Drought and Heat-Induced Tree Mortality

8/10/2019 Drought and Heat-Induced Tree Mortality

1/25

Forest Ecology and Managemem 259 (2010) 660 -684

Contents lists avai lable at Science Direct .... l ~

l ~ l

Forest Ecology and Management t ~ I I

j o u r n a l h o m e p a g e : w w w. e l s e v i e r . c o m l l o c a t e / f o r e c o

A global overview of drought and heat-induced tree mortality revealsemerging climate change risks for forests

Craig D. Allen a.*, Alison K. Macalady b, Haroun Chenchouni c, Dominique Bachelet d, Nate McDowell e,Michel Vennetier r Thomas Kitzberger g, Andreas Rigling 1\ David D. Breshears i, E.H. (Ted) Hoggj, Patrick Gonzalez 1< Rod Fensham t, Zhen Zhang m, Jorge Castro 11, Natalia Demidova 0 ,Jong-Hwan Lim p. Gillian Allard q, Steven W. Running r. Akkin Semerci 5 Neil Cobb t U,S Geological Survey, Fort Col/ins SCience Cepner, Jemez MounlOlns Field 51 non, Los Alamos, NM 87544, USAb School of Geography ond Development and Laboratory of Tree-Rmg Reseorch, University of Arizona, Tucson, AZ 85721, USA'Deporrment of Biology, University of 80rno, 05000 Borna, Algeriod Deparrmenr of BiolOgical and ecological engineering. Oregon State UniverSIl)', CorvoWis, OR 97330, USA

Eareh ond Environmental SCiences, MS J495, Los Alamos Notiollol Lobomrory. Los Alamos, NM 87544,USAr CEMAGREF, ECCOREV R 3098, Aix-MarseilJe University. Aix-ea-Pravence, Franceg Lobaratorio Ecotono, 1N/8IOMA-CONICETand Univ. Naaanal del Comalwe, Quinrra/ 1250. 8400 Boriloche, Argennnah Swiss Federa/lnstitute for Forest, Snow and Landscape Research VV iL. larellentr. // r CH-890J Birmensdorf. Switzerland'School of Natural Resources and rhe Environment, and Deporrment of Ecolagy and Eva/urianary Bia/agy, Universil)' af Arizona, Tucson, A l 85721. USAJ Northel7l Farestry Centre. Canadian Farest Service, 5320-/22 Street. Edmonton, A/berta T6H 355, Conodak Center for Forestry, Univenity of California, Berkeley, CA 94720, USAI QueellSland Herbarium, Envlronmenral Protection Agency, MI Caot-tha Raad, Taowallg. Queenslond 4066, AustralJOon Research Illstitute af Fores[ Ecolagy, Ellvironment alld Pra/eetian. Chinese Academy of Forestry.Key Labaratory of Faresr ProtectIOn af SlOte Faresrry Adminlstraliall, Berjmg 100091, Chinan Grupo de Ecolagra Terresrre, Departo men ro de Eealogio, Uni versidad de Granada, Granada E-1807f, Spa in NortheTil Research Insnrute of Faresny, NikilOV5r., 13 Arkhangelsk 1G3062, Russion FederanonP Division of Fore.


2/25

661D Allen et aL/Fore;t Ecology ond Management 259 2010) 660 684

1 Introduction

Forested ecosystems are being rapi dly and directly transformedby the land uses of our expanding human populations andeconomies. Currently less evident are the impacts of ongoingclimate change on the world's forests. Increasing emissions ofgreenhouse gases are now widely acknowledged by the scientificcommunity as a major cause of recent increases in global mean

temperature (about 0.5 cC since 1970) and changes in the world'shydrological cycle (lPCC, 2007a), ind uding a widening of theEarth's tropical belt (Seidel et aI., 2008; Lu et a ., 2009). Even underconservative scenarios, future climate changes are likely to includefnrther increases in mean temperature (about 2-4 C globally)with significant drying in some regions (Chrisrensen et al.. 2007:Seageret aI., 2007), as wel l as increases in frequency and severity ofextreme drollgllts, hot extremes, and heat waves (IPCC, 2007a:Sterl et aI., 2008).

Understand1l1g and predicting the consequences of theseclimatic cllanges on ecosystems is emerging as one of the grandchallenges for global change scientists, and forecasting the impactson forests is of particular importance (Boisvenue and Running,2006: Bonan, 2008). Forests, here hroadly defined to includewoodlands and savannas, cover 30% of the world's land surface(FAD, 2006). Around the globe societies rely on forests for essentialservices such as timber and watershed protection. and less tangiblebut equally important recreational, aesthetic, and spiritualbenefits. The effects of climate change on forests include bothpositive (e.g. increases in forest vigor and growth from CO 2fertilization. increased water use efficiency. and longer growingseasons) and negative effects (e.g. reduced growth and in creases instress and mortality due to the combined impacts of climatechange and climate-driven changes in the dynamics of forestinsects and pathogens) (Ayres and Lombardero, 2000: Bacheletet aI., 2003: Lucht ec aI., 2006; Scholze et aI., 2006: Lloyd and Bunn,2007). Furthermore, forests are subject to many other humaninfluences such as increased ground-level ozone and deposition(Fowler et a ., 1999; Karnosl{j' et aI., 2005; Ollinger ec aI., 2008).Considerable uncertainty remains in modeling how these andother relevant processes will affect the risk of future tree die-offevents, referred to hereafter as 'forest mortality', under a changingclimate (LoehJe and LeBlanc, 1996; Hanson and Weltzin, 2000:Bugmann et aI., 2001). Alchough a range of responses can andshould be expected, recent cases of increased tree mortality anddie-offs triggered by drought and/or high temperatures raise thepOSSibility that amplified forest mortality may already be occurringin some locations in response to global climate change. Examples ofrecent die-offs are particularly well documented for southern partsof Europe (Peiiuelas et al., 2001; Breda et aI., 2006; Bigleretal., 2006)and for temperate and boreal forests of western North America,where background mortality rates have increased rapidly in recentdecades (van Mantgem et aI., 2009) and widespread death of manytree species in multiple forest types has affected well over 1 0 millionha since 1997 (Raffa et aI., 2008). The common implicated causalfactor in these examples is elevated temperatures and/or waterstress. raising the possibility that the world's forests are increasinglyresponding to ong01l1g warming and drying.

This paper provides an overview of recent tree mortality due toclimatic water stress and warm temperatures in forests around theglobe. We idemify 88 well-documented episodes of increasedmortality due to drought and heat and summarize recem literatureon forest mortality and decline. From this review we examine thepossibility of emerging mortality risks due to increasing temperatures and drought. Climate as a driver of tree mortality is alsoreviewed, summarizing our scientific understanding of mortalityprocesses as context for assessing possible relationships betweenchanging climate and forest conditions. Note that while climatic

events can damage forests in many ways ranging from ice stormsto tornadoes and hurricanes, our emphasis is on climate-inducedphysiological stress driven by drought and warm temperatures.The ecological effects of increased mortality in forests and theassociated consequences for human society remain largelyunassessed. We conclude by outlining key information gaps andscientific uncertainties that currently limit our ability to determinetrends in forest mortality and predict future climate-induced forest

die-off. Addressing these gaps would provide improved information to support policy decisions and forest management worldwide.

2. Methods

This paper emerged in part from collaborations and presentations developed in special sessions on climate-related forestmortality at two international meetings: the 2007 annual meetingof the Ecological Society of America in San Jose, California (Allenand Breshears, 2007), and the 2008 international conferenceentitled Conference on Adaptation of Forests and Forest Management to Changi ng Climate with Emphasi s on Forest Health inUmea, Sweden (Allen, 2009). In addition to citing contributionsfrom these sessions, we conducted a systematic search forpublished accounts of climate-induced tree morrality since 1970using the ISrWeb of Science and Google Scholar. We used differentcombinations of the key words tree, forest, mortality, dieoff, dieback: ' decline, and drought in the searches. We alsoconsulted regional forestry experts to find examples recorded ingovernment documents and other sources outside the scientificliterature.

