Wildfires in northern Siberian larch dominated communities

This content has been downloaded from IOPscience Please scroll down to see the full text

Download details

IP Address 187111255130

This content was downloaded on 28092013 at 2104

Please note that terms and conditions apply

2011 Environ Res Lett 6 045208

(httpiopscienceioporg1748-932664045208)

View the table of contents for this issue or go to the journal homepage for more

Home Search Collections Journals About Contact us My IOPscience

IOP PUBLISHING ENVIRONMENTAL RESEARCH LETTERS

Environ Res Lett 6 (2011) 045208 (6pp) doi1010881748-932664045208

Wildfires in northern Siberian larchdominated communitiesViacheslav I Kharuk1 Kenneth J Ranson2 Maria L Dvinskaya1

and Sergey T Im1

1 V N Sukachev Institute of Forest Krasnoyarsk 660036 Russia2 NASArsquos Goddard Space Flight Center Code 618 Greenbelt MD 20771 USA

E-mail kharukksckrasnru

Received 14 June 2011Accepted for publication 26 October 2011Published 16 November 2011Online at stacksioporgERL6045208

AbstractThe fire history of the northern larch forests within the permafrost zone in a portion of northernSiberia (sim66N 100E) was studied Since there is little to no human activity in this area fireswithin the study area were mostly caused by lightning Fire return intervals (FRI) were estimated onthe basis of burn marks on tree stems and dates of tree natality FRI values varied from 130 to 350 yrwith a 200 plusmn 50 yr mean For southerly larch dominated communities FRI was found to be shorter(77 plusmn 20 yr at sim61N and 82 plusmn 7 at 64N) and it was longer at the northern boundary (sim71) oflarch stands (320 plusmn 50 yr) During the Little Ice Age period in the 16thndash18th centuries FRI wasapproximately twice as long those as recorded in this study Fire caused changes in the soilincluding increases in soil drainage and permafrost thawing depth and a radial growth increase toabout twice the background value (with more than six times observed in extreme cases) This effectmay simulate the predicted warming impact on the larch growth in the permafrost zone

Keywords wildfires larch forests fire return interval climate change

1 Introduction

Larch (Larix spp) forests comprise about 43 of Russianforests Larch stands are dominant from the Yenisei ridgeon the west (sim92E longitude) to the Pacific Ocean andfrom Baikal Lake to the south to the 73rd parallel tothe north Wildfires are typical for this area with themajority occurring as ground fires due to low forest crownclosure Larch is a pyrophytic species since fires promotethe establishment of larch regeneration and reduce between-species competition Mineralized burned surfaces enrichedwith nutrients are favorable for germination of the small andlow weight larch seeds The expanse of larch forests is atpresent considered to be a carbon sink (Shvidenko et al 2007)However an increase in fire frequency in response to observedclimate changes in the area may result in conversion of this areato a source for greenhouse gases (IPCC 2007) Changes in airtemperature permafrost depth and extent may affect wildfirefrequency (Kharuk et al 2008) In spite of the fact that larchdominated forests occupy about 70 of permafrost areas inSiberia data on fire occurrence in larch forests are presented

in only a few publications (Vaganov and Arbatskaya 1996Kovacs et al 2004 Kharuk et al 2005 2008 Schepaschenkoet al 2008 Wallenius et al 2011) For larch dominated areasof central Siberia the average fire return interval (FRI) wasfound to be 82 plusmn 7 yr (sim64N latitude) and 77 plusmn 20 yr forthe southward lsquolarchndashmixed taigarsquo ecotone For the northernboundary of larch forests (sim71N) the FRI value was estimatedto be 320 plusmn 50 yr (Kharuk et al 2011) For the north-eastlarch forests of Siberia FRI was found to be depending onthe site 50ndash80 and 80ndash120 yr (Schepaschenko et al 2008)For southern central Siberia Wallenius et al (2011) reporteda gradual increase of FRI from 52 yr in the 18th century to164 yr in the 20th century

The purpose of this work is to investigate wildfireoccurrence in the central part of larch dominated communities(figure 1)

2 The study area

The test sites were located within the Kochechum Riverwatershed The Kochechum River is a tributary of theNizhnyaya (Lower) Tunguska River which turns northwest and

1748-932611045208+06$3300 copy 2011 IOP Publishing Ltd Printed in the UK1

Environ Res Lett 6 (2011) 045208 V I Kharuk et al

Figure 1 Map of north central Siberia with the area of thisinvestigation shown as a rectangle Areas marked 1ndash3 are sites ofearlier studies (Kharuk et al 2007 2008 2011) Inset Landsat imageshowing locations (1ndash9) of test sites along the Kochechum RiverMeasurements of older trees from a study of long-term wildfiretrends were acquired in areas denoted by boxes

flows into the Yenisei River This area is the northern part ofthe central Siberian plateau with gentle hills with elevationsup to 1000 m (figure 1)

This is a permafrost area with a severe continentalclimate Mean summer (JJA) and winter (DJF) temperaturesare +12 C and minus35 C respectively Mean summer and annualprecipitation is 180 and 390 mm respectively (these values aremeans over the years 1997ndash2006 averaged for 15 times 25 gridcells and covering all test sites (Climate Explorer 2011))

The forests are composed of larch (Larix gmelinii Rupr)with a mixture of birch (Betula pendula Roth) Typicalground cover is composed of lichen and moss Bushes wererepresented by Betula nana Salix sp Ribes sp Rosa spJuniperus sp Vaccinium sp and Ledum palustre L (Labradortea)

The wildfires across this landscape occur as ground firesdue to low crown closure The seasonal fire distribution issingle-mode (late MayndashJune) with rare late (ie AugustndashearlySeptember) fires (Sofronov et al 1999 Kharuk et al 2007)Periodic stand-replacing fires cause a mosaic of (semi-)even-age stands embedded with older trees that survived the fire

3 The materials and method

31 Tree sampling

The samples were collected in 2001 and 2007 within theburned areas of larch forests along the river There are no

roads within the study area and rivers provide the best access(figure 1) Temporary test sites were established within burnedareas within about 10 km from the river and within a 160ndash420 m elevation range Disks of larch bole cross sections fortree ring analysis were cut at the root neck level Sampled treesincluded specimens of the dominant (even-age) trees as wellas older trees that survived stand-replacing fires

Trees within test sites 1ndash9 (figure 1) were sampled andused for the FRI analysis In addition to this older trees onsupplementary sites were sampled (figure 1) for the purpose ofestimation of long-term trends in fire frequency The sampleconsisted of 58 trees

32 Sample analysis

Tree ring widths were measured with a precision of 001 mmusing the well-known LINTAB-III instrument The dates offires were estimated on the basis of a master chronologyconstructed for northern larch forests (Naurzbaev et al 2004)The mean correlation with the master chronology was 054which is satisfactory for our purposes The COFECHA(Holmes 1983) and TSAP (Rinn 1996) programs were usedto detect double-counted and missing rings Fire-caused treering deletions were found only for three cases (out of about 75analyzed) Relevant dates of fire events were adjusted on thebasis of the master chronology

33 Fire return interval calculation

FRI were routinely estimated by tree ring calculation betweenconsecutive fire scars Di minus Diminus1 where Di and Diminus1 are thedates of fires i and i minus 1 Since many sampled trees have onlya single burn mark the dates of tree natality were included inthe FRI calculation where appropriate as described below

It is known that within larch dominated communities firesare mostly stand-replacing and promote the formation of even-aged stands Fresh burns with mineralized soil are quicklyoccupied by dense regeneration Over time the development ofa thick moss and lichen cover limits larch regeneration (Kharuket al 2008) Larch produce seeds annually with good harvestsoccurring on 5ndash7 yr cycles (Pleshikov and Novosibirsk 2002)Furthermore cones 2ndash3 yr old are known to be viable seedsources (Sofronov et al 1999) Importantly ground firesregularly do not damage cones leaving fire-killed stands as asource of seed Thus both tree natality dates and the dates offires were used for FRI calculations FRI was determined foreach individual tree within the test sites and then the mean FRIfor a given site was calculated

The long-term history was based on fire events were forthe period spanning the 17th century to 2007 To exclude theimpact of a decreasing sample size for older stands ie thelsquofading effectrsquo the sample size was adjusted by tree age withingroups Stand-replacing and non-stand-replacing fires wereinvestigated

Along the burn marks lsquotree ring width growthaccelerationsrsquo were also determined lsquoGrowth accelerationsrsquowere identified by the following procedure (1) Expert visualanalysis of increase in tree ring width (2) Calculation ofthe mean tree ring width 20 yr before and 20 yr after the

2

Environ Res Lett 6 (2011) 045208 V I Kharuk et al

Table 1 Mean FRI for nine study sites

Test sites Mean FRI Sample size (N)

1 134 52 203 23 138 94 165 25 349 56 278 177 142 48 256 39 131 4

Overall mean 200 plusmn 51 ( p 005)

increases began this reference period (20 yr) approximatelycorresponds to the period of post-fire tree ring incrementincreases (3) If the ratio (mean tree ring width after beginningof tree width increasemean tree ring width before treewidth increase) was gt20 the observed lsquogrowth accelerationrsquowas considered significant (4) The date of the lsquogrowthincreasersquo was compared with the air temperature record It isknown that for the boreal-forest zone radial growth has beenclosely connected with temperature variability (Esper et al2010) If that date coincided with air temperature increasethat particular lsquogrowth increment increasersquo was excludedfrom further analysis (because of the possibility of climate-driven tree ring increase) The above-mentioned lsquogrowthaccelerationrsquo dates were considered as a possible additionalindirect sign of wildfire impact It is known that trees thatsurvive fires (including those which do not have burn marks)experienced a period of growth increase (see eg Sofronovet al 1999)

34 Error analysis

Including the tree natality date into FRI analysis increasederrors due to the 0ndash5 yr lag of the post-fire tree establishmentThe natality date also has to be adjusted to the lsquostump agersquoie the difference of real and measured stump height tree ageEven if a tree was cut at the root neck level this differencecan be 2ndash5 yr In some cases disks sampled at the root necklevel were not suitable for analysis in these cases disks weresampled higher up the bole This procedure entailed about fiveadditional years of uncertainty The post-fire regeneration hada high growth increment the first 15ndash20 yr which graduallydecreased due to increasing competition and decreasing activeroot layer depth (Sofronov et al 1999) In summary themaximum total error was estimated to be 15 yr

