Climate change and forestry in Sweden – a literature revie...Climate change and forestry in Sweden probably face continuous, ongoing chang-es in climate, implying that conditions

K. Skogs-o. Lantbr.akad. Tidskr. 143:18, 2004

1

Kungl. Skogs- ochLantbruksakademiensÅrg. 143 • Nr 18 • År 2004

Climate change and forestryin Sweden– a literature review

TID

SK

RIF

T


2

Ansvarig utgivare: Akademiens sekreterare och VD: Bruno NilssonRedaktör: Gunilla Agerlid


3

Climate change and forestry in Sweden– a literature review

Report from Climate and the Forest Committee

Rapporten sammanställd under medverkan avSkog. dr Johan Sonesson


4


5

PrefaceThis is a report from the “Climate and theForest Committee”, appointed by the RoyalSwedish Academy of Agriculture and Forest-ry (KSLA) to promote the interest of climatechange issues among scientists and forestmanagers. The committee identified a needfor a comprehensive literature review as astarting point for their work, culminating inthis report.

Members of the committee are Kaj Rosén(chairman), Johan Sonesson (secretary), Jo-han Bergh, Christer Björkman, Kristina Blen-now, Hillevi Eriksson, Sune Linder, MarkkuRummukainen and Jan Stenlid, all of whomcontributed to this report, Johan Sonessonacting as editor.

SummaryThe scope of the study was to review therelevant literature regarding the impact ofclimate change on forestry in Sweden, to syn-thesise current knowledge, to draw conclu-sions on likely effects of climate change andto identify areas in which further researchand knowledge are required.We have limitedthe study to the effects over short and medi-um time spans (20–100 years), focussing ondirect climatic effects on the trees, and in-direct effects mediated by the climatic impacton soils, herbivores, insects, pests and dis-eases. We have largely ignored other aspectsof forests and climate change.

This literature review has revealed majordeficiencies in our knowledge about theeffects that expected climate change willhave on the forest ecosystems. For instance,the potential effects of climatic changes onthe structure and processes of forest eco-

systems are even less certain than the likelynature and magnitude of the climatic chang-es per se.

However, the most likely effects of climatechange can be predicted. They generallyinclude an increase in potential biomass pro-duction, possibilities to grow new speciescommercially and increased risk of severalkinds of damage. Climate change appears tooffer new opportunities to forestry, whileincreasing the risk of calamities. This calls forradical approaches to both forest- and risk-management.

The reviewed literature contains indica-tions that a better understanding of the linksbetween climate, the forest and forestry isrequired. However, the study also identifiedthree major obstacles that need to be over-come in order to improve our understandingof the issues, risks and possibilities associat-ed with the potential impact of continued cli-mate change on forests and forestry:

• Studies undertaken so far have generallyaddressed some specific aspect of theoverall forestry/forest system, instead ofadopting a more integrated approach inwhich the system as a whole and variousfeedback mechanisms are considered.

• The studies published so far differ in theirchoices of climate change scenarios. Thus,the findings refer to different shifts in tem-peratures, precipitation and other climatevariables, making it difficult to collate andintegrate the findings.

• The transience of the anticipated climatechanges have not been included in thestudies, as they typically refer to impactsunder a specific, static, new climatic regime.However, instead of switching instantane-ously to a new climatic regime sometimein the future, the forest and forestry will


6

Climate change and forestry in Sweden

probably face continuous, ongoing chang-es in climate, implying that conditions willconstantly change within a typical tree-crop rotation, and from one rotation to thenext.

Future research on the effects of climatechange on forestry and forest ecosystemshas to take account of a broad spectrum ofscientific fields, but a multidisciplinary scien-tific approach will probably be essential.

IntroductionInterest in the effects of future climate chang-es is increasing. Climate models, and thuspredicted climate scenarios are becomingmore regionally detailed, and are also improv-ing in their representation of variability andextremes of different climatic parameters.Studies of the impact of climate change onecosystems and human society are becomingsuccessively more relevant and interesting.

During the last decade Sweden has experi-enced somewhat warmer annual mean tem-peratures in comparison with the rest of the20:th century. In particular, the winters havebeen milder, typically with only temporarysnow cover in southern Sweden. A number ofnotably extreme weather events have alsooccurred, including both occasional droughtspells and heavy or persistent rains leadingto flooding in various parts of the country.Public opinion and the media often holdthese developments to be a sign of climatechange. This belief has contributed to in-creased interest in the future effects of cli-mate change. The question is simply: Will itbe worse in the future?A few years ago, theIntergovernmental Panel of Climate Change,IPCC, summarised current knowledge on thefuture of the world forest ecosystems interms of continued climate change during the21:st century (Watson et al. 1997). They pre-dicted that a substantial fraction (a global

average of one third) of the existing forestedarea of the world is likely to undergo majorchanges in broad vegetation types, with thegreatest changes occurring in high latitudesand the least in the tropics. In their regionaloutlook for Europe they stated that forestecosystems might expand at northern lati-tudes into previous tundra and permafrostareas. The effect on wood production in theproduction forests of Europe is expected tobe minor, or at least not threatening. Forestscover more than half the land area of Sweden.Forestry and the forest industry are of crucialimportance to the Swedish economy. Forestsalso supply the Swedes with recreationalfacilities and commodities, such as game, ber-ries and fuel wood. Trees are long-lived andforestry is a long-term business: rotations inSwedish forestry are normally between 60and 100 years. Trees that are young today aremost likely to live the later part of their livesin a different climate from the present. Thismakes the assessment of possible impacts ofclimate change very important, not only forthe forestry sector, but also to for Swedishsociety in general. Furthermore, irrespectiveof whether or not the climate will in factchange, the current debate on climate changehas already affected management of theSwedish forests. Eleven percent of southernSwedish non-industrial private forest ownerssurveyed in 1999 claimed to have changedtheir forest management practices because ofthe possibility of climatic changes (Blennowand Sallnäs, 2002).

To promote the awareness of climatechange issues among scientists and forestmanagers, the Royal Swedish Academy ofAgriculture and Forestry (KSLA) has appoint-ed a “Climate and the Forest Committee”. Thecommittee identified a need for a comprehen-sive literature review as a starting point fortheir work, culminating in this report, fundedby the Royal Swedish Academy of Agricultureand Forestry (KSLA) and the National Boardof Forestry (SKS). The Forestry Research In-


7


stitute of Sweden (Skogforsk), the SwedishUniversity of Agricultural Sciences (SLU) andthe Swedish Meteorological and HydrologicalInstitute (SMHI) have contributed in kind.The scope of the study is to review the rele-vant literature regarding the impact of climatechange on forestry in Sweden, to synthesisecurrent knowledge, to draw conclusions onlikely effects of climate change and to identi-fy areas in which further research and knowl-edge are required.We have limited the studyto the effects over short and medium timespans (20-100 years), focussing on the directeffects of climate change on the trees, andindirect effects mediated by their effects onsoil, herbivores, insects, pests and diseases.We have largely ignored other aspects of for-ests and climate change. The effects of forestdamages and management on the efficiencyof forests as carbon sinks have not been in-cluded in the study. Neither have the poten-tial effects of climate change on biodiversityand conservation, the possible effects on log-ging and transportation operations, or theglobal effects on forestry and other forms ofland use.

We have mainly focused on literature thathas reported results and predictions for Scan-dinavia, but material concerned with othertemperate and boreal forest regions has alsobeen considered, where appropriate.

Climate scenariosFuture climate scenarios are based on predic-tions of how the world will develop in thefuture. Modelled trends in technology, theeconomy, population social inequalities andother relevant factors profoundly affects pre-dicted anthropogenic emissions of green-house gases and other climate-forcing varia-bles. Where appropriate, especially for carbondioxide generated by the consumption of fos-sil fuels, the emission scenarios are enteredinto biogeochemical models, such as for the

carbon cycle. This leads to estimates ofchanges in atmospheric composition, andradiative forcing driving the enhanced green-house effect. The climatic consequences arethen studied through simulations with Gener-al Circulation Models (GCMs) at the globalscale. In order to examine climate change onnational and even finer scales, these are now-adays increasingly being followed by region-alisation using different techniques, such asregional climate modelling.In recent years,scenarios of both emissions and their climat-ic consequences have been steadily refined inattempts to reduce and understand remain-ing uncertainties (IPCC 2001). Uncertaintiesstill remain, but the basic message that theworld is warming is still valid. Indeed, thismessage has become better founded andmore detailed thanks to recent research ef-forts. Further change appears to be inevitableduring the 21st Century, and strategies tocope with it will have to be developed. Overan even longer time scale, the magnitude ofclimate change may still be limited, depend-ing on what actions are taken and globalsocio-economic trends. Climate change dur-ing the 21st Century is sometimes referred toas global warming. The anticipated rise in glo-bal mean temperature is, indeed, a centralissue. However, other changes in the climatesystem, not least those affecting the hydro-logical cycle and water availability, will alsobe very important. Climate change is alsovery likely to have different effects on differ-ent regions, and its consequences, on a re-gional basis, may be quite different than theglobal mean changes.The Swedish regionalclimate modelling programme, SWECLIM,(Rummukainen, 2003) developed in 1996–2003, enabled the climate of Northern Eu-rope, and the ways in which it might be affect-ed by global warming during the 21st Centuryto be adressed. Among other things, SWE-CLIM calculated a set of four detailed region-al climate change scenarios (Räisänen et al.2003, 2004) using an advanced regional cli-


8


mate modelling system. These calculationswere based on two different emission scenar-ios (dubbed “A2” and “B2” by Nakiceovic etal. 2000) and simulations by two different glo-bal climate models. In terms of global meanwarming, the SWECLIM regional scenarioscorresponded to a global mean temperaturerise of 2.5–3.5°C.

The SWECLIM scenarios should be under-stood as plausible descriptions of how theregional climate might have changed by thefuture period of 2071–2100, compared to therecent period of 1961–90. They are not fore-casts, as neither the underlying emissionscenarios nor the global simulations includeestimates of probabilities. Neither shouldthey be taken as best-case or worst-case alter-natives. Rather, they fall well within the rangeof recent global mean warming calculationsderived from several global models and addi-tional emission scenarios. For the periodfrom 1990 to 2100, this range is 1.4–5.8°C(IPCC, 2001).

The regional scale changes are dependent onthe global scale, but are more detailedAs is well-known on the global scale, theregional calculations indicate that the magni-tude of climate change will depend on thecumulative amount of greenhouse gas emis-sions. If there are large emissions in the fu-ture, major changes can be expected, but ifemissions are smaller, the magnitude ofclimate change will also be smaller. In theSWECLIM-scenarios, the results based onemission scenario A2 correspond to greateranthropogenic climate forcing than resultsbased on emission scenario B2. The sign ofthe simulated change in most climatic varia-bles is the same, regardless of which of thesetwo emission scenarios is used. Generally,however, the magnitude of the calculatedchanges increases with the forcing, so the A2-results highlight better the expected natureand direction of the changes than the B2-results.

This is especially true for aspects that nat-urally tend to be highly variable, such as pre-cipitation, and all kinds of extremes. It shouldbe noted, however, that the A2-results are notmore or less likely than the B2-results. Conse-quently, in assessing the possible consequenc-es of climate change, a range of scenariosshould be considered. However, the scenariosshould be internally consistent (as the differ-ent climate variables are interconnected) andin the interest of combining and comparingimpact assessment studies, there should besome common denominator as to the under-lying climate change scenarios used.

Compared to global simulations, the re-gional models address in more detail suchsmall-scale features as topography, land use,snow cover, lake ice and sea ice on the BalticSea. These are not well-represented in typicalglobal models. They do, however, have im-portant influences on the regional climateand the way it is projected to change due toglobal warming. Accounting for these factorsrequires regional studies. Some aspects of theregional scenarios by SWECLIM relevant toforests and forestry are summarised below.For clarity, only subsets of the available sce-narios are mentioned, depending on the par-ticular aspect being considered.

TemperatureIn the SWECLIM scenarios, the annual meantemperatures in different parts of the Nordicregion are projected to increase from 2 to5°C, depending not only on the emission sce-nario, but also on the underlying global sim-ulation. One example is depicted in Figure 1.The spatial pattern of change exhibits slightregional gradients from west to east and fromsouth to north. In our region, temperaturezones are predicted to move northwards (up-wards) in the order of 150 km (100–150 m) forevery 1°C rise in mean temperature. Wintertemperatures increase more than summertemperatures in Northern Europe. Thus thewinter warming is stronger and the summer

´ ´


9


warming weaker, than the annual meanchange. Extreme conditions will also change.The yearly minimum temperature mightincrease by 10–15°C, i.e. much more than themean winter temperature. The yearly maxi-mum temperature is calculated to increaseby about as much as the mean summer tem-perature, i.e. 1–5°C. In Central and SouthernEurope there is a similar large warming in

minimum temperatures. However, comparedto Northern Europe even the maximum tem-peratures change more than the summermean temperature. In Central and SouthernEurope the summertime warming might ex-ceed the warming during wintertime.

One consequence of the projected chang-es in temperature is that the vegetation peri-od will be prolonged (see Figure 2), starting a

Figure 1. Annual mean temperature corresponding to 1961–90 mean conditions (left), calculated regionalchanges under the A2 emission scenario by 2071–2100 (middle) and the projected mean condi-tions in 2071–2100.

Figure 2. The calculated mean length (in days) of the vegetation period in the control simulation (far leftpanel; nominally 1961–90 conditions) and in one of the future scenarios (second left panel). Therespective contributions of an earlier start and a later termination to the annual change in thelength of the vegetation period are illustrated in the two panels to the right.