From the extensive set of documents uncovered during thesesearches, we used two specific criteria to determine whetherthe reference was appropriate for this review. Criteria forinclusion were that the study included: 1) an estimate of areaaffected or amount of adult tree mortali ty at the stand orpopulation level. based on ground measurements, aerial photograp hy. or remote sensing, and (2) documentation of a strongcorrespondence between increases in mortality and increasedwater stress or lligh temperatures. We included examples wherebiotic agents were involved in the mortality, but eXcludedexamples of fire-driven death. Studies of forest decline or partialcanopy dieback without significanr increases in mortalitywere also excluded, as were studies that documented onlyseedling mortality. To simplify presentation, we standardizedstudy descriptors and combined references that describeimpacts of the same event on the same tree species but usedslightly different methods or were conducted at different spatialscales.

To estimate trends in the literature related to climate-inducedforest mortality, we searched the lSI Web of Science using the topicwords forest AND mortality AND drought over the availableinterval from 1985 to 2009. We then controll ed for increases in thegeneral scientific literature related to forests by srandardizing thenumber of target articles by the number of citations uncovered by asearch using only the topic word forest.

For each mortality event (listed as r ows in Appendix Tables A A6) we tested the association between the forest type affected bymortality and the categorized duration of the mortality-triggeringdrought (seasonal event vs. multi-year drought) with a Chi-squareanalysis, comparing number of observed triggering drouglltS (bydrought and foresr type) versus expected number of triggeringdroughts. Forest types were grouped into four major biome typesconsidering similar water limitations: (1) savannas, (2) coniferforests and Mediterranean woodlands, (3) temperate evergreenand deciduous forests, and (4) evergreen broadleaved tropicalforests.


3/25

662 CD. Allen et al./Forest Ecology and Management 259 2010) 660-684

Fig. 1. White dots indicate documented localities with rarest mortality related to climatic stress from drought and high temperatures. Background map shows potentialenvironmentallimirs co vegetation net primary production (Boisvenue and Running, 2005). Only the general areas documented in ehe tables are shown-many addinonallocalities are mapped more precisely on the continental-5cale maps. Droughe and heat-driven [otest mortality often is documented in relatively dry regions (-red/orangel

pink). bU also occurs outside ehese regions.

3. esults

3.1. Examples o f recent climate-induced forest mortality

More than 150 references that document 88 examples of forestmortality met our criteria of events that were driven by climaticwater/heat stress since 1970. The examples range from modest butsignificant local increases in background tree mortality rates toclimate-driven episodes of regional-scale forest die-off. We foundexamples from each of the wooded continents that collectivelyspan diverse forest types and climatic zones (Figs. 1-8 andTables A I-A6). Despite our collective efforts to secure referencesfrom non-English language sources, this review is clearly morecomprehensive for North America, Europe, and Australia, andobviously incomplete particularly for some regions, includingmainland Asia and Russia.

Our searches also reveal that published reports of climaterelated forest mortality in the scientific literature have increasedmarl


4/25

663D Allen et at/Forest Ecology and Management 259 2010) 660 684

Fig. 2. Satellite map or Arrica. with documented drought induced mortality areas indicated with numbers. tIed toTable AI references. Upper photo: Cedrus atlanl ica die-ofTlIlBelezma National Park. Algeria; 2007. by Har oun Chenchouni. Lower photo: quiver tre e Aloe dichatoma) morTality in Tir asberg Mountain s. Namibia; 2005. by Wendy Faden.

woody species in Spain (Penue las et al.. 2001 ; Mani nez- Vila lta and (Dobbert in and RigJing, 2006; Bigler et al.. 2006; Verrui andPinol. 2002), increased mortality of oak. fir, spruce, beech, and pine Tagliaferro, 1998). A severe drought in 2000 killed many biesspecies in france after tIle extreme heat wave and drought during cephalonica in mainland Greece (Tsopelas et al., 2004) and Pinusthe summer of 2003 (Breda et aI., 2006; Landmann et aI., 2006; IlQlapensis sub. bruda the most drought tolerant of the MediterVennetier et al.. 2007). and jncreases in mortality of Pinus sylv stris ranean pines in eastern Greece ([(orner et al .. 2005). Fa rrher north.near the species' southern range limits in SWitzerland and Italy summer drought paired with biotic stressors has been linked to

editerranean Sea

Codru Ia llCa chlnbulion

odnis 6taM

_ Oeaa

o LiveFig. 3. Map of northern Algeria climate zones and m ortali ty distrib ution of cedrus arlantica "Box I Atlas Cedar Die-ofT in Algeria" serves as Ihe rull caption.


5/25

664 CD Allen et aLI forest Ecology and Management 259 2010) 660-684

ox 1. Atlas Cedar Die-off in Algeria

Atlas cedar Cedrus atlantica occurs in northern Algeria, distributed in scattered montane populations near the limits of itsbioclimatic tolerance between the Sahara Desert and the Mediterranean Sea (Fig. 3 . Since the onset of severe drought from1999 to 2002 cedar forests have undergone mass mortality,affecting all age classes (Bentouati, 2008). While all Algerian

cedar forests are affected, the magnitude of mortality variesalong a steep moisture gradient (Fig. 3), with die-off greatest(up to 100%) in the drier mountains nearest [he Sahara, dropping to much lower mortality levels in the moister coastalmountains (Chenchouni et a ., 2008), Prolonged soil moisturedeficits lead to decline and progressive death of cedar treesover a period of 1-3 years; a variety of insects and fungi havecontinued to kill weakened cedar trees since the drought easedafter 2002 {Chenchouni et aI., 2008), The Cedrus mortalitybegan assmall patches on drier aspects in the arid near-Saharamountains, eventually coalescing into large patches affectingall ages on all exposures. In contrast, only small patches of oldtrees on dry aspects have died in more mesic regions near thecoast. This recent drought also triggered substantial mortalityin other Algerian tree species, including Pinus halapensis,Quercus ilex, Quercus suber, and Juniperus thurifera. Dendrochronological reconstructions of drought in Algeria show thatthis early 2000s dry period was the most severe drought sinceat least the middle of the 15th century (Touch an et aI., 2008),consistent with climate change projections for a trend ofincreasing aridity in this region (Seager et aI., 2007).

mortality of Quercus robur in Poland (Siwecki and Ufnall 10 million ha of Pinus contorta inBritish Columbia (Kurz et aI., 2008a), [ drought-inducedPopulus tremuloides mortality across a million hectares inSaskatchewan and Alberta (Hogg et al.. 2008), In [he southwestern U.S., die-off of Pinus edulis al l over a million hectareswas specifically linked to global-change-type drought (Breshears et aI., 2005). In the eastern portion of the continent,declines and increased mortality among oaks, particularly in thered oak family, have been reported from Missouri (Voelker et aI.,2008) to South Carolina (Clinton et aI., 1993) in relation tomulti-year and seasonal droughts in the 1980s-2000s. Droughtduring the 1980s followed by an unusual spring thaw in easternNorth America also contributed to decline and mortality ofmaples in Quebec (Hendershot and Jones, 1989). In addition,recent increases in background rates of tree mortality across the

Fig. 4. Satellite map or Asia, with documented drought-induced mon:ality localities indicated with numbers tied to Table A2 references. LowerR Photo: Dead Abie koreana,Mount Halla, South Korea; 2008, by Jong-Hwan Um.Upper Rpharo: Pinu5 tabulaeformis mortality in Shanxi Province, China; 2001, by Yugang Wang. Center photo: Dying PinusYUllnanensls in Yunnan Province, China; 2005, by Youqing Luo Upper Lphoto: Abies cilicicia marta lily in the Bozkir-I


