Forest Biomass Production Potential and Its Implications for Carbon Balance

Home , Biomass

Thesis for the degree of Licentiate of Philosophy

Forest biomass production potential and its implications for

carbon balance

Bishnu Chandra Poudel

Ecotechnology and Environmental Science

Department of Engineering and Sustainable Development

Mid Sweden University

Östersund, Sweden

2012

Mid Sweden University Licentiate Thesis

Forest biomass production potential and its implications for carbon balance

Bishnu Chandra Poudel

Ecotechnology and Environmental Science Department of Engineering and Sustainable Development Mid Sweden University SE-83125 Östersund Sweden

Mid Sweden University Licentiate Thesis ISBN 978-91-87103-27-8 ISSN 1652-8948

Cover photo: Managed spruce forest, Sweden, ©Bishnu Chandra Poudel Printed by: Kopieringen Mid Sweden University, Sundsvall, Sweden, 2012

Abstract

An integrated methodological approach is used to analyse the forest biomass production potential in the Middle Norrland region of Sweden, and its use to reduce carbon emissions. Forest biomass production, forest management, biomass harvest, and forest product use are analyzed in a system perspective considering the entire resource flow chains. The system-wide carbon flows as well as avoided carbon emissions are quantified for the activities of forest biomass production, harvest, use and substitution of non-biomass materials and fossil fuels. Five different forest management scenarios and two biomass use alternatives are developed and used in the analysis. The analysis is divided into four main parts. In the first part, plant biomass production is estimated using principles of plant- physiological processes and soil-water dynamics. Biomass production is compared under different forest management scenarios, some of which include the expected effects of climate change based on IPCC B2 scenario. In the second part, forest harvest potentials are estimated based on plant biomass production data and Swedish national forest inventory data for different forest management alternatives. In the third part, soil carbon stock changes are estimated for different litter input levels from standing biomass and forest residues left in the forest during the harvest operations. The fourth and final part is the estimation of carbon emissions reduction due to the substitution of fossil fuels and carbon-intensive materials by the use of forest biomass. Forest operational activities such as regeneration, pre-commercial thinning, commercial thinning, fertilisation, and harvesting are included in the analysis. The total carbon balance is calculated by summing up the carbon stock changes in the standing biomass, carbon stock changes in the forest soil, forest product carbon stock changes, and the substitution effects. Fossil carbon emissions from forest operational activities are calculated and deducted to calculate the net total carbon balance. The results show that the climate change effect most likely will increase forest biomass production over the next 100 years compared to a situation with unchanged climate. As an effect of increased biomass production, there is a possibility to increase the harvest of usable biomass. The annual forest biomass production and harvest can be further increased by the application of more intensive forestry practices compared to practices currently in use. Deciduous trees are likely to increase their biomass production because of climate change effects whereas spruce biomass is likely to increase because of implementation of intensive forestry practices.

Intensive forestry practices such as application of pre-commercial thinning, balanced fertilisation, and introduction of fast growing species to replace slow growing pine stands can increase the standing biomass carbon stock. Soil carbon stock increase is higher when only stem-wood biomass is used, compared to whole-tree biomass use. The increase of carbon stocks in wood products depends largely on the magnitude of harvest and the use of the harvested biomass. The biomass substitution benefits are the largest contributor to the total carbon balance, particularly for the intensive forest management scenario when whole- tree biomass is used and substitutes coal fuel and non-wood construction materials. The results show that the climate change effect could provide up to 104 Tg carbon emissions reduction, and intensive forestry practices may further provide up to 132 Tg carbon emissions reduction during the next 100 years in the area studied. This study shows that production forestry can be managed to balance biomass growth and harvest in the long run, so that the forest will maintain its capacity to increase standing biomass carbon and provide continuous harvests. Increasing standing biomass in Swedish managed forest may not be the most effective strategy to mitigate climate change. Storing wood products in building materials delays the carbon emissions into the atmosphere, and the wood material in the buildings can be used as biofuel at the end of a building life-cycle to substitute fossil fuels. These findings show that the forest biomass production potential in the studied area increases with climate change and with the application of intensive forestry practices. Intensive forestry practice has the potential for continuous increased biomass production which, if used to substitute fossil fuels and materials, could contribute significantly to net carbon emissions reductions and help mitigate climate change.

Sammanfattning

Ett integrerat tillvägagångssätt används för att analysera skogens potential för produktion av biomassa i mellersta Norrland i Sverige och dess användning för att minska koldioxidutsläppen. Produktion av skogsbiomassa, skogsskötsel, skörd av biomassa och användning av skogsprodukter analyseras i ett systemperspektiv med beaktande av hela resursflödeskedjan. Kolflöden i hela systemet och de koldioxidutsläpp som undviks kvantifieras för följande aktiviteter i resursflödeskedjan: Produktion av biomassa, skörd, användning samt substitution av icke biomassebaserade material och fossila bränslen. Fem olika skogsskötselscenarier och två alternativ för användning av biomassa tas fram och analyseras. Analysen är uppdelad i fyra huvuddelar. I den första delen uppskattas produktionen av trädbiomassa baserat på principer för växtfysiologiska processer och markvattendynamik. Produktion av biomassa jämförs för de olika skogsskötselscenarierna av vilka några innefattar de förväntade effekterna av klimatförändringar som bygger på IPCC B2-scenariot. I den andra delen uppskattas skogens produktion av biomassa och skördepotentialer utifrån produktionsdata för trädbiomassa från den svenska riksskogstaxeringen för olika skogsskötselalternativ. I den tredje delen beräknas förändringar i markens kolförråd för olika tillförselnivåer till förnan från trädbiomassa och skogsavfall som lämnas kvar efter skörd. I den fjärde och sista delen uppskattas minskningar av koldioxidutsläpp till följd av substitution av fossila bränslen och kolintensiva material mot skogsbiomassa. Skogsbruksaktiviteter såsom föryngring, röjning, gallring, gödsling och skörd ingår i analysen. Den totala kolbalansen beräknas genom summering av kolförrådets förändring i trädbiomassan, skogsmarken och träprodukter samt summering av substitutionseffekter. Fossila koldioxidutsläpp från skogsbruket beräknas och dras av från den totala kolbalansen för att beräkna nettokolbalansen. Resultaten visar att effekter av klimatförändring sannolikt kommer att öka produktionen av skogsbiomassa under de närmaste 100 åren jämfört med en situation med oförändrat klimat. Som en effekt av ökad produktion av biomassa är det möjligt att även öka skörden av användbar biomassa. Årlig produktion och skörd av skogsbiomassa kan ökas ytterligare genom användning av intensivare skogsbruksmetoder än nuvarande. Lövträd kommer sannolikt att öka sin produktion av biomassa till följd av klimatförändringens effekter, medan granbiomassan sannolikt kommer att öka till följd av intensivare skogsbruksmetoder. III

Intensivare skogsskötselmetoder såsom röjning, balanserad gödsling och introduktion av snabbväxande arter för att ersätta långsamt växande tallbestånd bidrar till att öka kolförrådet i trädbiomassan. Ökningen av kolförrådet i marken är högre när endast stambiomassa skördas jämfört med när hela trädet tas ut. Ökningen av kolförrådet i träprodukter beror till stor del på avverkningsnivåerna och vad den skördade biomassan används till. Substitution av fossila bränslen och råvaror mot biomassan ger det största bidraget till den totala kolbalansen, särskilt när hela trädbiomassan används och när den ersätter fossilt kol som bränsle och icke träbaserade byggnadsmaterial. Resultaten visar att effekten av klimatförändringar kan ge upp till 104 Tg minskade koldioxidutsläpp, och att intensivare skogsbruk därtill kan ge upp till 132 Tg minskning under de närmaste hundra åren i det studerade området. Denna studie visar att ett produktionsskogsbruk kan skötas så att tillväxt och skörd av biomassa balanseras långsiktigt, så att skogen kommer att bibehålla sin förmåga att öka kolförrådet i trädbiomassan och ge kontinuerliga skördar. Ökad trädbiomassa i Sveriges skogar är kanske inte den mest effektiva strategin för att motverka klimatförändringarna. Lagring av trä i byggmaterial fördröjer koldioxidutsläppen till atmosfären, och vid slutet av byggnaders livscykel kan träprodukterna användas som biobränsle och då ersätta fossila bränslen. Resultaten visar att potentialen för skogens produktion av biomassa i det studerade området ökar med klimatförändringarna och med ett intensivare skogsbruk. Intensivare skogsbruk har potential att kontinuerligt öka produktionen av biomassa som, om den används till att ersätta fossila bränslen och material, kan bidra till betydande nettoutsläppsminskningar och till att minska klimatförändringarna.

Preface

This work was performed as a doctoral research project in the Ecotechnology Research Group at Mid Sweden University in Östersund, Sweden. It is a part of an interdisciplinary programme to study “Forest as a resource in sustainable societal development”. The financial support of the European Union, Jämtland County Council, Sveaskog AB, and Swedish Energy Agency is gratefully acknowledged. I would like to thank my supervisor Professor Inga Carlman for her both professional and personal support that motivated me to continue this work. I would like to thank Professor Leif Gustavsson who supervised my work from March 2009 to 2011 December. His professional support helped to improve my knowledge on this work. I would like to thank my co-supervisors Prof. Johan Bergh, and Dr. Roger Sathre, for their invaluable guidance in research, without which this work could not have been completed. I acknowledge the fellow Ph.D. students and staffs for their cooperation and support. Particularly, I would like express my thanks to Nils Nilsson and Ulf Persson who has made my days in Östersund much easier by many reasons. Many discussions, including sharing ideas, with you all helped me to understand research life both professionally and personally. My parents have dreamed about my higher education and this work is a part of their invaluable support for long time. Finally, I thank my wife, Sabita, for her every support, without which this journey could not have been possible. My thanks to our sons, Shashwat and Saatvik, who has made our every day special in Östersund.

Bishnu Chandra Poudel Östersund, 2012

List of Papers

This licentiate thesis is based on the following papers:

1. Poudel, B.C., Sathre, R., Gustavsson, L., Bergh, J., Lundström, A., Hyvönen, R., 2011. Effects of climate change on biomass production and substitution in north-central Sweden. Biomass and Bioenergy 35(10), 4340- 4355.

2. Poudel, B.C., Sathre, R., Bergh, J., Gustavsson, L., Lundström, A., Hyvönen, R., 2012. Potential effects of intensive forestry on biomass production and total carbon balance in north-central Sweden. Environmental Science and Policy 15(1), 106-124.

3. Poudel, B.C., Sathre, R., Gustavsson, L., Bergh, J., 2011. Climate change mitigation through increased biomass production and substitution: A case study in north-central Sweden. Peer reviewed conference paper In: World Renewable Energy Congress 2011, 8-11 May 2011, Linköping, Sweden.

Contents

ABSTRACT ...... I SAMMANFATTNING...... III PREFACE ...... V LIST OF PAPERS ...... VI FIGURES ...... IX TABLES ...... X 1 INTRODUCTION ...... 1

1.1 CLIMATE CHANGE AND FORESTS ...... 1 1.2 FOREST MANAGEMENT AND CARBON BALANCE ...... 2 1.3 REVIEW OF PREVIOUS STUDIES ...... 6 1.3.1 Forest biomass production and harvest ...... 6 1.3.2 Soil carbon stock ...... 7 1.3.3 Substitution of fossil fuels and carbon-intensive products ...... 8 1.4 KNOWLEDGE GAPS ...... 9 1.5 STUDY OBJECTIVES ...... 10 1.6 THESIS ORGANISATION ...... 10 2 METHODS ...... 11

2.1 INTEGRATED APPROACH ...... 11 2.2 FUNCTIONAL UNIT ...... 12 2.3 REFERENCE SYSTEM ...... 12 2.4 ACTIVITIES INCLUDED IN SYSTEM ANALYSIS ...... 13 2.5 STUDY AREA ...... 14 2.6 SCENARIOS ...... 14 2.7 CLIMATE CHANGE ...... 15 2.8 FOREST BIOMASS PRODUCTION MODELLING WITH BIOMASS ...... 15 2.9 FOREST BIOMASS HARVEST MODELLING WITH HUGIN ...... 16 2.10 THE USE OF FOREST BIOMASS AS BIOENERGY AND CONSTRUCTION MATERIALS ... 17 2.11 CARBON BALANCE ...... 17 2.11.1 Total carbon balance ...... 17 2.11.2 Standing biomass carbon stock ...... 18 2.11.3 Soil carbon stock ...... 19 2.11.4 Wood product carbon stock ...... 19 2.11.5 Substitution of fossil fuels and carbon-intensive products ...... 20 2.11.6 Forestry operations and carbon emission ...... 20 2.12 DATA QUALITY ...... 20 3 RESULTS ...... 22

VII

3.1 FOREST BIOMASS PRODUCTION AND HARVEST ...... 22 3.2 STANDING BIOMASS CARBON STOCK ...... 24 3.3 SOIL CARBON STOCK...... 25 3.4 SUBSTITUTION OF FOSSIL FUELS AND CARBON-INTENSIVE PRODUCTS ...... 26 3.5 WOOD PRODUCT CARBON STOCK ...... 28 3.6 CARBON EMISSIONS FROM FOREST OPERATIONS ...... 28 3.7 TOTAL CARBON BALANCE ...... 29 4 DISCUSSIONS ...... 33

4.1 FOREST PRODUCTION AND HARVEST ...... 33 4.2 CARBON-INTENSIVE PRODUCTS SUBSTITUTION ...... 34 4.3 CARBON BALANCE ...... 34 4.4 UNCERTAINTIES AND LIMITATIONS ...... 36 5 CONCLUSIONS ...... 38 6 FUTURE RESEARCH ...... 39 7 REFERENCES ...... 40

VIII

Figures

Figure 1 Schematic diagram of carbon balances during forest production and biomass utilization ...... 12

Figure 2 Annual biomass productions and harvests at the beginning and end of the study period for all scenarios...... 22

Figure 3 Standing forest biomass carbon stock at the beginning (2010) and end (2109) of the study period in the different scenarios (Paper II, Figure 6)...... 25

Figure 4 Annual carbon stocks in forest soil in the different scenarios during 100- years with whole-tree biomass use...... 26

Figure 5 Average annual carbon emission reduction (Tg carbon year-1) for whole study area as a result of whole-tree (WT) and stem-wood (SW) biomass use to replace fossil fuels and non-wood construction materials...... 27

Figure 6 Wood product carbon stock (Tg carbon) due to non-wood material substitution (Paper III)...... 28

Figure 7 Carbon emissions due to forest operations and fertilisation for all scenarios for whole-tree biomass (Mg carbon year-1)...... 29

Figure 8 The average annual carbon emissions reduction (Tg carbon year-1) for different scenarios when whole-tree biomass was used to replace coal and fossil gas...... 30

Figure 9 The average annual carbon emissions reductions per hectare of forest land (Mg carbon ha-1 year-1) due to the benefits associated with the use of whole-tree (WT) and stem-wood (SW) biomass to replace fossil gas and coal...... 31

Figure 10 Differences in cumulative carbon emission reduction (Tg carbon) between the Climate, Environment, Production and Maximum scenarios and the Reference scenario for each 10-year period...... 32

Tables

Table 1 Activities and processes included in the life cycle carbon balance ...... 13