10


few weeks earlier and terminating a fewweeks later compared to present conditions.This implies an increase in the cumulativeamount of solar radiation available for photo-synthesis during the vegetation period.Changes in cloudiness are significant for thesame reason. These are projected to be of theorder of a few percent in Sweden, with lesscloud in southern Sweden but more in north-ern Sweden in the summer.

PrecipitationThe mean annual precipitation is calculatedto increase by 0–40%, with a higher increasein northern than in southern Sweden. Here,seasonal variations in the calculated changeare even more striking than for temperature(Figure 3). Precipitation increases in autumn,winter and spring in the whole country. Someof the precipitation increase arises from anincrease in the number of days with precipi-tation, but the amounts involved also in-crease. In the summer, on the other hand, themain feature is the decrease in precipitationby 0–40% in southern Sweden. There is also

an increase in precipitation intensity in thesummer, despite the reduction in the summerseason total.

Water balance / water resourcesChanges in precipitation inevitably affect theamount of water available at the earth’ssurface. However, temperature also plays arole. It dictates whether precipitation is tem-porarily stored as snow cover or finds its waydirectly to the soil and further to groundwater and runoff. Temperature also affectsthe amount of available soil moisture by con-trolling evaporation, and transpiration, viaphysical and biological effects. Increasingtemperature increases evaporation, even if ithas no further effect, and thus the net effectof climatic change on the simulated waterbalance (Figure 4) is not dictated by precipi-tation changes alone.

Soil moistureAnnual mean soil moisture is calculated todecrease in most of the country except thenorthernmost parts, mainly due to reduc-

Figure 3. Regionally simulated seasonal precipitation changes by 2071–2100 in the winter (panels to theleft) and summer (panels to the right) based on two different global models and the A2 emissionscenario. The changes are in percentages compared to simulated 1961–90 mean conditions. Thedifference in the predictions for the Norwegian west coast in the winter are to a large part attribu-table to the very different regional circulation changes described by global simulations. The re-sponse of regional circulation systems in the North Atlantic region to global warming is a majorarea of uncertainty in climate science.


11


tions in soil moisture in late summer in south-ern Sweden driven by the combination ofhigher evaporation, lower summer precipita-tion and earlier snowmelt. Potential changesin vegetation caused by increases in carbondioxide, the climate changes or adjustmentsin land use have not been included in theregional simulations so far. The possibilitythat the water use efficiency of plants couldchange has been discussed, leading to addi-tional effects on transpiration and soil mois-ture conditions.

Snow conditionsThe typical number of days with snow coveris calculated to decrease across the entirecountry as a consequence of the warming.Southern Sweden may have snow for only afew days per year during a typical winter inthe future. In central Sweden, lasting snowcover may persist for 1–4 months per yearand for 3–6 months in northern Sweden. Inaddition, the maximum snow depth is pre-

dicted to decrease, despite higher precipita-tion. The projected warming leads to more ofthe precipitation falling as rain rather than assnow.

WindThe calculated changes in average windsclose to the earth’s surface (10 m) differ be-tween the simulations, roughly in line withthe larger-scale atmospheric circulation re-sponses to global warming. In simulationswhere the winds do increase, they tend todo so during winter and spring. The meanwintertime changes span from +20% to un-changed conditions. In summer and autumnthe simulated changes are small in all casesstudied. Changes in extreme winds follow thechanges in mean wind speeds.

In the rest of this report, the results dis-cussed on how climate change might impactforests are drawn from published studies.The climate scenarios assumed in many ofthese studies do not conform to the regional

Figure 4. Calculated water balances (as the net of monthly precipitation minus evaporation) according tosome of the SWECLIM simulations. Possible biological changes of water use efficiency by treesand plants are not accounted for. The calculated start and end of the vegetation period (dashedlines) are also shown. Left: Sweden south of 59°N. Right: Sweden north of 59°N. The nominal1961–90 conditions are from a “control simulation” (CTRL). B2 and A2 are future simulations ba-sed on two different emission scenarios, which are predicted to be weaker in the B2 than in theA2 scenario (SWECLIM Årsrapport 2002.).


12


scenarios briefly described above. In anumber of cases, rather idealised scenarioshave been adopted, and furthermore theyonly address a few variables. Thus, it is diffi-cult to compare the studies or to combinetheir results in any quantitative manner.

The complexity of the systemClimatic changes will affect forests in manydifferent ways, via both their direct effects onthe climate on the trees and their indirect ef-fects on soil, pests and diseases affecting thetrees. In the literature many studies reportingresults and predictions on the effects of singleclimate variables and single species havebeen reported (Harrington et al. 1999). How-ever, the effects of climatic change on ecosys-tems will be much more complex and difficultto predict than such studies imply, since somany variables, species and their interac-tions have to be considered. Predictionsbased on comparisons between climaterecords and records of damage, species dis-

tributions and other relevant factors havealso been made in numerous studies. Somescientists interested in changes that are like-ly to occur in a specific area have examinedprocesses in other geographic areas, wherethe present climate matches the futureclimate predicted for it. The best general pre-diction for the impact of climate change onecosystems is probably obtained throughsynthesising results from as many studies aspossible, obtained using all of the differentapproaches mentioned above. A conceptualframework for analysing ecosystem respons-es to climate change has been suggested byShaver et al. (2000), who emphasise the com-plexity of the models needed to predict theimpact. In addition to these effects on the for-ests, the state of the forests is also affected bychanges in the way they are managed in re-sponse to possible climate changes (Blennowand Sallnäs, 2002), although this aspect is notwell covered in the literature.

This review has been essentially based onthe conceptual model of the direct and indirectimpacts on forests summarised in Figure 5.

• According to international climate re-search, anthropogenic global warmingis already occurring, and it will continueduring the 21st Century. The anticipatedchanges are large compared to naturalvariability over similar time periods.

• Global scenarios in general, and meas-ures of global mean warming in particu-lar, lack sufficient detail that is relevantto estimates of the impact of expected21st Century climate changes on forestsand forestry in Northern Europe. Betterdetail can be gained from regional stud-ies.

• A number of climatic aspects relevant toforests and forestry are likely to changedue to global warming. These includethe length of the vegetation period,

water availability, soil temperature andsnow conditions. Changes in variabilityand extremes are relevant as well aschanges in mean conditions.

• Due to the uncertainties related to futuresocio-economic development, futureemissions and sensitivity of the climatesystem, statements about climatechange, and consequently its effects, areessentially probabilistic. A range of plau-sible scenarios should be studied topromote better understanding of theproblems and new possibilities thatmight arise, and to assess the need foraction. It also is important to strive forcoherency across different studies, sothe results can be compared and usedto develop an integrated understanding.


13


Direct impact of climate ontreesTree species distributionsClimate is a major factor determining the ge-ographical distribution of tree species. Thegeneral effect of a warmer climate will be thatthe range-limits of individual species willmove northwards/upwards. At present thenatural southern limit of a number of speciesincluding Norway spruce (Picea abies) andgrey alder (Alnus incana) are considered topass through Sweden. The northern/upperlimits of a large group of species also occur inSweden, in fact the upper (altitudinal) limitsof all the tree species in Sweden occur here,since no trees grow above the treeline (bydefinition) in the Scandes.

Sykes and Prentice (1995) simulated thepotential distribution ranges of the mostcommon Swedish tree species in a climateresulting from a doubling of the atmospheric

CO2 –content (2×CO2-scenario). They used abioclimatic model (STASH) and predictedmajor changes in species distributions in re-sponse to large climate changes, especiallywinter warming. Some of the common borealspecies (e.g. Norway spruce, Scots pine (Pi-nus sylvestris) and grey alder) are predicted towithdraw to the far north.

A more recent simulation of the futurerange-limit of Norway spruce has been madeby Bradshaw et al. (2000), who also used theSTASH-model and a 2×CO2-scenario. Howeverthe climate scenarios have been modifiedsince the study by Sykes and Prentice (1995).The new prediction is that Norway sprucewill withdraw from the coastal areas of south-ern and central Sweden, but remain in theinterior parts of southern Scandinavia. Brad-shaw et al. (2000) have validated their predic-tion by simulating historical range-limits ofNorway spruce based on historical climaterecords and compared them to the species

Figure 5. Conceptual model underlying this review.


14


distribution as observed in pollen records.They found that the southern range-limit ofNorway spruce has been tracking climatechange closely during the last 1000 years(Figure 6).

The upper range-limits of tree species inthe Scandinavian mountains are likely to riseto higher altitudes. Large areas that are tree-less today are likely to be covered withforests in the future. The establishment ofnew forests in these areas is a slow process,but there are indications that it has alreadybegun. Kullman (2002) has registered newlyestablished seedlings of Norway spruce,Scots pine, hairy birch (Betula pubescens),rowan (Sorbus aucuparia) and willows (Salixspp) at altitudes 100–375 metres higher thanthe range-margins described in the 1950:s inthe studied area (Kilander 1955).

The future distributions of tree species arenot only dependent on the natural range-limitset by climate. Bond and Richardson (1990)have studied vegetation changes in the pre-Pleistocene past and invasions of pine-treesin present South Africa, comparing them toclimate records. They concluded that vegeta-tion change is most often caused by changesin disturbance regime or in competitive inter-actions, especially at the seedling stage, andnot primarily by direct effects of climate.They also state that climate-induced mortal-ity of established plants appears to be rare,especially for longer-lived species.

Human activities like forest management,fire prevention and management of mooseand deer populations have powerful effectson the species distribution and compositionof forests today, and their influence is likely tohave similar strength in the future. Indeed,simulations of the future species compositionof forests in north-eastern Germany (Lindner2000) indicate that forest management willhave a greater impact than the climate onspecies composition, even under future cli-mate change.

The use of tree species outside their natu-ral range is very common in plantation forest-ry. In Europe Norway spruce is widely usedfor plantations outside its natural range,mainly successfully. However, the frequencyof damages to Norway spruce is commonly

Figure 6. Observed and simulated Norway sprucedistributions during the past 1500 yearsand a predicted future distribution (afterBradshaw et al. 2000)


15


considered to be higher outside its naturalrange than within it. Norway spruce standsoutside the species present natural range arelikely to be the first to be severely damagedby a warmer climate. Redfern & Hendry(2002) predict that Norway spruce will ceaseto be a productive tree species over much ofEngland in a future warmer climate.

In conclusion, the most important predict-ed effect of climate change on species rangesis that the northern range limits of most na-tive broadleaves and many introduced treespecies will expand. In the Scandinavianmountains the upper-range limits of tree spe-cies will also rise, leading to forests develop-ing on previously treeless areas. The naturalrange-limit of Norway spruce may withdrawfrom the south Swedish coasts but the spe-cies will most likely continue to be a produc-tive plantation species even in southern Swe-den.

Wind and snow damagesThe predictions for future average and ex-treme winds are uncertain. However, in gener-al windiness is expected to increase. In awindier climate, the risk of wind damage totrees is also expected to increase. The fre-quency and magnitude of extreme winds willimportant, although other changes in varia-bles such as soil freezing and indirect effectsof climate change on the status of the forestsare also expected to affect the risk of wind

damage. The effects of climate change on therisk of wind damage to forests in Sweden arecurrently being studied (Blennow in prep.)using a system of models called WINDA (Blen-now and Sallnäs, in press) and the latest cli-mate scenarios (Räisänen et al 2003). Prelim-inary results for Asa Experimental Forest inSmåland indicate a slight increase in the prob-ability of wind damage due to changes in thewind regime. However, including changes inthe state of the forest in the simulations isexpected to further increase the probabilityof wind damage. Reductions in soil-freezingmay also contribute to increasing frequenciesof windthrow because of weaker tree anchor-age during frost-free soil conditions. Simula-tions for Scots pine in Finland (Peltola et al.1999) indicate that the frost-free period willincrease in length, and thus increase the like-lihood of winds causing damage. In southernFinland the calculated amount of felling windswith non-frozen soil will increase from 55% to80%, and in northern Finland from 40% to50%, in a climate with a 4°C higher averageannual mean temperature. In Sweden, winddamage is currently a bigger problem in thesouth of the country than in the north, andwindthrow is more common than stem break-age (Persson, 1975). For the southernmostpart, where soil freezing is rare or shallow, theeffect of reductions in soil freezing will likelybe marginal, while further north the effect willlikely be more important. On the other hand,

Damage by wind and snow generally varieswidely in space and time. Research is need-ed to characterise the regional distributionof wind and snow damage during thepresent and future climate. A tool, WINDA,has been developed to help assess the riskof wind damage (Blennow and Sallnäs, inpress). Work is underway to use this mod-el together with the SWECLIM scenarios toevaluate effects of climate change on the

risk of wind damage. This work should beextended to cover different parts of Swe-den, and the combined multiple effects ofclimate change need to be included in thesimulations. Efforts to use WINDA to pro-vide support for decision makers under thecurrent climate have started (Olofsson andBlennow, in press), but additional work isneeded where the effects of climate changeare included.


16


snow damage may decrease in large parts ofSweden due to reductions in snow, but theeffects of weather extremes make this pre-diction uncertain.

Frost and winter damagesThe phenology of the trees is mainly deter-mined by climatic factors and the photoperi-od. Understanding this interaction is of greatimportance for predicting how trees willrespond when climate changes but the pho-toperiod remains the same. This has pro-found implications for the effects of climatechange on growth, damage and range limitsof different species.