6/25

665D Allen er Forest Ecology and Management 2S9 (2010) 660-6 4

Fig. 5. Sarelhre map of Australasia, with documenred drought-induced mon:aliry areas Indicated wirh numbers, t ied to Table A3 references. R photo: Die-off of mulga, AcaCIaaneUIll, the dominant tree across large areas of semI-arid Australia: 2007, by Rod Fensham. photo: Eucalyptus xanthoclada mortaliry in Queensland, northeastern Australia:1996. by Rod Fenshaln.

western U.S. have been attributed to elevated temperatures van warm sea surface temperatures in the North Atlantic, has alsoMantgem et aI., 2009). recently been tied to regionally extensive increases in tree

mortality rates and subsequent aboveground biomass loss,3 1 1 6 Sourh and Central America [n Latin America Fig. 8; indicating vulnera bility of Amazonian forests to moisture suessTable A6 , ENSO-relaced seasonal droughcs have amplified back Phillips et aI., 2009 Fig. 9).ground uee mortality rates in tropical foresTs of Costa RicaChazdon et aI., 2005), Panama Condit et al., 1995), northwest 3 1 2 Spatial and temporal patterns of mortality

Brazil Williamson et a1.. 2000). and southeast Brazil Rolim et aI., Climate-induced mortality events in this review include2005). and caused extensive mortality of Nothofagus dombeyi in examples that span a broad gradient of woody ecosystems. fromPatagonian South America Suarez et aI., 2004). A hot and severe monsoonal savannas with mean precipitation


7/25

666 CD Allen et o/./Forest Ecology l1lld Monogement 259 2010) 660 684

Fig. 7 Satellite map of North Amenca. with documented drought-induced mortali(} localities indIcated with numbers. tied to Table AS references. Top pharo: Aenal viewshowIng severe moftality of aspen PopulllS cremu/oides) m the parkland zone of Alberta. Canada; 2004. by Michael MichaeJian.lower photo: Pinus ponderosa die-off. JemezMoumalns, New Mexico, USA; 2006. by Craig D Allen.

Fig. 8 SatellIte map of South and Central America. with documented drought-induced mortality localities indicated with numbers tied to Table A6 references. Phoco:NDtJiojagus dombeyi mortality at Rio Manso Inferior, norrhern PaL lgonia. Argemina; 2004. by Thoma s Kitzberger.


8/25

667D. Allen er a .fForest Ecology and Managemenr 259 2010) 660 684

05

040~ 10

U

S .

Sava nna Woodlllloc fConife7 l empeulile Tropical Savanna WoodlilndlComre( Tcmperele 7 ropiCalBroadleavad BrolJdleoved Broad -a vcd rordleOl't'dd

Mu IIi-yea r droug hI Seesonal droughl

Fig. 10. Dlfferences between observed and expected frequenCies of reponed foresl mortality cases listed in Tables A I-A6, sorted by duration of associated drought eWllls(seasonal VS. muln-year). with forests grouped into four major biomes. Monallty discriminated by forest type is dependent on drought dnratioll. with more drought-adaptedforest rypes showlIlg mortality durillg long droughts dnd less drought-adapted forest rypes shOWing more mortaliry cases during short-term seasonal droughts. Pearson (111-square; 23.46, df 3. p 0.000012.


9/25

668 CO. Allen et al /Foresr Ecology and Managemenc259 2010) 660-684

However, high severity drought can drive extensive forestmortality independenc of tree density (Floyd et a ., 2009). Highermortality rates can also occur on favorable sites where trees do notinvest in adequate root systems or where they otherwise becomehydraulically overextended (Ogle et aI., 2000; Fensham andFairfax, 2007: Nepstad et aI., 2007).

Spatial patterns of mortality at the stand and forest scale arealso heavily influenced by life-history traits and tolerances of

individual species within forests, with drought commonlytriggering differential mortality rates between co-occurring treespecies (Suarez et aI., 2004; Gitlin et a ., 2006; Fensham and Fairfax,2007: Newbery and Ungenfelder. 2009: Phillips et al.. 2009). Largerandfor older trees often appear more prone to drought-inducedmortality (Mueller et al.. 2005; Nepstad et aI., 2007: Floyd et aI.,2009), al though this relationship is species-dependent, and incases where stands are undergoing intense self-thinning, smallersub-dominant trees and saplings are often more affected (Kloeppelet aI., 2003: Elliott and Swank, 1994; Hanson and Weltzin. 2000).

Temporal patterns of drought-related tree mortality also can bedifficult to interpret due to lagged responses in some species, inwhich mortality has been shown to occur years or even decades afterdrought stress (Pedersen, 1998, 1999; Bigler et al.. 2007).Furthermore, the long-lived nature of trees and their ability to shift

allocation of resources and change their hydraulic architecturethroughout their lives can resu lt in non-linear responses to droughtstress in both space and time. Different sequences of climate eventsmay also affect the risk of mortality (Miao et aI., 2009).

4. Discussion

4 1 Climate induced forest mortality are new trends emerging?

The diverse instances of mortality reported here clearlyillustrate that drought and heat can impact trees in mauy foresttypes. However awareness of. and interest in, climate-inducedforest mortality and diebacl< is not new (Auclair, 1993; Ciesla andDonaubauer, 1994). Past die-offs have been extensively documented. Historic examples include: Widespread death of Euca-lyptus, Acacia and Callirris species in the early 1900s triggered bythe worst drought of the instrumental record in northeasternAustral1a (Fensham and Holman, 1999); Not/lOfagu.s mortalityduring 1914-1915 in New Zealand (Grant, 1984); Picea meyerimortality during the 1920s in northern China (Liang et al.. 2003):extensive tree mortality in the southern Appalachian Mountainsand the Great Plains during the dust-bowl droughts of the 1920s1930s (Hursh and Haasis, 1931; Albertson and Weaver. 1945);Pinus sylvestris death during 1940-1955 in Switzerland (Dobbertinec a ., 2007): oak mortality in many European countries followingsevere droughts episodes in 1892-1897, 1910-1917, 1922-1927,1946-1949, 1955-1961 (Delatour, 1983); extensive tree mortalityof Austrocedrus cl1ilens;s during El Nino droughts in the 1910s,1942-1943, and the 1950s in Argentina (Villalba and Veblen,1998); and die-off of multiple pine species during the 1950sdrought in the southwestern USA (Swetnam and Betancourt, 1998;Allen and Breshears, 1998). Furthermore, the overwroughtperception of unprecedented forest decline and impending deathdue to air pollution in central Europe (where it was referred to as'Waldsterben') and eastern North America that received muchattention in the 1980s provides a cautionary example ofexaggerated claims of widespread forest health risk in the absenceof adequate evidence (Skelly and [nnes, 1994).

So are recent occurrences of die-off simply well-documentedexamples ofa natural phenomenon linked to climate variability, orisglobal climate change driving increases in forest mortality? Werecognize that the available data on climate-induced forest mortalityhave many limitations: our examples represent a compilation of

idiosyncratic case studies with uneven geographic coverage. Thestudies differed greatly in their goals, methods. and definitions ofmortality, and inconsistently report mortalit y rates, spatial scale andpanerns of mortality, and severity parameters of climate stress. Therecent increase in forest mortality reports that we document couldmerely be an artifact of more scientific attention on climate change,perhaps in concert with a few high profile cases of climate-relatedforestdie-off.These limitations. and the lack of any systematic global

monitoring program, currently constrain our ability to determine ifglobal changes in forest mortality are emerging.Even though our review is insufficient to make unequivocal

causal attributions, our data are consist ent with the possibility thatclimate change is contributing to an increase in reported mortality.Documentation of climate-related forest mortality in associationwith recent warming and droughts is rising rapidly (fig. 9), and insome of these cases the droughts have been the most severe of thelast few centuries. Furthermore, recent research indicates thatwarmer temperatures alone can increase foresr water stressindependent of precipitation amount (Barber er al.. 2000). Inaddition, new experimental results show that warmer temperaturescan greatly accelerate drought-induced mortality (Adams et aI.,2009, and associated correspondence). If the recent increase inmortality reports is indeed driven in part by global clima te change.

far greater chronic forest stress and mortality risk should beexpected in coming decades due to tbe large increases in meantemperature and significant long-term regional drying projected insome places by 2100, in addition to projected increases in thefrequency of extreme events such as severe drougbts, hot extremes.and heat waves (IPCC. 2007a; Jentsch et aI., 2007: Stel'l et aI., 2008).