Table 2 Overview of scenarios ...... 14

Table 3 Differences in forest biomass production and harvest between the Reference scenario and other scenarios (Tg dry biomass) ...... 23

Table 4 Average annual tree biomass carbon stock changes during each 10-year period (Tg carbon year-1) for Jämtland and Västernorrland ...... 24

Table 5 Average annual carbon emissions reductions (Tg carbon year-1) due to fossil fuel and carbon-intensive material substitution by the use of stem-wood and whole-tree biomass during each 10-year period ...... 27

Abbreviations

C Carbon

CO2 Carbon dioxide CORRIM Consortium for Research on Renewable Industrial Materials CPF Collaborative Partnership in Forests EU European Union GHG Greenhouse Gas GJ Gigajoule (109 joules) GJe Gigajoule of electricity GPP Gross Primary Production Gt Gigatonne (109 tonne) IEA International Energy Agency IPCC Intergovernmental Panel on Climate Change ISO International Organization for Standardization Mg Megagram (106 grams, or 1 tonne) N Nitrogen NEP Net Ecosystem Productivity NFI Swedish National Forest Inventory NPK Nitrogen Phosphorus and Potassium NPP Net Primary Production OECD Organization for Co-operation and Development ppm Parts per million SMHI Swedish Meteorological and Hydrological Institute SOM Soil Organic Matter SOU Swedish Government’s Official Reports Tg Teragram (1012 grams, or 106 tonne) UNFCCC United Nations Framework Convention on Climate Change US United States

1 Introduction

1.1 Climate change and forests

Climate change is one of the most widely recognised environmental issues today. A consensus in the climate science research community has emerged that emissions of anthropogenic greenhouse gases (GHGs) into the atmosphere are altering the global climate system (2007c). Emissions of GHGs into the atmosphere are likely to increase the earth’s mean surface temperature, thereby affecting physical and biological systems (Rosenzweig et al., 2008; Trenberth et al., 2009; UNFCCC, 2011). Many effects of temperature increase have been observed, including threats to natural phenomena, societal disturbances, and threats to economic growth (IPCC, 2007a; UNFCCC, 2011). The atmospheric concentration of the most significant GHG, carbon dioxide (CO2), has increased substantially during the last 20 decades, primarily as a result of fossil fuel combustion (IPCC, 2007c). Currently, fossil fuels provide over 80% of the world’s primary energy and are responsible for over 50% of all anthropogenic GHG emissions (OECD/IEA, 2010). The global energy supply is largely dependent on fossil fuels of which 27% of the total primary energy is dependent on coal (OECD/IEA, 2010). The International Energy Agency (IEA) has examined the global energy scenarios for future energy supply and has indicated that coal fuel use is likely to increase until the year 2035 if the current energy policies are not changed (OECD/IEA, 2010). These findings may explain the severity of fossil fuel use in the future which may result in increasing total anthropogenic GHG emissions in the future (OECD/IEA, 2010). The likely increase of global mean surface temperature (Nakicenovic and Swart, 2000) is expected to be greater at higher latitudes (IPCC, 2001), which may result in a temperature increase of up to 1-2 °C in summer and 2-3 °C in winter in northern Europe in the next 50 years (Carter et al., 2005). The European Union (EU) suggests that limiting temperature increase to 2 °C, relative to pre-industrial levels, may avoid the risks of climate change (European-Commission, 2007). The projection of CO2 emissions reduction that is required to avoid climate change is difficult to predict because of the complexity of the global system. However, stabilising atmospheric GHG concentration levels below 450 ppm CO2-eq is likely to avoid a 50% chance of increasing temperature above 2 °C (EU, 2008). This stabilisation target is unlikely to be met unless strong and timely actions are taken

1 to reduce CO2 emissions, considering the current atmospheric CO2 concentration level of 394 ppm CO2-eq (NOAA/ESRL, 2012). The Intergovernmental Panel on Climate Change (IPCC) has presented several strategies that may help mitigate climate change (IPCC, 2007b). Various initiatives have been taken to reduce the atmospheric CO2 concentration such as the Kyoto protocol requiring industrial countries to reduce GHG emissions below 1990 levels by the year 2012. The EU has set an ambitious target of reaching a 20% share of energy from renewable sources by 2020 (European-Commission, 2009) in the pursuit of reducing CO2 emissions. Sweden, an EU member state, already obtains more than 20% of its energy from renewable sources (Swedish Forest Agency, 2011) and has a target of extending this share to 49% by the year 2020 (Swedish Energy Agency, 2008). The global society’s attempt to mitigate climate change may include reducing carbon emissions and increasing carbon sinks. Strategies to reduce CO2 emissions by reducing fossil fuel use and by increasing carbon sinks through a sustainable production forestry may be one of the practical solutions. A Sustainable production forestry in this context may be considered as a practice that maintains indefinitely the natural productive capacity of forest ecosystems, while providing an opportunity for harvesting and using at least a certain amount of forest product at regular intervals. Thus, such practice will ensure the continued existence of forest ecosystems and the availability of forest biomass for use in future. By doing so, forest biomass use for bioenergy may play a vital role in replacing fossil fuels and reaching the aim of increasing renewable energy use. Forest biomass can also replace carbon-intensive construction materials. Hence, the greater the forest biomass production, the greater the amount of harvest that is possible, and the earlier the biomass is harvested, the earlier the fossil fuels and carbon-intensive materials can be replaced (Malmsheimer et al., 2008). A strategy aimed to increase forest biomass production and use could thus be effective in mitigating climate change (Easterling et al., 2007).

1.2 Forest management and carbon balance

The role of a sustainable forest management in reducing atmospheric carbon emissions is very important. The IPCC suggests that the conservation of biological carbon in the forest ecosystem, the storage of carbon in wood products, and the use of forest biomass to replace fossil fuels and carbon-intensive materials would help to mitigate climate change (IPCC, 1996). The strategies of a forest

2 management for increasing carbon benefit needs to increase the carbon sink into the forest biomass by increasing biomass production, increase soil carbon stock, and increase the potential biomass harvest to contribute in obtaining higher carbon benefits. Forests constitute approximately 30% of the world’s land surface and contribute to the global net removal of three gigatonnes (Gt) of anthropogenic carbon per year (Canadell and Raupach, 2008). Forest ecosystems, including forest soil, store approximately 1200 Gt of carbon, which is considerably more than the 762 Gt of carbon present in the atmosphere (Freer-Smith et al., 2009). Annually, global forests turn over approximately one twelfth of the atmospheric stock of CO2 through gross primary production (GPP1), accounting for 50% of all terrestrial GPP (Malhi et al., 2002). It is estimated that, in 2005, forests contained 572 Gt of standing biomass (equivalent to 280 Gt of carbon) (IPCC, 2007b; CPF, 2008). The carbon stock increase in a forest ecosystem is limited. Phenomena such as plant respiration, soil decomposition processes, plant mortality, fire, and soil disturbance release carbon into the atmosphere. It is estimated that less than 1% of the carbon that is taken up by terrestrial ecosystems remains as a long-term terrestrial carbon sink (IPCC, 2000). Several studies suggested that increasing atmospheric CO2 and nitrogen (N) deposition rates are likely to increase the forest biomass in the future (Norby et al., 1999; Norby et al., 2005; Canadell et al., 2007a; IPCC, 2007c; Galloway et al., 2008; Malmsheimer et al., 2008; Dupre et al., 2010). Luo

(2007) observed that elevated atmospheric CO2 concentrations result in enhanced biomass production and subsequently increased carbon pools in living biomass, litter, and soil. Although uncertain, climate change projections predict temperature increases and longer growing seasons at higher latitudes, implying that more sunlight will be utilised for photosynthesis, thereby stimulating the biomass production of boreal forests (Bergh et al., 2003; Rosenzweig et al., 2008; Raupach and Canadell, 2010). In addition, biomass production can be further increased by various forest management practices (Bergh et al., 2005; Houghton, 2007; Skogsstyrelsen, 2008). Forest management involves the integration of silvicultural practices and economic alternatives (Bettinger et al., 2009) to achieve production and ecological objectives through a series of silvicultural operations, such as regeneration, fertilisation, tending operations (commercial and pre-commercial thinning),

1 Gross primary production is the sum of the photosynthesis by all leaves measured at the ecosystem scale, represented as biomass (Chapin et al. 2002). 3 harvesting, and reforestation (Buongiorno and Gilless, 2003; Bettinger et al., 2009). The scope of this thesis limits the definition of silvicultural activities to those that are relevant to forest production and management for CO2 emission reduction. Forest management strategies are associated with forest production and harvest goals, with defined goals for carbon sequestration and accumulation as above-and below-ground biomass, litter that falls to the ground, dead wood, soil organic matter, and harvested products and their utilisation. A harvest goal that has higher frequency and intensity of thinning and shorter rotation length may influence the biomass stock in the following generation (Liski et al., 2001; Kaipainen et al., 2004). The influence can be related to the types of forest products harvested. For instance, the removal of stem-wood-only biomass adds all litters to the soil, while whole- tree removal adds a part of litter to the soil. A forest sequesters carbon by removing carbon from the atmosphere through the photosynthesis process and storing the carbon in living biomass. A forest’s capacity for sequestrating carbon is a function of the productivity of the site and the potential sizes of the various carbon pools, including soil, litter, subterraneous woody materials, standing and fallen dead wood, live stems, branches, and foliage (Chapin et al., 2002; Malmsheimer et al., 2008). The carbon accumulation in biomass is typically higher when the tree is in its fast growing stage, and as the tree reaches maturity (older than 80 years in boreal forests) it becomes almost carbon neutral and may eventually become a carbon source if dies and decays (Gower et al., 1996; Gower et al., 2001; Nabuurs et al., 2003). Shortening rotation periods would encourage the growth of young stand forests in the long run and thus may result in the sequestering of carbon at higher rates. Soil type, texture, structure, and air and water availability inside the soil pores govern the soil carbon development process. The rates of accumulation and magnitudes of soil carbon stocks depend on the litter input into the soil and in part on nutrient management (Lal, 2004). Fertilisation may help to increase nutrient level in the soil thus may help to build up carbon stocks (Johnson, 1992). Changes in tree phenology, species composition, soil temperature, and water balance can change soil respiration and decomposition rates (Davidson et al., 2000; Luo, 2007). When the litter is fresh, nutrient cycling and oxidation stimulate its decomposition; however, later, as litter quality decreases, higher temperatures might increase decomposition and carbon cycling rates (Kirschbaum, 2004). Hyvönen et al. (2008) explained that elevated CO2 would encourage carbon sequestration by activating the carbon cycling process by adding litter to the soil, which can later assimilate with soil organic material (SOM), thereby increasing soil carbon stock. Ågren et al. 4

(2007) suggested that the soil carbon stocks in Swedish forests will increase but that the rate of increase will quickly slow down when the input is stopped. Ågren et al. (2007) warned that reducing litter input by harvesting more forest biomass would lead to a decline in soil carbon stocks. Nevertheless, increasing forest biomass production to allow more litterfall and input may reverse the positive feedback. Nitrogen is one of the limiting factors in boreal forest production (Pastor and Post, 1988; Tamm, 1991). If nutrients are a limiting factor, balanced fertilisation, which includes three major nutrients Nitrogen, Phosphorus and Potassium (NPK), may allow for increased productivity especially at nutrient-poor sites (Oren et al., 2001; Malmsheimer et al., 2008). Bergh et al. (2005) suggested that the forest production in northern Sweden could be increased by 300% if an additional nitrogen fertiliser is applied to address the nutrient deficiency. Providing a balanced nutrient supply to reduce the nitrogen deficiency, particularly in the young stands, could increase the production. Forest biomass production can be increased by improving site conditions for plantations and by selecting improved genetic plant materials and productive tree species suitable to local site conditions and microclimates. Density regulation through pre- commercial thinning reduces the competition between trees and shortens rotation periods, and shortened periods between two rotations will allow more time for biomass production. Forest biomass can be used to substitute fossil fuels to reduce GHG emissions (Zebre, 1983; Gustavsson et al., 1995; Bergman and Zebre, 2004). The IPCC (2007b) suggests that material substitution is an important part of climate change mitigation strategy through the production of forest biomass.

Eriksson et al. (2007) claimed that there could be significant reductions in CO2 emissions if forest biomass from sustainably managed forests replaced fossil fuels. The burning of biomass grown in sustainably managed forests is part of a cyclical flow, as carbon emitted during combustion will be absorbed by re-growth of the harvested forest stand. Using wood products from sustainably managed forests can reduce net CO2 emissions, as less energy is used to manufacture such materials compared to carbon-intensive materials. Substitution of carbon-intensive materials also avoids non-energy process emissions during the manufacture of non-wood- based products, such as cement and steel. The use of wood products may also delay the emission of biogenic carbon into the atmosphere by allowing the wood to be used for a longer period of time. Furthermore, biomass by-products of wood production chains and wood from construction work at the end of its useful life 5 can be used (Sirkin and Houten, 1994; Gustavsson et al., 2006b; Gustavsson and Sathre, 2006; Sathre et al., 2010). The wood product life cycle optimised for mitigating climate change would entail the use and possible reuse of wood products (Sathre, 2007). Nevertheless, the net carbon benefits depend on substitution efficiency, i.e., on how and where the wood products are used and on which products are substituted (Schlamadinger and Marland, 1996; Schlamadinger et al., 1996; Gustavsson et al., 2006a; Sathre and O'Connor, 2010).

1.3 Review of previous studies

Several studies have discussed potential forest biomass production and harvest over the past decades. Carbon balance studies in the past were largely focused on ecosystem production (Houghton, 2005, 2007) but recent studies have included the forest biomass use as material and bioenergy and their implications for carbon emissions reduction.

1.3.1 Forest biomass production and harvest

Studies to assess the effects of increased temperatures and CO2 concentrations on plant photosynthesis and biomass production began early in the 1980s. Solomon (1986) and Pastor and Post (1988) performed studies on the effect of increased atmospheric CO2 concentrations on forest growth and composition in North America. They found that the forest biomass carbon in the boreal forests would be increasing because of increased CO2 concentrations. Kauppi et al. (1992) explored the increasing forest biomass and carbon budget for most of western Europe. Prentice et al. (1993) developed a model to describe the transient effects of climate change-induced temperature and precipitation change on the dynamics of forest landscapes, growth processes, and species compositions. Swedish forest production in response to climate change was explored in several studies (Bergh et al., 1998; Bergh and Linder, 1999; Bergh et al., 1999). Furthermore, several studies explored forest productivities in boreal forests (Pussinen et al., 2002; Bergh et al., 2005; Briceño-Elizondo et al., 2006; Kirilenko and Sedjo, 2007; Eggers et al., 2008). Several studies suggested that forest biomass production might increase in northern Europe with increased atmospheric temperatures and CO2 concentrations. Swedish Forest Agency (2008) predicted a 25% increase in annual stem-wood production in Swedish forests due to the direct effects of climate change over the next 100 years. A larger amount of biomass production may

6 correspond to a larger biomass removal. Currently in Sweden, residues and stumps harvesting is in practice. In the year 2010, Sweden gave permission to harvest forest residues from 155 thousand hectares and stumps from 7.5 thousand hectares of forest (Swedish Forest Agency, 2011).