The general phenological cycle of temper-ate and boreal conifers has been reviewed byHannerz (1998). Budburst cannot occur untilthe rest phase is terminated, which is accom-plished by exposure to low temperatures,below +5°C to +10°C. Once the chilling re-quirement has been satisfied, budburst isinitiated by exposure to high temperatures,providing that the days are not too short tohinder growth. A deficit in chilling will nor-mally increase the temperature required toinitiate budburst. Photoperiod is a strong de-terminant of growth cessation, but tempera-ture and drought may provide additionalcues. The hardiness of the plant tissue is high-est during winter and dehardening is initiatedby temperature, starting several weeks beforebudburst. The lowest levels of frost hardinessoccur in new shoots immediately after bud-burst. Low hardiness then continues through-out the growing season. After budset, hardi-ness steadily develops during the autumn andis promoted by low night temperatures.

There is a great variation between speciesin the chilling requirements, temperaturesums and night hours required to initiate thedifferent phenological events during the year(Hannerz 1998). Even within species the re-quired amounts of initiating cues vary wide-ly, and these traits are often highly heritable(Hannerz 1998).

The phenology of young Norway spruceseedlings has been demonstrated to be de-pendent on the temperature during pollina-tion (Johnsen et al. 1995). The growth of seed-lings grown from seeds produced in a warm-er temperature cease later than seedlingsfrom the same families produced in a coolertemperature, and thus they are more suscep-tible to autumn frosts. This effect could bebeneficial for the adaptation of Norwayspruce to a warmer climate.

The natural southern range limit of Nor-way spruce is considered to pass throughsouthernmost Sweden, and the critical factordetermining this limit is believed to be thespecies chilling requirement. However, Nor-way spruce is successfully used in plantationforestry outside its natural range. Skre & Nes(1996) have grown Norway spruce seedlingsfor three years in normal (for Norway) andelevated (+4°C) winter temperatures, andconcluded that warmer winters could lead toincreased needle losses and reduced growththe following season, especially in northernprovenances. Top-dying sometimes occurs inyoung Norway spruce trees planted in theBritish Isles. The factors causing top-dyingare not fully understood, but climate isappears to play a role in its initiation, andRedfern & Hendry (2002) predict an increasein top-dying as a response to climate change.

Redfern & Hendry (2002) have also pre-dicted the effects of climate change on frostdamage to trees in the UK. They concludethat injuries due to winter cold are likely todecrease, and spring budflush will advance.Thus, the risk of spring frost is unlikely tochange, but autumn frosts may become moredamaging due to later hardening. The fre-quency and amplitude of future weather ex-tremes is very important for the occurrenceof frost damage, and must be accounted for.

In Scandinavia, ground frosts during clearnights in late spring or early summer fre-quently damage newly emerged Norwayspruce shoots. The risk of this type of frost


17


damage occurring in a warmer climate will bedetermined by a fine balance between risingtemperatures reducing the risk of frost dam-age, and the earlier start of dehardening in-creasing the risk (Jönsson et al. 2004).

Decreased vitality due to frost injuries isoften considered to be a decline-initiating fac-tor in several tree species. Auclair et al. (1996)have studied the dieback of birch and maplespecies in eastern Canada and found a strongcorrelation between dieback periods and anindex that is a function of winter frost, rootdamage due to lack of snow cover, summerdrought and heat stress. Dieback was foundto occur mainly in old trees, while youngertrees remained vital. They concluded that awarmer climate is likely to increase the inci-dence of dieback in northern hardwoods inCanada. Barklund (2002) identifies frost inju-ries, often combined with drought spells, asthe initiating factor for the decline of Europe-an oaks. Frost and drought makes the treesmore susceptible to secondary pathogens,and this leads to periods of tree decline overseveral years. Reductions in snow cover andmore frequent drought spells in the futurecould initiate tree decline, but on the otherhand periods of extremely cold winter tem-peratures are predicted to decrease, especial-ly in southern Sweden, and this could helpreduce the risk of tree decline.

Forest firesAn increase in summer temperature com-bined with lower precipitation and, perhaps,increased winds are likely to increase the riskof forest fires. Bergeron and Flannigan (1995)studied the effect of climate change (2×CO2-scenario) on the Canadian forest fire weatherindex (FWI), which is a function of tempera-ture, relative humidity, 24 h precipitation andwind speed. They predicted 1.5–5 fold in-creases in FWI for large regions, mainly inwestern Canada. Predictions for some re-gions in eastern Canada indicated that FWIwould be unaffected or decline. Thompson etal. (1998) predicted a 1.5–2 fold increase inFWI for Ontario in a 2×CO2-scenario.

Suffling (1992) has studied records offorest fires in Sweden, Finland, Norway andCanada and compared them to annual Julyand August temperatures. He found signifi-cantly higher numbers of forest fires andareas affected by fire during years with high-er late summer temperatures. Examination ofSwedish datasets from 1946–64 indicated thata 2°C increase in July–August temperaturemay increase the annual area affected byforest fire in Sweden fivefold.

Large regional differences in summer hu-midity are predicted for the future withinSweden. This has strong implications for thefrequency of forest fires, which may decrease

The magnitude of the risk of damage dueto late spring ground frosts might well besimilar in the future to the magnitudetoday, but predictions are complicated bya lack of relevant knowledge. There is anobvious need for more knowledge aboutthe combined effects of elevated mean tem-peratures and earlier dehardening andbudburst for different species and sites.The risk of autumn frosts is likely to

decrease. The risk of winter damage toNorway spruce due to unfulfilled chillingrequirements may increase in southern-most Sweden, but is unlikely to pose criti-cal threat. Periods of hardwood declinemay increase due to the risk of frost eventscombined with drought. There is a need forbetter understanding of the factors thatinitiate and promote decline and diebackperiods of different tree species.


18


in some regions and increase in others.Changes in forest fire patterns may occur ifclusters of dead wood in the forest increasedue to wind damage, pest outbreaks or peri-ods of decline.

Soil propertiesCarbon balanceClimate change can be expected to have afundamental effect on soil properties andprocesses, and a direct impact on water re-sources. This will have major implications forthe carbon fluxes between forest ecosystemsand the atmosphere, and thus influence therate of CO2 increase in the atmosphere. Forestecosystems have the capacity to store largeamounts of carbon, and appropriate manage-ment practices can mitigate increases of CO2

in the atmosphere, at least temporarily. How-ever, the focus of this literature study is onclimatic effects on damage to the trees andtheir productivity, so we will not considercarbon balance issues any further.

Soil water and frostsThe warmer climate will probably decreasethe length of the soil frost period. This hasbeen predicted for the whole of Finland byVenäläinen et al. (2001). They also predictedan increase in the probability of soil frost inthe middle of the winter in southern Finlanddue to the expected decrease in snow cover.These predictions are also likely to be validfor northern and central Sweden. In southernSweden soil frosts are likely to occur onlyoccasionally in a warmer future. The occur-rence of soil frost has implications for soilstructure, soil biology, weathering, flow ofprecipitation water and the windthrow riskfor trees.

Increased precipitation during winters,combined with less soil frost in some areas, islikely to increase soil water during theautumn, winter and spring. Nisbeth (2002)

has concluded that soil wetness, waterlog-ging and flooding are likely to increase in win-ter throughout the UK in a future climate.This also seems to be a likely developmentfor southern and central Sweden. The risks oferosion will then increase and root develop-ment and tree stability may be negativelyaffected by rise in the water table and in-creased incidence of waterlogging. In north-ern Sweden, where summer precipitation isalso predicted to increase somewhat, thismay lead to areas of productive forestlandturning into wetlands that are not suitable forforest production. The areas of wet forestsites may also increase.

In southern and central Sweden, wheresummers are predicted to be drier, the risksof permanent changes in site conditions arelower. However, more frequent and severesummer droughts may threaten seedling sur-vival on dry sites and contribute to tree de-cline. Similar developments have also beenpredicted for the UK (Nisbet 2002). The great-er winter rainfall will, however, decrease therisk of carry-over of soil moisture deficitsfrom one year to the next and thus help coun-teract damage caused by summer droughtspells. Wetter winters and drier summersmay lead to greater fluctuations of the groundwater table at some sites, and this may alsocontribute to drought stress and reduced treevigour.

Nutrient availabilityGrowth in Swedish forests is primarily limitedby nitrogen on most sites. Increases in tem-perature will also increase the rate of nutrientmineralisation in the soil (Rustad et al. 2001).Therefore, temperature increases due to cli-mate change may stimulate biomass produc-tion. To study the effects of increased temper-ature on forest productivity, a factorial soilwarming × fertilisation experiment was estab-lished in a 40-year-old Norway spruce standin northern Sweden (Strömgren & Linder2002). A 5°C increase in soil temperature


19


during the summer resulted in increases instemwood production of 115% on unfertilisedplots and 57% on fertilised plots. These re-sults indicate that in a future warmer climateincreased availability of nitrogen, combinedwith a longer growing season, may increasebiomass production substantially. How pro-longed these dramatic increases may be isnot clear, a major part of the observed in-creases in growth may be transient.

Biomass productionThe impact of climate change on tree growthis dependent on several factors. The mostobvious effect of a warmer climate is a prolon-gation of the growth season, enabling thetrees to increase their annual growth rates.The most important effect is the earlier startto the growing season, enabling an early andrapid change from loss of CO2 to gain (Jarvis& Linder 2000). Bergh (1997) has predictedthe net primary production (NPP) in a Nor-way spruce stand to increase with about 20%,mainly due to the earlier start of the growingseason, with earlier and more rapid recovery

of the winter-damaged photosynthetic appa-ratus. Predictions on the effects of longergrowth seasons in the Swedish forests indi-cate that NPP are likely to increase by 9–12%in southern Sweden and by 15–18% in north-ern Sweden (Bergh et al. 2000). NPP has alsobeen predicted to increase for a number ofother tree species in Sweden (Bergh et al.manuscript).

The growth of trees and forests is also pre-dicted to increase in other countries. In Fin-land the growth of both Norway spruce andScots pine is predicted to increase, especial-ly in the northern part of the country (Väisä-nen et al. 1994, Beuker et al. 1996). Simulati-ons for a Norway spruce stand in Norwaypredict an increase in NPP of 49% with cli-mate change (Zheng et al. 2002). Yield of Sit-ka spruce (Picea sitchensis) is predicted toincrease in Scotland (Proe et al. 1996) andforest growth in Oregon is predicted to in-crease by about 50% (Coops & Waring 2001).

Water has been identified as a limiting fac-tor for tree growth in southern Sweden, espe-cially for Norway spruce (Alavi 1996, Bergh etal. 1999), but also for Scots pine in some years(Cienciala et al. 1998). Water does not seem

For Nordic conditions, two of the mostimportant growth factors related to soilproperties are nitrogen availability and soilmoisture. In order to predict future treegrowth, more knowledge is needed con-cerning likely changes in soil moisture con-ditions under different climate changescenarios. Soil moisture, the C/N-ratio andtemperature are important variables fornitrogen mineralization. Climate changewill affect these variables, but we need toincrease our knowledge of the interactionsinvolved, especially on quantitative andtemporal effects over the growing and dor-mant seasons.

Effects of climate change on decomposi-tion and weathering need to be better un-derstood.

The length and pattern of the growingseason as well as the length and conditionsduring the non-growing season (with re-spect to snow and soil frost) will seriouslyaffect processes like mineralization, weath-ering and thus nutrient leaching. Increasedleaching of nutrients and other elementsmay have important effects on the biotaand surrounding environments, e.g. ground-water, lake and marine ecosystems.


20


to be an important growth-limiting factor forNorway spruce in northern Sweden (Bergh etal. 1999). A temperature increase that is notaccompanied by an increase in precipitationwill increase the frequency of periods ofdrought stress (Gärdenäs & Jansson 1995).The drier summers predicted for southernSweden in the most recent climate scenarios(Räisinen et al. 2003) may counteract thehigher growth rates predicted in previousstudies based on earlier climate scenariosthat forecast more summer precipitation insouthern Sweden. Simulations of forest pro-ductivity in Brandenburg, Germany, underclimate change predict a decrease in growthof Scots pine and deciduous tree species dueto drought stress (Lasch et al. 2002). In somecases growth may initially increase when theclimate starts changing, but decrease as theclimate becomes even warmer. This has beenpredicted for Norway spruce in parts of Ger-many by Pretzsch & Dursky (2002) in simulat-ed scenarios with +1°C and +3°C temperatureincreases.

It has been shown that elevated CO2 con-centrations can increase water use efficiencyin plants, largely through changes in stomat-al conductivity (Eamus & Jarvis 1989). Thiscould to some extent counteract the negativeeffects of drought stress if summer precipita-tion decreases in the future. For conifers,however, the stomatal conductance seems tobe less affected by increased CO2 concentra-tions (Medlyn et al. 2001). Elevated CO2 con-centrations may also increase tree growthdue to a carbon “fertilization” effect (Eamus &Jarvis 1989). Simulations for Scots pine for-ests in southern Finland (Väisänen et al 1994)indicate that an increase in temperature of+5°C will reduce the soil water. At the sametime photosynthesis will be enhanced by upto 6–8% under current CO2 concentrationsand up to 8–10% under doubled CO2 concen-trations. Total stem wood production in-creased up to 6% when all the effects wereincluded in these simulations. The water-use

efficiency of the Scots pine ecosystems waspredicted to increase by up to 3%. Simula-tions on a Norway spruce stand in Norwaypredicted an increase in NPP of 7% with atemperature increase of +4°C and of 36% witha doubled CO2 concentration (Zheng et al.2002). When both temperature and CO2 wereincreased the NPP increase was predicted tobe 49%.

