbon cycle. Chemosphere 49,845e863. https://doi.org/10.1016/S0045-6535(02)00385-5.

Carcaillet, C., Bergeron, Y., Richard, P.J.H., Fr�echette, B., Gauthier, S., Prairie, Y.T., 2001.Change of fire frequency in the eastern Canadian boreal forests during theHolocene: does vegetation composition or climate trigger the fire regime?J. Ecol. 89, 930e946. https://doi.org/10.1046/j.0022-0477.2001.00614.x.

Carcaillet, C., Blarquez, O., 2017. Fire ecology of a tree glacial refugium on a nunatakwith a view on Alpine glaciers. New Phytol. 216, 1281e1290. https://doi.org/10.1111/nph.14721.

Chaput, M.A., Gajewski, K., 2016. Radiocarbon dates as estimates of ancient humanpopulation size. Anthropocene 15, 3e12. https://doi.org/10.1016/j.ancene.2015.10.002.

Charles, D.F., Binford, M.W., Furlong, E.T., Hites, R.A., Mitchell, M.J., Norton, S.A.,Oldfield, F., Paterson, M.J., Smol, J.P., Uutala, A.J., 1990. Paleoecological investi-gation of recent lake acidification in the Adirondack Mountains, NY. J. Paleo-limnol 3, 195e241.

Clark, J.S., 1995. Climate and indian effects on southern Ontario forests - reply.Holocene 5, 371e379. https://doi.org/10.1177/095968369500500315.

Clark, J.S., Royall, P.D., 1995. Transformation of a northern hardwood forest byaboriginal (Iroquois) fire: charcoal evidence from Crawford Lake, Ontario,Canada. Holocene 5, 1e9. https://doi.org/10.1177/095968369500500101.

Clark, J.S., Royall, P.D., 1996. Local and regional sediment charcoal evidence for fireregimes in presettlement north-eastern North America. J. Ecol 84, 365e382.https://doi.org/10.2307/2261199.

Clark, J.S., Stocks, B.J., Richard, P.J.H., 1996. Climate implications of biomass burningsince the 19th century in eastern North America. Glob. Chang. Biol. 2, 433e442.https://doi.org/10.1111/j.1365-2486.1996.tb00093.x.

Colombaroli, D., Beckmann, M., van der Knaap, W.O., Curdy, P., Tinner, W., 2013.Changes in biodiversity and vegetation composition in the central Swiss Alpsduring the transition from pristine forest to first farming. Divers. Distrib. 19,157e170. https://doi.org/10.1111/j.1472-4642.2012.00930.x.

Dansgaard, W., White, J.W.C., Johnsen, S.J., 1989. The abrupt termination of theYounger Dryas climate event. Nature 339, 532e534. https://doi.org/10.1038/340301a0.

Day, G.M., 1953. The Indian as an ecological factor in the northeastern forest.Ecology 34, 329e346. https://doi.org/10.2307/1930900.

Dean Jr., Walter E., 1974. Determination of carbonate and organic matter incalcareous sediments and sedimentary rocks by loss on ignition: comparisonwith other methods. J. Sediment. Petrol 44, 242e248 doi: http://dx.doi.org/10.1306/74D729D2-2B21-11D7-8648000102C1865D.

Delignette-Muller, M.L., Dutang, C., 2015. Fitdistrplus: an R package for fitting dis-tributions. J. Stat. Softw. 64, 1e34. https://doi.org/10.18637/jss.v064.i04.

Denton, G.H., Karl�en, W., 1973. Holocene climatic variations-Their pattern andpossible cause. Quat. Res. 3 https://doi.org/10.1016/0033-5894(73)90040-9.

Dey, D.C., Guyette, R.P., 2000. Anthropogenic fire history and red oak forests insouth-central Ontario. For. Chron 76, 339e347.

Flannigan, M.D., Krawchuk B, M.A., De Groot, W.J., MikeWotton, B.A., Gowman, L.M.,2009. Implications of changing climate for global wildland fire. Int. J. Wildl. Fire18, 483e507. https://doi.org/10.1071/WF08187.