4.2. Climate and plant physiological interactions that driveforest mortality

Understanding complex spatial and temporal patterns ofclimate-induced tree death and forest die-off requires knowledgeof the physiological drivers of tree mortality. The fundamentalmechanisms underlying tree survival and mortality duringdrought remain poorly understood despite decades of researchwithin

the fieldsof

forestry, pathology, entomology, and ecology(Waring, 1987; Manion, 1991; Mueller-Dombois, 1986,1988;Breda et aI., 2006; Ogaya and Penuelas. 2007; McDowell et al..2008). Part of the challenge is that tree mortality commonlyinvolves multiple, interacting factors, ranging from parricularsequences of climate stress and stand i fe histories to insect pestsand diseases (Franklin et a ., 1987; Miao et aI., 2009). Based on thedecline spiral model (Manion, 1991: Manion and Lachance, 1992),drougbt can operate as a trigger ("inciting factor") that mayultimately lead to mortality in trees that are already under stress(by predisposing factors" such as old age. poor site conditions andair pollution) and succumb to subsequent stem and root damageby biotic agents ( contributing factors" such as wood-boringinsects and fungal pathogens). McDowell et al. (2008) build uponManion's framework to postulate three mutually non-exclusivemechanisms by which drought could lead to broad-scale forestmortaliry: (1) extreme drought and beat kill trees tbroughcavitation of water columns within the xylem (Rennenberget al., 2006; Zweifel and Zeugin, 2008); (2) protracted waterstress drives plant carbon deficits and metabolic limitations thatlead to carbon starvation and reduced ability to defend againstattack by biotic agents such as insects or fungi (McDowell et aI.,2008: Breshears et aI., 2009; Adams et aI., 200 9): and (3) extendedwarmth during droughts can drive increased population abundance in these biotic agents. allowing them to overwhelm theiralready stressed tree hosts (Desprez-Loustau et aI., 2006: Raffaet at., 2008; Wermelinger et al.. 2008). Although tbese hypmheseshave groWIng support, our physiological knowledge remains


10/25

669D. Allen et al . / Fares Ecology and Manag ement 259 (2010) 660 684

inadequate for confidently predicting patterns of regional die-off,as well as variation in survival for trees within the same stand.

The degree to which trees reg ulate water loss during drought mayexplain pattems of carbohydrate and resin) production andsubsequent susceptibility to drought or biotic attack (McDowellet aI. 2008; Zweifel et al.. 2009). A continuum of stomatal responsesto drought exist from drought avoidance (isohydlY), in whichstomata close at a threshold water potential to minimize further

transpiration, todrought

tolerance (anisohydry), inwhich stomatal

closure is less severe and transpiration continues at relatively highrates (McDowell et aI. 2008). The isohydric response protects xylemfrom cavitation through avoidance of severe low water potentials,but can cause eventual carbon starvation as stomatal closure shutsdown photosynthesis While respiration costs continue to depletecarbon stores. The anisohydric response can allow continued carbongain through maintaining open stomata but at greater risk ofcavitation, which might kill trees directly or could increase thelikelihood of future carbon deficits. Plants that typify eacl1 responsehave associated traits consIstent with heir mode of stomatalregulation. such as deep rootIng access to more reliable soil waterand cavitation-resistant xylem for drought-tolerant species.

In addition to hydraulic failure and carbon starvation. a thirdphysiological mechanism predisposing plants to mortality mayexist-cellular metabolism limitation. This hypothesis suggests thatlow tissue water potentials during drought may constrain cellmetabolism (WUrth et al.. 2005; Ryan et aI. 2006; Sala and Hoch,2009), thereby preventing the production and translocation ofcarbohydrates, resins, and other secondary metabolites necessa ry forplant defense against biotic attack. The common observation thattrees which succumb to insect attacks have weak resin flow and areunable to pitch out attacl


11/25

670 CD. Allen et al./Foresl Ecology ond Management 259 (2070) 660-684

I, I

IHala' I

8 "'\d r ~ i r 8 1 : t ~ : i,ro\A;th

ime - - -

Fig. 11. Abrupt reductions In forest biomass (or ecasysrem carbon) can result fromdrought-induced forest die-afT and occur more rapidly than the relatively slowcountervailing biomass increments from tree natality and growth. Trajectone5 ofchange vary wnh ecosystem, as do minimu III biomass and carbon val ues, and arenot to scale In this conceptual figure.

notably mountain pine beetle. have recently transformed someforests of interior British Columbia (Canada) from a net carbon sinkinto a net carbon source (Kurz et aI., 2008a). Similarly, it is possible

that widespread forest collapse via drought" could transform theworld's tropical forests from a net carbon sink into a large net sourceduring this century (Lewis. 2006, p. 19 5; cr Phillips etal., 2009:Joneset aI., 2009). Land-use impacts such as anthropogenic fires and forestfragmentation, interacting with climate-induced forest stress, arelikely to amplify these effects in some regions. including the AmazonBasin (Nepstad et aI., 2008). Overall, climate-induced forestmortality and related disturbances will increase global carbon fluxrates at least temporarily, potentially undermining the capacity ofthe world's forests to act as carbon sinks in the coming centuries.

Past forest management may have exacerbated recent mortality in some regions. [n portions of western North America, over acentuly of fire suppression has fostered the buildup of unusuallyhig!) tree densities. Trees in these unnaturally dense forests canhave decreased vigor, which can increase their vulnerability to

multiple mortality factors (Savage. 1997). Extensive reforestationswith pine plantations in regions such as China and theMediterranean Basin (e.g., -3 .5 million ha reforested with coniferssince 1940 in Spain alone; J Castro-from agency statisticalsources) may be particularly vulnerable, especially because someof these plantations are on marginal sites given the excessivedensities and unknown genetic provenances of the trees.

In summary, given the potential risks of climate-induced forestdie-off, forest managers need to develop adaptation strategies toimprove the resistance and resilience of forests to projectedincreases in climate stress (Seppala et aI., 2009). Options mightinclude thinning stands to reduce competition, selection ofappropriate genotypes (e.g., improved drought resiscance), andeven translocation of species to match expected climate changes(e.g. Millar et a1.. 2007b;Joyce et a1., 2008; Richardson et aI., 2009).

4.4. Key informarion gaps ond scientific uncertainties

The conclusions that can be drawn about recent trends in treemorrality and the predictions that can be made about futureclimate-induced forest die-off are limited by a number of keyinformation gaps and scientific uncertainties.

(1) Accurate documentation oj global Jorest mortality patterns anduends requires the establishment oj a worldwide monitoringprogram. Despite many national and regional forest-monitoringefforts (e.g .. the European Union's intensive forest healthmonitoring EU!lCP-Forests Levell! network), there is an absence

of adequate data on forest health status globally FAO, 2006,2007). Existing permanent sample plot networks can detect largescale events or a generalized background mortality increase, butare not designed to detect and assess patchy mortality. even atrather high rates, as is common when forest landscapes areheterogeneous and in mostof the cases of biotic agent outbreaks.Reliable, long -term, gl obal-scale forest healtb monitoring, likelycombining remote-sensing and ground-based measurements in

a mechodologically coordinated and consistent manner, isneeded to accurately determine che scatus and trends of foreststress and mortality on planec Earth, Regional and global maps ofaceual patterns of climate-induced cree mortality are also vicallyimportant for the development and validation of models forpredicting forest die-off in response to climate change.