Recent simulations of the possible effectsof SweClim’s B2- and A2-scenarios on the pro-duction of Scots pine and Norway spruce innorthern Europe have been made within theHeureka programme using the process-basedgrowth model BIOMASS. Simulations havebeen generated for three different age class-es: young, middle-aged and old stands. Re-sults for the A2-scenario indicate an increasein NPP of 30–50% in southern Sweden foryoung stands of Norway spruce and Scotspine and 20–30% in central and northern Swe-den (figure 7 a–c). The water availability inthe soil in young stands is sufficient and notlimiting for photosynthesis and growth, ac-cording to the model. However, for middle-aged stands of Norway spruce the increase inNPP is only 0–20% southern Sweden, mainlydue to increased water limitations, since thedemand from a middle-aged stand is consid-erably greater than that of young stands. Theeffect of water in middle-aged stands is weak-er for Scots pine than for Norway spruce. Theincrease in NPP in middle-aged stands for therest of Sweden is approximately 20–30%. Oldstands of Norway spruce also show a largeincrease in NPP in southern Sweden but lessthan young stands. Water also limits thegrowth in old stands, but to a lesser extentcompared with middle-aged stands. Oldstands of Scots pine have no water limitationsand the increase in NPP is 30–50% for most ofSweden. The increase in NPP for the A2- sce-nario is approximately 10% higher for thewhole of Sweden, compared with the B2-sce-nario. The BIOMASS model is adapted to aboreal climate, but includes no feedback


21


a b c

Figure 7. Relative change in NPP Norway spruce for the A2-scenario, compared with simulations of thepresent climate: (a) young stands, (b) middle-aged stands, (c) old stands.

Regional simulations of net primary pro-duction (NPP) in Scandinavia and theBaltic countries have been conductedpreviously for both Norway spruce andScots pine, using two of SweClim’s recentregional climate scenarios. Results fromthese simulations are preliminary and theparameterisation of models can be devel-oped further. Results from the simulationsare only valid for a mesic sandy-siltymoraine, and they cannot therefore begenerally applied to dry or moist soil con-ditions. Furthermore, the model used inthese earlier simulations includes no feed-back mechanisms to soil processes, suchas mineralization and nutrient availability,which might introduce biased estimates ofNPP. However, several models are nowavailable that include such feed-backmechanisms. Sensitivity analysis for differ-ent climatic and other variables (e.g. pre-cipitation, temperature and leaf area)

would also be valuable for interpreting theresults. Furthermore, the long-term effectson production of adverse events, such assevere outbreaks of harmful invertebrates,micro organisms and fungi or windthrows,extreme drought and frost damage in plan-tations can be incorporated in process-based models by integration with otherstudies in the abovementioned researchareas. Future research needs:• Further development of the parameter-

isation and validation of the model• Inclusion of different soil moisture con-

ditions (dry, moist and wet)• Inclusion of feed-back mechanisms to

soil nutrient dynamics in the models• Sensitivity analysis for important climat-

ic variables and parameters• Incorporation and evaluation of long-

term effects on calamities and produc-tion.


22


mechanisms, linking climatic variables to soilprocesses. This might have led to biased es-timates of NPP.

A higher temperature sum generally has apositive effect on wood density of Norwayspruce and Scots pine in Sweden (Wilhelms-son et al. 2002). Wood density is also knownto be relatively high in south-eastern Swedenwhere summers are dry. This implies that cli-mate change could lead to higher averagewood densities in our two major conifers,especially in southern and central Swedenwhere drier summers are predicted for thefuture.

In conclusion, the net primary productionfor Norway spruce and Scots pine is predictedto increase by 0–50%, depending on site andstand age, as a response to climate change.Other tree species will most likely react ina similar way. The importance of water asa growth-limiting factor may increase inmiddle-aged stands in southern Sweden.

Impact on other organisms ofimportance to the treesInvertebratesMost of the invertebrate species that areserious pests to trees are insects. Evans et al.(2002) have reviewed the effects of climatechange on forest insect pests in the UK andfound that several different effects are likelyto interact. Thus, making accurate predic-tions is not straightforward because of thecomplexity of the system. The diversity in life-history strategies among insect herbivoressuggests that they will be affected in differentways by global change (Bale et al. 2002). Cli-mate changes are likely to affect in diverseways the survival and reproduction of theinsects, the natural enemies of the insectpests, the nutrient content of the host trees,the vigour and defence capabilities of thehosts and the phenological synchrony of thepest and host (Evans et al. 2002).

The complexity of the effects of climatechange on the interactions between insectpests and trees has been illustrated by stud-ies on the spruce budworm (Choristoneurafumiferana) published by Fleming and Volney(1995). They have found, for instance, that ahigher CO2 concentration will reduce the ni-trogen concentration in the needles of thehost and thus the food quality for the insectwill deteriorate, while more drought stresswill increase the sugar concentration in theneedles and improve the food-quality.

Volney and Fleming (2000) have reviewedthe impact of climate change on three defoli-ating insect species that are major forestpests in North American boreal forests:spruce budworm, jack pine budworm (Chor-istoneura pinus) and forest tent caterpillar(Malacosoma disstria). They concluded thatoutbreaks of these species are likely to in-crease in frequency and intensity under cli-mate change, particularly in the margins ofthe host tree ranges.

The autumnal moth (Epirrita autumnata)has been subjected to several studies that arerelevant to insect herbivore’s responses toclimate change. The autumnal moth recur-rently defoliates vast areas of mountain birchforests in the Scandes and Northern Fenno-scandia, where it has been suggested thatwinter temperatures will be raised the mostby the anticipated climate warming. The spe-cies is also found outside the mountain birchforest but no outbreaks (defoliations) havebeen recorded in other habitats. The dynam-ics of northern populations of the autumnalmoth are heavily influenced by larval parasi-toids, which cause oscillations in the moth’spopulations that are more or less synchro-nised at a landscape or regional scale (Bylund1997). Defoliating outbreaks of the moth, to-gether with climate, have been suggested tobe the major driving forces of the dynamics ofthe mountain birch forest (Tenow et al 2001).It is well established that the winter minimumtemperature is a crucial factor delimiting the


23


moth’s distribution in the north. Incidencesof low winter temperatures and other types ofunusual weather patterns also affect the tem-poral population dynamics of the moth. It hasbeen suggested that a warmer climate mayboth increase the distribution of outbreaksinto new areas and affect the frequency ofpopulation peaks (Tenow 1996, Virtanen et al1998, Bylund 1999). Some studies suggestcounteracting effects may occur, e.g. higherwinter survival rates of eggs combined withlower survival rates of larvae due to highermortality rates caused by parasitoids andother natural enemies in the summer (Nie-melä et al 2001, Virtanen & Neuvonen 1999).

The importance of green spruce aphid(Elatobium abietinum) as a major pest to Sitkaspruce is predicted to increase in Scotland ifthe climate becomes warmer (Straw 1995). Ifthe climate becomes more humid, however,fungal diseases and pathogens attacking theaphid may have a controlling effect, counter-acting the possibility of damage increasing.Green spruce aphid is present in southernSweden, but the damage it does to Norwayspruce and Sitka spruce are very limited to-day. However, its seriousness could increasefor these reasons in a warmer climate.

Low winter temperatures limit the north-ern range of some insect pests. For instance,the critical winter temperature for egg mortal-ity of the European pine sawfly (Neodiprionsertifer) is -36°C (Austarå 1971). Virtanen et al.(1996) have predicted increases in the fre-quency of outbreaks in eastern and northernFinland as a response to warmer winters dueto climate change.

One of the major insect pests in Swedishforests is the spruce bark beetle (Ips typo-graphus), outbreaks of which occurred in theNordic countries after warm summers in thebeginning of the 1970:s (Löyttyniemi et al.1979) and in 1976 (Bakke 1983), but after cool-er summers in the late 1970:s the populationsdecreased (Bakke 1983). This implies thatclimate change with predicted warmer and

drier summers in southern Sweden couldincrease the frequency and severity of out-breaks of the spruce bark beetle. A decreasein the vigour of Norway spruce subjected tostress due to warm winters at its southernrange limit could further exacerbate the situ-ation. More frequent damage by the sprucebark beetle due to climate change has alsobeen predicted for the UK (Anderbrant 1986,Evans et al 2002).

The most serious insect pest in Sweden isthe large pine weevil (Hylobius abietis), due tothe mortality it causes to planted seedlings.No predictions have been published aboutthe possible impact of climatic changes onthe large pine weevil as a pest in forest plan-tations. However, the large pine weevil ismore common in areas with warmer anddrier summers. Bejer-Petersen et al. (1962)have studied this issue, and comment that itis difficult to draw general conclusions aboutthe climatic preferences of the species. Thelarval period varies between one and threeyears in Sweden, and since developmentaltime in insects is directly affected by temper-ature, a change in climate could alter thelength of the larval period. Possible conse-quences are difficult to predict.

Other invertebrates may also occur asforest pests. The pinewood nematode (Bur-saphelenchus xylophilus) is native to NorthAmerica and has caused pine wilt diseasewith severe tree mortality since it was intro-duced to Japan. The pinewood nematode hasrecently established populations in Portugaland it has the potential to spread over mostof Europe, but wilt disease on living trees areonly expected to occur in areas where themean July or August isotherm is >20°C andepidemic wilt is only likely at temperatures>24°C (Evans et al. 1996). If climate changeresults in a temperature increase of the mag-nitude predicted by the most extreme recentscenarios (Räisänen et al. 2003) pine wilt dis-ease could occur in southern Sweden. How-ever, this is also dependent on the pinewood


24


nematode spreading to Sweden and thefuture occurrence of beetles of the genusMonochamus, the vector that carries the nem-atode from tree to tree (Linit 1988).

Most insect species have good dispersalability and their range-limits can be predictedto change simultaneously with climaticchanges. Changes have already been ob-served in the distribution of native Europeanbutterfly populations, with northern rangesextending and southern ranges contracting(Virtanen & Neuvonen 1999). The same ef-fects are likely to be apply to forest insectpests (Evans et al. 2002). The range limits ofthe tree species hosting the insect pests willmove much more slowly, and these differenc-es in migration rates may cause directionalchanges with time in damage pattern. Theshort-term response to a warmer climate maybe that the ranges of the pest and host part-ly separate, with a consequent reduction in

pest damage, while in the long term, as thehost tree species migrate into the new rangeof the insect pests, the damage may increaseagain. The importance of this type of interac-tion due to different parts of the ecosystemresponding at different speeds to climatechange has been emphasised by Shaver et al.(2000).

Insect species, especially those occurringin periodic outbreaks, have a large potentialfor genetic adaptation to new environmentsdue to the huge numbers of individualsinvolved and their short generation times(Fleming and Volney 1995). Natural selectionis therefore likely to play a major role in adap-tation of insect species to climate change, forinstance in maintaining the phenological syn-chrony between insect pest and tree hosts.

As mentioned above, the complexity of theinsect pest systems makes it difficult to makeaccurate predictions or generalisations about

The main reason for the uncertaintiesabout the effects of climate change oninsect damage is that the host plant andthe pest insect’s natural enemies will beaffected by global change as well as theinsect itself (Bale et al. 2002). It is essential,therefore, to understand how importantinteractions – those determining densityand population trends – will be affected.Moreover, different types of pest insectsare likely to differ in their responses. Toaccount for these differences we need toidentify and pick out suitable model sys-tems, representing different categories ofinsects.

To improve predictions about possiblefuture insect pest problems there is a needto concentrate research efforts. We suggestthat future research should be focused ona set of selected insect species. Each select-ed species should represent a larger groupof known or potential pest species, and a

substantial body of basic biological know-ledge should be available about it. In addi-tion, the species selected should prefera-bly represent a wider range of taxonomicand ecological groups.

Regarding the types of studies thatshould be undertaken, we believe that thebest strategy is to include a combination oflong- and short-term studies complement-ed by modelling efforts. Further studiesshould include, for instance, investigationsof long-term changes in population densi-ties and their correlations to climatic data,replicated at several sites, and short-termexperiments focussing on important mech-anisms and interactions identified in empir-ical studies, are needed. The results fromthese experiments and observations shouldbe used to continuously develop and eval-uate population models. A valuable use forsuch models is to study changes in therisks associated with insect pests.


25


the impact of climate change on insects. Har-rington et al. (2001) reviewed this matter andfound that studies focusing on insect pestspecies often predict increases in the popula-tions, their range-limits and the damage theycause, while studies focusing on rare andendangered species often predict decline andincreased threats to the studied insect popu-lations. Harrington et al. (2001) also suggestthat studies reporting catastrophic scenariosare over-represented in the scientific litera-ture because studies reporting little or nochange receive less attention and they aremore difficult to get published. This calls forcaution when discussing the effects of insectpests on forests in a changing climate.

In conclusion, damage due to insect pestsmay increase somewhat in a warmer climate.The range limits of some insects may alsoexpand. However, it is difficult to make accu-rate predictions because of the complexity ofthe system, involving climate, host-trees, in-sect pests and their natural enemies.

VertebratesThe tree species composition of young foreststands in Sweden today is mainly dependenton two factors. First, forestry, which favourscertain species like Norway spruce and Scotspine in regeneration and early thinning. Sec-ond, the interaction between browsing pres-sure of moose and deer and the palatability ofthe different tree species. High browsing pres-sure during the last 20 years has promotedthe planting of low palatability species likeNorway spruce and lodgepole pine (P. contor-ta) on large areas that would otherwise beregenerated with Scots pine or broadleavedspecies. Browsing by moose in young Scotspine stands is considered to be the largestsilvicultural problem in Sweden today (Fri-berg pers. com.). Climate determines theranges of tree species, but plays a minor rolein determining the species composition of theforests established today. Continuing largepopulations of large herbivores are likely to

have a similar impact on species compositionin the future, so the impact of climate on theherbivores will be important.