1.3.2 Soil carbon stock

The influence of increased harvest level on soil carbon stock and nutrients have been discussed in the recent literature (Liski et al., 2001; Egnell and Valinger, 2003; Kaipainen et al., 2004; Egnell, 2011). The studies found that the effect of increased harvest level in soil carbon and nutrients are only temporary. Various studies have analysed the effects on forest soil carbon stock due to land use changes, forest management, climate change and other disturbances. Liski et al. (2002) found that the soil carbon stock in western Europe is increasing because of increased litter fall from living trees. Early studies reported that increased temperatures do not necessarily increase the nutrient cycling and oxidation of SOM and therefore decomposition (Lloyd and Taylor, 1994). However, later studies suggested that a warmer climate may govern a major role in the decomposition of SOM leading to an increase in CO2 release to the atmosphere and a decrease in soil carbon content (Davidson et al., 2000; Jones et al., 2005; Davidson and Janssens, 2006; Friedlingstein et al., 2006). Because of the release of soil carbon, positive feedback leading to additional climate warming of 0.1-1.5 °C may take place (Kirschbaum, 2000, 2004; Knorr et al., 2005; Friedlingstein et al., 2006; Luo, 2007; Piao et al., 2008). A significant loss of terrestrial carbon stock in the higher latitude is also possible because of the temperature-dependent net primary production system in the boreal forests compared to other parts of the world (Canadell et al., 2007b; Canadell and Raupach, 2008). However, Hyvönen et al. (2007) suggested that, as long as forest input is increasing, soil carbon loss will not be significant. Therefore, increasing the productivity of forests is a key factor. Moreover, reducing disturbances in forests may help retain soil carbon for a longer period of time (Karhu et al., 2010; O'Donnell et al., 2010). Davidson and Janssens (2006) suggested that a significant fraction of relatively unstable SOM is clearly subject to temperature-sensitive decomposition. Karhu et al. (2010) studied the temperature sensitivity of soil carbon and found that 30-45% more soil carbon will be lost in a warmer climate over the next few decades assuming that there will be no change in carbon input. To compensate for this loss, forest biomass productivity would need to increase by 100-120% (Karhu et al., 2010). O'Donnell et al. (2010)

7 performed an analysis on the sensitivity of soil to climate change and found severe losses of organic carbon from soil. Climate changes, particularly temperature increases, have effects on terrestrial ecosystems and have the potential to act as positive feedback in the climate change system (Friedlingstein et al., 2006; O'Donnell et al., 2010).

1.3.3 Substitution of fossil fuels and carbon-intensive products

In the past, forest production has been used mostly for wood products and pulp and paper. Recently, the concept of whole-tree utilisation has emerged, which focuses on the carbon benefits of forest management and product use. Recovered forest residues, tree stumps, wood chips, sawdust, and bark are important sources for bioenergy. In addition, wood material is widely used for building construction, and its use will be growing in the future (Ministry of Industry, 2004). Schulz (1993) speculated that dependency on wood would increase with increased interest in renewable resources to replace conventional materials and energy for environmental reasons. The use of forest biomass as a source of energy was considered during the global energy crisis in the 1970s. It was also in the 1970s when wood products were thought of as a sustainable source of materials for construction in place of energy-intensive products such as cement and steel (Boyd et al., 1976). A methodology for conducting energy analysis was developed in 1974 to provide consistency and comparability among studies (IFIAS, 1974). The Consortium for Research on Renewable Industrial Materials (CORRIM) was established by the National Academy of Sciences of United States (US) to study the effects of producing and using renewable materials (Lippke et al., 2004). CORRIM developed a comprehensive life cycle inventory of environmental inputs and outputs from forest regeneration through product manufacturing, building construction, use of wood in buildings and their maintenance, and disposal (Malmsheimer et al., 2008). The analysis focused on the energy impacts associated with various wood-based building materials and showed that wood-based materials are less carbon-intensive than other structural materials that perform the same function (Lippke et al., 2004). Methodologies for the analysis of forest biomass as an input in the energy system and for comparisons with carbon- intensive materials for carbon balance were developed in the US, New Zealand, Australia and in Europe. Some noted developments have been highlighted in several reports (Boustead and Hancock, 1979; Buchanan and Honey, 1994; Fossdal, 1995; Gustavsson et al., 1995; Schlamadinger and Marland, 1996; Börjesson et al.,

1997; Schlamadinger et al., 1997; Gustavsson and Karlsson, 2006; Eriksson et al., 2007). Schlamadinger et al. (1997) and Marland and Schlamadinger (1997) developed methodologies to assess forest biomass use for fossil fuel substitution. Börjesson and Gustavsson (2000) calculated GHG emissions from the life cycles of multi-story building construction. Pingoud and Lehtilä (2002) studied the energy efficiencies of different types of wood products. Lippke et al. (2004) compared energy consumption of energy-intensive materials to CORRIM’s life cycle assessment of wood-based building construction. Werner et al. (2005) explored GHG emission reduction as a result of the use of wood materials in construction and interior finishing. Gustavsson et al. (2006a) explored the possibility of wood recovery in different stages of the wood chain and its role in GHG emission mitigation. Eriksson et al. (2007) considered different scenarios for forest production and biomass use and found that forest stand fertilisation, residue and stump harvesting, and the use of wood as a construction material provided high reductions in CO2 emissions. Further studies have shown that the use of wood products and biomass residues significantly reduces net CO2 emissions, provided that the wood material used comes from sustainably managed forests and that biomass residues are used responsibly (Salazar and Meil, 2009; Gustavsson et al., 2010; Sathre and O'Connor, 2010).

1.4 Knowledge gaps

Previous studies have made significant contributions to assessing forest biomass production, soil carbon stock change, and substitution benefits for carbon emissions reductions. Nonetheless, biomass production projections in north- central Sweden due to the regional climate change effect have not been carried out before. Biomass growth projections in most cases include only the forest growth data (from inventory) instead of physiological processes of plant biomass production with the interaction to the environment. Studies of soil carbon stock changes may require a comprehensive approach such as input of organic materials, decomposition processes, and the climate change temperature increase effect to explain. Forest biomass production and product use were considered most often at the stand level. When this study began, there were no comprehensive studies that quantified the full chain of forest biomass production, harvest, effects in soil carbon stocks, and biomass use implications for carbon balances on a landscape level including climate change effects. In general, little work has been done using

9 an integrated methodological approach with a system perspective to draw a complete picture of carbon balance from forest biomass production and utilization.

1.5 Study objectives

This study explores carbon balances using an integrated analytical framework for the entire forest product chain, including forest management on a landscape level. The aims of the study are as follows: • To quantify the potential biomass production due to climate change in the Middle Norrland region of Sweden and the total carbon balance effects of biomass production and substitution of carbon-intensive materials. • To quantify the potential biomass production due to intensive forestry practices and the total carbon balance effects of biomass production and substitution of carbon intensive materials.

1.6 Thesis organisation

This thesis is based on three original papers and is divided into two parts. The first part includes an introduction, a discussion of methodology, a synthesis of the results presented in the papers, and a discussion and conclusion. The second part contains the three original papers. Paper I analyses the effects of climate change on forest biomass production and subsequent effects on carbon stock changes in standing biomass, soil carbon, and carbon emission reduction as a result of fossil fuel and building material substitution. Paper II analyses the effects of intensive forestry on biomass production, the carbon stock changes in forest biomass and forest soil, and carbon emission reduction as a result of fossil fuel and building material substitution. Paper III provides integrated results of total carbon balances in standing biomass, forest soil, and wood products as well as substitution effects. The major portions of the results and discussion sections are based on Paper I and II.

2 Methods

2.1 Integrated approach

An integrated carbon analysis includes modelling of forest generation, growth and harvest, the use of products, and the end-of-life management of products. The integrated approach used in this thesis consists of four different sub- models (BIOMASS, HUGIN, Q model, and SUBSTITUTION). A process-based model for forest biomass growth (see Paper I, II and III) provides biomass production values. These values serve as input for the empirical model HUGIN to estimate the harvest levels in the forests. The use of forest products as bioenergy and wood and its potential carbon reduction are calculated in a system analysis- based model (SUBSTITUTION). Soil carbon developments under different types of forest management practices are modelled using the Q model. The analysis is made for a 100-year period, starting from the year 2010 and ending at the year 2109. The flow chart used in this study is presented in Figure 1. The balance between net primary production (NPP2) and biomass loss during ecological processes determines the annual change in plant biomass (Chapin et al., 2002). At the same time, the changes in plant biomass have influence on changes in below- ground nutrient content. The analysis begins with a calculation of NPP in the BIOMASS model to provide NPP data, which is fed into the HUGIN system. The annual change in tree biomass based on all silvicultural activities, from regeneration to the final felling, is incorporated into HUGIN to calculate the amount of harvestable biomass. There are several possibilities for the utilisation of tree biomass after harvesting. For example, stem-wood may be utilised as construction materials, and tree residues may be utilised for bioenergy. The model outputs, in the form of different assortments of forest products from a landscape, are used in the SUBSTITUTION model to conduct a system analysis of forest biomass utilisation based on a life cycle perspective. Fossil carbon emissions from forest operations (CO2 emissions during regeneration, pre-commercial thinning, commercial thinning, fertilisation, and final felling) are included. Finally, carbon balances in standing biomass, harvested wood products, and forest soil are

2 NPP is the balance of carbon gained by GPP and the carbon lost by respiration of all plant parts (Chapin et al. 2002). 11 estimated. The carbon reduction associated with the use of harvested biomass to replace fossil fuel and carbon-intensive construction materials is also calculated. Photosynthesis GPP Plant Nutrient cycling NPP Litter and branch fall physiological Biomass increment processes Below-ground biomass

Annual increment, Silvicultural Standing biomass Forest growth activities and harvest Tree mortality Retained residues

Materials Residues Biomass utilisation Building Biofuel

Processing residues Construction and Demolition residues Soil carbon

Carbon balance Figure 1 Schematic diagram of carbon balances during forest production and biomass utilization

2.2 Functional unit

In this study, the carbon balance associated with forest production and the utilisation of wood and non-wood products is analysed. A “functional unit” is a key element for defining the function(s) to enable the objective comparison of different systems. This analysis focuses on the potential climate change mitigation effects of forest management and use. Hence, the overall net carbon emissions reduction in grams of carbon per unit of forest land is considered as a functional unit in this study.

2.3 Reference system

Forest management in this study is based on clear-cut forestry with one pre-commercial thinning and two to three commercial thinnings before the final harvest. The harvested biomass is assumed to be used as “‘stem-wood” and

“whole-tree”. It is assumed that stem-wood is used as construction material to replace carbon-intensive materials, while the remaining processing residues, small- diameter pine and spruce stem-wood, and all deciduous stem-wood are used as biofuels. The Wälludden building constructed in Växjö, Sweden, which is functionally equivalent to a concrete frame building, is used as a reference building for wood use. The Wälludden building is described in detail by Dodoo (2011), Sathre and Gustavsson (2007), Sathre (2007), Gustavsson et al. (2006b), and Gustavsson and Sathre (2006). Coal and fossil gas are considered as substituted reference fossil fuels, which currently covers 80% of the world’s primary energy (OECD/IEA, 2010).

2.4 Activities included in system analysis

Various activities and processes included in the study show the life cycle carbon balance (Table 1). The activities are defined to include the life cycle of forest biomass and in case of non-wood building materials, the analysis includes carbon emissions during the material production and its processes to be substituted by forest biomass. All energy inputs are accounted in the analysis. Table 1 Activities and processes included in the life cycle carbon balance Description Activities and processes Carbon implications included Production of Forest operation activities Carbon stock change in forest and materials (regeneration, pre-commercial building materials; (forest thinning, thinning, harvesting, Carbon stock changes in forest soil; biomass and fertilisation); Fossil fuel use for forest operation, non-wood) Wood processing, extraction of non-wood material production; process residues and Cement process. transportation; Non-wood material extraction, processing, and transportation. Use of Forest biofuel and Wood Forest residues and wood-process materials processing residues use in residues replaces fossil fuel; power plant; Wood materials replace non-wood Wood materials use in materials. buildings. End use of Building demolition, Wood use as biofuel replaces fossil materials Recovery of wood materials. fuel.

2.5 Study area

The study area (Paper I, Figure 1) encompasses the counties of Jämtland and Västernorrland in north-central Sweden, from 61° 33’ to 65° 07’ N latitude and from 12° 09’ to 19° 18’ E longitude. The elevation ranges from sea level to 1500 m in altitude. The annual precipitation ranges from 600-700 mm in the eastern part to 1500 mm in the western part (SMHI, 2009), and the duration of the growing season (mean temperature above +5 °C) is approximately 120-160 days per year (Sveriges- Nationalatlas, 1995). The average monthly temperatures during 1961-1990 varied in January between +1 °C to -12 °C and in July between 10 °C to 14 °C. The forest land areas of the counties of Jämtland and Västernorrland are 3.5 and 1.9 million hectare, respectively. The dominant tree species are Norway spruce (Picea abies (L.) Karst.), Scots pine (Pinus sylvestris L.) and birch (Betula spp.) (Swedish Forest Agency, 2010). The study area accounts for approximately 10% of the total forest land of Sweden. Detailed descriptions of the study area are given in Papers I and II.

2.6 Scenarios Five different forest management scenarios are formulated (Table 2) to describe the potential variability of climate and forest management practices through 2109. Table 2 Overview of scenarios Scenario Climate Forest management goals Biomass use Reference No change Reference (current management) Stem wood or whole tree Climate Change Current management Stem wood or whole tree Environment Change Fulfil environmental goals Stem wood or whole tree Production Change Increase biomass production Stem wood or whole tree Maximum Change Maximize biomass production Stem wood or whole tree

The Reference scenario in Paper I assumes an unchanged climate, and the Climate scenario assumes climate change as described in Section 2.7 below. These two scenarios assume current forest management practices. Paper II includes four scenarios, with the Climate scenario in Paper I serving as a Reference scenario in Paper II. The remaining scenarios assume different forest management practices to meet different objectives. The Environment scenario satisfies Swedish environmental goals by increasing the forest area allocated to natural set aside area. The Production and Maximum scenarios aim to meet higher production goals, with intensive forestry practices adopted to increase forest production. The

Production scenario assumes to include improved genetic material in seedling plantations, soil scarification, selection of tree species suitable for particular site conditions, and shorter time between clear-cut and re-planting. In addition, pre- commercial thinning and traditional fertilization area is increased. Balanced nutrient supply in young stands of Norway spruce and in former agricultural land forest are also included. The Maximum scenario includes all those activities, and also assumes to replace Scots pine with Lodgepole pine, and balanced nutrient supply in young stands of Lodgepole pine and Norway spruce forest land. Harvested biomass is used as “stem-wood” and “whole-tree”. Stem-wood harvest assumes the extraction of 95% of tree stems of >20 cm diameter with the rest of the biomass left on the forest floor. Whole-tree harvest assumes a harvest of 95% of the felled stem-wood, 75% of needles, branches and tops (referred to as “residue”), and 50% of stumps and coarse roots.