4 Results

41 Fire return intervals

Dates of tree natality and fire events are presented in figure 2FRI values for different test sites vary considerably ie from131 to 349 yr (with mean of 200 plusmn 51 yr table 1)

42 Long-term fire event history

The long-term fire event history (for the 17th through 20thcenturies) was developed on the basis of data for trees with

Figure 2 Fire event chronology for the test sites 1ndash9 (the locationsare shown in figure 1) Light bars indicate tree natality dates heavybars indicate fire events and diamonds indicate dates of initiation ofradial growth acceleration

natality dates before 1800 1700 and 1600 yr respectively(figure 3 table 2) To exclude the fading effect data wereadjusted for tree natality dates gt200 yr gt300 yr and gt400 yrfor 19thndash20th 18thndash20th and 17thndash20th century comparisonsrespectively

The number of wildfires in the 20th century increasedrelative to that for the 19th century (13 versus 8) (table 2N = 58) For the 18th 19th and 20th centuries (N = 33) thefire numbers were 1 6 and 9 respectively (table 2 N = 33)Results for the period of 17thndash20th centuries also showed an

3

Environ Res Lett 6 (2011) 045208 V I Kharuk et al

Figure 3 The data set (N = 58) for comparison of numbers of fireevents for different time periods Light and heavy bars show dates ofnatality and fire events respectively Synchronous fire events werecounted as a single fire event The vertical lines along the abscissadenote all fire occurrences

Table 2 Number of wildfires during 17thndash20th centuries

Century Number of wildfires during a given century

20th 13 9 619th 8 6 218th mdash 1 017th mdash mdash 2

Sample size (N) 58 33 14

increase in fire events from the 17th to 20th centuries (table 2N = 14) The minimum number of wildfires coincided withthe air temperature decrease during the Little Ice Age period(ie early 17th to early 19th centuries figure 4)

Figure 4 Individual (gray lines) and running mean (with a 50 yrwindow dense solid line) radial increment data for sampled trees(N = 58) dates of fire events (upper scale) and air temperaturedeviations for northern Siberia (grey solid line from Naurzbaev et al2003)

5 Discussion

51 FRI

The majority of the fires in the larch communities that westudied were stand-replacing This was revealed by thefine-grained mosaic of (semi-)even-age stands within forestedlandscapes For the kinds of forest communities the meanFRI can only be reconstructed using age-class analysis orfire records (Sannikov and Goldammer 1996 Johnson andMiyanishi 2001) However in our case the presence of firescars indicated that non-stand-replacing fires also occurred(figure 2)

FRI data for different sites within the study area (figure 1)differed by more than a factor of 2 from 131 to 349 yr with amean of 200 plusmn 51 yr (table 1) The reason for this is that firesare rare events within the area investigated The other causeis the non-uniformity of the topography and land cover withinthe study area It is known that fire frequency is different forsunlit and shadowed slopes as well as for bog areas (Beatyand Taylor 2001 Rollins et al 2002 Kharuk et al 2005 2008)Meanwhile even the lowest FRI value (131 yr) considerablyexceeds reported data for adjacent southward forests (80ndash90 yrVaganov and Arbatskaya 1996 Kharuk et al 2005) MeanFRI values (200 plusmn 51 yr) exceed published FRI values for theboreal conifer forests (60ndash150 yr Payette 1992 Larsen 1997Swetnam 1996 Sannikov and Goldammer 1996) Very longFRIs (up to 300 yr) were reported for fire-protected forestsin Europe and North America (Weir et al 2000 Heyerdahlet al 2001 Bergeron et al 2004 Buechling and Baker 2004)Evidently this is not the case here because within our studyarea fires were never suppressed Low fire frequency is notfavorable for the larch forests because fires promote larchregeneration growth ie larch is a lsquopyrophyticrsquo species Themain tree growth constraints are permafrost thawing depth andsoil drainage The depth of the seasonal thawing is dependenton the exposure moss and lichen layer thickness and firehistory Fires not only increase the permafrost thawing depthbut also increase the soil drainage which is very important for

4

Environ Res Lett 6 (2011) 045208 V I Kharuk et al

larch growth With time an increase in the thermal insulatorlayer composed of the on-ground moss and lichen cover causedupward migration of the permafrost layer and compression ofthe active root zone within a progressively decreasing upperlayer Fires also thin regeneration decreasing within-speciescompetition and thus promoting tree growth because larch isan extremely shade-intolerant species

52 Long-term trends in wildfire number

The advanced age of some sampled trees (gt400 yr figure 3)allows estimation of the fire history within the study areaback to the Little Ice Age period Dendrochronology datashowed that cooling during the Little Ice Age period causeddepression in the tree annual radial growth (figure 4) Fireevent numbers since the 17th century approximately coincidedwith air temperature deviations increasing with warming sincethe second half of 19th century (figure 4 table 2) For examplefire numbers increased from 8 in the 19th century to 13 inthe 20th century This phenomenon could not be attributed todecreased samples of older trees ie a lsquofading effectrsquo sincesample sizes were adjusted by natality date

The local asynchrony of the radial increment growth andtemperature deviations in figure 4 could be attributed to thefire-induced increase of the radial growth as well as to theincrementation decrease during the lag between the fire eventand incrementation increases beginning (which is about 5ndash7 yr figure 5) Thus in the middle of 20th century wildfireswere observed within the majority of the test sites (figure 2)The other reason is a lsquodivergence phenomenonrsquo ie growth-versus-temperature divergence (DrsquoArrigo et al 2008 Esperet al 2010)

These estimates coincide with earlier reported data on firefrequency increase in the 20th versus 19th centuries for thesoutherly larch and mixed forests (sites 2 and 1 in figure 1respectively Kharuk et al 2008) The causes of this trendcan be both natural and anthropogenic Earlier it was shownthat for site 1 (figure 1) the FRI decrease was caused byboth natural (warming) and anthropogenic causes For site2 (figure 1) the FRI decrease was attributed to temperatureincrease mainly because the leading factor in fire ignition inthe remote northern forests is lightning (gt90 of cases in thenorthern forests whereas within southern forests gt80 of firesare anthropogenically caused Kovacs et al 2004) This is alsotrue for our study site because of the remoteness of the areaas well as the low population density in general most of thefires are of natural origin Similar observations (FRI increasesince the Little Ice Age) were made for the northern boundaryof the larch forests (site 3 in figure 1 Kharuk et al 2011)All these data support the suggestion that observed climatechange will lead to an increase in fire frequency (Gillett et al2004 Bergeron et al 2004 Girardin et al 2009) Comparisonof FRI along the meridian (figure 1) showed a northwardincrease of FRI 77 plusmn 20 yr at sim61N (site 1) 82 plusmn 7 at 64N(site 2) 200 plusmn 51 yr at about 66N and 320 plusmn 50 yr at thenorthern boundary of the larch dominant communities (71Nsite 3) The main reason for the northward FRI increase is lessincoming solar radiation and consequently shortening of thefire-danger period

Figure 5 Diagrams for (1) individual tree ring widths (N = 36) and(3) mean tree ring widths before and after fire and (2) tree ring widthfor a specimen with an extremely high post-fire growth increment(inset) Data were compiled on the basis of fires in the 20th century(figure 2) Dates of fires were set to the lsquozerorsquo point Note thatpost-fire growth increase has a lag of about 5ndash7 yr

53 Burns as a lsquosimulatorrsquo of future warming

Areas experiencing wildfires may be considered as a simulatorof predicted warming impacts on northern forests Indeedin addition to soil enrichment with nutrients and decreasedcompetition fires cause an increase of permafrost thawingdepth by a factor of 3ndash5 (Kharuk et al 2008) and increasedsoil drainage Trees that survived wildfires showed anapproximately double (193 p gt 095) increase of the radialincrement (figure 5) Comparison was made for the period for25 yr before and after the fire (periods of plusmn5 yr around zerowere deleted to exclude the effects of direct fire damage ontrees) Some trees showed an extremely high response to fireeffects For example the tree cross section shown in figure 5showed an increase of radial growth by about a factor of 10in comparison with background observations This tree wassampled at a latitude near the Polar Circle

Generally speaking growth increases following fire scarsshould be measured along radii very much distant from thewound itself But in our case we compare 37 specimenswith the same pattern of fire damage one of which showsoutstanding increment growth (about ten times that of thebackground set figure 5) The basic difference of thisspecimen from the others sampled was in the depth of soilthawing (about 15 m versus lt03ndash05 m for the otherspecimens) and good drainage since that tree was growing onthe southern river bank It is known that larch prefer drainedsoils (Schepaschenko et al 2008) The observed radial growthincrease (and consequently increased carbon sequestration)may be an alternative to the scenario of forests in climate-induced permafrost transformation areas becoming greenhousegas sources (IPCC 2007) Thus the vegetation dynamics andproductivity of the burned areas deserve future investigation

5

Environ Res Lett 6 (2011) 045208 V I Kharuk et al

54 Indirect signs of wildfires

The tree ring growth history showed periods of radial growthaccelerations (ie tree ring width increase figures 2ndash4)These increases are commonly considered to be climate driven(eg Shiyatov 2003) and may also contain information onfire events The observed growth accelerations may also becaused by the above-mentioned fire-caused soil meliorationDifferentiation of these effects could be carried out on the basisof the comparison with air temperature anomalies The fire-induced origin of the acceleration is supported by the fact thatin some cases the date of accelerations coincides with the burnmarks on the trees from the same test site (figure 2) Theseeffects deserve future investigation based on larger samplesizes

6 Conclusions

FRI in the study area is considerably longer than in southerlyterritories Along the 100 meridian (figure 1) FRI valuesincreased northward 77 plusmn 20 yr at sim61N (site 1) 82 plusmn 7 yr at64N (site 2) 200plusmn51 yr at about 66N and 320plusmn50 yr at thenorthern boundary of the larch stands (sim71N) The numberof fire events during the Little Ice Age period (17thndash18thcenturies) was approximately half the number observed in the19thndash20th centuries Fire-caused soil melioration (basicallysoil drainage and thawing depth increases) caused a radialgrowth increase to about twice the background value (up togt6 times at extremes) This effect may simulate the predictedimpact of warming on the larch growth in the permafrost zone