Moose (Alces alces) has a circumpolar dis-tribution, and its southern range-limit is main-ly determined by temperature. In wintercoats moose become stressed by tempera-tures higher than 5°C, while in summer coatsthey experience stress at temperatures of14°C or higher (Karns 1997). During warmweather they normally seek shade and water.The southern range-limit of moose in Europetoday is mainly influenced by human inter-vention, but the climatic limit is considered tobe at approximately latitude 50°N. In a futurewarmer climate, moose may very well suffersevere temperature stress in southern Swe-den and gradually disappear from theseareas. The decline of moose populations inOntario due to climate change has beenpredicted by Thompson et al. (1998).

Roe deer (Capreolus capreolus) is distribut-ed over the whole of Sweden, but populationsin northern Sweden have a low density and inmany areas they are highly dependent onwinter feed supplied by humans. A warmerclimate with shorter winters and thinnersnow cover will most likely increase the pop-ulation densities in central and northern Swe-den. Red deer (Cervus elaphus) and fallowdeer (Dama dama) have scattered distribu-tions in southern and central Sweden, but areslowly increasing and may in the future havea continuous distributions over large parts ofSweden. The effects of a warmer climate onthese species can be expected to be similar tothose for roe deer. However, roe deer havehigher dispersal abilities and may respondmore rapidly to the changing climate.

The future population densities of largeherbivores will also be strongly affected bypopulation management measures taken byhumans, and changes in the populations oflarge carnivores. The amount of damage bylarge herbivores on trees may also change ifclimate change affects the quantity and qual-


26


ity of alternative forage species like herbs anddwarf-shrubs. The browsing by moose androe deer on trees is partly determined by thelength and depth of the snow cover. With lesssnow the animals feed more on field-layervegetation and less on trees (Cederlund etal.1980). Snow cover is also the critical factorinitiating winter migration of moose in north-ern Sweden, and reductions in the duration ofsnow cover could make browsing damage inthe winter ranges less severe (Ball et al. 1999).

In conclusion, the ranges of roe, red andfallow deer are likely to expand northwardswhile moose populations may decline insouthernmost Sweden if the climate becomeswarmer. The impact of populations of largeherbivores is likely to be at least as importantfor the regeneration success of trees in thefuture as it is today. Population patterns oflarge carnivores and management by huntingare the two main factors influencing deer andmoose populations besides the climate.

Micro-organisms and FungiFungi comprise the major group of micro-organisms that cause severe disease to forest

trees. In addition, certain bacteria and, to alesser extent, viruses and mycoplasma-likeorganisms are known to be disease agents.There are also a number of complex diseasesor decline syndromes of trees where thecausal factors include several interactingdisease agents, both biotic and abiotic. Thenutritional mode of the microorganisms is ofimportance when considering their impact inforests and the potential influence of climat-ic change on disease development. Biotroph-ic pathogens, such as rusts and mildews, relyon living host cells for their nutrition, andthus may frequently be favoured by environ-mental conditions promoting vigorousgrowth of the host. Necrotrophic pathogens,on the other hand, kill host tissues beforefeeding on dead cells. These pathogens areoften favoured in their development by con-ditions that stress the host organism. This isespecially important when considering op-portunistic, relatively weak pathogens thatcan cause major disease symptoms whendeveloping on hosts stressed by environmen-tal conditions, while perhaps only causing neg-ligible effects to a vigorously growing host.

In the context of forestry and potentialdamage to the trees, the most importanteffects of climate change on herbivores arelikely to be changes to the populations andenvironments of cervids, wild boar and(possibly) hares. From the present climat-ic scenarios, several aspects of the tree-large herbivore system can be identifiedthat deserve special attention:

• The impact of expanding or shrinkingtree species ranges (with accompanyingvegetation changes) and the implica-tions for the distribution of herbivoresand population densities.

• The impact of summer weather on forestrotation time, food quantity, food qualityand the implications for herbivore popu-lation dynamics and damage patterns.

• The impact of winter weather on migra-tion, habitat use and foraging patterns,together with the implications foranimal distributions at regional andlandscape scale, the relative amount offoraging in different vegetation layersand the impact of large herbivores onthe migration rate of woody species.

Man and other predators are confoundingfactors.


27


The principal assumptions about the com-plexity of the pest-host-environment inter-actions discussed above are also valid formicrobial and fungal pathogens. Most patho-gens with that disperse via spores will rapid-ly respond to climate change via adjustmentof their range limits and frequency of out-breaks.

The fungus Heterobasidion annosum (rootand butt rot) is the most economically seri-ous disease in Swedish forests. Its spore pro-duction and dispersal are generally promotedby increased temperatures. Dry weather alsofavours dispersal. Low winter temperaturesand snow decrease spore production and dis-persal (Redfern & Stenlid 1998). This impliesthat the spread of H. annosum after thinningwill be favoured by climate change, especial-ly in southern Sweden. It is possible that se-vere damage, which is currently restricted tothe coasts of southern Sweden, may start toaffect larger areas of southern and centralSweden (Stenlid pers. com.).

A number of different inter-sterility groupsof H. annosum are known to occur, with differ-ing host and regional ranges. The S-group,which mainly attacks Norway spruce, is dis-tributed over the whole of Sweden. In south-ern Sweden, the P-group attacks Norwayspruce, Scots pine and other tree species(Korhonen et al 1998). The absence of the P-group in northern Sweden and the northernparts of neighbouring countries implies thatthe range limit may be set mainly by climaticfactors. A warmer climate could thus imply anorthward spread of the P-group with Scotspine being attacked, especially on sandy soilswith relatively high alkalinity (Stenlid pers.com.).

Root disease cause by Armillaria spp. oc-curs across the entire country on several treespecies, including the economically impor-tant conifers. This pathogen mainly infectstrees that have a reduced vitality due to somekind of stress. Drought stress is reported tomake trees more susceptible to Armillaria

infection (Wargo & Harrington 1991). A futureclimate with warmer and drier summers islikely to promote Armillaria and cause in-creased damage by this pathogen.

Outbreaks of Gremmeniella abietina haveoccurred regularly in Sweden. Two differenttypes of disease can be distinguished. Insouthern Sweden attacks are mainly confinedto middle-aged stands of Scots pine, while innorthern Sweden it occurs in young stands ofScots pine and lodgepole pine, mainly attack-ing the lower parts of the tree that are cov-ered by snow during winter. There are alsoindications of corresponding ecotypic differ-entiation within G. abietina (Hellgren 1995).

The main factor promoting outbreaks of G.abietina is rainy, cool and cloudy weatherduring the growing season, which increasesspore dispersal and survival rates of the path-ogen. The opposite conditions, warm andsunny summers, are not conducive for infec-tion and promote canker healing (Hellgren1995). This implies that outbreaks of G. abie-tina will be less frequent in the future warm-er and drier climate predicted by the currentscenarios. The northern disease type is de-pendent on deep and long-lasting snow cov-er (Hellgren 1995), and a warmer climate islikely to contribute to a reduction in damagecaused by G abietina in large areas of centraland northern Sweden.

Phacidium infestans cause damage toneedles and mortality, mainly to Scots pine,but also to lodgepole pine. The pathogen iscommon in northern Sweden. Deep and softsnow is a prerequisite for this pathogen, andwith warmer winters and less snow the dam-age it causes is likely to decrease in Swedenand totally disappear from some areas.

Some pathogens exploit frost damage toinfect trees. Lachnellula willkommii is knownto attack larch (Larix spp.) trees in this way. Ifthe frequency of frost damage changes in afuture climate scenario, the damage causedby this type of pathogen will probably de-cline.


28


Pathogen species that are not currentlypresent in Sweden may be able to invadesouthern Sweden in a future climate. Severalimportant species that cause severe damageare present in southern and central Europe,like some oomycete species of the genus Phy-tophthora, and have the potential to causediseases in a future, warmer Sweden.

The following general effects of changedclimatic conditions on the flora of micro-organisms affecting forests in Sweden can bepredicted.

1. Organisms for which dispersal is favouredby mild whether can be predicted to in-crease in importance. For example, thewindow for spore dispersal of Hetereoba-sidion annosum is expected to expand intothe winter period under milder climaticconditions.

2. Organisms that are disfavoured by thewarmer conditions are likely to decline inimportance, e.g. the snow mould Phacid-ium infestans, which requires snow coverin order to infect conifer seedlings effi-ciently.

3. Organisms (e.g. Armillaria spp) that arefavoured by the stress conditions createdby extreme weather conditions, such asdrought will tend to increase in impor-tance.

4. The distribution ranges of some organ-isms that require warmer conditions thanare currently found in Sweden may expandfrom continental Europe into Scandinavia.This group includes organisms like Spha-ropsis sapinea, which causes shoot blight,and Phytophthora spp which causes rootrot.

Ground vegetationGround vegetation of herbs, grasses, fernsand shrubs is an important factor determin-ing the regeneration success of tree species,for natural regeneration as well as for plant-ing or seeding in silvicultural systems. Com-petition between ground vegetation and treeseedlings is generally more severe on fertilesites than on poorer ones. There is also anorth-south gradient, where the heaviestcompetition from ground vegetation occursin southern Sweden and declines to thenorth. In a warmer climate, both the climateitself as well as possible increases in nutrientavailability may increase the competitivepressure on tree seedlings. Intensified compe-tition may be due to increased pressure frompresent ground vegetation species, as well asthe invasion of new species promoted by thewarmer climate.

Research should be directed towardsunderstanding how trees react to milderwinter conditions. Several shoot- andneedle- infecting fungi developing duringthe period when the trees are dormant.This is likely to be affected under thecurrently considered climatic scenarios.The interplay between dormancy of thetissues and fungal development is thus of

great interest for predicting future diseaseoutbreaks.

Another area for future research is theinterplay between stress episodes and thedevelopment of decline symptoms ontrees. It is of crucial importance to under-standing how pathogenic fungi are trig-gered under these conditions.


29


Tree and stand factors ofimportance to the impact ofclimate changeTree speciesThe range limits of tree species will change asclimate changes, as discussed above. In plan-tation forestry, tree species are often usedoutside their natural range and this will prob-ably also be the case in the future. The effectsof climate change on the suitability of fourplantation tree species in the UK have beenreviewed by Ray et al. (2002). They concludethat the suitability for plantation forestry ofSitka spruce,

Corsican pine (Pinus nigra), Douglas fir(Pseudotsuga menzieesii), and beech (Fagussylvatica) will all increase in Scotland. In largeparts of England, in contrast, Sitka spruceprobably will cease to be suitable, whileDouglas fir will remain suitable and the suita-bility of Corsican pine will increase. They alsopredict that beech may not be suitable as atimber crop in parts of southern England.Norway spruce is predicted to be more proneto “top-dying” damage in Scotland and tocease being a productive species in England(Redfern & Hendry 2002). Thompson (1998)predicts that the effects of climate change onthe suitability of Sitka spruce as a plantationspecies in Ireland will be minimal. Other spe-cies, like Norway spruce and Scots pine, willprobably be adversely affected.

Norway spruce is currently the main plan-tation species in southern Sweden. The rangelimit of Norway spruce is predicted to move

northwards in response to climate change(Bradshaw et al. 2000). Whether Norwayspruce will continue to be a suitable planta-tion species in the southernmost parts ofSweden and in coastal areas cannot be pre-dicted with confidence, but it is likely to con-tinue to be suitable somewhat outside its nat-ural range. Other species may very well proveto be more suitable than Norway spruce insome parts of southern Sweden in the future.

The number of species both native andintroduced that are suitable for plantationforestry is highest today in southern Sweden,while there are fewer species to choose fromin northern Sweden. A warmer climate is like-ly to make a number of species, like nativehardwoods and exotic conifers, more suitableand potentially interesting as forestry speciesin a larger part of Sweden. This may increasethe species diversity in managed forests, atleast at the landscape scale.

In both naturally regenerated forests andplanted forests the competition between dif-ferent tree species may be affected by climatechange. The relative competitiveness of somespecies will change, and this has implicationsfor future species composition and standstructure of mixed forest stands. Kellomäki &Kolström (1992) have predicted that the ear-ly height growth of silver birch (B. pendula)will increase more than that of Scots pine insouthern Finland as a response to climatechange. This implies that early managementwith pre-commercial thinning will be moreimportant in the future to promote the devel-opment of stands.

There is a need for more knowledge aboutbiomass production and management of anumber of species of potential interest forforestry in a warmer climate. Knowledgeabout altered competition patterns be-tween species are also of great importance.

A better understanding of the physiologicaland ecological factors determining thesouthern range limit of Norway spruce aswell as range limits for other species wouldalso be desirable.


30


Genetic variationMost of the studies reviewed above make pre-dictions about the effects of climate changeson different species. However, not only areresponses of different species likely to differ,the responses of different populations andgenotypes within species will also differ,although these differences are likely to begenerally smaller than between species.Large genetic variations between populations(provenances) of forest trees clearly occurfor many traits of adaptive significance. With-in populations, large genetic variations ingrowth and biomass traits in response to dif-ferent temperatures and water availabilityhave been reported for young seedlings ofScots pine (Sonesson & Eriksson 2000) andNorway spruce (Sonesson et al. 2002).