2.7 Climate change

This study considers climate change based on the IPCC SRES B2 climate change scenario (Nakicenovic and Swart, 2000). This scenario corresponds to moderate emissions of greenhouse gases leading to an atmospheric CO2 concentration of 621 ppm by the year 2100 (Levy et al., 2004). The simulations of temperature change were conducted using the Rossby Centre’s regional atmospheric model based on IPCC global scenarios with the downscaled climate model RCA3 with a grid of approximately 50 km x 50 km (SMHI, 2009). In the calculations, RCA3 uses global driving variables from the general circulation model ECHAM4/OPYC3 (Kjellström et al., 2006). Climate projections for north- central Sweden are for an increase in average regional annual temperatures of more than 4 °C by 2100 for SRES B2 (see Figure 3 in Paper I). Soil carbon modelling was performed with the assumption of a shift in latitude to 58-59 °N instead of the actual latitude of 62-63 °N because the temperature change due to climate change is expected to make the climatic conditions at 62-63 °N similar to those at 58-59 °N.

2.8 Forest biomass production modelling with BIOMASS

The process-based growth model BIOMASS was used to estimate NPP. BIOMASS uses canopy characteristics, foliage photosynthetic characteristics, and daily meteorological conditions as inputs and calculates daily evapotranspiration from the stand. Later, foliage, stem and branch respiration values are subtracted

15 from the total CO2 uptake by the canopy to obtain daily NPP values. Both litterfall and stand water balances are incorporated into the model. BIOMASS consists of a series of equations based on established theories of physiological plant processes and soil-water dynamics. This is summarized by the simple equation

NPP= GPP − Rplant

where GPP is the gross primary production, and Rplant is respiration by tree components. All sub-models for the calculation of carbon accumulation in foliage, branches, stems, bark, and roots, including below-ground components, were formulated, and respiration in each component was subtracted from the total accumulation to obtain NPP. A detailed description of the BIOMASS model for canopy photosynthesis and water balance was given by McMurtrie et al. (1990) and McMurtrie and Landsberg (1992). BIOMASS uses climatic data from the SRES B2 scenario. These climatic data were taken from transient simulations of 1961- 2100 with reference to 1961-1990 climatic data. The simulations were performed for a 30-year cycle to determine the effects of temperature change on NPP increment which was found to be 21.6%, 11.0%, and 13.0% greater for Norway spruce, Scots pine, and Silver birch, respectively compared to unchanged climate. The NPP output data were first estimated and then assimilated with the HUGIN model to determine the forest growth function at a county level.

2.9 Forest biomass harvest modelling with HUGIN

HUGIN is a model system for long-term forecasts of timber yields and potential harvest levels that is used for planning at a regional level and for strategic planning for large forestry companies. The model uses data from sample plots from the Swedish National Forest Inventory (NFI) to define initial forest conditions. HUGIN describes all stages in plant development from stand establishment through treatments, thinning, and final fellings. The growth simulators in HUGIN are valid for all forest land in Sweden, for all types of stands, and within a wide range of management practices. The details on stand establishment, treatments, thinnings, and final fellings are given in Paper I (Section 2.4.2), and further details are provided by Lundström and Söderberg (1996).

2.10 The use of forest biomass as bioenergy and construction materials

Forest biomass harvested in the whole-tree scenario is classified as large diameter stem-wood and residues. Large-diameter stem-wood is defined as stems with a minimum diameter of 20 cm and is assumed to be used to produce wood construction materials to substitute conventional reinforced concrete. The material substitution effects are based on a multi-story apartment building (described in Section 3.3). The details of wood use for material substitution in building construction are given by Gustavsson et al. (2006b). Large-diameter pine and spruce stem-wood is used for the production of wood construction materials, while the processing residues (bark, sawdust, chips) are used for bioenergy. In the whole-tree scenario, tops, branches, needles, and stumps including coarse roots are also used for bioenergy. Small-diameter pine and spruce stem-wood, and all deciduous stem-wood are used for bioenergy. In this study, small-diameter stem-wood is not used for pulp and paper production because we have not analysed whether the demand for additional pulpwood in the pulp industry in Sweden would increase. The wood materials in the buildings are assumed to be recovered and used for energy purposes as biofuels at the end of the building service life. All biofuels are assumed to substitute either coal or fossil gas in stationary plants with conversion efficiencies of 100% and 96% relative to the conversion efficiencies of the respective fossil fuel-fired plants (Gustavsson et al., 2006b). Fossil energy inputs for the recovery and transport of biofuels and forest management practices are analysed according to Eriksson et al. (2007) and Berg and Lindholm (2005).

2.11 Carbon balance 2.11.1 Total carbon balance

In this study, total carbon balance is defined as the reduced net carbon emission into the atmosphere due to forest biomass production and their uses. Hence, a positive carbon balance refers to a net carbon benefit, and a negative carbon balance refers to net carbon emission. The activities included in the system analysis (Section 2.5) are categorised so as to represent either carbon emissions reductions or carbon emissions. The carbon emissions reductions are expressed as a positive change as a result of increased standing biomass carbon stock, changes in forest soil carbon stock, changes in wood product carbon stocks in the buildings, substitution carbon benefits achieved due to the replacement of fossil fuels by 17 recovered forest biofuels and wood processing residues, and substitution carbon benefits due to the avoidance of non-wood material production and process emissions. The carbon emissions are expressed as the fossil fuel emissions during forest operation activities including fertilisation. Avoided emissions from production of non-wood materials are included in the calculation of substitution carbon benefits. The parameters included for material production and process emissions include fossil-fuel-cycle carbon emission, end-use fossil fuel carbon emission, emissions due to end-use electricity to extract, process, and transport the materials, and emissions from industrial process reactions such as the calcination reaction during manufacture of the cement. The total carbon balance for the Jämtland and Västernorrland forest landscape is calculated as:

CB=∑[ ∆ Csb +∆ C fs +∆ C wp + C scb −( C fo + C f )] where CB is the total carbon balance (gram C);

ΔCsb is the change in standing forest biomass carbon stock (gram C);

ΔCfs is the change in forest soil carbon stock (gram C);

ΔCwp is the change in carbon stock in wood products (gram C);

Cscb is the substitution carbon benefits due to the replacing of fossil fuel with recovered biofuel (gram C) and avoidance of material production and process emissions (gram C) due to use of wood materials;

Cfo is the carbon emissions due to forestry operation activities (gram C);

Cf is the equivalent carbon emissions due to fertilization in the forest (gram C); Emissions of nitrous oxide and methane also occur during fertilization. Global Warming Potentials of nitrous oxide (298) and methane (25) are used to

convert them into equivalent CO2 emissions based on their relative radiative

efficiencies and atmospheric residence times compared to CO2 (IPCC, 2007c). A positive carbon balance signifies reduction in carbon emission to the atmosphere, while a negative carbon balance signifies increase in carbon emission to the atmosphere.

2.11.2 Standing biomass carbon stock

In this study, the annual increase in standing biomass carbon stock is estimated after the deduction of harvest, mortality, and disturbances from the total

18 annual biomass production of the forest. On average, approximately 95% of the annual production of forest biomass is felled. HUGIN calculates the standing biomass carbon stock based on an evaluation of forest production, disturbances, harvests, and retained standing biomass in the forest. HUGIN uses spruce stumps as a proxy for the calculation of deciduous and pine tree stumps.

2.11.3 Soil carbon stock

In this study, soil carbon stock changes due to different forest management practices and removal strategies are estimated. The estimates are based on the amount of organic carbon available in the SOM, soil carbon development due to litterfall from the canopy of the standing biomass and below ground carbon stock development, and carbon development due to biomass residues left in the forest after harvest and thinning. A fraction of leaves/needles, branches, and fine roots die and fall in a standing stand, and a fraction of stems and parts of stumps and coarse roots are left at the site during harvest and thinning, contributing to soil carbon stock development. The decomposition of the different litter fractions is calculated using the Q model, which incorporates the invasion rates of different litter types (Ågren, 1983; Ågren and Bosatta, 1998; Hyvönen and Ågren, 2001). This study excluded the soil carbon stock available in mineral soil because the land use history of the landscape was unknown. The details of soil carbon modelling are given in Paper I (Section 2.5).

2.11.4 Wood product carbon stock

Wood products stored in the buildings are considered for wood product carbon stock estimation. Increasing the total amount of wood product stock allows for the storage of more carbon in the products. The carbon content in wood was assumed to be 49% of the dry weight of the wood (Swedish Forest Agency, 2010). Wooden buildings store carbon for the lifetime of the buildings but not beyond that time. Therefore, it was assumed that when the building is demolished, 100% of the used wood is recovered and used as biofuel to replace fossil fuels. The carbon in the wood products returns to the atmosphere as CO2 when the wood is used as biofuel.

2.11.5 Substitution of fossil fuels and carbon-intensive products

Carbon emissions reduction due to substitution of fossil fuels and carbon- intensive products by the use of forest biomass is estimated. This substitution is calculated by estimating fossil fuel emissions during material production and during process reactions (e.g., cement calcination) of material production. These emissions are compared with emissions from wood construction material production and processes. Fossil fuel substitution by the use of biomass residues as bioenergy is estimated separately. A full fuel-cycle emission is considered to estimate carbon emissions reduction. This study assumes that the biomass considered in this study is harvested from sustainably managed forests thus the combustion of biofuel is assumed to contribute a net emission of zero. However, emissions from fossil fuels used to produce, transport, and process the biofuels contribute net emissions. Detailed methods and processes for bioenergy use, conversion, and co-generation opportunities are discussed elsewhere (Schlamadinger and Marland, 1996; Schlamadinger et al., 1997). This study assumes that construction of the reference building is substituted by a wood- framed building that is replaced with another wood-framed building using similar wood products indefinitely.

2.11.6 Forestry operations and carbon emission

Emissions from fossil fuels used for forest operations (stand establishment, thinnings, final harvest of roundwood, and forwarding and transport of roundwood to mills and fertilisation) are calculated based on Berg and Lindholm (2005). Energy use data for forestry operations for northern Sweden are used to calculate carbon emissions. The details of management activities and fertilisation and carbon emissions are described in Paper II (Section 3.6).

2.12 Data quality

Many factors can affect the data used in carbon balance calculations. Different values of solar radiation, water amount in the soil, plant numbers and their competition in the forest stand, the photosynthesis capability of a single plant, and growing capacity in terms of carbon content storage capacity may result in different values in biomass production, thereby affecting the final result. Forest inventory data collected by the Swedish Forest Agency has been used to model

20 forest growth and harvest by HUGIN. Any error in the data bank may also affect the final results of this study. This study assumes the existence of only three tree species; therefore, the existence of any other species in the forest stand would change the values and results. Forest product removals may not match current practices, and there will be different results based on different systems of harvest, removal, and use. Furthermore, the differences in physical properties of raw materials, processes of material production, types of fossil fuel use, and the amount of wood to be utilised in the buildings may lead to different results.

3 Results

3.1 Forest biomass production and harvest

Figure 2 shows the annual forest biomass production and harvest at the year 2010 and year 2109 for the five scenarios for both whole-tree and stem-wood use. The annual biomass in the Reference scenario increased by ~17% in the year 2109 for whole-tree biomass and by ~8% for stem-wood biomass compared to the base year 2010 (Paper I). More biomass was produced in the other scenarios compared to the Reference scenario, with values increasing up to ~75% for whole- tree biomass and ~65% for stem-wood biomass in the Maximum scenario during the study period from 2010 to 2109 (Paper III). 25 Whole-tree production Whole-tree harvest 1

- Stem-wood production Stem-wood harvest 20

5 Dry biomass biomass Dry TgYear

0 2010 2109 2010 2109 2010 2109 2010 2109 2010 2109

Reference Climate Environment Production Maximum

Scenario and year Figure 2 Annual biomass productions and harvests at the beginning and end of the study period for all scenarios.

The potential amount of biomass harvested during the study period increased significantly when taking into account the effect of climate change and intensive forestry practices. However, the harvest was only slightly larger in the Environment scenario compared to the Reference scenario. The Climate, Production and Maximum scenarios all resulted in larger biomass harvests than the Reference scenario (Figure 2). Harvests were greater in the Climate, Environment, Production, and Maximum scenarios than in the Reference scenario (Table 1 in Paper I and Table 3 in Paper II). The differences in forest production and harvest between the different scenarios and the Reference scenario for both whole-tree biomass and stem-wood biomass represent a net increase during the study period (Table 3). The differences

22 gradually increased throughout the entire study period (Paper I, Table 1 and Paper II, Table 3).

Table 3 Differences in forest biomass production and harvest between the Reference scenario and other scenarios (Tg dry biomass) Cumu 2010- 2020- 2030- 2040- 2050- 2060- 2070- 2080- 2090- 2100- lative 2019 2029 2039 2049 2059 2069 2079 2089 2099 2109 (Tg) Differences between the Climate and Reference scenarios Biomass production Stem-wood 0.18 0.27 0.50 0.70 0.84 1.10 1.3 1.46 1.85 2.23 104 Whole-tree 0.35 0.51 0.94 1.35 1.57 2.01 2.4 2.72 3.43 4.11 193 Biomass harvest Stem-wood 0.06 0.18 0.34 0.42 0.52 0.66 1.14 1.21 1.56 1.8 79 Whole-tree 0.10 0.26 0.55 0.68 0.79 1.07 1.77 1.87 2.33 2.81 123 Differences between the Environment and Reference scenarios Biomass production Stem-wood 0.18 0.35 0.53 0.72 0.70 0.92 1.12 1.3 1.44 1.76 89 Whole-tree 0.31 0.66 1.04 1.32 1.29 1.54 1.93 2.25 2.53 2.97 157 Biomass harvest Stem-wood 0.30 0.11 0.04 0.10 0.15 0.25 0.47 0.57 1.27 1.03 33 Whole-tree 0.47 0.27 0.00 0.15 0.10 0.25 0.64 0.86 1.91 1.50 47 Differences between the Production and Reference scenarios Biomass production Stem-wood 0.28 0.55 1.13 1.42 1.70 1.82 1.92 2.10 2.24 2.66 158 Whole-tree 0.51 1.16 2.24 2.92 3.29 3.44 3.63 3.95 4.13 4.77 298 Biomass harvest Stem-wood 0.00 0.19 0.54 0.70 0.85 1.35 1.77 1.87 2.17 2.33 118 Whole-tree 0.03 0.33 0.80 1.25 1.40 2.15 2.74 2.86 3.41 3.70 187 Differences between the Maximum and Reference scenarios Biomass production Stem-wood 0.28 0.55 1.13 1.82 2.10 2.42 2.82 3.20 3.54 3.96 218 Whole-tree 0.51 1.16 2.44 3.72 4.09 4.54 5.33 6.05 6.63 7.37 418 Biomass harvest Stem-wood 0.10 0.29 0.64 0.9 0.95 1.35 1.97 2.17 2.47 2.83 136 Whole-tree 0.13 0.53 1.10 1.45 1.40 2.25 3.04 3.36 3.81 4.50 217

Paper I and II discuss differences in biomass production and harvest in different tree species (Norway spruce, Scots pine and birch) due to climate change and intensive forestry practices. The production and harvest of biomass are classified into different types of forest products. Stem-wood and stump biomass yield a greater amount of biomass than other biomass components. However, the stump biomass may be overestimated because the HUGIN model uses spruce stumps as a proxy for the calculation of deciduous and pine tree stumps. Furthermore, because of very few deciduous trees in the past, forest management practices in the past did not consider promoting their production. But today, the number of deciduous trees in forests has increased more than the model expected, which may result in an overestimation of deciduous tree biomass production. It

23 may be also because the model considers deciduous trees to live longer than coniferous trees, so the model could have considered the larger deciduous tree sizes compared to coniferous trees.