Acknowledgments

This work was supported by the Siberian Branch of theRussian Academy of Sciences Project No 2733 and NASAHQ Terrestrial Ecology Program

ReferencesBeaty R M and Taylor A H 2001 Spatial and temporal variation of

fire regimes in a mixed conifer forest landscape SouthernCascades California USA J Biogeogr 28 955ndash66

Bergeron Y Gauthier S Flannigan M and Kafka V 2004 Fireregimes at the transition between mixed wood and coniferousboreal forest in Northwestern Quebec Ecology 85 1916ndash32

Buechling A and Baker L 2004 A fire history from tree rings in ahigh-elevation forest of Rocky Mountain National Park Can JForest Res 34 1259ndash73

Climate Explorer 2011 httpclimexpknminlDrsquoArrigo R Wilson R Liepert B and Cherubini P 2008 On the

lsquoDivergence Problemrsquo in Northern Forests a review of thetree-ring evidence and possible causes Glob Planet Change60 289ndash305

Esper J Frank D Buentgen U Verstege A Hantemirov R andKirdyanov A 2010 Trends and uncertainties in Siberianindicators of 20th century warming Glob Change Biol16 386ndash98

Gillett N P Weaver A J Zwiers F W and Flannigan M D 2004Detecting the effect of climate change on Canadian forest firesGeophys Res Lett 31 L18211

Girardin M P Ali A A Carcaillet C Mudelsee M Drobyshev IHely C and Bergeron Y 2009 Heterogeneous response ofcircumboreal wildfire risk to climate change since the early1900s Glob Change Biol 15 2751ndash69

Heyerdahl E K Beubaker L B and Agee J K 2001 Spatial controls ofhistorical fire regimes a multiscale example from the interiorwest USA Ecology 82 660ndash78

Holmes R L 1983 Computer-assisted quality control in tree-ringdating and measurement Tree-Ring Bull 43 69ndash78

IPCC 2007 Climate Change 2007 Synthesis Report Contribution ofWorking Groups I II and III to the Fourth Assessment Report ofthe Intergovernmental Panel on Climate Changeed Core Writing Team R K Pachauri and A Reisinger (GenevaIPCC)

Johnson E A and Miyanishi K 2001 Forest FiresmdashBehavior andEcological Effects (San Diego CA Academic)

Kharuk V I Dvinskaya M L and Ranson K J 2011 FRI within theNorthern Boundary of the Larch Forest Int J Wildland Firesin review

Kharuk V I Dvinskaya M L Ranson K J and Im S T 2005 Expansionof evergreen conifers to the larch-dominated zone and climatictrends Russ J Ecol 36 164ndash70

Kharuk V I Kasischke E S and Yakubailik O E 2007 The spatial andtemporal distribution of fires on Sakhalin Island Russia Int JWildland Fire 16 556ndash62

Kharuk V I Ranson K J and Dvinskaya M L 2008 Wildfires dynamicin the larch dominance zone Geophys Res Lett 35 L01402

Kovacs K Ranson K J Sun G and Kharuk V I 2004 The relationshipof the Terra MODIS fire product and anthropogenic features inthe Central Siberian landscape Earth Interact 8 (available athttpearthinteractionsorg)

Larsen C P S 1997 Spatial and temporal variations in boreal forestfire frequency in northern Alberta J Biogeogr 24 663ndash73

Naurzbaev M M Hughes M K and Vaganov E A 2004 Tree-ringgrowth as sources of climatic information Quat Res 62 126ndash33

Naurzbaev M M Vaganov E A and Sidorova O V 2003 Thevariability of on-ground air temperature in northern Eurasiabased on millennium tree ring chronology Earth Criosphere 7(2) 84ndash91 (in Russian)

Payette S 1992 Fire as a controlling process in the North Americanboreal forest A Systems Analysis of the Boreal Forested H H Shugart R Leemans and G B Bonan (CambridgeCambridge University Press) pp 144ndash69

Pleshikov F I (ed) 2002 Forest Ecosystems of the Yenisei Meridian(Novosibirsk Novosibirsk Publ House of SB RAS) p 356(in Russian)

Rinn F 1996 Tsap V 36 Reference Manual Computer Program forTree-ring Analysis and Presentation (RINNTECH Hardstr20ndash22 D-69124 Heidelberg Germany)

Rollins M G Morgan P and Swetnam T 2002 Landscape-scalecontrols over 20th century fire occurrence in two large RockyMountain (USA) wilderness areas Landscape Ecol 17 539ndash57

Sannikov S N and Goldammer J G 1996 Fire ecology of pine forestsof Northern Eurasia Fire in Ecosystems of Boreal Eurasia vol48 ed J G Goldammer and V V Furyaev (Dordrecht KluwerAcademic) pp 151ndash67

Schepaschenko D G Shvidenko A Z and Shalaev V S 2008Biological Productivity and Carbon Budget of Larch Forests ofNorthern-East Russia (Moscow Moscow State ForestUniversity) p 296 (in Russian)

Shiyatov S G 2003 Rates of change in the upper treeline ecotone inthe Polar Ural Mountains Pages News 11 8ndash10

Shvidenko A Schepaschenko D McCallum I and Nilsson S 2007Russian Forests and Forestry (Laxenburg International Institutefor Applied Systems Analysis and the Russian Academy ofScience) (CD-ROM)

Sofronov M A Volokitina A V and Kajimoto T 1999 Ecology ofwildland fires and permafrost their interdependence in theNorthern part of Siberia Proc 8th Symp on the Joint SiberianPermafrost Studies Between Japan and Russia in 1999 pp 211ndash8

Swetnam T W 1996 Fire and climate history in the central YeniseyRegion Siberia Fire in Ecosystems of Boreal Eurasiaed J G Goldammer and V V Furyaev (Dordrecht KluwerAcademic) pp 90ndash104

Vaganov E A and Arbatskaya M K 1996 Climate history and wildfirefrequency in the central part of Krasnoyarsky krai I Climaticconditions in growing season and seasonal wildfires distributionSiberian J Ecol 3 9ndash18

Wallenius T Larjavaara M Heikkinen J and Shibistova O 2011Declining fires in Larix-dominated forests in northern Irkutskdistrict Int J Wildland Fire 20 248ndash54

Weir J M H Johnson E A and Miyanishi K 2000 Fire frequency andthe spatial age mosaic of the mixed-wood boreal forest inwestern Canada Ecol Appl 10 1162ndash77

6

Environ Res Lett 6 (2011) 045208 V I Kharuk et al

Table 1 Mean FRI for nine study sites

Test sites Mean FRI Sample size (N)

1 134 52 203 23 138 94 165 25 349 56 278 177 142 48 256 39 131 4

Overall mean 200 plusmn 51 ( p 005)

increases began this reference period (20 yr) approximatelycorresponds to the period of post-fire tree ring incrementincreases (3) If the ratio (mean tree ring width after beginningof tree width increasemean tree ring width before treewidth increase) was gt20 the observed lsquogrowth accelerationrsquowas considered significant (4) The date of the lsquogrowthincreasersquo was compared with the air temperature record It isknown that for the boreal-forest zone radial growth has beenclosely connected with temperature variability (Esper et al2010) If that date coincided with air temperature increasethat particular lsquogrowth increment increasersquo was excludedfrom further analysis (because of the possibility of climate-driven tree ring increase) The above-mentioned lsquogrowthaccelerationrsquo dates were considered as a possible additionalindirect sign of wildfire impact It is known that trees thatsurvive fires (including those which do not have burn marks)experienced a period of growth increase (see eg Sofronovet al 1999)

34 Error analysis

Including the tree natality date into FRI analysis increasederrors due to the 0ndash5 yr lag of the post-fire tree establishmentThe natality date also has to be adjusted to the lsquostump agersquoie the difference of real and measured stump height tree ageEven if a tree was cut at the root neck level this differencecan be 2ndash5 yr In some cases disks sampled at the root necklevel were not suitable for analysis in these cases disks weresampled higher up the bole This procedure entailed about fiveadditional years of uncertainty The post-fire regeneration hada high growth increment the first 15ndash20 yr which graduallydecreased due to increasing competition and decreasing activeroot layer depth (Sofronov et al 1999) In summary themaximum total error was estimated to be 15 yr

4 Results

41 Fire return intervals

Dates of tree natality and fire events are presented in figure 2FRI values for different test sites vary considerably ie from131 to 349 yr (with mean of 200 plusmn 51 yr table 1)

42 Long-term fire event history

The long-term fire event history (for the 17th through 20thcenturies) was developed on the basis of data for trees with

Figure 2 Fire event chronology for the test sites 1ndash9 (the locationsare shown in figure 1) Light bars indicate tree natality dates heavybars indicate fire events and diamonds indicate dates of initiation ofradial growth acceleration

natality dates before 1800 1700 and 1600 yr respectively(figure 3 table 2) To exclude the fading effect data wereadjusted for tree natality dates gt200 yr gt300 yr and gt400 yrfor 19thndash20th 18thndash20th and 17thndash20th century comparisonsrespectively

The number of wildfires in the 20th century increasedrelative to that for the 19th century (13 versus 8) (table 2N = 58) For the 18th 19th and 20th centuries (N = 33) thefire numbers were 1 6 and 9 respectively (table 2 N = 33)Results for the period of 17thndash20th centuries also showed an

3

Environ Res Lett 6 (2011) 045208 V I Kharuk et al

Figure 3 The data set (N = 58) for comparison of numbers of fireevents for different time periods Light and heavy bars show dates ofnatality and fire events respectively Synchronous fire events werecounted as a single fire event The vertical lines along the abscissadenote all fire occurrences

Table 2 Number of wildfires during 17thndash20th centuries

Century Number of wildfires during a given century

20th 13 9 619th 8 6 218th mdash 1 017th mdash mdash 2

Sample size (N) 58 33 14

increase in fire events from the 17th to 20th centuries (table 2N = 14) The minimum number of wildfires coincided withthe air temperature decrease during the Little Ice Age period(ie early 17th to early 19th centuries figure 4)