The genetic variation within species willallow seedling material to e selected for re-forestation in the future that is better adapt-ed to the future climates (Thompson 1998).However, trees that have already been plant-ed today will probably be growing in climatesthat they are less well adapted to in the laterpart of their rotation than the earlier stages.Studies on provenance trials indicate that thegrowth reduction, compared to the optimalprovenance, in a future warmer climate willbe in the range of 5–10% for Norway spruce(Schmidtling 1994) and Scots pine (Persson1998). The non-optimal adaptation of oldtrees in the future may also cause a loss ofvigour, resulting in increased susceptibility topests and diseases. The magnitude of thisrisk cannot be easily predicted, but it will in-crease if the climate changes are rapid.

Evolution based on natural selection in re-sponse to environmental changes tends to beslow because of the long generation time oftrees. Trees therefore have to rely on theirphenotypic plasticity to withstand variationsin weather, which can be considerable be-tween years. Wind-pollinated tree specieswith a large distribution range, like Norwayspruce and Scots pine are considered to havehigh phenotypic plasticity for many adaptivetraits (Eriksson & Ekberg 2001). The largephenotypic plasticity will be beneficial for thecapacity of trees to survive and thrive in achanged future climate, within certain limits.

Management strategiesRisk managementAn analytical framework for risk managementin relation to climate change is needed to helpdecision makers attain their goals in accord-ance with their values and to help the re-search community provide useful decisionsupport. Some components of such an analyt-ical framework will be described below, but amore detailed description requires addition-al research. For example, it is important tounderstand how different groups of decisionmakers perceive uncertainties and risks relat-ed to climate change. At least two levels ofuncertainty can be identified. First there isuncertainty about whether or not the climateis changing and what the climate will be likein the future. Secondly, there are uncertain-ties about how climate changes will affect theforest and what forestry measures will be

Research should focus on identifying adap-tive traits of importance in a changingclimate, and the genetic variation andphenotypic plasticity associated withthese traits. Studies are needed on themost important forestry species today and

on new species of potential interest for thefuture climate. The genetic control of pestresistance also needs to be urgentlyresearched, both for the common conifersas well as a broader range of potentialforestry species.


31


useful. Decision makers’ perceptions of theseuncertainties affect their forestry decisionsand, consequently, the state of the forests.Thus, it is important to distinguish betweendirect effects of the climate and indirecteffects of forest management in order tounderstand the effects of a changing climateand, not least, to allow the decisions to beimproved.

The analytical framework is intended toassist decision makers in their risk analysisand to help them choose the appropriatestrategy to attain their goals, in accordancewith their values. For this, the risk factorsperceived as being most important by eitherdecision makers or the research communityneed to be identified. Research is needed tohelp identify such risk factors and the ques-tions that most need to be answered. Therisks may be due to deficiencies in the know-ledge base or to uncertain events. For each ofthe identified risk factors the research com-munity can help answer questions. To maxim-ise their utility, the answers should be pre-sented in a format that facilitates decisionmaking. The Nature of such decision supportwill depend on the risk factor under consider-ation. Research is needed to assist the com-munication of risk between the research com-munity and the decision makers, as well asbetween groups of researchers and betweengroups of decision makers.

Genetic materialThe period that is most important for treeseedlings is in the first years during and

immediately after establishment. Therefore,the general recommendation for selectinggenetic material for regeneration purposes isto use material that is well adapted to theclimate it will experience during the firstyears (Sonesson 2001). Planting seedlingsthat are adapted to the predicted warmerclimate may lead to high rates of mortalityand few seedlings living to experience theanticipated future climate.

In order to provide forestry with well-adapted material for future regeneration inthe rapidly changing climate, it is of crucialimportance to have a well-designed breedingprogramme. The Swedish breeding pro-gramme is integrated with a dynamic geneconservation programme and provides scopeto maintain and develop the amount andstructure of genetic variation needed for fu-ture selections (Danell 1993). The Swedishbreeding programme is focused on Norwayspruce, Scots pine, lodgepole pine and silverbirch. Other species are subjected to minorintermittent breeding efforts.

Besides managing genetic variation andstructure, the breeding programme has otherfeatures that enhance the scope for develop-ing and selecting well adapted genetic mate-rial for the future.• The generation time in tree breeding is

generally 20–30 years, which is less thanhalf the generation time in nature, so theadaptation process is faster in the breed-ing population than in nature.

• Nature always selects for the currentclimate. In breeding programmes we can

In conclusion, the principal motive forconducting research related to risk man-agement in the context of climate changeis to help make better forestry decisions.For this we need to better understandboth the direct effects of climate changeon the bio-physical system and its indirect

effects mediated via the social system. Aset of key terms summarising the issuesinvolved would include: analytical frame-work for risk management, risk perception,understanding effects of climate change,risk factors, decision support, risk commu-nication and decision strategy.


32


select for a future predicted climate bytesting in artificial or natural conditionsthat resemble the predicted climate.

• In the breeding programme each genotypeis tested in the field at several sites withdifferent climates. This provides know-ledge that can be used for future recom-mendations on the transfer of geneticmaterial.

Species selectionIf drastic climatic changes occur in the futureit may be impossible to find well adaptedmaterial of our main forestry species. Switch-ing to species better adapted to the new cli-mate would probably be the best option un-der such circumstances. Southernmost Swe-den is the area where this developmentwould be most likely to occur, and Norwayspruce the most likely species to be affected.To supply the spruce-based industry in theregion with similar raw material, plantations

of Sitka spruce could be appropriate options.Other alternative tree species could be foundamong native hardwoods.

Another relevant reason for switching thespecies used is that climate change may alterthe relative competitiveness among species.This could help make species other thanthose used today commercially attractive.Changes in soil moisture due to changes inprecipitation and evaporation could changethe choice of species on certain sites. Lessprecipitation in the summer in southern Swe-den could, for instance, make Scots pinemore suitable on some sites that at presentwould be best suited for Norway spruce.

Regeneration practicesThe likelihood that summer droughts willbecome more frequent has implicationsfor regeneration practices. Site preparationmeasures involving the creation of patches ormounds of mineral soil are commonly used

Our most economically important treespecies are managed in the breedingprogramme. Therefore, they offer the bestopportunities for rapid adaptation andknowledge-based material transfer. Treespecies of minor economic importancereceive less attention from forestry andmay have greater difficulties in adapting toa rapidly changing climate. Since climatechange will probably make forestry withsome alternative species a viable option,at least in southern Sweden, it could be

The use of a broader range of species couldoffer a means for forest landowners tospread the risks of an uncertain future.A warmer climate will likely increase thenumber of species that are suitable forforest production in most of Sweden. Trees

appropriate to allocate more of the breed-ing efforts to these species.

Breeding-related research should focus on.• A wider range of species than today• Methods for genetic testing for future

climate scenarios• Studies of transfer effects and G × E inter-

actions• Methods and scope for resistance

breeding

can be mixed within stands or as a mosaicof monocultures of different species. Re-search should consequently allocate moreresources to forestry species that may beinteresting in the future.


33


today. One of the beneficial effects of thispractice is that it improves water availabilityfor the seedlings. In a future, drier summerclimate thorough site preparation would be-come even more important than today. Moreintensive site preparation may also be need-ed to control ground vegetation since it mayotherwise compete increasingly intenselywith the seedlings for water, nutrients andlight. Chemical control of competing vegeta-tion is commonly used in plantation forestryin warmer countries and this could be neces-sary in Sweden as well, if alternative methodsare not developed.

Direct seeding is used in some areas today.The germination of the seeds are very poorduring droughts and the method is currentlynot recommended on dry sites in south-east-ern Sweden. If future summers will becomedrier, this could lead to a decrease in the areathat is suitable for direct seeding.

The future need to establish forests withspecies and genetic materials that are adapt-ed to the new climate will probably lead to anincreased use of artificial regeneration, main-ly planting. Natural regeneration with the oldmaladapted seeds from the old stands willprobably not be a viable alternative. How-ever, natural regeneration will probably bepossible with several species in areas whereclimate currently limits seed production, e.g.Scots pine in high altitudes in northernSweden and beech in southern and centralSweden.

Stand managementClimate change will change the site condi-tions on forestland over the whole of Sweden.This will change the relative competitiveness

of the tree species. For southern Finland,Kellomäki & Kolström (1992) have predictedthat silver birch will become more competi-tive relative to Scots pine in a future climateand they conclude that more intense cleaningof broadleaves will be necessary if Scots pineis to be used as a production species in thefuture. Changes in competitiveness amongspecies would likely have effects on standmanagement in most mixed species stands.

The most significant change in stand man-agement is likely to be towards shorter rota-tions, for several reasons. First, the increasedproductivity of the forest stands will make theoptimal rotation time shorter. Second, thegenetic material planted today will not beoptimally adapted to the future climate and itcould be desirable to replace it with betteradapted material to exploit the productivityof the site. Third, the non-optimal adaptationof the trees in the later part of the rotationmay lead to damage that necessitates an ear-lier final harvest than planned. Shorter rota-tions due to increased productivity can in-crease the economic return of investments insilviculture. This may result in more intensivestand establishment and management meth-ods being deployed in some regions and bysome landowners.

Stand management may also be affectedby calamities like windthrow and attacks bybark beetles or root rot becoming morefrequent, as expected in some areas. Man-agement measures to foster forest health mayinclude early and hard thinning to promotestability and vigour in the trees, stump treat-ments against rots, and restrictions on theamount of wood left in the forest afterharvest.

Research should focus on regenerationmethods adapted to dry summers anddrought spells. This includes developmentof site preparation methods, seedling mate-

rial and after-planting treatments. Methodsshould also be adapted to the possibilitiesof increased competition from ground veg-etation.


34


Health monitoringIn a rapidly changing climate, monitoring thehealth of the forest is very important in orderto take necessary actions to prevent adverseeffects. The health of the Swedish forests iscontinuously monitored by the SwedishNational Forest Inventory. A system of sampleplots is assessed at regular intervals and dataabout the sites and trees are recorded. Healthis also monitored by the National Board ofForestry through surveys among their localforestry districts. The monitoring system isprobably satisfactory, but there may be aneed to monitor additional traits and to fol-low up carefully the results of the inventoriesin the future.

QuarantineTo prevent new pests and pathogens enteringthe country that could establish in Sweden ifthe climate becomes warmer, quarantine reg-ulations can be used. Quarantine regulationsare more likely to be effective in Sweden,which has a fortunate geographical locationin this respect, with a sea border separatingit from neighbouring countries to the south,than they would probably be in many othercountries.

ConclusionsThe most likely effects of climate change ontrees and forests are summarised in Table 1.The general prediction is that potential bio-mass production, the opportunities to grownew species commercially and the risks ofseveral kinds of damages will all increase. Itseems that climate change will offer new

opportunities to forestry, while increasing therisks of calamities occurring (and/or makingthem more difficult to anticipate and thushandle) demanding new approaches toforest- as well as risk-management.

It should be considered that most of thestudies referred to in this report are based onrather old climate scenarios and, further-more, the scenarios used vary considerablyfrom one study to another. The studies oftenconsider only one factor, often temperature.In rare cases a few factors and their interac-tions have been considered, but virtually allof the studies published so far have failed (orbeen unable) to consider all of the climaticfactors expected to change with increasedCO2 concentrations in the atmosphere. Some

More research in this area is needed on:• The impact on tree health of different

management methods and logging oper-ations.

• Methods promoting intensive manage-ment and shorter rotations.

• Forest management in stands of alterna-tive species and mixtures of species.

Table 1. Likely effects of climate change on forestproductivity and health in Sweden in ashort to medium time span (20–100 years).+ = increase, - = decrease.

Southern Central Northern Changes in: Sweden Sweden Sweden

Tree species diversity ++ +++ +Wood production +++ +++ +++Damages caused by: Windthrow + + + Snow breakage - - - Forest fires + + ± Spring frost ± ± ± Autumn frost - - - Winter damage + ± ± Hardwood decline + + ± Drought ++ + ± Waterlogging + + ++ Invertebrates + + + Vertebrates ± + + Microorganisms

and fungi ± ± ± Ground vegetation + + +


35


of the more recent studies that have focusedspecifically on Swedish forests have startedusing the more consistent set of regionalclimate scenarios provided by SWECLIM.There have also been very few attempts tomodel the effects of transient climate changescenarios, despite the fact that in practice theclimate will change gradually.

Research needsThis literature review has revealed majorshortcomings in our knowledge about im-pacts that an expected climate change willhave on the forest ecosystems. Its potentialeffects on the structure and processes of theforest ecosystems are even more uncertainthan the nature and magnitude of the climatechange per se.

The reviewed literature contains indica-tions of the relevance and importance of abetter understanding of the climate-forest/forestry linkages. However, the study alsoidentifies three major obstacles that need tobe overcome in order to improve our under-standing of the issues, risks and possibilitiesrelated to the potential impact of continuedclimate change on forests and forestry:

1) Studies so far have generally addressedspecific aspect of the overall forestry/forest system, instead of exploring aspectsof the system as a whole and feedbackmechanisms (cf. Figure 5).

2) Studies published to date have differed intheir choice of climate change scenario.Thus, they refer to different changes intemperature, precipitation etc. making itdifficult to collate the findings and tosynthesis a coherent body of knowledge.

3) The transience of the anticipated climatechange has not been included in the stud-ies, as they typically refer to the effects ofa specified new climate regime. However,rather than switching instantaneously to a

new static climate sometime in the future,the forest and forestry will be experiencingcontinuous, ongoing changes in climate,implying that conditions will be constant-ly changing within a typical rotation peri-od, and from one rotation to the next.