(2) Undersranding tile mechanisms by which climare change mayafJectJorests requires quantitative knowledge oJ the physiologicalthresholds oj individual u mortality under chronic or acutewarer stress (Fig, 12). With the exception of information for afew tree species (McDowell et aI., 2008; Zweifel et al.. 2009),there is surprisingly little species-specific Imowledge onregulation of xylem water potentials; therefore, placing variousspecies on the continuum of isohydly-anisohydry is difficult,and predicting how diverse species differentially experience

carbon starvation or hydraulic failure is curremly impossible.Similarly, there is almost no knowledge on the patterns ormechanisms of carbohydrate storage in response to droughcand heat. The potential effects of other components of changingatmospheric chemistry (e.g .. elevated levels of nitrogendeposition and ground-level ozone) on the sensitivity of treesco drought remain inadequately known (Grulke et a1., 2009),Research is also needed on how tree phenologies will respondto climate warming, because increasing winter temperaturesmay contribuce to depletion of carbohydrate reserves relevantto carbon starvation thresholds. In addition. better knowledgeis needed on Within-species genetic variability and selection oftrees related to drought and heat stress,

(3) More accurate global vegetation maps are needed as essenrialinputs to calibrate and validare dynamicgfobal vegetation models.

The extent of foresc mortality can only be documented ormodeled if there is precise informacion on the locations andextent of pre-die-off forests.

Drought DurationShOl'l

,'-- f .cMortaJlly MORT UTYjJ r hord

:;,

IJ

Q.E

I , i.. NO MORTAliTYo

Uo _ 6...lWetter Drier

Precipitation

Fig. 12. Conceplual dia gram. showing ran ge of vanabilit y of "Curreut Climate"parameters for precipitation and temperature, or alteroanvely for drought durationand Illtensity, WIth only a small portion of the climate "space" currently exceeding aspecies-specific tree morrallty threshold. "Future Climate" shows increases In extremedraught and temperarure events associaled witl) praJecred global climate change.indicating heightened risks of drought-lnduced die-ofT for current tree populations


12/25

671D AI/en l ll Forest Ecology and onogemerrr 59 2010) 6 6 ~ G 8

(4) Spatially e ~ p i c i t documentotion oj environmental conditions inareas oj Joresr die-off is necessary to IlJlk mortality to callsalclimare dlivers. including precipitation, temperature, and vaporpressure deficir Given the difficulties in measuring precipitation and the absence of reliable soil datasets at adequateresolutions for continental-scale studies. a robust wateravailability index. possibly derived from remote sensing. isneeded to help modelers simulate water stress in trees. In order

to disentangle moisture deficit from temperature effects ontree mortality. more research is also needed to relate spatialgradients of mortality to variation in temperature. Thisresearch might utilize historical and dendrochronologicalrecords across spatial and temporal gradients where variationsin rainfall deficit and temperature increase are expressed.

(5) Mechanistic understanding oj climare-induced tree mortalityrequires unproved knowledge oj belowground processes and soilmoisture conditiolls (e.g. Brunner et al.. 2009). Models ofteninclude detailed algorithms describing aboveground physiological process es but treat belowground processes as a "blact< box".Understanding of the impacts of increasing atmospheric CO 2 nitrogen deposition. ground-level ozone, and drought on rootdynamics, productivity, exudation fluxes, and mycorrhizalinteractions would particularly improve belowground modeling.

(6) TIle directeJJecrs oj climate on rhe population dynamics oj almost allJoresrinsect pescs and other biotic disturbance agents remain poorlyunderstood but are importallr to modeling climare-induced Joresrmortaliry (Wermelinger and Seiffert. 1999; Logan et aI., 2003:Desprez-Loustau et al.. 2006; Bredaet aI., 2006; Bentzetal.. 2009).Generalization through synthesis of current knowledge on (hedynamICS of damaging biotic agents and tree response to attackscould improve existing mortality functions in forest models.

(7) Feedbacks between physiological stress and rree morrolity) drivenby climore and orller Joresr disturbance processes (e.g., inseceoutbreaks,jire) are poorly undersrood (Allen, 2007). These majordisturbance processes may increasingly drive the mortalitydynamics of forests in a rapidly changing climate, necessitatingimproved modeling of their cumulative and collective effects(Nepstad et a1.. 2008).

Current models of vegetation response to climate change shareweaknesses associated with the knowledge gaps idenrified here.including individual tree-based process models (Keane et a .. 2001),species-specific empirical models (climate envelope models, e.g ..Hamann and Wang, 2005; Thuiller et aI., 2008), climate envelopethreshold models linked to plant functional types in dynamic globalveget atio n models (Scholze et al., 2006). and earth system model s (Ciaiset aI., 2005; Huntingford et aI., 2008). The significant uncertaintiesassociated with modeling tree mortality are reflected in ongoingdebates about the magnitude of die-off risk to Amazon rainforests andboreal forests from climate change this century, the potential for dieofTs in forests more genera lly (Loehle and LeBlanc, 1996; Phillips et aI.,2008; Sqja et aI., 2007), and the degree to which forests worldwide arelikely to become a net carbon source or sink (e.g .. Kurz et al.. 2008b).

5.Conclusions

Thls overVlew illustrates the complex impacts of droug ht andheat stress on patterns of tree mortality. and hints at the myriadways in which changes in drought and/or heat severity, duration,and frequency may lead to gradually increasing background treemortality rates and even rapid die-off events. Many recentexamples of drought and heat-related tree mortality from aroundthe world suggest that no forest type or climate zone isinvulnerable to anthropogenic climate change. even in environments not normally considered water-limited. Current obseTVations of forest mortality are insufficient to determine if worldwidetrends are emerging in part due to the lack of a reliable, consistent.

global monitoring system. Although the effects of climate changecannot be isolated in these studies and clearly episodic fQrest rreemortality occurs in the absence of climate change. the globallyextensive studies identified here are consistent with projections ofincreased forest mortality and suggest that some forestedecosystems may already be shifting in response to climate.

There are major scientific uncertainties in our understanding ofclimate-induced tree mortality. particularly regarding the mechan

isms that drive mortality, including physiological thresholds of treedeath and interacrions with biotic agents. Recent advances 111 theunderstanding of tree mortality mechanisms suggest that forestscould be particularly sensitive to increases in temperature illaddition to drought alone, especially in cases where carbonstarvation rather than hydraulic failure is the plimary mechanismof tree mortality. However, we currently lack the ability to predictmortality and die-off of tree species and forest types based onspecific combinations of climatic events and their interactions witilbiotic stressors and place-specific site conditions. The potential forbroad-scale climate-induced tree mortality can beconsidered a nonlinear tipping element in the Earth's climate system (Lenton et al..2008). because forest die-offs from drought can emerge abruptly at aregional scale when climate exceeds species-specific physiologicalthresholds. or ifclimate triggers associated irrupDons of insect pests

in weakened forests. Such cross-scale mortality processes in forestsremain poorly understood.

Collectively. rhese uncertainties currently prevent reliabledetermination of actual mortality trends in forests worldwide.and also hinder model projections of future forest mortality Inresponse to climate change. As one consequence, the potencial forclimate change to trigger widespread forest die-off may be underrepresented in important assessments to dare, norably includingthe latest major (PCe report (2007b). If extensive climate-inducedtree mortaliry occurs, thell substantial negative ecological andsocietal consequences can be expected. Detennining the potentialfor broad-scale. climate-induced tree mortality is therefore a keyresearch priority for ecologists and global change scientists. and isessential for informing and supporting policy decisions and forestmanagement practices.