Figure 4 Individual (gray lines) and running mean (with a 50 yrwindow dense solid line) radial increment data for sampled trees(N = 58) dates of fire events (upper scale) and air temperaturedeviations for northern Siberia (grey solid line from Naurzbaev et al2003)

5 Discussion

51 FRI

The majority of the fires in the larch communities that westudied were stand-replacing This was revealed by thefine-grained mosaic of (semi-)even-age stands within forestedlandscapes For the kinds of forest communities the meanFRI can only be reconstructed using age-class analysis orfire records (Sannikov and Goldammer 1996 Johnson andMiyanishi 2001) However in our case the presence of firescars indicated that non-stand-replacing fires also occurred(figure 2)

FRI data for different sites within the study area (figure 1)differed by more than a factor of 2 from 131 to 349 yr with amean of 200 plusmn 51 yr (table 1) The reason for this is that firesare rare events within the area investigated The other causeis the non-uniformity of the topography and land cover withinthe study area It is known that fire frequency is different forsunlit and shadowed slopes as well as for bog areas (Beatyand Taylor 2001 Rollins et al 2002 Kharuk et al 2005 2008)Meanwhile even the lowest FRI value (131 yr) considerablyexceeds reported data for adjacent southward forests (80ndash90 yrVaganov and Arbatskaya 1996 Kharuk et al 2005) MeanFRI values (200 plusmn 51 yr) exceed published FRI values for theboreal conifer forests (60ndash150 yr Payette 1992 Larsen 1997Swetnam 1996 Sannikov and Goldammer 1996) Very longFRIs (up to 300 yr) were reported for fire-protected forestsin Europe and North America (Weir et al 2000 Heyerdahlet al 2001 Bergeron et al 2004 Buechling and Baker 2004)Evidently this is not the case here because within our studyarea fires were never suppressed Low fire frequency is notfavorable for the larch forests because fires promote larchregeneration growth ie larch is a lsquopyrophyticrsquo species Themain tree growth constraints are permafrost thawing depth andsoil drainage The depth of the seasonal thawing is dependenton the exposure moss and lichen layer thickness and firehistory Fires not only increase the permafrost thawing depthbut also increase the soil drainage which is very important for

4

Environ Res Lett 6 (2011) 045208 V I Kharuk et al

larch growth With time an increase in the thermal insulatorlayer composed of the on-ground moss and lichen cover causedupward migration of the permafrost layer and compression ofthe active root zone within a progressively decreasing upperlayer Fires also thin regeneration decreasing within-speciescompetition and thus promoting tree growth because larch isan extremely shade-intolerant species

52 Long-term trends in wildfire number

The advanced age of some sampled trees (gt400 yr figure 3)allows estimation of the fire history within the study areaback to the Little Ice Age period Dendrochronology datashowed that cooling during the Little Ice Age period causeddepression in the tree annual radial growth (figure 4) Fireevent numbers since the 17th century approximately coincidedwith air temperature deviations increasing with warming sincethe second half of 19th century (figure 4 table 2) For examplefire numbers increased from 8 in the 19th century to 13 inthe 20th century This phenomenon could not be attributed todecreased samples of older trees ie a lsquofading effectrsquo sincesample sizes were adjusted by natality date

The local asynchrony of the radial increment growth andtemperature deviations in figure 4 could be attributed to thefire-induced increase of the radial growth as well as to theincrementation decrease during the lag between the fire eventand incrementation increases beginning (which is about 5ndash7 yr figure 5) Thus in the middle of 20th century wildfireswere observed within the majority of the test sites (figure 2)The other reason is a lsquodivergence phenomenonrsquo ie growth-versus-temperature divergence (DrsquoArrigo et al 2008 Esperet al 2010)

These estimates coincide with earlier reported data on firefrequency increase in the 20th versus 19th centuries for thesoutherly larch and mixed forests (sites 2 and 1 in figure 1respectively Kharuk et al 2008) The causes of this trendcan be both natural and anthropogenic Earlier it was shownthat for site 1 (figure 1) the FRI decrease was caused byboth natural (warming) and anthropogenic causes For site2 (figure 1) the FRI decrease was attributed to temperatureincrease mainly because the leading factor in fire ignition inthe remote northern forests is lightning (gt90 of cases in thenorthern forests whereas within southern forests gt80 of firesare anthropogenically caused Kovacs et al 2004) This is alsotrue for our study site because of the remoteness of the areaas well as the low population density in general most of thefires are of natural origin Similar observations (FRI increasesince the Little Ice Age) were made for the northern boundaryof the larch forests (site 3 in figure 1 Kharuk et al 2011)All these data support the suggestion that observed climatechange will lead to an increase in fire frequency (Gillett et al2004 Bergeron et al 2004 Girardin et al 2009) Comparisonof FRI along the meridian (figure 1) showed a northwardincrease of FRI 77 plusmn 20 yr at sim61N (site 1) 82 plusmn 7 at 64N(site 2) 200 plusmn 51 yr at about 66N and 320 plusmn 50 yr at thenorthern boundary of the larch dominant communities (71Nsite 3) The main reason for the northward FRI increase is lessincoming solar radiation and consequently shortening of thefire-danger period

Figure 5 Diagrams for (1) individual tree ring widths (N = 36) and(3) mean tree ring widths before and after fire and (2) tree ring widthfor a specimen with an extremely high post-fire growth increment(inset) Data were compiled on the basis of fires in the 20th century(figure 2) Dates of fires were set to the lsquozerorsquo point Note thatpost-fire growth increase has a lag of about 5ndash7 yr

53 Burns as a lsquosimulatorrsquo of future warming

Areas experiencing wildfires may be considered as a simulatorof predicted warming impacts on northern forests Indeedin addition to soil enrichment with nutrients and decreasedcompetition fires cause an increase of permafrost thawingdepth by a factor of 3ndash5 (Kharuk et al 2008) and increasedsoil drainage Trees that survived wildfires showed anapproximately double (193 p gt 095) increase of the radialincrement (figure 5) Comparison was made for the period for25 yr before and after the fire (periods of plusmn5 yr around zerowere deleted to exclude the effects of direct fire damage ontrees) Some trees showed an extremely high response to fireeffects For example the tree cross section shown in figure 5showed an increase of radial growth by about a factor of 10in comparison with background observations This tree wassampled at a latitude near the Polar Circle

Generally speaking growth increases following fire scarsshould be measured along radii very much distant from thewound itself But in our case we compare 37 specimenswith the same pattern of fire damage one of which showsoutstanding increment growth (about ten times that of thebackground set figure 5) The basic difference of thisspecimen from the others sampled was in the depth of soilthawing (about 15 m versus lt03ndash05 m for the otherspecimens) and good drainage since that tree was growing onthe southern river bank It is known that larch prefer drainedsoils (Schepaschenko et al 2008) The observed radial growthincrease (and consequently increased carbon sequestration)may be an alternative to the scenario of forests in climate-induced permafrost transformation areas becoming greenhousegas sources (IPCC 2007) Thus the vegetation dynamics andproductivity of the burned areas deserve future investigation

5

Environ Res Lett 6 (2011) 045208 V I Kharuk et al

54 Indirect signs of wildfires

The tree ring growth history showed periods of radial growthaccelerations (ie tree ring width increase figures 2ndash4)These increases are commonly considered to be climate driven(eg Shiyatov 2003) and may also contain information onfire events The observed growth accelerations may also becaused by the above-mentioned fire-caused soil meliorationDifferentiation of these effects could be carried out on the basisof the comparison with air temperature anomalies The fire-induced origin of the acceleration is supported by the fact thatin some cases the date of accelerations coincides with the burnmarks on the trees from the same test site (figure 2) Theseeffects deserve future investigation based on larger samplesizes

6 Conclusions

FRI in the study area is considerably longer than in southerlyterritories Along the 100 meridian (figure 1) FRI valuesincreased northward 77 plusmn 20 yr at sim61N (site 1) 82 plusmn 7 yr at64N (site 2) 200plusmn51 yr at about 66N and 320plusmn50 yr at thenorthern boundary of the larch stands (sim71N) The numberof fire events during the Little Ice Age period (17thndash18thcenturies) was approximately half the number observed in the19thndash20th centuries Fire-caused soil melioration (basicallysoil drainage and thawing depth increases) caused a radialgrowth increase to about twice the background value (up togt6 times at extremes) This effect may simulate the predictedimpact of warming on the larch growth in the permafrost zone

Acknowledgments

This work was supported by the Siberian Branch of theRussian Academy of Sciences Project No 2733 and NASAHQ Terrestrial Ecology Program

ReferencesBeaty R M and Taylor A H 2001 Spatial and temporal variation of

fire regimes in a mixed conifer forest landscape SouthernCascades California USA J Biogeogr 28 955ndash66

Bergeron Y Gauthier S Flannigan M and Kafka V 2004 Fireregimes at the transition between mixed wood and coniferousboreal forest in Northwestern Quebec Ecology 85 1916ndash32

Buechling A and Baker L 2004 A fire history from tree rings in ahigh-elevation forest of Rocky Mountain National Park Can JForest Res 34 1259ndash73

Climate Explorer 2011 httpclimexpknminlDrsquoArrigo R Wilson R Liepert B and Cherubini P 2008 On the

lsquoDivergence Problemrsquo in Northern Forests a review of thetree-ring evidence and possible causes Glob Planet Change60 289ndash305

Esper J Frank D Buentgen U Verstege A Hantemirov R andKirdyanov A 2010 Trends and uncertainties in Siberianindicators of 20th century warming Glob Change Biol16 386ndash98

Gillett N P Weaver A J Zwiers F W and Flannigan M D 2004Detecting the effect of climate change on Canadian forest firesGeophys Res Lett 31 L18211

Girardin M P Ali A A Carcaillet C Mudelsee M Drobyshev IHely C and Bergeron Y 2009 Heterogeneous response ofcircumboreal wildfire risk to climate change since the early1900s Glob Change Biol 15 2751ndash69