Swedish forests are the prime raw materialfor the forest-based industry that is very im-portant nationally. The need for managementand the long rotation period for this renewa-ble raw material makes its increasingly im-portant to improve our knowledge baseabout effects of climate change, our scope totake measures early enough to counteractnegative effects and to take advantage ofopportunities for improved growth.

In parallel with the concern related to theraw material base, various human activities,such as land-use (including forestry) and airpollution have profound effects on the forestecosystems, severely impacting the structureand function of the forest ecosystems. Theimpact of climate change on factors such asbiodiversity and biogeochemical processeshas to be included in the future research.

Future research into the effects of climatechange on forest ecosystems has to take ac-count of a broad spectrum of scientific fields,but a multidisciplinary scientific approach isprobably essential. Obvious areas whichhave to be interactively considered include:

• The development of climate scenarios,including climate-forest interactions

• The effect of climate change on soil phys-ical, chemical and biological processes,including their relationships to treegrowth

• Biomass production and forest manage-ment strategies, including developmentand selection of genetic material.

• The effect of climate change on verte-brates, invertebrates, microorganisms (in-cluding fungi) and ground vegetation

• Risk and risk management


36


References

Alavi, G. 1996. Radial stem growth of Piceaabies in relation to spatial variation in soilmoisture conditions. Scand. J. For. Res. 11:209–219.

Anderbrant, O. 1986. A model for the temper-ature and density dependent re-emer-gence of the bark beetle Ips typographus.Entomologia Experimentalis et Applicata40: 81–88.

Auclair, A.N.D., Lill, J.T. & Revenga, C. 1996. Therole of climate variability and global warm-ing in the dieback of northern hardwoods.Water, Air and Soil Pollution 91: 163–186.

Austarå, Ö. 1971. Cold hardiness in eggs ofNeodiprion sertifer (Geoffroy)(Hym., Dipri-onidae) under natural conditions. NorskEntomologisk Tidskrift 18: 45–48.

Bakke, A. 1983. Host tree and bark beetleinteraction during a mass outbreak of Ips ty-pographus. Z. Angew. Entompl. 99:118–125.

Bale, J.S. et al. 2002. Herbivory in globalchange research: direct effects of risingtemperature on insect herbivores. GlobalChange Biology 8: 1–16.

Ball, J.P., Ericsson, G. & Wallin, K. Climatechange, moose and their human preda-tors. In: Animal responses to globalchange in the north. Hofgaard, A., Ball, J.P.,Danell, K. & Callaghan, T. (eds.). EcologicalBulletins 47. Lund. Sweden.

Barklund P. 2002 Ekskador I Europa. Skogs-styrelsen, Jönköping. Rapport No 1. InSwedish.

Bejer-Petersen, B., Juutinen, P., Kangas, E.,Bakke, A., Butovitsch, V. Eidmann, H.Hedqvist, K.J. & Lekander, B. 1962. Studieson Hylobius abietis L. I. Development andlife cycle in the Nordic countries. ActaEntomologica Fennica 17: 1–106.

Bergeron, Y. & Flannigan, M.D. 1995. Predict-ing the effect of climate change on firefrequency in the southeastern Canadianboreal forest. Water, Air and Soil Pollution82: 437–444.

Bergh, J., Freeman, M. & Räisänen, J. Effects ofglobal change on net primary production innorthern Europe – a model based analysiswith a regional resolution. Manuscript.

Bergh, J., Linder, S., Lundmark, T. & Elfving, B.1999. The effect of water and nutrientavailability on the productivity of Norwayspruce in northern and southern Sweden.For. Ecol. Manage. 119: 51–62.

Bergh, J. 1997. Climatic and nutritional con-straints to productivity of Norway spruce.Acta Universitatis Agriculturae Suecica.Silvestria 37. PhD-thesis. 34 pp.

Bergh, J., Flemberg, S., Kindberg, J., Linder, S.,Räisänen, J., Strömgren, M. & Wallin, J. 2000.Framtida klimatförändringar – tänkbara ef-fekter på den svenska skogen. Fakta Skog,SLU, Uppsala. Nr 13. 4 pp. In Swedish.

Bergh, J., Freeman, M., Sigurdsson, B., Kel-lomäki, S., Laitinen, K., Niinistö, S., Peltola,H. & Linder, S. Modelling the short-termeffects of climate change on the productiv-ity of selected tree species in Nordic coun-tries. Manuscript.

Beuker, E., Kolström, M. & Kellomäki, S. 1996.Changes in wood production of Picea abiesand Pinus sylvestris under a warmer cli-mate: Comparison of field measurementsand results of a mathematical model. SilvaFennica 20(2–3): 239–246.

Blennow, K., & Sallnäs, O. WINDA – A systemof models for assessing the probability ofwind damage to forest stands within alandscape. Ecological Modelling. In press.

Blennow, K., & Sallnäs, O., 2002. Risk percep-tion among non-industrial private forestowners. Scandinavian Journal of ForestResearch 17: 472–479.

Bond, W.J. & Richardson, D.M. 1990. What canwe learn from extinctions and invasionsabout the effects of climate change? SouthAfrican Journal of Science: 86: 429–433.

Bradshaw, H.W., Holmqvist, B.H., Cowling,S.A. & Sykes, M.T. 2000. The effect of cli-mate change on the distribution andmanagement of Picea abies in southern


37


Scandinavia. Can. J. For. Res. 30: 1992–1998.

Bylund, H. 1997. Stand age-structure influencein a low population peak of Epirrita autum-nata in a mountain birch forest. – Ecogra-phy 20:319–326.

Bylund, H. 1999. Climate and the populationdynamics of two insect outbreak speciesin the north. – Ecol. Bull. 47:54–62.

Cederlund. G., Ljungqvist, H., Markgen, G. &Stålfelt, F. 1980. Foods of moose and roedeer at Grimsö in central Sweden, resultsof rumen content analysis. Swedish Wild-life Research. 11(4): 169–247.

Ciencialia, E., Kucera, J., Ryan, M.G & Lind-roth, A. 1998. Water flux in boreal forestduring two hydrologically contrastingyears; species specific regulation of cano-py conductance and transpiration. Ann.Sci. For. 55: 47–61.

Coops, N.C. & Waring, R.H. 2001. Assessing for-est growth across south-western Oregonunder a range of current and future globalchange scenarios using a process model, 3-PG. Global Change Biology. 7: 15–29.

Danell, Ö. 1993. Breeding programs in Swe-den, general approach. In Progeny testingand breeding strategies. Proceedings ofthe Nordic group of forest tree breeding.Lee S.J. (ed). Forestry Commission. Edin-burgh, U.K. 184 p.

Eamus, D & Jarvis, P.G. 1989. The directeffects of increase in the global atmospher-ic CO2 concentration on natural and com-mercial temperate trees and forests. Ad-vances in Ecological Research 19: 1–53.

Eriksson, G. & Ekberg, I. 2001 An introductionto forest genetics. Swedish Univeristy ofAgricultural Scienses. Uppsala. pp 165.

Evans, H.F., McNamara, D.G., Braasch, H.,Chadoeuf, J. & Magnusson, C. 1996. Pestrisk analysis (PRA) for the territories of theEuropean Union (as PRA area) on Bursa-phelenchus xylophilus and its vectors inthe genus Monochamus. EPPO Bulletin 26.199–249.

Evans, H., Straw, N. & Watt, A. 2002. Climatechange: implications for insect pests. For-estry Commission Bulletin, Forestry Com-mission, Edinburgh, UK. No 125. 99–118.

Fleming, R.A. & Volney, W.J.A. 1995. Effects ofclimate change on insect defoliator popu-lation processes in Canada’s boreal forest:some plausible scenarios. Water, Air andSoil Pollution. 82: 445–454.

Gärdenäs, A. & Jansson, P-E. 1995. Simulatedwater balance of Scots pine stands in Swe-den for different climate change scenarios.J. Hydrol. 166: 107–125.

Hannerz 1998. Genetic and seasonal variationin hardiness and growth rhythm in borealand temperate conifers – a review andannotated bibliography. The Forestry Re-search Institute of Sweden. Report No. 2:pp.140.

Harrington, R, Fleming, R.A. & Woivod, I.P.2001. Climate change impacts on insectmanagement and conservation in temper-ate regions: can they be predicted? Agri-cultural and Forest Entomology 3: 233-240.

Harrington, R., Woiwod, I. & Sparks, T. 1999.Climate change and trophic interactions.TREE 14(4): 146-150.

Hellgren, M. 1995. Gremmeniella abietina –disease biology and genetic variation with-in Fennoscandia. Swedish University ofAgricultural Sciences. Dissertation.

IPCC 2001. Climate Change 2001: The Scientif-ic Basis. Contribution of Working Group I tothe Third Assessment Report of the Inter-governmental Panel on Climate Change[Houghton, J. T., Ding, Y., Griggs, D. J., Nogu-er, M., van der Linden, P. J., Dai, X., Maskell,K., and Johnson, C. A. (eds.)]. CambridgeUniversity Press, Cambridge, United King-dom and New York, NY, USA, 881 p.

Jarvis, P. & Linder, S. 2000. Constraints togrowth in boreal forests. Nature. 405: 904–905.

Johnsen, Ø., Skrøppa, T., Haug, G., Apeland, I.& Østreng, G. 1995. Sexual reproduction ina greenhouse and reduced autumn frost


38


hardiness of Picea abies progenies. TreePhysiology 15:551–555.

Jönsson, A.M., Linderon, M-L., Stjernquist, I.And Schlyter, P. 2004. Climate change andthe effect of temperature backlashes caus-ing frost damage in Picea abies. Submittedmanuscript.

Karns, P.D. 1997. Population distribution, den-sity and trends. In: Ecology and manage-ment of the north America moose. Franz-mann, A.W. & Schwartz, C. (eds.). Smith-sonian Institution Press. Washington andLondon. 125–127.

Kellomäki, S. & Kolström, M. 1992. Computa-tions on the management of seedlingstands of Scots pine under the influence ofchanging climate in southern Finland. SilvaFennica 26(2): 97–110.

Kilander, S. 1955. Kärlväxternas övre gränserpå fjäll i sydvästra Jämtland samt angrän-sande delar av Härjedalen. Acta Phytogeo-graphica Suecica 35: 1–98. In Swedish.

Korhonen, K. Stenlid, J. Capretti, P. & Kar-jalainen, R. 1998. Distribution of Heteroba-sidion annosum intersterility groups inEurope. In: Woodward, S., Stenlid, J., Kar-jalainen, R. & Hüttermann, A. A. Heteriba-sidion annosum biology, ecology, impactand control. CAB International, Walling-ford, UK pp. 93–104.

Kullman, L. 2002. Rapid recent range-marginrise of tree and shrub species in the Swed-ish Scandes. J. Ecol. 2002. 90: 68–77.

Lasch, P., Lindner, M., Erhard, M., Suckow, F. &Wenzel, A. 2002. Regional impact assess-ment on forest structure and functions un-der climate change – the Brandenburg casestudy. For. Ecol. Manage. 162: 73–86.

Lindner, M. 2000. Developing adaptive forestmanagement strategies to cope with cli-mate change. Tree Physiology 20: 299–307.

Linit, M.J. 1988. Nematode-vector relation-ships in the pine wilt disease system. Jour-nal of Nematology 20: 227–235.

Löyttyniemi, K. et al. 1979. Insect pests inforests of the Nordic countries 1972–1976.Folia Forest. 395:1–13.

Medlyn, B.E., Barton, C.V.M., Broadmeadow,M. Ceulemans, R., de Angelis, P., Forstreu-ter, M., Freeman, M., Jackson, S.B., Kello-mäki, S., Laitat, E., Rey, A., Roberntz, P.,Sigurdsson, B., Strassenmeyer, J., Wang, K.,Curtis, P.S. & Jarvis, P.G. 2001. Elevated CO2

effects on stomatal conductance in Euro-pean forest species: a synthesis of experi-mental data. New Phytologist 149: 247–264.

Nakicovic N, Alcamo J, Davis G, de Vries B, Fen-hann J, Gaffin S, Gregory K, Grübler A, JungTY, Kram T, La Rovere EL, Michaelis L, MoriS, Morita T, Pepper W, Pitcher H, Price L,Raihi K, Roehrl A, Rogner H-H, Sankovski A,Schlesinger M, Shukla P, Smith S, Swart R,van Rooijen S, Victor N, Dadi Z (2000) Emis-sion Scenarios. A Special Report of WorkingGroup III of the Intergovernmental Panel onClimate Change, Cambridge UniversityPress, 599 pp.

Niemelä, P., Chapin III, F.S., Danell, K. andBryant, J.P. 2001. Herbivory-mediated re-sponses of selected boreal forests to clim-atic change. – Climatic Change 48:427–440.

Olofsson, E. and Blennow, K. Decision sup-port for identifying spruce stand edgeswith high probability of wind damage. For-est Ecology and Management, In press.

Peltola, H., Kellomäki, S. & Väisänen. H. 1999.Model computations of the impact ofclimatic change on the windthrow riskof trees. Climatic Change 41: 17–36.

Persson, P., 1975. Stormskador på skog – Upp-komstbetingelser och inverkan på skogligaåtgärder. (Windthrow in forests – It’s caus-es and the effect of forestry measures).Department of Forest Yield Research,Royal College of Forestry, Stockholm.Research Notes 36. In Swedish with Eng-lish summary.

Persson, B. 1998. Will climate change affectthe optimal choice of Pinus sylvestris prov-enances? Silva Fennica 32: 121–128.

Pretzsch, H. & Dursky, J. 2002. Growth reac-tion of Norway spruce (Picea abies (L.)