AcJ nowledgements

We thank Rebecca Oertel, Andrew Goumas. Angeles G. Mayor,Russell Fairfax. and Megan Eberhardt Franl< for literature reviewassistance; Jennifer Shoemaker for graphics support; and JulioBe(ancourt, Adrian Das. Dan Fagre. Brian Jacobs. Francisco Lloret.Cynthia Melcher, Catherine Parks. Tom Veblen. and ConnieWoodhouse. and two anonymous reviewers for comments on thispaper. Support was provided by the U.s. Geological $uTVey.Biological Resourc es Discipl ine, Global Change Program (CDA); theNational Science Foundation and Science Foundation Arizona(AKM); US DOE NICCR DE-FC02-06ER64159 and Biosphere 2Philecology (DDB); and Chinese Special Research Program forPublic-Welfare Forestry 2007BAC03A02 an d 200804001 (ZZ). Thi s

work is a con[ribution of the Western Mountain Initiative, a USGSglobal change research project.

ppendix A

These appendix tables (Tables A 1-A6) accompany the continental-scale maps and associated text descriptions. and are the corecompilation of documented examples of drought and heat-inducedtree mortality. Organized by continem and year of mortality event.concisely listing key information for each documemed example,including an idemification number allowing easy visual linkage to thecontinental-scale map locations.


13/25

en...'

Table l

Documented ~ e s of dronght and/or heat-induced forest mortal>ty from Afnc.l, 1 9 7 0 p r e ~ e n t . 10 numbers refer lu l o c a t l o n ~ m pped in FIg. 2,

10 Wc.l I On Year(s) of Forest type/mean precip. ' Domlu nl tree l.lXa Spatial concentration Climate anumaly Stand/ Scale of Biotic agents Reference(s)dmortalitY of mortality Within Huked to mort litY population- impact/ rea associated with

geogr phic or level affected morta Iity?'elevational rauge mortalitY

(%)"

Senegal 1972-1973 Sav nna (300) Acacia ~ e n e g o lGuiero Middle-lower edges of Multi-year drought 50 Regional None Poupon (1980)senegolensis elevational range:

arid edge of geogra ph icrange

2 Sou th Africa 1988-1992 Savanna (366) Caiophospermum mopane Patchy within r nge Multi-year drought 1'1-87 Not reported Not reported MacGregor and(NorthernProvince)

(basal area) O'Connur (2002)np

3 Zimbabwe 1970-1982, .av,anna IJrachysteg;o glalicescens; Not repo rted Multi-year d r o u g h t ~ Nor reported Subregional; Elephants, Tafangenyasha.

::::

"Southeast) 1991-1992 other savanna species ~ 5 0 0 0 0 0 h a scale insecls (2001. 1998, 1997) ::I~

affected ::;

4 Senegal 1945-1993 Savanna. deciduous Anocoruium aCCldenrale, Arid edges of geographic Multi-year drought 23 Regional None Gonzalez (2001)...~

btoadleaf woodland Cordylo pinnow Ficus ingens range[240-560) many orhers

~

5 South Africa[NonheroProvince)

1991-1993 Woodland, decidUOUSbroad leaf (500-600)

i c h r o s t a c l l Y ~ cinereaI'teracarplls angoieflsis.Strychnos modagoscoriensis,Terminatia sericea C. mopane

Patchy willun range Multi-year drought 1-78(speciesdependent)

NOt reported None Vlfjoen (1995)gQ

g~

many others ::I

~6 Sourh Africa 1982-1997 Savanna (24D-500) C mopone Combrecum Patchy within range Multi-year drought 7 Not reported None O'Connor (1999) 3

(NorthernProvince)

opiallarum, Grewia spp.,X/menta meric n

E..Ln

'Uganda 1999 TropIcal Rainforest (1492) Uvonap is . 'pp .. Celtis spp. Not repo rted Seasonal drought 19 No r reported Not reported l w n g ~ (/.003) >:(Western) ~

-38 Namibia,

South Africa

1904-2002 S a v ~ n n a 10D-200) loe dlchoroma Arid edge of geographic

range

Multi-year drought.

high temperatures

2-71 Subregional None Faden et a . (2007) CT \

cI

9 Algeria 2000-2008 Med. conifer 348-3S6) [eunJS arlanrica Arid edge of geographic Mulri-year drought 40-80 SubregIOnal Insects Bentouati (2008): ~

range Benlouatl andB a r i t e ~ u (2006);Chenchouniet al. (2008)

10 Morocco 2002-2008 Med. montane COnifer Cedl1J.S atlonrica Arid edge of geographic Multi-year drought 10-40 Subregional Not reported 1 Ahidine (2003);(300-600) r ~ n g Adil (2008)

M e d i t e r r a n e ~ n forest types are abbreviated as Med. n Ih column. A n n u ~ 1 precipitation IS in mmlyr in parentheses If reported.h Severily of mortality IS reported ~ t 1he stand or population level as percentage of e ~ d lrees (depending on study design). unless otherwise noted n the entry. Other common fl l IS are annual mortality r ~ l e during drought (%/

year). percent dead b a ~ l area, and dead wood volume ,n mete,. , '.< If biotIC agents arc thought to have played a p f l m ~ r y role in tree morlalitY. thiS IS noted in bold type. Ifbiotic agents were involved 10 mortality bUllheir role was nOl evaluated or IS secondary to Climate, Ihe agents are SImply

lisled.d CitatIOns from which reporred mortality dala is deflved are written In bold lype. Other citarions provide corroborating or secondary eVIdence. If there are multiple citatIOns without no bold type, reported data rellects number=(Eastern mIxed conifer of ranges drnllght, high palchy across Ips, fusodes) Departmenr (2003-2008) :toPyrenees) (800-1000) tempefJtures -150,000ha ;>

25 France 2006-2008 Mediterranea n uercus SUber Northern edge to MultI-year 10-70 Subregion.ll; Insects Platypus spp Vennet,er el al (2008)(Provence.Maures

broadleaf middle of geographicrange

drought patchy across-120,000ha

Coroebus spp.) ...ti'

Mount.:>ins)

Footnotes as In Table A 1 8'fc

Table AS ;>QDocumented cases of drollgbt andlor heal-induced forest mortality from North Amenc.,. I 970-present. ID numbers rerer to lootlol\< mapped in r.g. 7

ID location Ye.,(s) of Forest type' Dominant tree spatial Climate anomaly Standi Scale 01 Bio tic agents Reference(s)ii

mortality mean precip.' taxa co I centr lionof mortality

linked to mortality population-level

impact/areaaffected

associa red wllhmonaliry?L

tv

''wHniugeographic or

mortality (%) ~2

USA (50lltheast, late 1970s Upland Quercus sPp

elevational range

NOl reported Mufti-year droughts; 16.6 in stands Regional Wood borers Stringer et al. (1989); Starkey

I

~Northeast, Midwest) 1980s temperate Cayro spp. high remperatures across Southeast; Agrilus and Oak (1989); Starkey et aL ...mixed precede d hy severe 1.2-6.3 in MIssouri bilineacus); (1989): Clinton et al. (1993);

winters fungi; insecr Millers et al. (1989); Tainterdefnhalors er al (1983); law Jnd Gort

(1987); Ke


18/25

Table AS Continued)

ID Location Yearts) ofmortality

Foresr typeme n prccip.'

Dominan{ lfr tXa

Spatialconcentrationofmort.>litywirhingeographic urelevarional range

Chmate anomalylinked to morLllily

Standpopularionlevelmortality (X)

Scale ofImpact/;"eaaffected

Biotic agemsassocl;lted withmottality?'

Referencer S)d

5 USA (Minnesola) 1987-1989 Savanna (726) QlJercus ellipsoidalis,Q mocrocarpa

Not reported Mulll-yeal druught 18,2 Not reported Not reported Faber-Iansendocn andTester l 9 9 ~

6 East ern North Amenca 1980s Temperatedeciduous(900-1200)

Ace saccharum Patchy withinranges

Drought. highlemperarurespreceded by wlntelthaw

10-15 Subregional:parchyacross>1 Mba

Insect defoliatorMalocosomadisstria)

Hendershot and Jones (1989);Payette et al. (1996): Auclairel at (1996): Kay et al (2004):Robilaille el at (1982)

7 USA and Mexico

(C.lhfomia and BajaCalifornia)

8 USA (Califorma)

1985-early

1990s

1986-1992

Montane

mixed conifer(-600-800)

Montanemixed cunifer(945)

Pinus jeffreyl,

Abies co ncOlor

Pinus ponderosa,Caiocedf1Js decurrens,Abies cOllc%r

Not reported

Not repOited

Multi-year drollght

Multi-year drought.high spring andsummer remperatures

4-15

23,3-69,2

Landscape-

subregional

Landscape

Bark beetles

Dendroceonusspp,)

Bark beetles(Dendroceonusspp.)