3.2 Standing biomass carbon stock

The average annual carbon stock in the tree biomass for all scenarios increased continuously during the study period (Table 4). Overall, the Maximum and Environment scenarios had higher rates of carbon stock increase than the Reference and Production scenarios.

Table 4 Average annual tree biomass carbon stock changes during each 10-year period (Tg carbon year-1) for Jämtland and Västernorrland (Paper I and II) 2010- 2020- 2030- 2040- 2050- 2060- 2070- 2080- 2090- 2100- 2019 2029 2039 2049 2059 2069 2079 2089 2099 2109 Reference scenario 0.38 0.76 0.22 0.55 0.34 0.20 0.35 0.40 0.67 0.45 Climate scenario 0.49 0.85 0.35 0.80 0.63 0.53 0.45 0.61 0.96 0.75 Environment scenario 0.81 1.25 0.75 1.08 0.90 0.81 0.93 0.97 0.76 0.98 Production scenario 0.61 1.08 0.79 1.22 1.07 0.59 0.47 0.59 0.65 0.53 Maximum scenario 0.54 1.01 0.73 1.52 1.44 1.08 1.07 1.34 1.58 1.31

The total standing biomass at the beginning and end of the study period for each scenario was calculated (Figure 3). All scenarios had larger standing biomasses at the end of the study period than at the beginning. The stock increases for the Maximum and Environment scenarios were 56% and 44%, respectively, compared to a 21% increase in the Reference scenario. The production increase in the Reference scenario is because the large areas with low productive forests are replaced with new planted forests with improved genetic material. The greater standing biomass in the Production and Maximum scenario may be explained as increased production due to climate change and intensive forestry practices, while it may be explained in the Environment scenario as a result of an increase in the amount of land set aside and a lower harvest intensity.

Stemwood Stump Branch Bark Needle 700

600

500

400

Drymass Tg 300

200

100

0 Year Reference Climate Environment Production Maximum 2010 2109 2109 2109 2109 2109 Scenario and year Figure 3 Standing forest biomass carbon stock at the beginning (2010) and end (2109) of the study period in the different scenarios (Paper II, Figure 6).

3.3 Soil carbon stock

Soil carbon changes due to different types of forest management and residues harvest were analysed. The results for soil carbon changes are presented in Table 2 of Paper I and Table 5 of Paper II, and the total soil carbon stock are presented in Figure 3 of Paper III. Figure 4 shows that the soil carbon changes occurred throughout the entire study period. The soil carbon stock increased more following stem-wood biomass harvest than following whole-tree biomass harvest. Total soil carbon stock was the highest for Maximum scenario and the lowest for Climate scenario compared to all other scenarios.

20 Forest soil carbon stock Reference 18 Climate 16 Environment 14 Production

Carbon stock (Tg C) (Tg stock Carbon 12 Maximum

10 2009 2019 2029 2039 2049 2059 2069 2079 2089 2099 2109 ------

2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100 Figure 4 Carbon stocks in forest soil in the different scenarios during 100-years with whole-tree biomass use.

3.4 Substitution of fossil fuels and carbon-intensive products

The average annual carbon emissions reductions due to the use of forest biomass in place of fossil fuel and materials during each 10-year period are given in Table 5. Carbon emission reduction amounts were smaller when only biofuel substitution or non-wood construction material substitution was considered. The carbon emission reductions were smaller when biofuel only considered for fossil fuel substitution compared to both biofuel and construction material considered for fossil fuel substitution.

Table 5 Average annual carbon emissions reductions (Tg C year-1) due to fossil fuel and carbon-intensive material substitution by the use of stem-wood and whole- tree biomass during each 10-year period (emissions reduction is shown with coal as a reference fuel, with values in parentheses representing substitutions using fossil gas as a reference fuel) (Paper I and II) 2010- 2020- 2030- 2040- 2050- 2060- 2070- 2080- 2090- 2100- 2019 2029 2039 2049 2059 2069 2079 2089 2099 2109 Reference scenario Stem-wood 4.9(3.4) 4.9(3.4) 4.8(3.2) 4.9(3.3) 5.3(3.6) 5.2(3.5) 5.4(3.7) 5.3(3.7) 5.7(4.0) 6.1(4.3) Whole-tree 3.8(2.9) 3.8(2.8) 3.7(2.6) 3.7(2.7) 4.1(3.0) 4.0(2.9) 4.3(3.1) 4.2(3.1) 4.6(3.4) 5.0(3.6) Climate scenario Stem-wood 3.9(2.9) 3.8(2.8) 3.8(2.7) 3.9(2.8) 4.3(3.1) 4.4(3.2) 4.7(3.4) 4.7(3.4) 5.1(3.7) 5.6(4.1) Whole-tree 5.0(3.5) 4.9(3.4) 5.0(3.4) 5.1(3.5) 5.6(3.8) 5.7(3.9) 6.1(4.2) 6.1(4.2) 6.5(4.5) 7.1(4.9) Environmental scenario Stem-wood 3.6(2.7) 3.6(2.7) 3.6(2.6) 3.7(2.7) 4.0(2.9) 4.1(3.0) 4.3(3.2) 4.5(3.3) 4.8(3.5) 5.0(3.7) Whole-tree 4.6(3.2) 4.6(3.2) 4.7(3.2) 4.9(3.3) 5.2(3.5) 5.3(3.6) 5.5(3.8) 5.7(4.0) 6.1(4.2) 6.4(4.4) Production scenario Stem-wood 3.8(2.9) 3.8(2.9) 4.0(2.9) 4.2(3.0) 4.6(3.3) 5.0(3.6) 5.3(3.9) 5.4(4.0) 5.8(4.3) 6.2(4.6) Whole-tree 4.9(3.5) 5.0(3.5) 5.2(3.5) 5.5(3.7) 6.0(4.1) 6.4(4.4) 6.7(4.7) 6.8(4.8) 7.3(5.1) 7.8(5.5) Maximum Scenario Stem-wood 3.9(2.9) 3.9(2.9) 4.1(2.9) 4.3(3.1) 4.7(3.4) 5.0(3.6) 5.5(4.0) 5.7(4.2) 6.0(4.4) 6.6(5.0) Whole-tree 5.0(2.9) 5.0(2.9) 5.4(2.9) 5.6(3.1) 6.0(3.4) 6.4(3.6) 7.0(4.0) 7.2(4.2) 7.6(4.4) 8.4(5.0)

Production increase associated with climate change (see Paper I) could increase carbon reduction by 10% using coal as a reference fuel and by 8% using fossil gas as a reference fuel compared to the Reference scenario. With intensive forestry practices, the Maximum scenario yielded a larger carbon emission reduction. The average annual carbon emission reductions as a result of whole -tree and stem-wood biomass use to replace fossil fuels and non-wood construction materials are shown in Figure 5. For all scenarios, the substitution of coal fuel with whole-tree biomass provided the greatest carbon emission reduction.

7 Emission reduction by WT 6 biomass use replacing coal fuel

5 1 - Emission reduction by SW 4 biomass use replacing coal fuel

3 Emission reduction by WT 2 biomass use replacing fossil gas Tg carbon year carbon Tg

1 Emission reduction by SW 0 biomass use replacing fossil gas

Reference Climate Environment Production Maximum Scenario a

Climate change and intensive forestry significantly increased carbon emission reductions. Carbon emission reductions due to whole-tree use were compared to reductions due to the use of stem-wood. The greater carbon emission reduction occurred with whole-tree use. The Production and Maximum scenarios yielded the largest carbon reductions due to the greater amounts of biomass production and harvest in these scenarios compared to other scenarios. Overall carbon emission reductions due to substitution were greater compared to reductions due to standing biomass increases. Forest operation emissions were negligible compared to the carbon emission reductions.

3.5 Wood product carbon stock

Harvested wood products used as construction materials in buildings store carbon until the building is demolished. The carbon stocks calculated to be in the buildings are shown in Figure 6. The carbon stocks increased in all scenarios throughout the study period. A larger increase was observed in the Maximum and Production scenarios than in the other scenarios. 120 Wood product C stock 100 Reference

80 Climate

60 Production 40 Maximum

Carbonstock (Tg C) 20 Environment 0 2009 2019 2029 2039 2049 2059 2069 2079 2089 2099 2109 ------2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100

Figure 6 Wood product carbon stock (Tg carbon) due to non-wood material substitution (Paper III).

3.6 Carbon emissions from forest operations

Intensive forestry requires additional operational activities in the forest. In addition, a larger biomass production requires additional work for e.g., thinning and harvest of the biomass. This study includes the calculation of carbon emissions

28 during forest operational activities from forest regeneration to the final harvest. The Maximum, Production, and Climate scenarios yielded greater carbon emissions than the Reference scenario. The greater emissions in these scenarios are due in part to the fertiliser production and application, and greater amounts of biomass to be harvested and hauled to the road head (Figure 7). The largest percentage of the emissions comes from round wood harvest during thinning and final felling. Stump harvest and haulage result in carbon emissions comparable to those associated with residue collection and haulage. 0.25 ) 1 -

0.20

0.15

0.10 Reference Climate Environment

Carbon emission, (Mg, year emission,(Mg, Carbon 0.05 Production Maximum

0.00 2019 2029 2039 2049 2059 2069 2079 2089 2099 2109 ------2010 2020 2030 2040 2050 2060 2070 2080 2090 2100 Year Figure 7 Carbon emissions due to forest operations and fertilisation for all scenarios for whole-tree biomass (Mg carbon year-1).

3.7 Total carbon balance

Table 6 shows the total carbon balance, which is reduced net carbon emissions into the atmosphere due to forest biomass production and their uses. The calculation of carbon emissions reductions takes into account the carbon stock increases in living biomass, in forest soil, in wood products, and substitution carbon benefit due to the substitution of non-wood materials and fossil fuels by the use of forest biomass. Forest operation emissions were deducted from the carbon emissions reductions due to substitution effects to obtain net substitution values. Thus, the total carbon balance is the net carbon emission reduction resulting from forest production and product use. The carbon balance is increasing every year because of carbon stock increases in trees, soil, and products, as well as increasing substitution effects.

Table 6 The total carbon balance including forest production, forest biomass harvest, soil carbon stock change, and stem-wood and whole-tree biomass use to replace of fossil fuels and carbon-intensive materials (Tg carbon year-1) during each 10-year period in Jämtland and Västernorrland. The total carbon balance is for coal as a reference fuel, with values in parentheses representing fossil gas as a reference fuel (Paper I and II) 2010- 2020- 2030- 2040- 2050- 2060- 2070- 2080- 2090- 2100- 2019 2029 2039 2049 2059 2069 2079 2089 2099 2109 Reference scenario Stem-wood 5.5(4.5) 5.4(4.4) 4.6(3.5) 4.9(3.8) 4.9(3.8) 5.0(3.9) 5.1(4.0) 5.1(4.3) 5.4(4.3) 5.4(5.0) Whole-tree 6.3(4.8) 6.2(4.8) 5.5(4.0) 5.9(4.3) 5.9(4.3) 6.0(4.4) 6.1(4.5) 6.4(4.8) 6.4(4.8) 6.5(4.8) Climate scenario Stem-wood 5.5(4.5) 5.5(4.6) 4.9(3.8) 5.4(4.3) 5.7(4.5) 5.8(4.5) 6.1(4.8) 6.3(5.0) 7.1(5.8) 7.5(6.0) Whole-tree 6.4(4.9) 6.5(5.1) 6.0(4.4) 6.6(4.9) 6.9(5.1) 7.1(5.2) 7.4(5.5) 7.6(5.7) 8.5(6.5) 9.0(6.8) Environmental scenario Stem-wood 5.5(4.5) 5.7(4.7) 5.0(4.0) 5.4(4.4) 5.6(4.5) 5.7(4.5) 6.1(5.0) 6.4(5.2) 6.5(5.2) 7.0(5.7) Whole-tree 6.3(4.9) 6.6(5.2) 6.1(4.6) 6.6(5.0) 6.8(5.2) 6.9(5.2) 7.3(5.6) 7.6(5.8) 7.8(5.9) 8.4(6.4) Production scenario Stem-wood 5.6(4.6) 5.8(4.8) 5.5(4.4) 6.2(5.0) 6.6(5.3) 6.5(5.1) 6.8(5.4) 7.1(5.6) 7.6(6.1) 8.1(6.5) Whole-tree 6.5(5.0) 6.8(5.3) 6.7(5.0) 7.4(5.7) 7.9(6.0) 7.9(5.9) 8.3(6.2) 8.5(6.4) 9.1(6.0) 9.7(7.3) Maximum Scenario Stem-wood 5.6(4.6) 5.8(4.8) 5.6(4.4) 6.6(5.4) 7.0(5.7) 7.0(5.6) 7.7(6.2) 8.2(6.7) 8.8(7.3) 9.4(7.7) Whole-tree 6.5(5.0) 6.8(5.3) 6.8(5.1) 7.9(6.1) 8.4(6.5) 8.5(6.4) 9.2(7.0) 9.7(7.5) 10.4(8.2) 11.2(8.7)

Figure 8 shows the average annual carbon emission reductions (Tg carbon year-1) over the 100-year period for all scenarios for whole-tree biomass use to replace coal and fossil gas. The largest carbon emission reduction was achieved in the Maximum scenario (8.53 Tg carbon year-1) when compared to the Reference scenario (6.07 Tg carbon year-1) while replacing coal fuel. The average annual carbon emission reductions were smaller when fossil gas was replaced. 9 8 Coal whole-tree

1 7 - 6 Fossil gas whole-tree 5 4 Coal stem-wood 3 Tg carbon Tgcarbon year 2 1 Fossil gas stem-wood 0 Reference Climate Environment Production Maximum Scenario Figure 8 The average annual carbon emissions reduction (Tg carbon year-1) for different scenarios when whole-tree biomass was used to replace coal and fossil gas.

The total carbon balance is highly influenced by substitution effect. The carbon balance increased for both stem-wood and whole-tree biomass during the study period, with the greatest increase occurring when whole-tree biomass was used as an alternative to fossil fuel. The use of whole-tree biomass resulted in lower soil carbon stock increases than the use of stem-wood biomass, though the increased substitution benefits more than compensated. Figure 9 shows the average annual carbon emission reductions per hectare of forest (Mg ha-1 year-1) for all scenarios for whole-tree biomass use during the study period. The reduction was greatest in the Maximum scenario. When coal was replaced with biofuel instead of fossil gas, the overall carbon emission reduction increased by ~30-35% for whole-tree biomass. Forest carbon stock increases were greater for the Environment and Maximum scenarios than for the other scenarios. The increase in the amount of set aside areas and lower harvest levels in the Environment scenario led to an increase in the forest carbon stock. Greater carbon emissions reductions in the Maximum scenario compared to other scenarios were due to forest management practices that promoted biomass production.

1.8 Coal whole-tree Coal stem-wood 1.6 Fossil gas whole-tree Fossil gas stem-wood 1

- 1.4 1.2 year 1 - 1 0.8 0.6 0.4 Mg carbon ha carbon Mg 0.2 0 Reference Climate Environment Production Maximum Scenario Figure 9 The average annual carbon emissions reductions per hectare of forest land (Mg carbon ha-1 year-1) due to the benefits associated with the use of whole-tree (WT) and stem-wood (SW) biomass to replace fossil gas and coal.