Fleming, R.A., Candau, J.N., McAlpine, R.S., 2002. Landscape-scale analysis of in-teractions between insect defoliation and forest fire in central Canada. Clim.Change 55, 251e272. https://doi.org/10.1023/A:1020299422491.

Fordham, D.A., Saltr�e, F., Brown, S.C., Mellin, C., Wigley, T.M.L., 2017a. Why decadalto century timescale paleoclimate data is needed to explain present-day pat-terns of biological diversity and change. Glob. Chang. Biol. https://doi.org/10.1111/gcb.13932.

Fordham, D.A., Saltr�e, F., Haythorne, S., Wigley, T.M.L., Otto-Bliesner, B.L., Chan, K.C.,Brook, B.W., 2017b. PaleoView: a tool for generating continuous climate pro-jections spanning the last 21 000 years at regional and global scales. Ecogr. 40,1348e1358. https://doi.org/10.1111/ecog.03031.

Fuller, J.L., 1998. Ecological impact of the mid-Holocene hemlock decline insouthern Ontario, Canada. Ecology 79, 2337e2351.

Gill, J.L., Williams, J.W., Jackson, S.T., Lininger, K.B., Robinson, G.S., 2009. Pleistocenemegafaunal collapse, novel plant communities, and enhanced fire regimes innorth America. Sci. 326, 1100e1103. https://doi.org/10.1126/science.1179504.

Girardin, M.P., Ali, A.A., Carcaillet, C., Blarquez, O., H�ely, C., Terrier, A., Genries, A.,Bergeron, Y., 2013. Vegetation limits the impact of a warm climate on borealwildfires. New Phytol. 199, 1001e1011. https://doi.org/10.1111/nph.12322.

Grimm, E.C., 1987. Constrained cluster analysis by the method of incremental sumof squares. Comput. Geosci. 13, 13e35. https://doi.org/10.1016/0098-3004(87)90022-7.

Guyette, R.P., Dey, D.C., Stambaugh, M.C., Muzika, R.M., 2006. Fire scars revealvariability and dynamics of eastern fire regimes. In: Fire in Eastern Oak Forests:Delivering Science to Land Managers. Proceedings of a Conference. GeneralTechnical Report NRS-P-1, pp. 20e39.

H�ely, C., Bergeron, Y., Flannigan, M.D., 2000. Effects of stand composition on firehazard in mixed-wood Canadian boreal forest. J. Veg. Sci. 11, 813e824. https://doi.org/10.2307/3236551.

Higuera, P.E., Brubaker, L.B., Anderson, P.M., Hu, F.S., Brown, T.A., 2009. Vegetationmediated the impacts of postglacial climate change on fire regimes in thesouth-central Brooks Range, Alaska. Ecol. Monogr. 79, 201e219. https://doi.org/10.1890/07-2019.1.

Higuera, P.E., Peters, M.E., Brubaker, L.B., Gavin, D.G., 2007. Understanding the originand analysis of sediment-charcoal records with a simulation model. Quat. Sci.

O. Blarquez et al. / Quaternary Science Reviews 180 (2018) 63e74 73

Rev. 26, 1790e1809. https://doi.org/10.1016/j.quascirev.2007.03.010.Higuera, P.E., Sprugel, D.G., Brubaker, L.B., 2005. Reconstructing fire regimes with

charcoal from small-hollow sediments: a calibration with tree-ring records offire. Holocene 15, 238e251. https://doi.org/10.1191/0959683605hl789rp.

Johnson, E.A., Gutsell, S.L., 1994. Fire frequency models, methods and in-terpretations. Adv. Ecol. Res. Vol. 25, 239e287. https://doi.org/10.1016/S0065-2504(08)60216-0.

Juggins, S., 2015. Rioja: Analysis of Quaternary Science Data, R Package.Kelly, R.F., Higuera, P.E., Barrett, C.M., Hu, F.S., 2011. A signal-to-noise index to

quantify the potential for peak detection in sediment-charcoal records. Quat.Res. 75, 11e17. https://doi.org/10.1016/j.yqres.2010.07.011.