Savage (1 ~ 9 7

Guarin and Taylor (2005) ".

9 USA (Califonlla) 1986-1992 Montanemixed conifer

Not reporred Not repotted Multi-year droughl 13 (basal area) Landsc.,pesubregional:56,OOOhaaffected

Engraverbeetles Scolycusspp.)

Macomber and Woodcock(1994) ....;;>

10 USA (Califorma) 1986-1992 Montanemixed cunifer

Pinus spp., Ah,es spp. Drier edge oflocal range:lower edgesof elevariona'ranges

Multi-year drought Nor reporled Landscape-subregional

Primary role,bark beetlesDendrarlonus

spp.): engraverbeetles (Scolytusspp.)

Ferrell et "I (1994):Ferrell (1996)

00~

Q

;::0:J

~

11 USA (Cahforn,a) 1 9 8 5 - 1 9 ~ 5 Montanemixed conifer

PinllS flexills Lower edgesof elevationalrange

Multi-year drollght,high temperatures

50-75 Stand-landscape

Mistletoe(ArceutllObium)bark beetles(Dendrocronuspanderosne)

Millar et ~ L (200ia)

"'.nI Patchy withlOranges

Drought preced I llSwarm winter andspring

18-47 Subreglunal:patchy across-1 Mha

Insect defolialorMa/ocosama

disstria)

Hogg et aL (2002)

14 USA (Midwest.Southeast)

1990-2002 Uplandtemperatemixed

Quercus spp. Patchy withinrang'

Multi-year dluuglll 1S-50 basalarea reduction

Regluna):-1 .8 M haaffecred

Wood borers(fnaph%desrututus. Agrilus,pp.): fungi

Starkey et at (2004):O.lk et al. (2004): VnelkPlet at. (200H): Heilzmanet ai, (2004): Lawrenceet aL (2002)

15 USA (California) 1983-2004 Monlane mixedcOlllfer(1100-1400)

Pinus spp.Abies spp.

Lnwer edgesof elevationalrange

Drought, hightern peratures

63 increasein nnu lmortality rotc

LandscapesUbreglUl Ll

Insects,pathogens

van Mantgcm andStephenson (2007)

16 USA and Canada(Alaska, Yukon)

1989-20 04 Coasral rainforest,boreal (4HS)

Picco spp. Palchy withinranges

Oruught, high summertemperatures

Not reported Subregional:> 1.2 M ha

Primary role,bark beetle(Dcndroctonusrujipenni., )

Derg et aL (2006)

a>......


19/25

T ~ b l e AS (Collflllued).

...co

10 Location Y e a ~ sof forest lype/ Dominant tree Spatial Climate anomaly Stand/ Scale of Biotic agents Reference(s)mortality mean precip.' taxa concentr tion linked to mortahty population Impact/are> associated wi th

of mortality level affected mortality?Cwithin mortality ( }bgeographic ore1evational rilnge

17 USA (Southwe..24 Wester North 1997-2007 Conirerous inUS spp., Picea Not report ed Drought, high NOl repo rted Regional; Primary role, Bentz et al. (2009)

America spp., Ables spp. temperatures 60.7Mha bark andPseudotsugo affected engrave r beetlesmenzies (DendroctotlUs.

Ips. DryocoeresScolytus spp.)

25 USII (Mmnesola) 2 0 0 4 ~ 2 0 0 7 Boreal and l opulus Lower edges Drought Not report ed Nol report ed Insect defoliat ors Minnesot a Dept. Na r.temperate mixed eremuloldes, and mir1dle of Resou rreS (200 7)(480-900) fraxillus spp. r nges

26 USII (California) 1998-2001, NOl reported Not re ported No t leported Droughl preceded 423,000 dead Iress Landscape- Primary role. GMletl el al (2006)2005-2008 by wet, warm episodes in northern California s\lbregional pathogen

(Phytophrllororanorum)

27 Canada and USA Long-term Temperate coastal Cilamuecypal;s Middle armer winters 70 of basal area lost Snbregional: None Beier et al. (2008); Hennon(Alaska. 1880-2008 rainrotest nootkatenSlS and springs -200.000 ha and Shaw (1997); HennonBtltish Columbia) (1300-4000) allecled el al. (2005)

foolnotes as ,n rabl. AI.


20/25

679D. Allel et al./Forest Ecology and Management 259 2010) 660-684

c:.- '~

~ ' . c- ~ C l'"c:

U'l.".

ilt:Rf:'(5z

onoo

'j;j

'

:o'0

~

.cU

0

foCo~

C c: oz Z Z

ilto:::

o

oZ

'

oSu

'" 0.'"" IV'l::

IN.

c:

"" o0-

1:


21/25


22/25

CD Allen et al / Foresr Ecology and Managemenr 259 2010) 660-684 68 I

Guarin. A.. Taylor, A.H., 2005. Drought triggered tree mortality In mixed coniFerforests in Yosemire National Parle, CaliFornia. USA. Foresr Ecology and Management 218, 229-244.

Gutschick. V.P., BassiriRad, H .. 2003. Extreme events as shaping physIOlogy, ecol ogy,and evolU(ion of plants: toward a unified definit ion and evaluation of theirconsequences. New Phyrologist 160.21-42.

Halnann. A, Wang, T.. 200S. Models of climatic nonnals for genecology and climatechange studies in BritISh Colnmbia. Agricultural and Forest Meteorology 128.211-221.

Hanson, P.I., Weltzin,J,F., 2000. Drought dISturbance from chmare change: respOllseof Un ired States forests. SCience of the Total Environment 262. 20S-220.

HellZlnan, E .. Muzika. R.M .. Kabricl(.j .. Guldin,J,M., 2004. Assessment of oak declineIn Missouri, ArkallSas, and Oklahoma. In: Yaussy, DA., et al. (Eds.), Proceedingsof the 14rh Central Hardwood Forest ConFerence, Wooster, OH. 16-19 March2004. Gen. Tech. Rep. NE-316. USDA Forest Service. Northeasrern ResearchStatIon, Newlown Square. PA pp. 510.

Hendershot, W.H., jones, AR.C, 1989. Maple decline In Quebec: a dIscussion ofpossible causes and ehe use oFFe,tllizers to limit damage. Forestry Chromcle 65(4). 280-287.

Hennon, P.E .. Shaw. CG . 1997. Th e enigma of yellow-cedar decl ine-What IS killongrhese long-lived, deFensive trees? journal of Forestry 95, 4-10 .

Hennon, P.E .. D'Amore. D.V .. Zeglen, S.. Grainger, M., 2005. Yellow-cedar decline ,nthe Nonh Coast Forest District of British Columbia. In: USDA Forest ServiceResearch Note PNW-RN-549. Pacific NOrlhwest Research Station, 16 pp.

HlCke,].A .. Logan,I.A .. Powell,J., Ojima. D.S., 2006. Changing temperarures influencesuitabilicy for modeled mounrall1 pine beetle (Dendrocronus ponderosae) outbreaks in the WeStern UnIted Staees. Journal of GeophYSical Research 111,G02019 doi:1 0.1029j2005jGOOOl 01.

Hogg. E.H .. Brande,J,P., Kochruba,lda, B, 2002. Gro .....1:h and dieback of Aspen rorests IIIn o r e h w e s t e n l ~ l h e n : aCanada, m rela tlon to cl nate and inseets Canad,all joumalor Forest Research-Revue Callad,enlle De Recherche Foresnere 32, 823-832.