The carbon emissions reductions were greatest in the Maximum scenario. The cumulative carbon emissions reductions in different scenarios of whole-tree biomass use with coal as a reference fuel are presented in Paper I and II. The largest carbon reduction was found in the Maximum scenario, and the lowest reduction occurred in the Reference scenario. Figure 10a and 10b shows the

31 differences in cumulative carbon emission reduction between the Reference scenario and the Climate, Environment, Production, and Maximum scenarios for whole-tree biomass use (Figure 10a) and for stem-wood biomass use (Figure 10b) with coal and fossil gas as reference fuels. The difference is largest between the Maximum and Reference scenarios for both whole-tree and stem-wood biomass use with coal as a reference fuel. A greater amount of carbon emissions was reduced when coal was used as a reference fuel than when fossil gas was used as a reference fuel. a. 250 Whole tree biomass use Clim minus Ref Coal

200 Clim minus Ref Fossil gas Env minus Ref Coal 150 Env minus Ref Fossil gas

Prod minus Ref Coal 100 Prod minus Ref Fossil gas

50 Max minus Ref Coal

Max minus Ref Fossil gas 0 Reduced carbon emission (Tg carbon) (Tg emission Reducedcarbon 2069 2009 2019 2029 2039 2049 2059 2079 2089 2099 2109 ------

b. 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100 250 Stem wood biomass use Clim minus Ref Coal

200 Clim minus Ref Fossil gas

Env minus Ref Coal 150 Env minus Ref Fossil gas

100 Prod minus Ref Coal Prod minus Ref Fossil gas

50 Max minus Ref Coal

Reduced carbon emisson (Tg (Tg carbon) emisson Reducedcarbon Max minus Ref Fossil gas 0 2009 2019 2029 2039 2049 2059 2069 2079 2089 2099 2109 ------2050 2090 2000 2010 2020 2030 2040 2060 2070 2080 2100 Year Figure 10 Differences in cumulative carbon emission reduction (Tg carbon) between the Climate, Environment, Production and Maximum scenarios and the Reference scenario for each 10-year period.

4 Discussions

4.1 Forest production and harvest

The current average annual forest productivity for the Jämtland and Västernorrland forests is about 2.1 Mg (dry mass) ha-1 year-1 (Swedish Forest Agency, 2010). In this study, the forest biomass production started at 2.09 Mg (dry mass) ha-1 year-1 in the year 2010 and reached up to 3.14 Mg (dry mass) ha-1 year-1 in 2109 due to the effects of climate change. Levels of biomass production were larger when the effects of climate change and intensive forestry were included compared to results from a Reference scenario that did not include these effects. The largest biomass productions at the end of the study period were 3.3 and 3.74 Mg (dry mass) ha-1 year-1 in the Production and Maximum scenarios, respectively, which included both climate change and intensive forestry effects. Increasing forest production increases the potential harvest, substitution potential, standing biomass and soil carbon. The harvesting of forest residues from forest land during final felling is a common practice in Sweden, while stump harvesting is not. The Swedish forest agency recommends that at least 20% of logging residues and 15-20% of stumps should be left at the site of the final felling (Swedish Forest Agency, 2008, 2009). The practice of whole-tree harvest is debated because harvesting residues and stumps may reduce forest productivity in the following generation because of tree litters removals. This issue has been discussed in earlier studies (Jacobson et al., 2000; Egnell and Valinger, 2003; Smolander et al., 2010; Egnell, 2011). However, results from other studies, such as that performed by Egnell (2011), show that these effects are only temporary and can be compensated for by fertilisation. Nutrient deficiencies are assumed to be satisfied because this study includes balanced fertilisation (except for the Reference scenario). The forest management practices considered in this study assume that the biomass production and harvest is continuous without losing the forest’s capacity to produce and deliver at least the same amount of product every growth cycle. This study also follows criteria for increased environmental ambitions, for example increased set-aside areas for special environmental care proposed by the Swedish Environmental Agency. All scenarios follow, at a minimum, the specified environmental goals set by current Swedish forest and environmental policy.

4.2 Carbon-intensive products substitution

Substituting fossil fuels with forest biomass has a significant impact on the life cycle carbon balance. Biomass residues left in the forest after harvest release carbon into the atmosphere for some period of time as they decay; however, recovering residues to replace fossil fuels reduces net carbon emissions due to avoided fossil emissions and forest regrowth. Carbon emissions resulting from forest operations are less for stem-wood harvest than for whole-tree harvest because of the smaller amount of biomass to be harvested compared to whole-tree biomass. Although carbon emissions resulting from forest operations are greater for whole-tree harvest, the overall carbon emission reduction is greater with whole-tree biomass harvest compared to stem-wood biomass harvest, because of the increased substitution benefits. Substituting coal provides greater carbon emission reductions than substituting fossil gas, because coal emits more CO2 per unit of energy than does fossil gas. The production of non-wood construction materials such as cement and steel uses a greater amount of fossil energy compared to the amount needed to produce wood products. Therefore, replacement of carbon-intensive fuels and materials with wood products will generally reduce carbon emissions. However, there may be a question as to whether all of the wood material produced can be used in building construction or to replace fossil fuels. Jonsson (2009) claimed that about 30% of multi-storey houses constructed in Sweden by 2015 will be wood- framed buildings. When the total number of wood-framed buildings increases, the carbon benefit will increase (Gustavsson et al., 2006b). Wood products can also be exported to substitute for non-wood products in other countries. The use of bioenergy is also increasing, and international trade in bioenergy is made feasible by efficient long-distance transport methods (Junginger et al., 2008). Gustavsson et al. (2011) calculated that woody biofuels can be economically transported internationally with little loss of net energy. If additional wood biomass is exported and used in place of non-wood building materials in other countries, high emission reductions per unit of biomass could be obtained using a large amount of the forest biomass, thus resulting in a greater overall emission reduction globally.

4.3 Carbon balance This study considers carbon reduction benefits in different stages of forest production and utilisation. The overall carbon balance depends on the intensity of

34 forest management practices, what reference system is selected for biomass utilisation, which reference fuel is considered to be replaced during substitution, and which materials are compared. Sweden has adopted advanced forest management practices for decades. Continuously increasing the carbon stock in living forest biomass is not a viable long-term option for the mitigation of climate change because the carbon stock starts to be released into the atmosphere after its equilibrium level is reached, thereby limiting total carbon sequestration (Schlamadinger and Marland, 1996). Furthermore, the dead biomass left in the forest eventually decomposes and most of its stored carbon is released into the atmosphere. The calculated carbon emission reductions are large enough to contribute significantly to the carbon emission reduction target in Sweden. In the year 2008, the total amount of CO2 emissions in Sweden was 50 million tonnes (Swedish Energy Agency, 2010). Based on the calculations performed in this study to determine the additional potential contribution of forest production and product use to the reduction of carbon emissions, forest management in Jamtland and Vasternorrland counties based on the Maximum scenario would reduce annual average CO2 emissions corresponding to 18% and 15% of the total 2008 carbon emissions of Sweden if the biomass is used to replace coal fuel and fossil gas fuel respectively. The CO2 emissions reduction due to forest management based on the Production scenario corresponds to a 13% and 11% reduction in the total 2008 carbon emissions of Sweden if the biomass is used to replace coal and fossil gas respectively. In this study, it was not possible to calculate the total carbon benefit for all of Sweden due to the differences in forest growth, soil carbon, and standing biomass in the different parts of the country. However, extending this study to all of Sweden may provide a more comprehensive picture of possible carbon emission reductions. In Sweden, forest residue is mainly used within the forest industry for process heat, for district heating, and in the residential sector. District heating plants used forest biofuels for the production of 20 TWh of energy out of the total 36 TWh of energy supplied by bioenergy in 2006 (Swedish Energy Agency, 2009). The share of biofuel used in total energy production in Sweden increased from 10% in 1980 to 22% in 2009, and an increase is expected in the future due to higher demands for biofuels in heating systems (Swedish Energy Agency, 2010). Biofuel not utilised in Sweden can be transported and used in other countries in Europe. The increasing use of wood as a construction material is an important option for reducing net CO2 emissions because of the low energy requirements for 35 processing, the increased wood product stock, and the increased availability of biofuel as by-products (Gustavsson and Sathre, 2006). The soil carbon stock was found to increase, but the increase occurred at a slower rate in the later stages of the study period, assuming whole-tree harvest. The reduction in soil nutrients and forest soil carbon due to residue harvest may be only temporary if the forest is enriched by fertilisation (Paper I and II) (Egnell, 2011).

4.4 Uncertainties and limitations

The integrated analysis method used in this study contains inherent uncertainties because of the large number of factors to be considered and assumptions to be made. In addition, the carbon balance analysis is carried out over long time periods, resulting in a greater number of factors to be considered and inherent uncertainty about future events. One of the basic assumptions of this study was the temperature change due to climate change effect. As discussed by the IPCC, climate change and subsequent temperature changes will depend on changes in global population, energy demand, climate change mitigation activities that are implemented, and the additional need for resources, leading to changes in land use patterns and an increase in the number of methods developed to improve industrial efficiency to reduce CO2 emissions. This thesis uses the SRES B2 scenario of climate change, which assumes an increase of temperature of 4 °C, but the actual future increase is uncertain. A number of factors in plant growth and nutrient cycling could not be taken into account and might result in different NPP results. The elevated temperature may encourage the earlier onset of bud-burst in the spring, which may increase the risk of frost injury during the early growth stage of planted seedlings (Cannell and Smith, 1986; Hänninen, 1991; Kramer, 1994; Krasowski et al., 1995). Potential injuries and shoot mortality due to frost were not included in this study. The increased demand for nutrients and increased biomass production must be met by increased mineralisation, nutrient availability, and uptake by roots; otherwise, the growth response will stagnate at a lower level (Bonan and Cleve, 1991; Houghton et al., 1998; McMurtrie et al., 2001). In the Reference and Environment scenarios, traditional fertilisation was assumed to supply additional nutrients in some part of forest land, whereas in the other two scenarios, balanced fertilisation was provided to assist soil nutrient dynamics in increased forest area. Changes in the amount of fertilisation and the availability of water resources during growth periods might result in slightly different values of NPP. In this

36 landscape study, the site fertility was not known in detail, and the decision to fertilise forest stands may yield different results in a site-specific analysis. Increased mineralisation and nutrient availability as effects of increased soil temperatures have been observed in earlier studies (Jarvis and Linder, 2000). In a soil-warming experiment in northern Sweden, stem-wood production was greater for heated plots compared to unheated plots (Strömgren and Linder, 2002). In this study, a residue harvest of 75% and a 50% stump harvest in the whole-tree harvest scenario did not affect plant growth in the subsequent years. The amount of soil nitrogen and plant growth might be affected to some degree in these scenarios, which could affect biomass production. Any error in BIOMASS would be multiplied in HUGIN and in the other models as well. BIOMASS, however, is a suitable model for predicting NPP as forest productivity in boreal and cold-temperate environments because low- temperature effects have been included, and the response of elevated temperature is predicted in a realistic way. HUGIN uses national forest inventory data for harvest projections. Any error in NFI data would yield different values for the production and harvest of products. The Q model used for soil carbon estimation uses slightly different forest growth functions than HUGIN, so the soil carbon numbers might have been slightly different than they would have been had they been calculated with a function similar to that used by HUGIN. Substitution modelling is based on a case study of a single building, and may give slightly different results if the architectural or engineering designs of the building were different. However, climate benefits of wood substitution in general are quite robust (Sathre and O'Connor, 2010). Climate change may result in wind damages, fire risk, pathogens, and insect outbreaks, which may affect forest growth and mortality (Battles et al., 2008; Kurz et al., 2008; Blennow et al., 2010; Lindner et al., 2010). Such risks have not been considered in this study but could be substantial in Sweden. A Swedish governmental survey (SOU, 2007) noted these uncertainties and risks as possible threats to future forestry in Sweden. The carbon balance analysis is based on both the forest products derived from a sustainably managed forest and the Swedish forest production and utilisation recommendations and guidelines. The risks and advantages of intensive fertilisation in the forest and the use of whole-tree biomass need to be compared.

5 Conclusions

Climate change could significantly increase the forest biomass production in Jämtland and Västernorrland. Forest productivity could further be increased by applying intensive forestry practices. Intensive forestry practices are possible for all of the forest types in Jämtland and Västernorrland and in some cases can be applied on a large scale, such as soil scarification during regeneration, increased pre-commercial thinning, and fertilisation. Increased forest biomass production increases the potential harvest of biomass. The harvest of residues and stumps could increase the carbon benefit through the substitution of carbon-intensive fossil fuels and materials. The harvesting of residues and stumps does not significantly reduce soil carbon, but available biomass could be used to achieve a greater reduction in carbon emissions. Carbon emission reductions as a result of forest biomass use can be achieved during each rotation period, and the reductions would be cumulative in the long run. Carbon emission reduction benefits would be lower if the forest is not managed to increase biomass production and harvest. The carbon emission reduction benefits may be greater if the biomass harvest is larger. In such cases, forests can only sequester carbon until the natural equilibrium of biomass stock is reached and it then either remains near constant or is reduced due to mortality. Leaving residues and stumps in the forest during harvest operations would help increase soil carbon by a small amount compared to the carbon emission reductions achieved by the use of residues and stumps to substitute fossil fuels. Forest management strategies that promote an increase in sustainable biomass production and use would provide a reduced net carbon emission.

6 Future research

Previous studies of the climate implications associated with forest production have generally focused on ecosystem carbon budgets, bioenergy use and material substitution. It would be useful to design complete carbon balance analysis models that include parameters such as forest production, utilisation, carbon emissions, and radiative forcing effects from different types of forest management practices. The concept of forest conversion into continuous cover forestry has emerged, and studies on the standing biomass carbon stock and soil carbon stock in continuous cover forestry would allow for a comparison of the carbon balance benefits between continuous cover forestry and clear-cut forestry. Carbon balance differences due to changes in shorter and longer rotations should be a topic of future research. Generally, carbon balance studies do not include economic valuations of the forest production and carbon benefits. Adding economic analyses to carbon balance studies would provide a clearer picture of the economic costs and benefits. Sweden has made commitments to reduce carbon emissions by 2020. It would be interesting to study carbon storage in forest ecosystems and wood products for this very short time perspective. It is also necessary to examine the current physical quantities of forest products and their use so that future plans can be developed for the energy and construction sectors. Studies on wood consumption in the building sector would provide information about the potential demand for wood in the building sector in the future. Sweden produces large amounts of forest biomass (a recent annual harvest is 90 million m3), but it still imports biofuels from other countries. The tradeoffs between the production, harvest, export and import of biomass could be useful. This study provides a brief assessment of forest production and utilisation at the landscape level in only part of Sweden. Extending this study over all of Sweden may provide a clearer picture of the potential effects of forest production and product use for the entire country. In such studies, the specific quantity of biomass use in the energy sector, the specific quantity of wood use in the building sector, and the amounts of energy recovery can be analysed to provide more precise results for the carbon balance for the entire country. This may be of interest to the Swedish government and to forest owners for planning their forest management, harvest and product business both in the national and international contexts. Future studies should also focus on estimating the value of the substitution effect that is lost if nothing is harvested from the forest. A study on the 39 history of carbon storage and the substitution effect when the forest biomass stock was smaller than what it is today would provide a reference for forest product use in the future.