Klein Goldewijk, K., Beusen, A., Janssen, P., 2010. Long-term dynamic modeling ofglobal population and built-up area in a spatially explicit way: HYDE 3.1. Ho-locene 20, 565e573. https://doi.org/10.1177/0959683609356587.

Lalibert�e, M., 2002. Projet Kabeshinan: recherches arch�eologiques de 2002 sur lessites BiFw-6 et BiFw-26. Ottawa.

Lalibert�e, M., 2000. Synth�ese: Recherches arch�eologiques dans le Parc du Lac-Leamy1993-1999. Hull.

Lauriol, B., Deschamps, �E., Carrier, L., Grimm, W., Morlan, R., Talon, B., 2003. Caveinfill and associated biotic remains as indicators of Holocene environment inGatineau Park (Quebec, Canada). Can. J. Earth Sci. 40, 789e803. https://doi.org/10.1139/e03-015.

Lavoie, M., Richard, P.J.H., 2016. The 8200-year vegetation history of an urbanwoodland as reconstructed from pollen and plant remains. Botany 95, 163e172.https://doi.org/10.1139/cjb-2016-0182 cjb-2016-0182.

Levine, J.S., 1990. Global biomass burning: atmospheric, climatic and biosphericimplications. Eos, Trans. Am. Geophys. Union 71, 1075e1077. https://doi.org/10.1029/90EO00289.

Lewis, H.T., 1982. Fire technology and resource management in Aboriginal NorthAmerica and Australia. AAAS Sel. Symp. 45e68.

Leys, B., Carcaillet, C., Dezileau, L., Ali, A.A., Bradshaw, R.H.W., 2013. A comparison ofcharcoal measurements for reconstruction of Mediterranean paleo-fire fre-quency in the mountains of Corsica. Quat. Res. 79, 337e349. https://doi.org/10.1016/j.yqres.2013.01.003.

Liu, Z., Otto-Bliesner, B.L., He, F., Brady, E.C., Tomas, R., Clark, P.U., Carlson, A.E.,Lynch-Stieglitz, J., Curry, W., Brook, E., Erickson, D., Jacob, R., Kutzbach, J.,Cheng, J., 2009. Transient simulation of last deglaciation with a newmechanismfor bolling-allerod warming. Sci. 325, 310e314. https://doi.org/10.1126/science.1171041.

Mack, M.C., Bret-Harte, M.S., Hollingsworth, T.N., Jandt, R.R., Schuur, E.A.G.,Shaver, G.R., Verbyla, D.L., 2011. Carbon loss from an unprecedented Arctictundra wildfire. Nature 475, 489e492. https://doi.org/10.1038/nature10283.

Mann, M.E., Zhang, Z., Rutherford, S., Bradley, R.S., Hughes, M.K., Shindell, D.,Ammann, C., Faluvegi, G., Ni, F., 2009. Global signatures and dynamical originsof the Little ice age and medieval climate anomaly. Sci. 326, 1256e1260. https://doi.org/10.1126/science.1177303.

Marlon, J.R., Bartlein, P.J., Gavin, D.G., Long, C.J., Anderson, R.S., Briles, C.E.,Brown, K.J., Colombaroli, D., Hallett, D.J., Power, M.J., Scharf, E.A., Walsh, M.K.,2012. Long-term perspective on wildfires in the western USA. Proc. Natl. Acad.Sci. 109, E535eE543. https://doi.org/10.1073/pnas.1112839109.

Marlon, J.R., Kelly, R., Daniau, A.L., Vanni�ere, B., Power, M.J., Bartlein, P., Higuera, P.,Blarquez, O., Brewer, S., Brücher, T., Feurdean, A., Romera, G.G., Iglesias, V., YoshiMaezumi, S., Magi, B., Mustaphi, C.J.C., Zhihai, T., 2016. Reconstructions ofbiomass burning from sediment-charcoal records to improve data-modelcomparisons. Biogeosciences 13, 3225e3244. https://doi.org/10.5194/bg-13-3225-2016.

Martindale, A., Morlan, R., Betts, M., Blake, M., Gajewski, K., Chaput, M., Mason, A.,Vermeersch, P., 2016. Canadian Archaeological Radiocarbon Database (CARD2.1). . (Accessed 15 September 2017).