Hogg, E.H., Brande, J.P., Michaelilan. M. 2008. Impacts of a regIOnal drought on theproductivley, dleback, and hlOmass of west em CanadIan aspen (oresls. CanadianJournal or Forest Research-Revue Canadlenne De Recherche Forestlere 38,1373-1384.

Holst,J .. Barnard. R., Brandes, E .. Buchmann, N Gessler. A.. jaeger, L. 2008,Impactsof summer water limitatiolJ on the carbon balance of a Scots pine forest in thesouthern upper Rhine plaIn. Agricultural and Forest Meteorology 148, 18151826.

Horner, G,J., Baker, P.J .. MacNally, R., Cunningham. S.c., Thomson,].R .. Hamilton. F .2009. Mortah(Y of developing Ooodplaln forests subjected Q a drying climateand water extraction. Global Change Biology 15.2176-2186.

Hosking, G.P., Hutcheson, J,A., 1988, Mountain beech (NochoJagus-Solandri var.liJJortioldes) dechne in rhe Kaweka Rallge. North Island, New Zealalld. New

Zealand Joumal of Botany 26, 393-400.Hosking. G.p .. Kershaw, D.J .. 1985. Red Beech deaeh 111 the Maruia Valley South

Island, New Zealand. New Zealand Journal or Botany 23,201-211.Huntingford, c., Fisher. R.A., Mercado, L.. BOOlh, B.B.B.,Sitch. 5., Hams, P.P .. Cox, P.M ..

Jones, e.D .. Betts, R.A., Mall, i Y., Harris. G .. Col hns. M., MoorcroFt, p .. 2008,Towards quantifying uncertaimy in predictions of Amazon Die-back . Philosophical TransaCtIons of the Royal Society B-Biological Sciences 363 (1498),1857-1864.

Hursh, e.R .. Haasis, F.W., 1931. E[fects of 1925 summer drought on SouehernAppalachian hardwoods, Ecology 12,380-386.

Huxman, T.E .. Wilcox, 8.P .. Breshears, D.D .. Scott, R .. Snyder. K.. Small. E.A., Hultine,K. Pock man, W.,]acllson, R,B.. 2 0 05 , Wo o dy p l an t encroachmentand the watercycle: an ecohydrological Framework. Ecology 86. 308-319.

IPCC. 2007a. Climate change 2007: the physical science basis. I,,: Solomon,S .. Qin,D.. Manning. M., Chen, Z., Matquis, M., Averyt, IbJlley. Parry.M.L.. Canzlalll. D.F .. Palutd


23/25

682 CD. Aliell et ol./Foresr Ecology and Management 259 2010) 660 684

Lim,j.H., Chun,j. H .. Woo, S.Y.. KlIn, Y.K .. 2008. Increa sed decli nes oflCorean fir forestcaused by climate change in Mountain Halla. Korea. I,,: Oral Presentation M:Imernational Conference Adaptation of Forests and Forest ManagemelH toChanging Climate with Emphasis on Forest Health: A Reviewal ' Science.Policies. and Praences , Umea, Sweden. FAOIIUFRO. 25-28 August 2008.

Lingenfelder. M., Newbery. D.M., 2009. On the detectIOn of dynamic responses in adrought-perturbed tropical rainforeS[ in Borneo. Plane Ecology 201,267-290.

lIoret. F . Siscart. D., 1995. Las efecror demogralicos de la ;equia en poblacione; deencina. Cuademos de la Sociedad Esp",ola de C,enclas Fore;rales 2, 77-81.

lIoret, F . Siscarr, D., Dalmase;. c.. 2004. Canopy recovery afrer drought dieback inholm-oa k Mediterranean foresrs of Caraloma (NE Spam). Global Change 8iology

10.2092-2099lloyd. AH .. Bunn. A,G.. 2007. Responses of the circumpolar boreal foresr to 20th

cencury climate vanab,liry. Environmemal Research Letrers 2, 045013,dOl: 10: 1 08811748-9326/2/4/045013.

Loehle, c.. LeBlanc, D . 1996. Madel-based assessmentS of cI imate change effeces anforests: a enrical review. Ecological Modelling 90. 1-31.

Logan.J,. Regn,ere,j,. Powell,J,A .. 2003. Assessmg che impacts of global warming anforest pesl dynamics. Frontiers in Ecology and the Environment 1, 130-137.

Lu.).. Deser, C. ReichleI'. T .. 2009. Cause of the widening of the tropical belt since1958. Geophysical Research Letters 36. l03803.

Lucht. W .. Schaphoff, 5., Erbrecht, T., HeydeI'. U . Cramer. W .. 2006. Terrestrialvegetation rediStribution and carbon balance under climate change. CarbonBalance and Management 1 . 6 dOl: 10.1186/1750-0680-1-6.

Lwanga.j,S., 2003. Localized tree mortality follOWing the drought of 1999 at Ngogo.ICibale National ParI


24/25

683D. Allen et al . / Foresl Ecology and Management 259 2010) 660 684

R ~ f f ~ . I


25/25

684 CD, Allen et al.jForesl Ecology ond Manoge,nen 259 2010) 660-684

Viljoen. A.j 1995,The influence of the 1991/92 drought on the woody vegetation ofthe Kruger National Park Koedoe 32, 85-97.

Villalba, R . Veblen. TT, 199B. Influences of large-scale climatic variability onepisodic tree mortality ill northern Patagonia. Ecology 79, 2624-2640.

Voelker. S.L.. Muz.ika. R Guyette, R.P., 200B. Individual tree and stand levelIIlfluences on the growth, vigor, and decline of Red Oaks in the Oz.arks. ForesrScience 54. 8-20,

Walther. G.-R.. Berger. S Sykes. M.T 2005. An ecological footprint of climatechauge, Proceedings of the Royal Society B 272. 1427-1432.

Wang, H.B .. Zhang, Z"Kong, X.B., Lui, S.C.. Shell, Z.R.. 2007. Preliminary deduction ofpotential distribution and alternative hosts of invasive pest. Dendrocconusvolens (Coleopcera: Scolytidae), Scientia Silvae Sillicae \43 .71-76 .

Waring. R.H., 1987, Characterisllcs of lrees predisposed to die. BiOSCience 37. 569S77

Wermelinger, B.. Seifert, M .. 1999. Tempelature-dependent reproduction of rhespruce barl' beetle Ips cypographus, and analysis of the porentlal papulationgrowth. EcologICal Entomology 24, 103-110

Wennelinger. B Rlghng.A .. Schneider. M .. Dobbertin. M .. 200B.Assesslng the role ofbar ,- and wood-botlng insects in the decline of Scots pille Pinus sylvestris) Inthe SWISS Rhone valley. EcologICal Entomology 33, 239-249.

Werner, W.L" 1988. Canopy diebacl< in the upper montane rain forests of Sri Lanlta.Geojournal 17,245-248.

Willian150n, G.B .. Laurance, W.F" Oliveira. A.A., Delamouica, P Gascon, c. Lovejoy.T.E Pohl, L 2000. Amazonian tree mortality during the 1997 EI Nino droughtConservation Biology 14, 1538-1542.

Woo, S.-y,. Lim, ].H .. ]e, S.M., Lee. D.K., Kwon. M.j., Ryang. S., 2007. Decline in Mr.Halla-A Linkage with Physiological Changes Caused by Climate Change.ln:Fourth USDA Creenhouse Cas Conference. Baltllnore. MD, 6 February 2007,

Woods. P .. 1989. EIfects of logging. drought. and fire on structure and compoSltionof tropical forests in Sabah, MalaYSia. BlOrropica 21. 290-298.

Worrall. n Egeland. L . Eager. T Mask, R.A.. Johnson. E.W Kemp, P.A .. Shepperd.W.D .. 2008. Rapid mortality of Populus tremuJo/des in southwesteru Colorado,USA Forest Ecology and Management 255, 686-696,

Wurth. M.I

Documents

Drought and Heat-Induced Tree Mortality