Heyerdahl E K Beubaker L B and Agee J K 2001 Spatial controls ofhistorical fire regimes a multiscale example from the interiorwest USA Ecology 82 660ndash78

Holmes R L 1983 Computer-assisted quality control in tree-ringdating and measurement Tree-Ring Bull 43 69ndash78

IPCC 2007 Climate Change 2007 Synthesis Report Contribution ofWorking Groups I II and III to the Fourth Assessment Report ofthe Intergovernmental Panel on Climate Changeed Core Writing Team R K Pachauri and A Reisinger (GenevaIPCC)

Johnson E A and Miyanishi K 2001 Forest FiresmdashBehavior andEcological Effects (San Diego CA Academic)

Kharuk V I Dvinskaya M L and Ranson K J 2011 FRI within theNorthern Boundary of the Larch Forest Int J Wildland Firesin review

Kharuk V I Dvinskaya M L Ranson K J and Im S T 2005 Expansionof evergreen conifers to the larch-dominated zone and climatictrends Russ J Ecol 36 164ndash70

Kharuk V I Kasischke E S and Yakubailik O E 2007 The spatial andtemporal distribution of fires on Sakhalin Island Russia Int JWildland Fire 16 556ndash62

Kharuk V I Ranson K J and Dvinskaya M L 2008 Wildfires dynamicin the larch dominance zone Geophys Res Lett 35 L01402

Kovacs K Ranson K J Sun G and Kharuk V I 2004 The relationshipof the Terra MODIS fire product and anthropogenic features inthe Central Siberian landscape Earth Interact 8 (available athttpearthinteractionsorg)

Larsen C P S 1997 Spatial and temporal variations in boreal forestfire frequency in northern Alberta J Biogeogr 24 663ndash73

Naurzbaev M M Hughes M K and Vaganov E A 2004 Tree-ringgrowth as sources of climatic information Quat Res 62 126ndash33

Naurzbaev M M Vaganov E A and Sidorova O V 2003 Thevariability of on-ground air temperature in northern Eurasiabased on millennium tree ring chronology Earth Criosphere 7(2) 84ndash91 (in Russian)

Payette S 1992 Fire as a controlling process in the North Americanboreal forest A Systems Analysis of the Boreal Forested H H Shugart R Leemans and G B Bonan (CambridgeCambridge University Press) pp 144ndash69

Pleshikov F I (ed) 2002 Forest Ecosystems of the Yenisei Meridian(Novosibirsk Novosibirsk Publ House of SB RAS) p 356(in Russian)

Rinn F 1996 Tsap V 36 Reference Manual Computer Program forTree-ring Analysis and Presentation (RINNTECH Hardstr20ndash22 D-69124 Heidelberg Germany)

Rollins M G Morgan P and Swetnam T 2002 Landscape-scalecontrols over 20th century fire occurrence in two large RockyMountain (USA) wilderness areas Landscape Ecol 17 539ndash57

Sannikov S N and Goldammer J G 1996 Fire ecology of pine forestsof Northern Eurasia Fire in Ecosystems of Boreal Eurasia vol48 ed J G Goldammer and V V Furyaev (Dordrecht KluwerAcademic) pp 151ndash67

Schepaschenko D G Shvidenko A Z and Shalaev V S 2008Biological Productivity and Carbon Budget of Larch Forests ofNorthern-East Russia (Moscow Moscow State ForestUniversity) p 296 (in Russian)

Shiyatov S G 2003 Rates of change in the upper treeline ecotone inthe Polar Ural Mountains Pages News 11 8ndash10

Shvidenko A Schepaschenko D McCallum I and Nilsson S 2007Russian Forests and Forestry (Laxenburg International Institutefor Applied Systems Analysis and the Russian Academy ofScience) (CD-ROM)

Sofronov M A Volokitina A V and Kajimoto T 1999 Ecology ofwildland fires and permafrost their interdependence in theNorthern part of Siberia Proc 8th Symp on the Joint SiberianPermafrost Studies Between Japan and Russia in 1999 pp 211ndash8

Swetnam T W 1996 Fire and climate history in the central YeniseyRegion Siberia Fire in Ecosystems of Boreal Eurasiaed J G Goldammer and V V Furyaev (Dordrecht KluwerAcademic) pp 90ndash104

Vaganov E A and Arbatskaya M K 1996 Climate history and wildfirefrequency in the central part of Krasnoyarsky krai I Climaticconditions in growing season and seasonal wildfires distributionSiberian J Ecol 3 9ndash18

Wallenius T Larjavaara M Heikkinen J and Shibistova O 2011Declining fires in Larix-dominated forests in northern Irkutskdistrict Int J Wildland Fire 20 248ndash54

Weir J M H Johnson E A and Miyanishi K 2000 Fire frequency andthe spatial age mosaic of the mixed-wood boreal forest inwestern Canada Ecol Appl 10 1162ndash77

6

Environ Res Lett 6 (2011) 045208 V I Kharuk et al

Figure 3 The data set (N = 58) for comparison of numbers of fireevents for different time periods Light and heavy bars show dates ofnatality and fire events respectively Synchronous fire events werecounted as a single fire event The vertical lines along the abscissadenote all fire occurrences

Table 2 Number of wildfires during 17thndash20th centuries

Century Number of wildfires during a given century

20th 13 9 619th 8 6 218th mdash 1 017th mdash mdash 2

Sample size (N) 58 33 14

increase in fire events from the 17th to 20th centuries (table 2N = 14) The minimum number of wildfires coincided withthe air temperature decrease during the Little Ice Age period(ie early 17th to early 19th centuries figure 4)

Figure 4 Individual (gray lines) and running mean (with a 50 yrwindow dense solid line) radial increment data for sampled trees(N = 58) dates of fire events (upper scale) and air temperaturedeviations for northern Siberia (grey solid line from Naurzbaev et al2003)

5 Discussion

51 FRI

The majority of the fires in the larch communities that westudied were stand-replacing This was revealed by thefine-grained mosaic of (semi-)even-age stands within forestedlandscapes For the kinds of forest communities the meanFRI can only be reconstructed using age-class analysis orfire records (Sannikov and Goldammer 1996 Johnson andMiyanishi 2001) However in our case the presence of firescars indicated that non-stand-replacing fires also occurred(figure 2)

FRI data for different sites within the study area (figure 1)differed by more than a factor of 2 from 131 to 349 yr with amean of 200 plusmn 51 yr (table 1) The reason for this is that firesare rare events within the area investigated The other causeis the non-uniformity of the topography and land cover withinthe study area It is known that fire frequency is different forsunlit and shadowed slopes as well as for bog areas (Beatyand Taylor 2001 Rollins et al 2002 Kharuk et al 2005 2008)Meanwhile even the lowest FRI value (131 yr) considerablyexceeds reported data for adjacent southward forests (80ndash90 yrVaganov and Arbatskaya 1996 Kharuk et al 2005) MeanFRI values (200 plusmn 51 yr) exceed published FRI values for theboreal conifer forests (60ndash150 yr Payette 1992 Larsen 1997Swetnam 1996 Sannikov and Goldammer 1996) Very longFRIs (up to 300 yr) were reported for fire-protected forestsin Europe and North America (Weir et al 2000 Heyerdahlet al 2001 Bergeron et al 2004 Buechling and Baker 2004)Evidently this is not the case here because within our studyarea fires were never suppressed Low fire frequency is notfavorable for the larch forests because fires promote larchregeneration growth ie larch is a lsquopyrophyticrsquo species Themain tree growth constraints are permafrost thawing depth andsoil drainage The depth of the seasonal thawing is dependenton the exposure moss and lichen layer thickness and firehistory Fires not only increase the permafrost thawing depthbut also increase the soil drainage which is very important for

4

Environ Res Lett 6 (2011) 045208 V I Kharuk et al

larch growth With time an increase in the thermal insulatorlayer composed of the on-ground moss and lichen cover causedupward migration of the permafrost layer and compression ofthe active root zone within a progressively decreasing upperlayer Fires also thin regeneration decreasing within-speciescompetition and thus promoting tree growth because larch isan extremely shade-intolerant species

52 Long-term trends in wildfire number

The advanced age of some sampled trees (gt400 yr figure 3)allows estimation of the fire history within the study areaback to the Little Ice Age period Dendrochronology datashowed that cooling during the Little Ice Age period causeddepression in the tree annual radial growth (figure 4) Fireevent numbers since the 17th century approximately coincidedwith air temperature deviations increasing with warming sincethe second half of 19th century (figure 4 table 2) For examplefire numbers increased from 8 in the 19th century to 13 inthe 20th century This phenomenon could not be attributed todecreased samples of older trees ie a lsquofading effectrsquo sincesample sizes were adjusted by natality date

The local asynchrony of the radial increment growth andtemperature deviations in figure 4 could be attributed to thefire-induced increase of the radial growth as well as to theincrementation decrease during the lag between the fire eventand incrementation increases beginning (which is about 5ndash7 yr figure 5) Thus in the middle of 20th century wildfireswere observed within the majority of the test sites (figure 2)The other reason is a lsquodivergence phenomenonrsquo ie growth-versus-temperature divergence (DrsquoArrigo et al 2008 Esperet al 2010)

These estimates coincide with earlier reported data on firefrequency increase in the 20th versus 19th centuries for thesoutherly larch and mixed forests (sites 2 and 1 in figure 1respectively Kharuk et al 2008) The causes of this trendcan be both natural and anthropogenic Earlier it was shownthat for site 1 (figure 1) the FRI decrease was caused byboth natural (warming) and anthropogenic causes For site2 (figure 1) the FRI decrease was attributed to temperatureincrease mainly because the leading factor in fire ignition inthe remote northern forests is lightning (gt90 of cases in thenorthern forests whereas within southern forests gt80 of firesare anthropogenically caused Kovacs et al 2004) This is alsotrue for our study site because of the remoteness of the areaas well as the low population density in general most of thefires are of natural origin Similar observations (FRI increasesince the Little Ice Age) were made for the northern boundaryof the larch forests (site 3 in figure 1 Kharuk et al 2011)All these data support the suggestion that observed climatechange will lead to an increase in fire frequency (Gillett et al2004 Bergeron et al 2004 Girardin et al 2009) Comparisonof FRI along the meridian (figure 1) showed a northwardincrease of FRI 77 plusmn 20 yr at sim61N (site 1) 82 plusmn 7 at 64N(site 2) 200 plusmn 51 yr at about 66N and 320 plusmn 50 yr at thenorthern boundary of the larch dominant communities (71Nsite 3) The main reason for the northward FRI increase is lessincoming solar radiation and consequently shortening of thefire-danger period