McParland, L.C., Collinson, M.E., Scott, A.C., Campbell, G., Veal, R., 2010. Is vitrifi-cation in charcoal a result of high temperature burning of wood? J. Archaeol.Sci. 37, 2679e2687. https://doi.org/10.1016/j.jas.2010.06.006.

Millar, C.I., Stephenson, N.L., 2015. Temperate forest health in an era of emergingmegadisturbance. Sci. 349, 823e826. https://doi.org/10.1126/science.aaa9933.

Monnin, E., Steig, E.J., Siegenthaler, U., Kawamura, K., Schwander, J., Stauffer, B.,Stocker, T.F., Morse, D.L., Barnola, J.M., Bellier, B., Raynaud, D., Fischer, H., 2004.Evidence for substantial accumulation rate variability in Antarctica during theHolocene, through synchronization of CO2 in the Taylor Dome, Dome C andDML ice cores. Earth Planet. Sci. Lett. 224, 45e54. https://doi.org/10.1016/j.epsl.2004.05.007.

Mott, R.J., Farley-Gill, L., 1981. Two Late Quaternary Pollen Profiles from GatineauPark, Qu�ebec. Geological Survey of Canada. Paper no. 80e31.

Munoz, S.E., Gajewski, K., 2010. Distinguishing prehistoric human influence on late-Holocene forests in southern Ontario, Canada. Holocene 20, 967e981. https://doi.org/10.1177/0959683610362815.

Mustaphi, C.J.C., Pisaric, M.F.J., 2014. A classification for macroscopic charcoalmorphologies found in Holocene lacustrine sediments. Prog. Phys. Geogr. 38,734e754. https://doi.org/10.1177/0309133314548886.

Myers, R.L., Peroni, P.A., 1983. Approaches to determining aboriginal fire use and itsimpact on vegetation. Bull. Ecol. Soc. Am. 64, 217e218. https://doi.org/10.2307/20166351.

O'Brien, S.R., Mayewski, P.A., Meeker, L.D., Meese, D.A., Twickler, M.S., Whitlow, S.I.,1995. Complexity of Holocene climate as reconstructed from a Greenland icecore. Sci. 270, 1962e1964. https://doi.org/10.1126/science.270.5244.1962.

Occhietti, S., 2007. The Saint-Narcisse morainic complex and early Younger Dryas

events on the southeastern margin of the Laurentide Ice Sheet. G�eogr. Phys.Quat. 61, 89e117. https://doi.org/10.7202/038987ar.

Occhietti, S., Richard, P.J.H., 2003. Effet r�eservoir sur les ages C de la Mer deChamplain �a la transition Pl�eistoc�ene-Holoc�ene: r�evision de la chronologie de lad�eglaciation au Qu�ebec m�eridional. G�eogr. Phys. Quat. 57, 115e138. https://doi.org/10.7202/011308ar.

Oris, F., Ali, A.A., Asselin, H., Paradis, L., Bergeron, Y., Finsinger, W., 2014. Charcoaldispersion and deposition in boreal lakes from 3 years of monitoring: differ-ences between local and regional fires. Geophys. Res. Lett. 41, 6743e6752.https://doi.org/10.1002/2014GL060984.

Parshall, T., Foster, D.R., 2002. Fire on the New England landscape: regional andtemporal variation, cultural and environmental controls. J. Biogeogr. 29,1305e1317. https://doi.org/10.1046/j.1365-2699.2002.00758.x.

Patterson III, W.A., Sassaman, K.E., 1988. Indian fires in the prehistory of new en-gland. In: Holocene Human Ecology in Northeastern North America. Springer,Boston, MA, pp. 107e135. https://doi.org/10.1007/978-1-4899-2376-9_6.

Pausas, J.G., Keeley, J.E., Schwilk, D.W., 2017. Flammability as an ecological andevolutionary driver. J. Ecol. 105, 289e297. https://doi.org/10.1111/1365-2745.12691.