7 References

Ågren, G., Hyvönen, R., Nilsson, T., 2007. Are Swedish forest soils sinks or sources for CO2—model analyses based on forest inventory data. Biogeochemistry 82, 217-227. Ågren, G.I., 1983. Nitrogen productivity of some conifers. Canadian Journal of Forest Research 13, 494-500. Ågren, G.I., Bosatta, E., 1998. Theoretical ecosystem ecology : understanding element cycles. Cambridge University Press, Cambridge ; New York, NY. Battles, J., Robards, T., Das, A., Waring, K., Gilless, J., Biging, G., et al., 2008. Climate change impacts on forest growth and tree mortality: a data-driven modeling study in the mixed-conifer forest of the Sierra Nevada, California. Climatic Change 87, 193-213. Berg, S., Lindholm, E.-L., 2005. Energy use and environmental impacts of forest operations in Sweden. Journal of Cleaner Production 13, 33-42. Bergh, J., Freeman, M., Sigurdsson, B., Kellomäki, S., Laitinen, K., Niinistö, S., et al., 2003. Modelling the short-term effects of climate change on the productivity of selected tree species in Nordic countries. Forest Ecology and Management 183, 327-340. Bergh, J., Linder, S., 1999. Effects of soil warming during spring on photosynthetic recovery in boreal Norway spruce stands. Global Change Biology 5, 245- 253. Bergh, J., Linder, S., Bergström, J., 2005. Potential production of Norway spruce in Sweden. Forest Ecology and Management 204, 1-10. Bergh, J., Linder, S., Lundmark, T., Elfving, B., 1999. The effect of water and nutrient availability on the productivity of Norway spruce in northern and southern Sweden. Forest Ecology and Management 119, 51-62. Bergh, J., McMurtrie, R.E., Linder, S., 1998. Climatic factors controlling the productivity of Norway spruce: A model-based analysis. Forest Ecology and Management 110, 127-139. Bergman, R.D., Zebre, J., 2004. Primer on wood biomass for energy. In. USDA Forest service, State and Private Forestry Technology Marketing Unit, Madison, WI. Bettinger, P., Boston, K., Siry, J., Grebner, D.L., 2009. Forest management and planning. Academic Press, New York.

Blennow, K., Andersson, M., Bergh, J., Sallnäs, O., Olofsson, E., 2010. Potential climate change impacts on the probability of wind damage in a south Swedish forest. Climatic Change 99, 261-278. Bonan, B.G., Cleve, K.v., 1991. Soil temperature, nitrogen mineralisation, and carbon source-sink relationships in boreal forests. Canadian Journal of Forest Research 22. Börjesson, P., Gustavsson, L., 2000. Greenhouse gas balances in building construction: wood versus concrete from life-cycle and forest land-use perspectives. Energy Policy 28, 575-588. Börjesson, P., Gustavsson, L., Christersson, L., Linder, S., 1997. Future production and utilisation of biomass in Sweden: Potentials and CO2 mitigation. Biomass and Bioenergy 13, 399-412. Boustead, I., Hancock, G.F., 1979. Handbook of Industrial Energy Analysis, Ellis Horwood Publishers, Chichester, UK. Boyd, C., Koch, P., McKean, H., Morschauser, C., Preston, S., Wangaard, F., 1976. Wood for Structural and Architectural Purposes. Wood and Fiber Science 8, 3-72. Briceño-Elizondo, E., Garcia-Gonzalo, J., Peltola, H., Kellomäki, S., 2006. Carbon stocks and timber yield in two boreal forest ecosystems under current and changing climatic conditions subjected to varying management regimes. Environmental Science & Policy 9, 237-252. Buchanan, A.H., Honey, B.G., 1994. Energy and carbon dioxide implications of building construction. Energy and Buildings 20, 205-217. Buongiorno, J., Gilless, J.K., 2003. Decision Methods for Forest Resource Management. Academic Press, New York. Canadell, J.G., Kirschbaum, M.U.F., Kurz, W.A., Sanz, M.-J., Schlamadinger, B., Yamagata, Y., 2007a. Factoring out natural and indirect human effects on terrestrial carbon sources and sinks. Environmental Science & Policy 10, 370-384. Canadell, J.G., Le Quéré, C., Raupach, M.R., Field, C.B., Buitenhuis, E.T., Ciais, P., et al., 2007b. Contributions to accelerating atmospheric CO2 growth from economic activity, carbon intensity, and efficiency of natural sinks. Proceedings of the National Academy of Sciences 104, 18866-18870. Canadell, J.G., Raupach, M.R., 2008. Managing Forests for Climate Change Mitigation. Science 320, 1456-1457. Cannell, M.G.R., Smith, R.I., 1986. Climatic Warming, Spring Budburst and Forest Damage on Trees. Journal of Applied Ecology 23, 177-191. Carter, T.R., Jylhä, K., Perrels, A., Fronzek, S., Kankaanpää, S., 2005. FINADAPT scenarios for the 21st century: alternative futures for considering adaptation to climate change in Finland. In. Finnish Environment Institute, Helsinki, p. 42 Mimeographs 332.

Chapin, F.S.I., Matson, P.A., Mooney, H.A., 2002. Principles of terrestrial ecosystem ecology Springer-Verlag, New York, USA. CPF, 2008. Strategic framework for forests and climate change. A proposal by the Collaborative Partnership on Forests for a coordinated forest-sector response to climate chnage. The Collaborative Partnership on Forests (CPF). Davidson, E.A., Janssens, I.A., 2006. Temperature sensitivity of soil carbon decomposition and feedbacks to climate change. Nature 440, 165-173. Davidson, E.A., Trumbore, S.E., Amundson, R., 2000. Biogeochemistry: Soil warming and organic carbon content. Nature 408, 789-790. Dodoo, A., 2011. Life-cycle pimary energy use and carbon emission of residential buildings. Doctoral Thesis. Department of Engineering and Sustainable Development. Mid Sweden University, Östersund, p. 71. Dupre, C., Stevens, C.J., Ranke, T., Bleeker, A., Peppler-Lisbach, C., Gowing, D.J.G., et al., 2010. Changes in species richness and composition in European acidic grasslands over the past 70 years: the contribution of cumulative atmospheric nitrogen deposition. Global Change Biology 16, 344-357. Easterling, W.E., Aggarwal, P.K., Batima, P., Brander, K., Erda, L., Howden, M., et al., 2007. Chapter 5: Food, Fibre, and Forest Products. In: Parry, M.L., Canziani, O.F., Palutikof, J.P., Linden, P.J.v.d., Hanson, C.E. (Eds.), Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK, 273-313. Eggers, J., Lindner, M., Zudin, S., Zaehle, S., Liski, J., 2008. Impact of changing wood demand, climate and land use on European forest resources and carbon stocks during the 21st century. Global Change Biology 14, 2288- 2303. Egnell, G., 2011. Is the productivity decline in Norway spruce following whole-tree harvesting in the final felling in boreal Sweden permanent or temporary? Forest Ecology and Management 261, 148-153. Egnell, G., Valinger, E., 2003. Survival, growth, and growth allocation of planted Scots pine trees after different levels of biomass removal in clear-felling. Forest Ecology and Management 177, 65-74. Eriksson, E., Gillespie, A.R., Gustavsson, L., Langvall, O., Olsson, M., Sathre, R., et al., 2007. Integrated carbon analysis of forest management practices and wood substitution. Canadian Journal of Forest Research 37, 671-681. EU, 2008. EU Climate Change Expert Group: Background on Impacts, Emission Pathways, Mitigation Options and Costs.Task Group 3: Construction and Demolition Waste. . In, Final report of the construction and emolition waste working group, Brussels. Aavailable at: www.ec.europa.eu [Accessed on May 2011].

European-Commission, 2007. Limiting Global Climate Change to 2 Degrees Celcius: The way for 2020 and beyond. In, Communication from the Commission to the Council, teh European Parliament, the European Economic and Social Committee and the Committee of the Regions. Available at: www.eur-lex.europa.eu [Accessed on: May 2011]. European-Commission, 2009. Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC (Text with EEA relevance). In: European Parliament, C.o.t.M.S. (Ed.). European Union. Fossdal, S., 1995. Energi- og Miljøregnskap for bygg (Energy and environmental accounts of building construction). In. The Norwegian Institute of Building Research, Oslo. (In Norwegian). Freer-Smith, P.H., Broadmeadow, M.S.J., Lynh, J.M., 2009. Forests and Climate Change: the Knowledge-base for Action. In: Freer-Smith, P.H., Broadmeadow, M.S.J., Lynh, J.M. (Eds.), Forestry and Climate Change. OECD, Forest Research the research agency of the forestry commission, and CABI Publishing, Cambridge, p. 253. Friedlingstein, P., Cox, P., Betts, R., Bopp, L., von Bloh, W., Brovkin, V., et al., 2006. Climate–Carbon Cycle Feedback Analysis: Results from the C4MIP Model Intercomparison. Journal of Climate 19, 3337-3353. Galloway, J.N., Townsend, A.R., Erisman, J.W., Bekunda, M., Cai, Z., Freney, J.R., et al., 2008. Transformation of the nitrogen cycle: Recent trends, questions, and potential solutions. Science 320, 889-892. Gower, S.T., Krankina, O., Olson, R.J., Apps, M., Linder, S., Wang, C., 2001. Net primary production and carbon allocation patterns of boreal forest ecosystems. Ecological Applications 11, 1395-1411. Gower, S.T., McMurtrie, R.E., Murty, D., 1996. Aboveground net primary production decline with stand age: potential causes. Trends in Ecology & Evolution 11, 378-382. Gustavsson, L., Börjesson, P., Johansson, B., Svenningsson, P., 1995. Reducing CO2 emissions by substituting biomass for fossil fuels. Energy 20, 1097-1113. Gustavsson, L., Eriksson, L., Sathre, R., 2011. Costs and CO2 benefits of recovering, refining and transporting logging residues for fossil fuel replacement. Applied Energy 88, 192-197. Gustavsson, L., Joelsson, A., Sathre, R., 2010. Life cycle primary energy use and carbon emission of an eight-storey wood-framed apartment building. Energy and Buildings 42, 230-242. Gustavsson, L., Karlsson, Å., 2006. CO2 Mitigation: On Methods and Parameters for Comparison of Fossil-Fuel and Biofuel Systems. Mitigation and Adaptation Strategies for Global Change 11, 935-959.

Gustavsson, L., Madlener, R., Hoen, H.F., Jungmeier, G., Karjalainen, T., Klöhn, S., et al., 2006a. The Role of Wood Material for Greenhouse Gas Mitigation. Mitigation and Adaptation Strategies for Global Change 11, 1097-1127. Gustavsson, L., Pingoud, K., Sathre, R., 2006b. Carbon Dioxide Balance of Wood Substitution: Comparing Concrete-and Wood-Framed Buildings. Mitigation and Adaptation Strategies for Global Change 11, 667-691. Gustavsson, L., Sathre, R., 2006. Variability in energy and carbon dioxide balances of wood and concrete building materials. Building and Environment 41, 940-951. Hänninen, H., 1991. Does climatic warming increase the risk of frost damage in northern trees? Plant, Cell & Environment 14, 449-454. Houghton, R.A., 2005. Aboveground Forest Biomass and the Global Carbon Balance. Global Change Biology 11, 945-958. Houghton, R.A., 2007. Balancing the Global Carbon Budget. Annual Review of Earth and Planetary Sciences 35, 313-347. Houghton, R.A., Davidsson, E.A., Woodwell, G.M., 1998. Missing sinks, feedbacks, and understanding the role of terrestrial ecosystems in the global carbon balance. Global Biogeochemical Cycles 12. Hyvönen, R., Ågren, G.I., 2001. Decomposer invasion rate, decomposer growth rate, and substrate chemical quality: how they influence soil organic matter turnover. Canadian Journal of Forest Research 31, 1594-1601. Hyvönen, R., Ågren, G.I., Linder, S., Persson, T., Cotrufo, M.F., Ekblad, A., et al., 2007. The likely impact of elevated [CO2], nitrogen deposition, increased temperature and management on carbon sequestration in temperate and boreal forest ecosystems: a literature review. New Phytologist 173, 463-480. Hyvönen, R., Persson, T., Andersson, S., Olsson, B., Ågren, G., Linder, S., 2008. Impact of long-term nitrogen addition on carbon stocks in trees and soils in northern Europe. Biogeochemistry 89, 121-137. IFIAS, 1974. Energy Analysis (International Federation of Institutes for Advanced Study). In. Energy Analysis Workshop on Methodology and Conventions, 25-30 August, Guldsmedshyttan, Sweden, . IPCC, 1996. Climate Change 1995 — Impacts, Adaptations and Mitigation of Climate Change: Scientific-Technical Analyses. Cambridge University Press, Cambridge, UK. IPCC, 2000. Land Use, Land Use Change, and Forestry. Cambridge University Press, Cambridge, UK. IPCC, 2001. The Scientific Basis: Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, New York , USA. IPCC, 2007a. Climate Change 2007 : Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the fourth assessment report of the

Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, U.K. ; New York . IPCC, 2007b. Climate Change 2007: Mitigation of Climate Change. Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, United Kingdom and New York, USA. IPCC, 2007c. Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, United Kingdom and New York, USA. Jacobson, S., Kukkola, M., Mälkönen, E., Tveite, B., 2000. Impact of whole-tree harvesting and compensatory fertilization on growth of coniferous thinning stands. Forest Ecology and Management 129, 41-51. Jarvis, P.G., Linder, S., 2000. Constraints to growth of boreal forests. Nature 405, 904-905. Johnson, D.W., 1992. Effects of forest management on soil carbon storage. Water, Air, and Soil Pollution 64, 83-120. Jones, C., McConnell, C., Coleman, K., Cox, P., Falloon, P., Jenkinson, D., et al., 2005. Global climate change and soil carbon stocks; predictions from two contrasting models for the turnover of organic carbon in soil. Global Change Biology 11, 154-166. Jonsson, R., 2009. Prospects for timber frame in multi-storey house building in England, France, Germany, Ireland, the Netherlands and Sweden. Växjö University, School of Technology and Design, Växjö, Sweden, p. 30. Junginger, M., Bolkesjø, T., Bradley, D., Dolzan, P., Faaij, A., Heinimö, J., et al., 2008. Developments in international bioenergy trade. Biomass and Bioenergy 32, 717-729. Kaipainen, T., Liski, J., Pussinen, A., Karjalainen, T., 2004. Managing carbon sinks by changing rotation length in European forests. Environmental Science & Policy 7, 205-219. Karhu, K., Fritze, H., Hämäläinen, K., Vanhala, P., Jungner, H., Oinonen, M., et al., 2010. Temperature sensitivity of soil carbon fractions in boreal forest soil. Ecology 91, 370-376. Kauppi, P.E., Mielikdinen, K., Kuusela, K., 1992. Biomass and Carbon Budget of European Forests,1971 to 1990. Science 256. Kirilenko, A.P., Sedjo, R.A., 2007. Climate change impacts on forestry. In. Proceedings of the National Academy of Sciences of the USA, PNAS, pp. 19697-19702. Kirschbaum, M.U.F., 2000. Will changes in soil organic carbon act as a positive or negative feedback on global warming? Biogeochemistry 48, 21-51.