´ ´


39


Karst.) and European beech (Fagus syl-vatica L.) to possible climatic changes inGermany. A sensitivity study. Forstw. Cbl.121(1): 145–154.

Ray, D., Pyatt, G. & Broadmeadow, M. 2002.Modelling the future suitability of planta-tion forest tree species. Forestry Commis-sion Bulletin, Forestry Commission, Edin-burgh, UK. No 125. 151–167.

Redfern, D. & Hendry, S. 2002. Climate changeand damage to trees caused by extremesof temperature. Forestry Commission Bul-letin, Forestry Commission, Edinburgh,UK. No 125. 29–39.

Redfern, D.B. & Stenlid, J. 1998. Spore disper-sal and infection. In: Woodward, S., Stenlid,J., Karjalainen, R. & Hüttermann, A. A.Heteribasidion annosum biology, ecology,impact and control. CAB International,Wallingford, UK pp. 105–124.

Rummukainen, M. 2003. The Swedish region-al climate modeling program, SWECLIM,1996–2003. Final report. Reports Meteorol-ogy and Climatology 104, SMHI, Norr-köping, Sweden, 47 pp.

Rustad, L.E., Campbell, J.L., Marion, G.M. et al.2001. A meta-analysis of the response ofsoil respiration, net nitrogen mineraliza-tion, and aboveground plant growth toexperimental ecosystem warming. Oecolo-gia 126: 543–562.

Räisänen, J., Hansson, U., Ullerstig, A., Dösh-er, R., Graham, L.P., Jones,C., Meier, M.,Samuelsson, P. & Willén, U. 2003. GCM driv-en simulations of recent and future climatewith the Rossby centre coupled atmos-phere – Baltic Sea regional climate modelRCAO. SMHI, Reports Meteorology andClimatology. No 101. 61 p

Räisänen, J., Hansson, U., Ullerstig, A., Dösh-er, R., Graham, L.P., Jones,C., Meier, M.,Samuelsson, P. & Willén, U. 2004. Europeanclimate in the late 21st century: regionalsimulations with two driving global mod-els and two forcing scenarios. ClimateDynamics 22: 13–31.

Schmidtling, R.C. 1994. Use of provenancetests to predict response to climatechange: Loblolly pine and Norway spruce.Tree Physiol. 14: 805–817.

Shaver, G.R., Canadell, J., Chapin III, F.S. Gure-vitch, J., Harte, J., Henry, G., Ineson, P., Jo-nasson, S., Mellilo, J., Pitelka, L. & Rustad,L. 2000. Global warming and terrestrialecosystems: a conceptual framework foranalysis. Bioscience 50(10) 871–882.

Skre, O. & Nes, K. 1996. Combined effects ofelevated winter temperatures and CO2 onNorway spruce seedlings. Silva Fennica30(2–3): 135–143.

Sonesson, J. 2001. Skogsodlingsmaterial förframtidens klimat. Skogforsk. Resultat No6. 4 p, In Swedish with English summary

Sonesson, J. & Eriksson, G. 2000. Genotypicstability and genetic parameters for growthand biomass traits in a water?temperaturefactorial experiment with Pinus sylvestris L.seedlings. For. Sci. 46(4): 487–495.

Sonesson, J., Jansson, G. & Eriksson, G. 2002.Retrospective genetic testing of Picea abiesunder controlled temperature and mois-ture regimes. Can. J. For. Res. 32(1) 81–91.

Straw, N.A. 1995. Climate change and theimpact of green spruce aphid, Elatobiumabietinum (Walker), in the UK. Scottish For-estry. 49(3): 134–145.

Strömgren, M & Linder, S. 2002. Effects ofnutrition and soil warming on stemwoodproduction in a boreal Norway sprucestand. Global Change Biology 8: 1195–1204.

Suffling, R. 1992. Climate change and borealforest fires in Fennoscandia and centralCanada. In: Boer, M. & Koster, E. (eds.).Greenhouse impact on cold-climate eco-systems and landscapes. Catena Supple-ment No 22: 111–132.

Sykes, M.T. & Prentice, I.C. 1995. Boreal forestfutures: modelling the control on speciesrange limits and transient response to cli-mate change. In: Apps, M.J., Price, D.T. &Wisniewski, J. (eds.). Boreal forests andglobal change: peer-reviewed manuscripts


40


selected from the International BorealForest Research Association conference,held in Saskatoon, Saskatchewan, Canada,September 25–30 1994. Water, Air and SoilPollution 82(1–2). 415–428.

Tenow, O. 1996. Hazards to a mountain birchforest-Abisko in a perspective. – Ecol. Bull.45:104–114.

Tenow, O. 1975. Topographical dependence ofan outbreak of Oporinia autumnata Bkh.(Lep., Geometridae) in a mountain birch for-est in northern Sweden. – ZOON 3:85–110.

Tenow, O., Bylund, H. and Holmgren, B. 2001.Impact on mountain birch forests in thepast and the future of outbreaks of twogeometrid insects. – In F.E. Wielgolaski(ed.), Nordic mountain birch ecosystems.Man and Biosphere series vol. 27., Parthe-non Publishing group, New York, pp. 223–239.

Thompson, D.G. 1998. Getting the species andprovenances right for climate change.Irish Forestry. 55(2): 114–121.

Thompson, I.D., Flannigan, M.D., Wotton, B.M.& Suffling, R. 1998. The effect of climatechange on landscape diversity: an exam-ple in Ontario forests. Environmental Mon-itoring and Assessment 49: 213–233.

Venäläinen, A., Tuomenvirta, H., Heikinheimo,M., Kellomäki, S., Peltola, H., Strandman, H,& Väisänen, Hannu. 2001. Impact of cli-mate change on soil frost under snowcover in a forested landscape. ClimateResearch. 17: 63–72.

Virtanen, T., Neuvonen, S., Nikula, A., Varama,M. & Niemelä, P. 1996. Climate change andthe risks of Neodiprion sertifer outbreaks onScots pine. Silva Fennica 30(2–3): 169–177.

Virtanen, T., Neuvonen, S. and Nikula, A. 1998.Modelling topoclimatic patterns of eggmortality of Epirrita autumnata (Lepidop-tera: Geometridae) with Geographical In-formation System: predictions for currentand warmer climate scenarios. – Journal ofAppl. Ecology 35:311–322.

Virtanen, T. & Neuvonen, S. 1999. Climate

change and macrolepidopteran biodiversi-ty in Finland: Global Change Science 1:439–448.

Virtanen, T. and Neuvonen, S. 1999. Perform-ance of moth larvae on birch in relation toaltitude, climate, host quality and parasi-toids. – Oecologia 120:92–101.

Volney, W.J.A. & Fleming, R.A. 2000. Climatechange and impacts of boreal forest in-sects. Agriculture, Ecosystems and Envi-ronment 82: 283–294.

Väisänen, H., Strandman, H. & Kellomäki, S.1994. A model for simulating the effects ofchanging climate on the functioning andstructure of the boreal forest ecosystem:an approach based on object-orienteddesign. Tree Physiology. 14: 1081–1095.

Watson, R.T., Zinyowera, M.C., Moss, R.H. &Dokken, D.J. 1997. The Regional Impacts ofClimate Change: An Assessment of Vulner-ability. Special Report of IPCC Workinggroup II. Cambridge University Press, UK.pp 517.

Wargo, P.M. & Harrington, T.C. 1991. Hoststress and susceptibility . I: Armillaria rootdisease. CG Shaw III and G.A Kile (eds.)USDA forest service agricultural handbookNo 691. Washington DC pp 88–101.

Wilhelmsson, L., Arlinger, J., Spångberg, K.,Lundqvist, S-O., Grahn, T., Hedenberg, Ö. &Olsson, L. 2002. Models for predictingwood properties in stems of Picea abiesand Pinus sylvestris in Sweden. Scand. J.For. Res. 17: 330–350.

Zheng, D., Freeman, M., Bergh, J., Rösberg, I.& Nilsen, P. 2002. Production of Picea abiesis south-east Norway in response to climatechange: A case study using process-basedmodel simulation with field validation.Scand. J. For. Res. 17: 35–46.

Personal communicationsRagnar FribergGöran Örlander


41



42



43

Förteckning över tidigare utgivna nummerÅr 2003; Årgång 142Nr 1 Det sydsvenska landskapet, framtidsvisioner och framtidssatsningar SAMT

Idéer för framtidens skogslandskapNr 2 Viltets positiva värdenNr 3 Inför toppmötet i JohannesburgNr 4 Kapital för landsbygdsföretagareNr 5 Kompetensförsörjningen i svenskt jordbrukNr 6 Fiskets miljöeffekter – kan vi nå miljömålen?Nr 7 Verksamhetsberättelse 2002Nr 8 De glesa strukturerna i den globala ekonomin – kunskapsläge och forskningsbehovNr 9 Tro och vetande om husdjurens välfärd (Enbart publicerad på www.ksla.se)Nr 10 Svenska satsningar på ökad träanvändning (Enbart publicerad på www.ksla.se)Nr 11 Kapital för landsbygdsföretagare (Enbart publicerad på www.ksla.se)Nr 12 Feminisering av Moder natur? Östrogener i naturen och i livsmedelNr 13 Crop and Forest Biotechnology for the FutureNr 14 Landskap och vindkraft – i medvind eller motvind (Enbart publicerad på www.ksla.se)Nr 15 Lantbrukskooperationen – Hållbar företagsidé eller historisk parentesNr 16 Utvecklingen i PolenNr 17 Mid Term Review Vad händer i Sverige när EU ändrar jordbrukspolitik?Nr 18 Soil and surface water acidification in theory and practiceNr 19 Skogsindustrinsråvaruförsörjningskedja – pågående utveckling och utblickar mot andra branscherNr 20 CAP och folkhälsanNr 21 Vilda djur i stadsmiljö – Tillgång eller problem? –Nr 22 Översvämningar och deras orsakerNr 23 Sötvattenfisk – Framtidens resurs (Enbart publicerad på www.ksla.se)Nr 24 Mat med mervärden – Goda affärer (Enbart publicerad på www.ksla.se)Nr 25 Hur planeras boendet på landsbygden? OCH Trädgården som rekreation och terapi

(Enbart publicerad på www.ksla.se)Nr 26 Det nya uppdraget – högre utbildning för landsbygd och landskap (Enbart publicerad på www.ksla.se)Nr 27 Hör göken han gal – Hur kan ekologiskt lantbruk och samhället gynna den biologiska mångfalden?

(Enbart publicerad på www.ksla.se)Nr 28 Den svenska modellen för husdjursavel och dataregistrering samt datautnyttjande inom husdjurs-

och växtodlingsområdet (Enbart publicerad på www.ksla.se)

År 2004; Årgång 143Nr 1 Ecosystem services in European agriculture – theory and practiceNr 2 Vad gör vi när vi inte vet? (Enbart publicerad på www.ksla.se)Nr 3 Ekologiskt Forum mönstrar köttet och marknaden (Enbart publicerad på www.ksla.se)Nr 4 Jordbruk – ekologi – samhälle (Enbart publicerad på www.ksla.se)Nr 5 LBU-programmet, skogen och landsbygden – till vilken nytta? (Enbart publicerad på www.ksla.se)Nr 6 Framtida forskning inom den gröna sektorn (Enbart publicerad på www.ksla.se)Nr 7 Ekologisk produktion med miljönytta (Enbart publicerad på www.ksla.se)Nr 8 Har jord- och skogsbruk en roll i framtidens kultur? (Enbart publicerad på www.ksla.se)Nr 9 Aktuella trämekaniska alternativ (Enbart publicerad på www.ksla.se)Nr 10 Mat för livet eller dödlig diet? (Enbart publicerad på www.ksla.se)Nr 11 Verksamhetsberättelse 2003Nr 12 Kvävehushållning och kväveförluster – förbättringsmöjligheter i praktiskt jordbrukNr 13 Sigvard Andersson – En nyskapande forskar- och lärargärning i markfysikNr 14 Vem köper biologisk mångfald? (Enbart publicerad på www.ksla.se)Nr 15 Svenskt jordbruk i EU-25 – Östutvidgningens betydelse för jordbruks- och livsmedelsnäringen i Sverige

(Enbart publicerad på www.ksla.se)Nr 16 Vad är tvivelaktigt med vattenorganisationens direktiv? What’s fishy about the Water Framwork Directive?

(Enbart publicerad på www.ksla.se)Nr 17 Behövs en tredje aktionsplan för ekologisk produktion? (Enbart publicerad på www.ksla.se)Nr 18 Climate change and forestry in Sweden – a literature review


44

Kungl. Skogs- och Lantbruksakademiens Tidskrift (KSLAT) har, under olika namn, utkommit sedan1813, då akademien grundades. Från och med 1994 utges KSLAT som en numrerad serie av skrifter

(15–20 häften/år) med egna titlar. Innehållet består huvudsakligen av dokumentering frånakademiens sammankomster och seminarier – även debattnummer förekommer –

och speglar akademiens verksamhetsområde; de areella näringarnaoch till dessa knutna verksamheter.

Prenumerationspris 350 kr/år.

Kungl. Skogs- och LantbruksakademienDrottninggatan 95 B, Box 6806, 113 86 Stockholm

Tel 08-54 54 77 00, Fax 08-54 54 77 10, Postgiro 18 32 80 - 7

ISSN 0023–5350 ISBN 91–89379-94-2 Eski

lstun

a O

ffset

AB,

Esk

ilstu

na 2

004

Pris 75:-

Documents

Climate change and forestry in Sweden – a literature revie...Climate change and forestry in Sweden probably face continuous, ongoing chang-es in climate, implying that conditions