Pedersen, M.W., Ruter, A., Schweger, C., Friebe, H., Staff, R.A., Kjeldsen, K.K.,Mendoza, M.L.Z., Beaudoin, A.B., Zutter, C., Larsen, N.K., Potter, B.A., Nielsen, R.,Rainville, R.A., Orlando, L., Meltzer, D.J., Kjær, K.H., Willerslev, E., 2016. Post-glacial viability and colonization in North America's ice-free corridor. Nature537, 45e49. https://doi.org/10.1038/nature19085.

Pilon, J.-L. (Ed.), 1999. Ottawa Valley Prehistory. Outaouais Historical Society, Hull.Power, M., Mayle, F., Bartlein, P., Marlon, J., Anderson, R., Behling, H., Brown, K.,

Carcaillet, C., Colombaroli, D., Gavin, D., Hallett, D., Horn, S., Kennedy, L., Lane, C.,Long, C., Moreno, P., Paitre, C., Robinson, G., Taylor, Z., Walsh, M., 2013. Climaticcontrol of the biomass-burning decline in the Americas after ad 1500. Holocene23, 3e13. https://doi.org/10.1177/0959683612450196.

Power, M.J., Marlon, J., Ortiz, N., Bartlein, P.J., Harrison, S.P., Mayle, F.E., Ballouche, A.,Bradshaw, R.H.W., Carcaillet, C., Cordova, C., Mooney, S., Moreno, P.I.,Prentice, I.C., Thonicke, K., Tinner, W., Whitlock, C., Zhang, Y., Zhao, Y., Ali, A.A.,Anderson, R.S., Beer, R., Behling, H., Briles, C., Brown, K.J., Brunelle, A., Bush, M.,Camill, P., Chu, G.Q., Clark, J., Colombaroli, D., Connor, S., Daniau, A.L.,Daniels, M., Dodson, J., Doughty, E., Edwards, M.E., Finsinger, W., Foster, D.,Frechette, J., Gaillard, M.J., Gavin, D.G., Gobet, E., Haberle, S., Hallett, D.J.,Higuera, P., Hope, G., Horn, S., Inoue, J., Kaltenrieder, P., Kennedy, L., Kong, Z.C.,Larsen, C., Long, C.J., Lynch, J., Lynch, E.A., McGlone, M., Meeks, S., Mensing, S.,Meyer, G., Minckley, T., Mohr, J., Nelson, D.M., New, J., Newnham, R., Noti, R.,Oswald, W., Pierce, J., Richard, P.J.H., Rowe, C., Sanchez Go~ni, M.F., Shuman, B.N.,Takahara, H., Toney, J., Turney, C., Urrego-Sanchez, D.H., Umbanhowar, C.,Vandergoes, M., Vanniere, B., Vescovi, E., Walsh, M., Wang, X., Williams, N.,Wilmshurst, J., Zhang, J.H., 2008. Changes in fire regimes since the last glacialmaximum: an assessment based on a global synthesis and analysis of charcoaldata. Clim. Dyn. 30, 887e907. https://doi.org/10.1007/s00382-007-0334-x.

Pyne, S.J., 1997. Fire in America: a Cultural History of Wildland and Rural Fire.University of Washington Press.

R Core Team, 2017. R: a language and environment for statistical computing. RFound. Stat. Comput. doi:ISBN: 3-900051-07-0.

Reimer, P.J., Bard, E., Bayliss, A., Beck, J.W., Blackwell, P.G., Ramsey, C.B., Buck, C.E.,Cheng, H., Edwards, R.L., Friedrich, M., Grootes, P.M., Guilderson, T.P.,Haflidason, H., Hajdas, I., Hatt�e, C., Heaton, T.J., Hoffmann, D.L., Hogg, A.G.,Hughen, K.A., Kaiser, K.F., Kromer, B., Manning, S.W., Niu, M., Reimer, R.W.,Richards, D.A., Scott, E.M., Southon, J.R., Staff, R.A., Turney, C.S.M., van derPlicht, J., 2013. IntCal13 and Marine13 radiocarbon age calibration curves0e50,000 Years cal BP. Radiocarbon 55, 1869e1887. https://doi.org/10.2458/azu_js_rc.55.16947.