Figure 5 Diagrams for (1) individual tree ring widths (N = 36) and(3) mean tree ring widths before and after fire and (2) tree ring widthfor a specimen with an extremely high post-fire growth increment(inset) Data were compiled on the basis of fires in the 20th century(figure 2) Dates of fires were set to the lsquozerorsquo point Note thatpost-fire growth increase has a lag of about 5ndash7 yr

53 Burns as a lsquosimulatorrsquo of future warming

Areas experiencing wildfires may be considered as a simulatorof predicted warming impacts on northern forests Indeedin addition to soil enrichment with nutrients and decreasedcompetition fires cause an increase of permafrost thawingdepth by a factor of 3ndash5 (Kharuk et al 2008) and increasedsoil drainage Trees that survived wildfires showed anapproximately double (193 p gt 095) increase of the radialincrement (figure 5) Comparison was made for the period for25 yr before and after the fire (periods of plusmn5 yr around zerowere deleted to exclude the effects of direct fire damage ontrees) Some trees showed an extremely high response to fireeffects For example the tree cross section shown in figure 5showed an increase of radial growth by about a factor of 10in comparison with background observations This tree wassampled at a latitude near the Polar Circle

Generally speaking growth increases following fire scarsshould be measured along radii very much distant from thewound itself But in our case we compare 37 specimenswith the same pattern of fire damage one of which showsoutstanding increment growth (about ten times that of thebackground set figure 5) The basic difference of thisspecimen from the others sampled was in the depth of soilthawing (about 15 m versus lt03ndash05 m for the otherspecimens) and good drainage since that tree was growing onthe southern river bank It is known that larch prefer drainedsoils (Schepaschenko et al 2008) The observed radial growthincrease (and consequently increased carbon sequestration)may be an alternative to the scenario of forests in climate-induced permafrost transformation areas becoming greenhousegas sources (IPCC 2007) Thus the vegetation dynamics andproductivity of the burned areas deserve future investigation

5

Environ Res Lett 6 (2011) 045208 V I Kharuk et al

54 Indirect signs of wildfires

The tree ring growth history showed periods of radial growthaccelerations (ie tree ring width increase figures 2ndash4)These increases are commonly considered to be climate driven(eg Shiyatov 2003) and may also contain information onfire events The observed growth accelerations may also becaused by the above-mentioned fire-caused soil meliorationDifferentiation of these effects could be carried out on the basisof the comparison with air temperature anomalies The fire-induced origin of the acceleration is supported by the fact thatin some cases the date of accelerations coincides with the burnmarks on the trees from the same test site (figure 2) Theseeffects deserve future investigation based on larger samplesizes

6 Conclusions

FRI in the study area is considerably longer than in southerlyterritories Along the 100 meridian (figure 1) FRI valuesincreased northward 77 plusmn 20 yr at sim61N (site 1) 82 plusmn 7 yr at64N (site 2) 200plusmn51 yr at about 66N and 320plusmn50 yr at thenorthern boundary of the larch stands (sim71N) The numberof fire events during the Little Ice Age period (17thndash18thcenturies) was approximately half the number observed in the19thndash20th centuries Fire-caused soil melioration (basicallysoil drainage and thawing depth increases) caused a radialgrowth increase to about twice the background value (up togt6 times at extremes) This effect may simulate the predictedimpact of warming on the larch growth in the permafrost zone

Acknowledgments

This work was supported by the Siberian Branch of theRussian Academy of Sciences Project No 2733 and NASAHQ Terrestrial Ecology Program

ReferencesBeaty R M and Taylor A H 2001 Spatial and temporal variation of

fire regimes in a mixed conifer forest landscape SouthernCascades California USA J Biogeogr 28 955ndash66

Bergeron Y Gauthier S Flannigan M and Kafka V 2004 Fireregimes at the transition between mixed wood and coniferousboreal forest in Northwestern Quebec Ecology 85 1916ndash32

Buechling A and Baker L 2004 A fire history from tree rings in ahigh-elevation forest of Rocky Mountain National Park Can JForest Res 34 1259ndash73

Climate Explorer 2011 httpclimexpknminlDrsquoArrigo R Wilson R Liepert B and Cherubini P 2008 On the

lsquoDivergence Problemrsquo in Northern Forests a review of thetree-ring evidence and possible causes Glob Planet Change60 289ndash305

Esper J Frank D Buentgen U Verstege A Hantemirov R andKirdyanov A 2010 Trends and uncertainties in Siberianindicators of 20th century warming Glob Change Biol16 386ndash98

Gillett N P Weaver A J Zwiers F W and Flannigan M D 2004Detecting the effect of climate change on Canadian forest firesGeophys Res Lett 31 L18211

Girardin M P Ali A A Carcaillet C Mudelsee M Drobyshev IHely C and Bergeron Y 2009 Heterogeneous response ofcircumboreal wildfire risk to climate change since the early1900s Glob Change Biol 15 2751ndash69

Heyerdahl E K Beubaker L B and Agee J K 2001 Spatial controls ofhistorical fire regimes a multiscale example from the interiorwest USA Ecology 82 660ndash78

Holmes R L 1983 Computer-assisted quality control in tree-ringdating and measurement Tree-Ring Bull 43 69ndash78

IPCC 2007 Climate Change 2007 Synthesis Report Contribution ofWorking Groups I II and III to the Fourth Assessment Report ofthe Intergovernmental Panel on Climate Changeed Core Writing Team R K Pachauri and A Reisinger (GenevaIPCC)

Johnson E A and Miyanishi K 2001 Forest FiresmdashBehavior andEcological Effects (San Diego CA Academic)

Kharuk V I Dvinskaya M L and Ranson K J 2011 FRI within theNorthern Boundary of the Larch Forest Int J Wildland Firesin review

Kharuk V I Dvinskaya M L Ranson K J and Im S T 2005 Expansionof evergreen conifers to the larch-dominated zone and climatictrends Russ J Ecol 36 164ndash70

Kharuk V I Kasischke E S and Yakubailik O E 2007 The spatial andtemporal distribution of fires on Sakhalin Island Russia Int JWildland Fire 16 556ndash62

Kharuk V I Ranson K J and Dvinskaya M L 2008 Wildfires dynamicin the larch dominance zone Geophys Res Lett 35 L01402

Kovacs K Ranson K J Sun G and Kharuk V I 2004 The relationshipof the Terra MODIS fire product and anthropogenic features inthe Central Siberian landscape Earth Interact 8 (available athttpearthinteractionsorg)

Larsen C P S 1997 Spatial and temporal variations in boreal forestfire frequency in northern Alberta J Biogeogr 24 663ndash73

Naurzbaev M M Hughes M K and Vaganov E A 2004 Tree-ringgrowth as sources of climatic information Quat Res 62 126ndash33

Naurzbaev M M Vaganov E A and Sidorova O V 2003 Thevariability of on-ground air temperature in northern Eurasiabased on millennium tree ring chronology Earth Criosphere 7(2) 84ndash91 (in Russian)

Payette S 1992 Fire as a controlling process in the North Americanboreal forest A Systems Analysis of the Boreal Forested H H Shugart R Leemans and G B Bonan (CambridgeCambridge University Press) pp 144ndash69

Pleshikov F I (ed) 2002 Forest Ecosystems of the Yenisei Meridian(Novosibirsk Novosibirsk Publ House of SB RAS) p 356(in Russian)

Rinn F 1996 Tsap V 36 Reference Manual Computer Program forTree-ring Analysis and Presentation (RINNTECH Hardstr20ndash22 D-69124 Heidelberg Germany)

Rollins M G Morgan P and Swetnam T 2002 Landscape-scalecontrols over 20th century fire occurrence in two large RockyMountain (USA) wilderness areas Landscape Ecol 17 539ndash57

Sannikov S N and Goldammer J G 1996 Fire ecology of pine forestsof Northern Eurasia Fire in Ecosystems of Boreal Eurasia vol48 ed J G Goldammer and V V Furyaev (Dordrecht KluwerAcademic) pp 151ndash67

Schepaschenko D G Shvidenko A Z and Shalaev V S 2008Biological Productivity and Carbon Budget of Larch Forests ofNorthern-East Russia (Moscow Moscow State ForestUniversity) p 296 (in Russian)

Shiyatov S G 2003 Rates of change in the upper treeline ecotone inthe Polar Ural Mountains Pages News 11 8ndash10

Shvidenko A Schepaschenko D McCallum I and Nilsson S 2007Russian Forests and Forestry (Laxenburg International Institutefor Applied Systems Analysis and the Russian Academy ofScience) (CD-ROM)

Sofronov M A Volokitina A V and Kajimoto T 1999 Ecology ofwildland fires and permafrost their interdependence in theNorthern part of Siberia Proc 8th Symp on the Joint SiberianPermafrost Studies Between Japan and Russia in 1999 pp 211ndash8

Swetnam T W 1996 Fire and climate history in the central YeniseyRegion Siberia Fire in Ecosystems of Boreal Eurasiaed J G Goldammer and V V Furyaev (Dordrecht KluwerAcademic) pp 90ndash104

Vaganov E A and Arbatskaya M K 1996 Climate history and wildfirefrequency in the central part of Krasnoyarsky krai I Climaticconditions in growing season and seasonal wildfires distributionSiberian J Ecol 3 9ndash18

Wallenius T Larjavaara M Heikkinen J and Shibistova O 2011Declining fires in Larix-dominated forests in northern Irkutskdistrict Int J Wildland Fire 20 248ndash54

Weir J M H Johnson E A and Miyanishi K 2000 Fire frequency andthe spatial age mosaic of the mixed-wood boreal forest inwestern Canada Ecol Appl 10 1162ndash77