Rose, N.L., 2015. Spheroidal carbonaceous fly ash particles provide a globally syn-chronous stratigraphic marker for the anthropocene. Environ. Sci. Technol. 49,4155e4162. https://doi.org/10.1021/acs.est.5b00543.

Rose, N.L., 1994. A note on further refinements to a procedure for the extraction ofcarbonaceous fly-ash particles from sediments. J. Paleolimnol. 11, 201e204.https://doi.org/10.1007/BF00686866.

Rose, N.L., Harlock, S., Appleby, P.G., Battarbee, R.W., 1995. Dating of recent lakesediments in the United Kingdom and Ireland using spheroidal carbonaceousparticle (SCP) concentration profiles. Holocene 5, 328e335. https://doi.org/10.1177/095968369500500308.

Russell, E., 1983. Indian-Set Fires in the forests of the Northeastern United States.Ecology 64, 78e88. https://doi.org/10.2307/1937331.

Sanchez-Cabeza, J.A., Ruiz-Fern�andez, A.C., Ontiveros-Cuadras, J.F., P�erezBernal, L.H., Olid, C., 2014. Monte Carlo uncertainty calculation of 210Pb chro-nologies and accumulation rates of sediments and peat bogs. Quat. Geochronol.23, 80e93. https://doi.org/10.1016/j.quageo.2014.06.002.

Stocks, B.J., Mason, J.A., Todd, J.B., Bosch, E.M., Wotton, B.M., Amiro, B.D.,Flannigan, M.D., Hirsch, K.G., Logan, K.A., Martell, D.L., Skinner, W.R., 2002.Large forest fires in Canada, 1959e1997. J. Geophys. Res. 108, 8149. https://doi.org/10.1029/2001JD000484.

Swindles, G.T., Watson, E., Turner, T.E., Galloway, J.M., Hadlari, T., Wheeler, J.,Bacon, K.L., 2015. Spheroidal carbonaceous particles are a defining stratigraphicmarker for the Anthropocene. Sci. Rep. 5, 10264. https://doi.org/10.1038/srep10264.

Th�ery-Parisot, I., Chabal, L., Chrzavzez, J., 2010. Anthracology and taphonomy, fromwood gathering to charcoal analysis. A review of the taphonomic processesmodifying charcoal assemblages, in archaeological contexts. Palaeogeogr.

O. Blarquez et al. / Quaternary Science Reviews 180 (2018) 63e7474

Palaeoclimatol. Palaeoecol. 291, 142e153. https://doi.org/10.1016/j.palaeo.2009.09.016.

Turner, M.G., Romme, W.H., 1994. Landscape dynamics in crown fires ecosystems.Landsc. Ecol. 9, 59e77.

Vanni�ere, B., Blarquez, O., Rius, D., Doyen, E., Brücher, T., Colombaroli, D., Connor, S.,Feurdean, A., Hickler, T., Kaltenrieder, P., Lemmen, C., Leys, B., Massa, C.,Olofsson, J., 2015. 7000-year human legacy of elevation-dependent Europeanfire regimes. Quat. Sci. Rev. 132, 206e212. https://doi.org/10.1016/j.quascirev.2015.11.012.

Williams, G., 2000. Introduction to aboriginal fire use in North America. Fire Manag.

Today 60, 8e12.Willard, D.A., Bernhardt, C.E., Korejwo, D.A., Meyers, S.R., 2005. Impact of

millennial-scale Holocene climate variability on eastern North Americanterrestrial ecosystems: pollen-based climatic reconstruction. Glob. Planet.Change 47, 17e35. https://doi.org/10.1016/j.gloplacha.2004.11.017.

Willis, K.J., Birks, H.J.B., 2006. What is Natural? The need for a long-termperspective in biodiversity conservation. Sci. (80-. ) 314, 1261e1265. https://doi.org/10.1126/science.1122667.

Wotton, B.M., Martell, D.L., Logan, K.A., 2003. Climate change and people-causedfire occurrence in Ontario. Clim. Change 60, 275e295.