6

Environ Res Lett 6 (2011) 045208 V I Kharuk et al

larch growth With time an increase in the thermal insulatorlayer composed of the on-ground moss and lichen cover causedupward migration of the permafrost layer and compression ofthe active root zone within a progressively decreasing upperlayer Fires also thin regeneration decreasing within-speciescompetition and thus promoting tree growth because larch isan extremely shade-intolerant species

52 Long-term trends in wildfire number

The advanced age of some sampled trees (gt400 yr figure 3)allows estimation of the fire history within the study areaback to the Little Ice Age period Dendrochronology datashowed that cooling during the Little Ice Age period causeddepression in the tree annual radial growth (figure 4) Fireevent numbers since the 17th century approximately coincidedwith air temperature deviations increasing with warming sincethe second half of 19th century (figure 4 table 2) For examplefire numbers increased from 8 in the 19th century to 13 inthe 20th century This phenomenon could not be attributed todecreased samples of older trees ie a lsquofading effectrsquo sincesample sizes were adjusted by natality date

The local asynchrony of the radial increment growth andtemperature deviations in figure 4 could be attributed to thefire-induced increase of the radial growth as well as to theincrementation decrease during the lag between the fire eventand incrementation increases beginning (which is about 5ndash7 yr figure 5) Thus in the middle of 20th century wildfireswere observed within the majority of the test sites (figure 2)The other reason is a lsquodivergence phenomenonrsquo ie growth-versus-temperature divergence (DrsquoArrigo et al 2008 Esperet al 2010)

These estimates coincide with earlier reported data on firefrequency increase in the 20th versus 19th centuries for thesoutherly larch and mixed forests (sites 2 and 1 in figure 1respectively Kharuk et al 2008) The causes of this trendcan be both natural and anthropogenic Earlier it was shownthat for site 1 (figure 1) the FRI decrease was caused byboth natural (warming) and anthropogenic causes For site2 (figure 1) the FRI decrease was attributed to temperatureincrease mainly because the leading factor in fire ignition inthe remote northern forests is lightning (gt90 of cases in thenorthern forests whereas within southern forests gt80 of firesare anthropogenically caused Kovacs et al 2004) This is alsotrue for our study site because of the remoteness of the areaas well as the low population density in general most of thefires are of natural origin Similar observations (FRI increasesince the Little Ice Age) were made for the northern boundaryof the larch forests (site 3 in figure 1 Kharuk et al 2011)All these data support the suggestion that observed climatechange will lead to an increase in fire frequency (Gillett et al2004 Bergeron et al 2004 Girardin et al 2009) Comparisonof FRI along the meridian (figure 1) showed a northwardincrease of FRI 77 plusmn 20 yr at sim61N (site 1) 82 plusmn 7 at 64N(site 2) 200 plusmn 51 yr at about 66N and 320 plusmn 50 yr at thenorthern boundary of the larch dominant communities (71Nsite 3) The main reason for the northward FRI increase is lessincoming solar radiation and consequently shortening of thefire-danger period

Figure 5 Diagrams for (1) individual tree ring widths (N = 36) and(3) mean tree ring widths before and after fire and (2) tree ring widthfor a specimen with an extremely high post-fire growth increment(inset) Data were compiled on the basis of fires in the 20th century(figure 2) Dates of fires were set to the lsquozerorsquo point Note thatpost-fire growth increase has a lag of about 5ndash7 yr

53 Burns as a lsquosimulatorrsquo of future warming

Areas experiencing wildfires may be considered as a simulatorof predicted warming impacts on northern forests Indeedin addition to soil enrichment with nutrients and decreasedcompetition fires cause an increase of permafrost thawingdepth by a factor of 3ndash5 (Kharuk et al 2008) and increasedsoil drainage Trees that survived wildfires showed anapproximately double (193 p gt 095) increase of the radialincrement (figure 5) Comparison was made for the period for25 yr before and after the fire (periods of plusmn5 yr around zerowere deleted to exclude the effects of direct fire damage ontrees) Some trees showed an extremely high response to fireeffects For example the tree cross section shown in figure 5showed an increase of radial growth by about a factor of 10in comparison with background observations This tree wassampled at a latitude near the Polar Circle

Generally speaking growth increases following fire scarsshould be measured along radii very much distant from thewound itself But in our case we compare 37 specimenswith the same pattern of fire damage one of which showsoutstanding increment growth (about ten times that of thebackground set figure 5) The basic difference of thisspecimen from the others sampled was in the depth of soilthawing (about 15 m versus lt03ndash05 m for the otherspecimens) and good drainage since that tree was growing onthe southern river bank It is known that larch prefer drainedsoils (Schepaschenko et al 2008) The observed radial growthincrease (and consequently increased carbon sequestration)may be an alternative to the scenario of forests in climate-induced permafrost transformation areas becoming greenhousegas sources (IPCC 2007) Thus the vegetation dynamics andproductivity of the burned areas deserve future investigation

5

Environ Res Lett 6 (2011) 045208 V I Kharuk et al

54 Indirect signs of wildfires

The tree ring growth history showed periods of radial growthaccelerations (ie tree ring width increase figures 2ndash4)These increases are commonly considered to be climate driven(eg Shiyatov 2003) and may also contain information onfire events The observed growth accelerations may also becaused by the above-mentioned fire-caused soil meliorationDifferentiation of these effects could be carried out on the basisof the comparison with air temperature anomalies The fire-induced origin of the acceleration is supported by the fact thatin some cases the date of accelerations coincides with the burnmarks on the trees from the same test site (figure 2) Theseeffects deserve future investigation based on larger samplesizes

6 Conclusions

FRI in the study area is considerably longer than in southerlyterritories Along the 100 meridian (figure 1) FRI valuesincreased northward 77 plusmn 20 yr at sim61N (site 1) 82 plusmn 7 yr at64N (site 2) 200plusmn51 yr at about 66N and 320plusmn50 yr at thenorthern boundary of the larch stands (sim71N) The numberof fire events during the Little Ice Age period (17thndash18thcenturies) was approximately half the number observed in the19thndash20th centuries Fire-caused soil melioration (basicallysoil drainage and thawing depth increases) caused a radialgrowth increase to about twice the background value (up togt6 times at extremes) This effect may simulate the predictedimpact of warming on the larch growth in the permafrost zone

Acknowledgments

This work was supported by the Siberian Branch of theRussian Academy of Sciences Project No 2733 and NASAHQ Terrestrial Ecology Program

ReferencesBeaty R M and Taylor A H 2001 Spatial and temporal variation of

fire regimes in a mixed conifer forest landscape SouthernCascades California USA J Biogeogr 28 955ndash66

Bergeron Y Gauthier S Flannigan M and Kafka V 2004 Fireregimes at the transition between mixed wood and coniferousboreal forest in Northwestern Quebec Ecology 85 1916ndash32

Buechling A and Baker L 2004 A fire history from tree rings in ahigh-elevation forest of Rocky Mountain National Park Can JForest Res 34 1259ndash73

Climate Explorer 2011 httpclimexpknminlDrsquoArrigo R Wilson R Liepert B and Cherubini P 2008 On the

lsquoDivergence Problemrsquo in Northern Forests a review of thetree-ring evidence and possible causes Glob Planet Change60 289ndash305

Esper J Frank D Buentgen U Verstege A Hantemirov R andKirdyanov A 2010 Trends and uncertainties in Siberianindicators of 20th century warming Glob Change Biol16 386ndash98

Gillett N P Weaver A J Zwiers F W and Flannigan M D 2004Detecting the effect of climate change on Canadian forest firesGeophys Res Lett 31 L18211

Girardin M P Ali A A Carcaillet C Mudelsee M Drobyshev IHely C and Bergeron Y 2009 Heterogeneous response ofcircumboreal wildfire risk to climate change since the early1900s Glob Change Biol 15 2751ndash69

Heyerdahl E K Beubaker L B and Agee J K 2001 Spatial controls ofhistorical fire regimes a multiscale example from the interiorwest USA Ecology 82 660ndash78

Holmes R L 1983 Computer-assisted quality control in tree-ringdating and measurement Tree-Ring Bull 43 69ndash78

IPCC 2007 Climate Change 2007 Synthesis Report Contribution ofWorking Groups I II and III to the Fourth Assessment Report ofthe Intergovernmental Panel on Climate Changeed Core Writing Team R K Pachauri and A Reisinger (GenevaIPCC)

Johnson E A and Miyanishi K 2001 Forest FiresmdashBehavior andEcological Effects (San Diego CA Academic)

Kharuk V I Dvinskaya M L and Ranson K J 2011 FRI within theNorthern Boundary of the Larch Forest Int J Wildland Firesin review

Kharuk V I Dvinskaya M L Ranson K J and Im S T 2005 Expansionof evergreen conifers to the larch-dominated zone and climatictrends Russ J Ecol 36 164ndash70

Kharuk V I Kasischke E S and Yakubailik O E 2007 The spatial andtemporal distribution of fires on Sakhalin Island Russia Int JWildland Fire 16 556ndash62

Kharuk V I Ranson K J and Dvinskaya M L 2008 Wildfires dynamicin the larch dominance zone Geophys Res Lett 35 L01402

Kovacs K Ranson K J Sun G and Kharuk V I 2004 The relationshipof the Terra MODIS fire product and anthropogenic features inthe Central Siberian landscape Earth Interact 8 (available athttpearthinteractionsorg)

Larsen C P S 1997 Spatial and temporal variations in boreal forestfire frequency in northern Alberta J Biogeogr 24 663ndash73

Naurzbaev M M Hughes M K and Vaganov E A 2004 Tree-ringgrowth as sources of climatic information Quat Res 62 126ndash33

Naurzbaev M M Vaganov E A and Sidorova O V 2003 Thevariability of on-ground air temperature in northern Eurasiabased on millennium tree ring chronology Earth Criosphere 7(2) 84ndash91 (in Russian)

Payette S 1992 Fire as a controlling process in the North Americanboreal forest A Systems Analysis of the Boreal Forested H H Shugart R Leemans and G B Bonan (CambridgeCambridge University Press) pp 144ndash69

Pleshikov F I (ed) 2002 Forest Ecosystems of the Yenisei Meridian(Novosibirsk Novosibirsk Publ House of SB RAS) p 356(in Russian)

Rinn F 1996 Tsap V 36 Reference Manual Computer Program forTree-ring Analysis and Presentation (RINNTECH Hardstr20ndash22 D-69124 Heidelberg Germany)

Rollins M G Morgan P and Swetnam T 2002 Landscape-scalecontrols over 20th century fire occurrence in two large RockyMountain (USA) wilderness areas Landscape Ecol 17 539ndash57

Sannikov S N and Goldammer J G 1996 Fire ecology of pine forestsof Northern Eurasia Fire in Ecosystems of Boreal Eurasia vol48 ed J G Goldammer and V V Furyaev (Dordrecht KluwerAcademic) pp 151ndash67

Schepaschenko D G Shvidenko A Z and Shalaev V S 2008Biological Productivity and Carbon Budget of Larch Forests ofNorthern-East Russia (Moscow Moscow State ForestUniversity) p 296 (in Russian)

Shiyatov S G 2003 Rates of change in the upper treeline ecotone inthe Polar Ural Mountains Pages News 11 8ndash10

Shvidenko A Schepaschenko D McCallum I and Nilsson S 2007Russian Forests and Forestry (Laxenburg International Institutefor Applied Systems Analysis and the Russian Academy ofScience) (CD-ROM)

Sofronov M A Volokitina A V and Kajimoto T 1999 Ecology ofwildland fires and permafrost their interdependence in theNorthern part of Siberia Proc 8th Symp on the Joint SiberianPermafrost Studies Between Japan and Russia in 1999 pp 211ndash8

Swetnam T W 1996 Fire and climate history in the central YeniseyRegion Siberia Fire in Ecosystems of Boreal Eurasiaed J G Goldammer and V V Furyaev (Dordrecht KluwerAcademic) pp 90ndash104

Vaganov E A and Arbatskaya M K 1996 Climate history and wildfirefrequency in the central part of Krasnoyarsky krai I Climaticconditions in growing season and seasonal wildfires distributionSiberian J Ecol 3 9ndash18

Wallenius T Larjavaara M Heikkinen J and Shibistova O 2011Declining fires in Larix-dominated forests in northern Irkutskdistrict Int J Wildland Fire 20 248ndash54

Weir J M H Johnson E A and Miyanishi K 2000 Fire frequency andthe spatial age mosaic of the mixed-wood boreal forest inwestern Canada Ecol Appl 10 1162ndash77

6

Environ Res Lett 6 (2011) 045208 V I Kharuk et al

54 Indirect signs of wildfires

The tree ring growth history showed periods of radial growthaccelerations (ie tree ring width increase figures 2ndash4)These increases are commonly considered to be climate driven(eg Shiyatov 2003) and may also contain information onfire events The observed growth accelerations may also becaused by the above-mentioned fire-caused soil meliorationDifferentiation of these effects could be carried out on the basisof the comparison with air temperature anomalies The fire-induced origin of the acceleration is supported by the fact thatin some cases the date of accelerations coincides with the burnmarks on the trees from the same test site (figure 2) Theseeffects deserve future investigation based on larger samplesizes

6 Conclusions

FRI in the study area is considerably longer than in southerlyterritories Along the 100 meridian (figure 1) FRI valuesincreased northward 77 plusmn 20 yr at sim61N (site 1) 82 plusmn 7 yr at64N (site 2) 200plusmn51 yr at about 66N and 320plusmn50 yr at thenorthern boundary of the larch stands (sim71N) The numberof fire events during the Little Ice Age period (17thndash18thcenturies) was approximately half the number observed in the19thndash20th centuries Fire-caused soil melioration (basicallysoil drainage and thawing depth increases) caused a radialgrowth increase to about twice the background value (up togt6 times at extremes) This effect may simulate the predictedimpact of warming on the larch growth in the permafrost zone

Acknowledgments

This work was supported by the Siberian Branch of theRussian Academy of Sciences Project No 2733 and NASAHQ Terrestrial Ecology Program

ReferencesBeaty R M and Taylor A H 2001 Spatial and temporal variation of

fire regimes in a mixed conifer forest landscape SouthernCascades California USA J Biogeogr 28 955ndash66

Bergeron Y Gauthier S Flannigan M and Kafka V 2004 Fireregimes at the transition between mixed wood and coniferousboreal forest in Northwestern Quebec Ecology 85 1916ndash32

Buechling A and Baker L 2004 A fire history from tree rings in ahigh-elevation forest of Rocky Mountain National Park Can JForest Res 34 1259ndash73

Climate Explorer 2011 httpclimexpknminlDrsquoArrigo R Wilson R Liepert B and Cherubini P 2008 On the

lsquoDivergence Problemrsquo in Northern Forests a review of thetree-ring evidence and possible causes Glob Planet Change60 289ndash305

Esper J Frank D Buentgen U Verstege A Hantemirov R andKirdyanov A 2010 Trends and uncertainties in Siberianindicators of 20th century warming Glob Change Biol16 386ndash98

Gillett N P Weaver A J Zwiers F W and Flannigan M D 2004Detecting the effect of climate change on Canadian forest firesGeophys Res Lett 31 L18211

Girardin M P Ali A A Carcaillet C Mudelsee M Drobyshev IHely C and Bergeron Y 2009 Heterogeneous response ofcircumboreal wildfire risk to climate change since the early1900s Glob Change Biol 15 2751ndash69

Heyerdahl E K Beubaker L B and Agee J K 2001 Spatial controls ofhistorical fire regimes a multiscale example from the interiorwest USA Ecology 82 660ndash78

Holmes R L 1983 Computer-assisted quality control in tree-ringdating and measurement Tree-Ring Bull 43 69ndash78

IPCC 2007 Climate Change 2007 Synthesis Report Contribution ofWorking Groups I II and III to the Fourth Assessment Report ofthe Intergovernmental Panel on Climate Changeed Core Writing Team R K Pachauri and A Reisinger (GenevaIPCC)

Johnson E A and Miyanishi K 2001 Forest FiresmdashBehavior andEcological Effects (San Diego CA Academic)

Kharuk V I Dvinskaya M L and Ranson K J 2011 FRI within theNorthern Boundary of the Larch Forest Int J Wildland Firesin review

Kharuk V I Dvinskaya M L Ranson K J and Im S T 2005 Expansionof evergreen conifers to the larch-dominated zone and climatictrends Russ J Ecol 36 164ndash70

Kharuk V I Kasischke E S and Yakubailik O E 2007 The spatial andtemporal distribution of fires on Sakhalin Island Russia Int JWildland Fire 16 556ndash62

Kharuk V I Ranson K J and Dvinskaya M L 2008 Wildfires dynamicin the larch dominance zone Geophys Res Lett 35 L01402

Kovacs K Ranson K J Sun G and Kharuk V I 2004 The relationshipof the Terra MODIS fire product and anthropogenic features inthe Central Siberian landscape Earth Interact 8 (available athttpearthinteractionsorg)

Larsen C P S 1997 Spatial and temporal variations in boreal forestfire frequency in northern Alberta J Biogeogr 24 663ndash73

Naurzbaev M M Hughes M K and Vaganov E A 2004 Tree-ringgrowth as sources of climatic information Quat Res 62 126ndash33

Naurzbaev M M Vaganov E A and Sidorova O V 2003 Thevariability of on-ground air temperature in northern Eurasiabased on millennium tree ring chronology Earth Criosphere 7(2) 84ndash91 (in Russian)

Payette S 1992 Fire as a controlling process in the North Americanboreal forest A Systems Analysis of the Boreal Forested H H Shugart R Leemans and G B Bonan (CambridgeCambridge University Press) pp 144ndash69

Pleshikov F I (ed) 2002 Forest Ecosystems of the Yenisei Meridian(Novosibirsk Novosibirsk Publ House of SB RAS) p 356(in Russian)

Rinn F 1996 Tsap V 36 Reference Manual Computer Program forTree-ring Analysis and Presentation (RINNTECH Hardstr20ndash22 D-69124 Heidelberg Germany)

Rollins M G Morgan P and Swetnam T 2002 Landscape-scalecontrols over 20th century fire occurrence in two large RockyMountain (USA) wilderness areas Landscape Ecol 17 539ndash57

Sannikov S N and Goldammer J G 1996 Fire ecology of pine forestsof Northern Eurasia Fire in Ecosystems of Boreal Eurasia vol48 ed J G Goldammer and V V Furyaev (Dordrecht KluwerAcademic) pp 151ndash67

Schepaschenko D G Shvidenko A Z and Shalaev V S 2008Biological Productivity and Carbon Budget of Larch Forests ofNorthern-East Russia (Moscow Moscow State ForestUniversity) p 296 (in Russian)

Shiyatov S G 2003 Rates of change in the upper treeline ecotone inthe Polar Ural Mountains Pages News 11 8ndash10

Shvidenko A Schepaschenko D McCallum I and Nilsson S 2007Russian Forests and Forestry (Laxenburg International Institutefor Applied Systems Analysis and the Russian Academy ofScience) (CD-ROM)

Sofronov M A Volokitina A V and Kajimoto T 1999 Ecology ofwildland fires and permafrost their interdependence in theNorthern part of Siberia Proc 8th Symp on the Joint SiberianPermafrost Studies Between Japan and Russia in 1999 pp 211ndash8

Swetnam T W 1996 Fire and climate history in the central YeniseyRegion Siberia Fire in Ecosystems of Boreal Eurasiaed J G Goldammer and V V Furyaev (Dordrecht KluwerAcademic) pp 90ndash104

Vaganov E A and Arbatskaya M K 1996 Climate history and wildfirefrequency in the central part of Krasnoyarsky krai I Climaticconditions in growing season and seasonal wildfires distributionSiberian J Ecol 3 9ndash18

Wallenius T Larjavaara M Heikkinen J and Shibistova O 2011Declining fires in Larix-dominated forests in northern Irkutskdistrict Int J Wildland Fire 20 248ndash54

Weir J M H Johnson E A and Miyanishi K 2000 Fire frequency andthe spatial age mosaic of the mixed-wood boreal forest inwestern Canada Ecol Appl 10 1162ndash77

6