Kirschbaum, M.U.F., 2004. Soil respiration under prolonged soil warming: are rate reductions caused by acclimation or substrate loss? Global Change Biology 10, 1870-1877. Kjellström, E., Bärring, L., Gollvik, S., Hansson, U., Jones, C., Samuelsson, P., et al., 2006. A 140-year simulation of European climate with the new version of the Rossby Centre regional atmospheric climate model (RCA3). In, SMHI Reports in Meteorology and Climatology, No. 108. SMHI, Norrköping, Sverige., p. 54. Knorr, W., Prentice, I.C., House, J.I., Holland, E.A., 2005. Long-term sensitivity of soil carbon turnover to warming. Nature 433, 298-301. Kramer, K., 1994. A modelling analysis of the effects of climatic warming on the probability of spring frost damage to tree species in The Netherlands and Germany. Plant, Cell & Environment 17, 367-377. Krasowski, M.J., Letchford, T., Caputa, A., Bergerud, W.A., 1995. Desiccation of white spruce seedlings planted in the southern boreal forest of British Columbia. Water, Air, & Soil Pollution 82, 133-146. Kurz, W.A., Dymond, C.C., Stinson, G., Rampley, G.J., Neilson, E.T., Carroll, A.L., et al., 2008. Mountain pine beetle and forest carbon feedback to climate change. Nature 452, 987-990. Lal, R., 2004. Soil Carbon Sequestration Impacts on Global Climate Change and Food Security. Science 304, 1623-1627. Levy, P.E., Cannell, M.G.R., Friend, A.D., 2004. Modelling the impact of future changes in climate, CO2 concentration and land use on natural ecosystems and the terrestrial carbon sink. Global Environmental Change 14, 21-30. Lindner, M., Maroschek, M., Netherer, S., Kremer, A., Barbati, A., Garcia-Gonzalo, J., et al., 2010. Climate change impacts, adaptive capacity, and vulnerability of European forest ecosystems. Forest Ecology and Management 259, 698-709. Lippke, B., Wilson, J., Perez-Garcia, J., Boyer, J., Meil, J., 2004. CORRIM: life-cycle environmental performance of renewable building materials. Forest Products Journal 54, 8–19. Liski, J., Perruchoud, D., Karjalainen, T., 2002. Increasing carbon stocks in the forest soils of western Europe. Forest Ecology and Management 169, 159- 175. Liski, J., Pussinen, A., Pingoud, K., Kip, R.M., Karjalainen, T., 2001. Which rotation length is favourable to carbon sequestration? Canadian Journal of Forest Research 31, 2004. Lloyd, J., Taylor, J.A., 1994. On the temperature dependence of soil respiration. Functional Ecology 8, 315-323. Lundström, A., Söderberg, U., 1996. Outline of the Hugin system for longterm forecasts of timber yields and possible cut. In, Large-Scale Forestry Scenario Models: experiences and requirements. EFI proceeding, pp. 63-77.

Luo, Y.Q., 2007. Terrestrial carbon-cycle feedback to climate warming. Annual Review of Ecology Evolution and Systematics 38, 683-712. Malhi, Y., Meir, P., Brown, S., 2002. Forests, carbon and global climate. Philosophical Transactions of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences 360, 1567-1591. Malmsheimer, R.W., Heffernan, P., Brink, S., Crandall, D., Deneke, F., Galik, C., et al., 2008. Forest Management Solutions for Mitigating Climate Change in the United States. Journal of Forestry 106, 115-118. Marland, G., Schlamadinger, B., 1997. Forests for carbon sequestration or fossil fuel substitution? A sensitivity analysis. Biomass and Bioenergy 13, 389-397. McMurtrie, R.E., Landsberg, J.J., 1992. Using a simulation model to evaluate the effects of water and nutrients onthe growth and carbon partitioning of Pinus radiata. Forest Ecology and Management 52, 243-260. McMurtrie, R.E., Medlyn, B.E., Dewer, R.C., 2001. Increased understanding of nutrient immobilisation in soil organic matter is critical for predicting the carbon sink strength of forest ecosystem over the next 100 years. Tree Physiology 21, 831-839. McMurtrie, R.E., Rook, D.A., Kelliher, F.M., 1990. Modelling the yield of Pinus radiata on a site limited by water and nitrogen. Forest Ecology and Management 30, 381-413. Ministry of Industry, 2004. Mer trä i byggandet: Underlag för en nationell strategi att främja användning av trä i byggandet [More wood in construction: Basis for a national strategy to promote the use of wood in construction]. In. Swedish Government, Stockholm. (In Swedish). Nabuurs, G.-J., Schelhaas, M.-J., Mohren, G.M.J., Field, C.B., 2003. Temporal evolution of the European forest sector carbon sink from 1950 to 1999. Global Change Biology 9, 152-160. Nakicenovic, N., Swart, R., 2000. Emission Scenarios, A special report of Working Group III of the Intergovernmental Panel on Climate Change. In. Cambridge University Press, Cambridge, United Kingdom. NOAA/ESRL, 2012. Mauna Loa Observatory (NOAA/ESRL), monthly and annual mean CO2 concentration (ppm) March 1958 to present. Available at: ftp://ftp.cmdl.noaa.gov/ccg/co2/trends/co2_mm_mlo.txt. [Accessed: August 2012] Norby, R.J., DeLucia, E.H., Gielen, B., Calfapietra, C., Giardina, C.P., King, J.S., et al., 2005. Forest response to elevated CO2 is conserved across a broad range of productivity. Proceedings of the National Academy of Sciences of the United States of America 102, 18052-18056. Norby, R.J., Wullschleger, S.D., Gunderson, C.A., Johnson, D.W., Ceulemans, R., 1999. Tree responses to rising CO2 in field experiments: implications for the future forest. Plant, Cell & Environment 22, 683-714.

O'Donnell, J.A., Harden, J.W., McGuire, A.D., Romanovsky, V.E., 2010. Exploring the sensitivity of soil carbon dynamics to climate change, fire disturbance and permafrost thaw in a black spruce ecosystem. Biogeosciences Discussions 7, 8853-8893. OECD/IEA, 2010. World Energy Outlook 2010. OECD Publishing. Oren, R., Ellsworth, D.S., Johnsen, K.H., Phillips, N., Ewers, B.E., Maier, C., et al., 2001. Soil fertility limits carbon sequestration by forest ecosystems in a CO2-enriched atmosphere. Nature 411, 469-472. Pastor, J., Post, W.M., 1988. Response of northern forests to CO2-induced climate change. Nature 334, 55-58. Piao, S., Ciais, P., Friedlingstein, P., Peylin, P., Reichstein, M., Luyssaert, S., et al., 2008. Net carbon dioxide losses of northern ecosystems in response to autumn warming. Nature 451, 49-52. Pingoud, K., Lehtilä, A., 2002. Fossil carbon emissions associated with carbon flowsof wood products. Mitigation and Adaptation Strategies for Global Change 7, 63-83. Prentice, I.C., Sykes, M.T., Cramer, W., 1993. A simulation model for the transient effects of climate change on forest landscapes. Ecological Modelling 65, 51- 70. Pussinen, A., Karjalainen, T., Mäkipää, R., Valsta, L., Kellomäki, S., 2002. Forest carbon sequestration and harvests in Scots pine stand under different climate and nitrogen deposition scenarios. Forest Ecology and Management 158, 103-115. Raupach, M.R., Canadell, J.G., 2010. Carbon and the Anthropocene. Current Opinion in Environmental Sustainability 2, 210-218. Rosenzweig, C., Karoly, D., Vicarelli, M., Neofotis, P., Wu, Q., Casassa, G., et al., 2008. Attributing physical and biological impacts to anthropogenic climate change. Nature 453, 353-357. Salazar, J., Meil, J., 2009. Prospects for carbon-neutral housing: the influence of greater wood use on the carbon footprint of a single-family residence. Journal of Cleaner Production 17, 1563-1571. Sathre, R., 2007. Life-Cycle Energy and Carbon Implications of Wood-Based Products and Construction. In, Department of Engineering, Physics and Mathematics. Mid Sweden University, Östersund, Sweden, p. 113. Sathre, R., Gustavsson, L., 2007. Effects of energy and carbon taxes on building material competitiveness. Energy and Buildings 39, 488-494. Sathre, R., Gustavsson, L., Bergh, J., 2010. Primary energy and greenhouse gas implications of increasing biomass production through forest fertilization. Biomass and Bioenergy 34, 572-581. Sathre, R., O'Connor, J., 2010. Meta-analysis of greenhouse gas displacement factors of wood product substitution. Environmental Science & Policy 13, 104-114.

Schlamadinger, B., Apps, M., Bohlin, F., Gustavsson, L., Jungmeier, G., Marland, G., et al., 1997. Towards a standard methodology for greenhouse gas balances of bioenergy systems in comparison with fossil energy systems. Biomass and Bioenergy 13, 359-375. Schlamadinger, B., Marland, G., 1996. The role of forest and bioenergy strategies in the global carbon cycle. Biomass and Bioenergy 10, 275-300. Schlamadinger, B., Marland, G., Apps, M.J., Price, D.T., 1996. Carbon implications of forest management strategies. Springer-Verlag, Berlin, Germany. Schulz, H., 1993. Entwicklung der Holzverwendung im 19., 20. und 21. Jahrhundert (The history of wood utilisation in the 19th, 20th and 21st Century). Holz als Roh- und Werkstoff 51, 75-82 (In German). Sirkin, T., Houten, M.t., 1994. The cascade chain: A theory and tool for achieving resource sustainability with applications for product design. Resources, Conservation and Recycling 10, 213-276. Skogsstyrelsen, 2008. Skogliga konsekvensanalyser – SKA-VB 08. Swedish Forest Agency Rapport. Skogsstyrelsen, Sweden. SMHI, 2009. Climatic data, Swedish Meteorological and Hydrological Institute. In. http://www.smhi.se/cmp/jsp/polopoly.jsp?d=8785&l=sv Accessed on 2009/08/02. Smolander, A., Kitunen, V., Tamminen, P., Kukkola, M., 2010. Removal of logging residue in Norway spruce thinning stands: Long-term changes in organic layer properties. Soil Biology and Biochemistry 42, 1222-1228. Solomon, A.M., 1986. Transient response of forests to CO2-induced climate change: simulation modeling experiments in eastern North America. Oecologia 68, 567-579. SOU, 2007. Sweden facing climate change - threats and opportunities. Final Report of the Climate and Vulnerability. In. Statens offentliga utredningar, Stockholm. Strömgren, M., Linder, S., 2002. Effects of nutrition and soil warming on stem wood production in a boreal Norway spruce stand. Global Change Biology 8, 1194-1204. Sveriges-Nationalatlas, 1995. Sveriges Nationalatlas: Klimat, Sjöar och Vattendrag [Swedish National Atlas: Climate, Lakes and Rivers]. Bra Böcker, Höganäs, Sweden. Swedish Energy Agency, 2008. Energy Indicators 2008 Theme: Renewable energy. Swedish Energy Agency, Eskilstuna. Swedish Energy Agency, 2009. Energy Outlook. In, Bioenergy- an important piece of the climate change puzzle. Energimyndigheten, Eskilstuna, p. 60. Swedish Energy Agency, 2010. Energy in Sweden 2010. Swedish Energy Agency, Eskiltuna.

Swedish Forest Agency, 2008. Rekommendationer vid uttag av avverkningsrester och Askåterföring [Recommendations for extraction of logging residue and ash recycling]. In, Meddelande. Skogsstyrelsen, Jönköping, p. 33. Swedish Forest Agency, 2009. Stubbskörd - kunskapssammanställning och Skogsstyrelsens rekommendationer (Stump harvesting - knowledge compilation and Forestry Board's recommendations). In, Meddelande. Skogsstyrelsen, Jönköping, p. 40. Swedish Forest Agency, 2010. Swedish Statistical Yearbook of Forestry. Swedish Forest Agency, Jonköping. Swedish Forest Agency, 2011. Swedish Statistical Yearbook of Forestry. Swedish Forest Agency, Jonköping. Tamm, C.O., 1991. Nitrogen in terrestrial ecosystems. Ecological Studies 81, 1-115. Trenberth, K.E., Fasullo, J.T., Kiehl, J., 2009. Earth's Global Energy Budget. Bulletin of the American Meteorological Society 90, 311-323. UNFCCC, 2011. Fact sheet: Climate change science - the status of climate change science today. Available at http://unfccc.int/press/items/2794txt.php. Werner, F., Taverna, R., Hofer, P., Richter, K., 2005. Carbon pool and substitution effects of an increased use of wood in buildings in Switzerland: first estimates. Ann. For. Sci. 62, 889-902. Zebre, J.I., 1983. Energy properties of wood. In, Fuelwood management and utilization seminar. Michigan State University, Michigan State University.

Paper I

biomass and bioenergy 35 (2011) 4340e4355

Available at www.sciencedirect.com

http://www.elsevier.com/locate/biombioe

Effects of climate change on biomass production and substitution in north-central Sweden

Bishnu Chandra Poudel a, Roger Sathre a,b, Leif Gustavsson b,*, Johan Bergh a,c, Anders Lundstro¨m d, Riitta Hyvo¨nen e a Ecotechnology, Mid Sweden University, 83125 O¨ stersund, Sweden b Linnaeus University, Va¨xjo¨, Sweden c Southern Swedish Forest Research Centre, Swedish University of Agricultural Sciences, Alnarp, Sweden d Department of Forest Resource Management, Swedish University of Agricultural Sciences, Umea˚, Sweden e Department of Ecology, Swedish University of Agricultural Sciences, Uppsala, Sweden article info abstract

Article history: In this study we estimate the effects of climate change on forest production in Received 3 June 2010 north-central Sweden, as well as the potential climate change mitigation feedback effects Received in revised form of the resulting increased carbon stock and forest product use. Our results show that an 1 August 2011 average regional temperature rise of 4 C over the next 100 years may increase annual Accepted 4 August 2011 forest production by 33% and potential annual harvest by 32%, compared to a reference Available online 10 September 2011 case without climate change. This increased biomass production, if used to substitute fossil fuels and energy-intensive materials, can result in a significant net carbon emission Keywords: reduction. We find that carbon stock in forest biomass, forest soils, and wood products also Climate change increase, but this effect is less significant than biomass substitution. A total net reduction Forest production in carbon emissions of up to 104 Tg of carbon can occur over 100 years, depending on Biomass substitution harvest level and reference fossil fuel. Climate feedback ª 2011 Elsevier Ltd. All rights reserved. Sweden

1. Introduction conditions for photosynthesis [5,6]. Increased soil temperature may also lead to increased soil biological activity and The emissions of greenhouse gases, primarily carbon dioxide therefore increased mineralization and nitrogen (N) avail-

(CO2), into the atmosphere increased during the 20th century ability in the soil [7,8]. [1]. This has led to an increasing global surface mean Several researchers have investigated the anticipated temperature, which is projected to continue to rise further [2]. changes in boreal forest production due to climate change. Continuing changes to the global climate patterns are likely Pussinen et al. [9] concluded that boreal forests will likely during the 21st century [2]. The temperature increase is increase their productivity and reduce the length of rotation expected to be greater at higher latitudes [3], where northern periods by 5e10 years, as an effect of anticipated climate Europe may face temperature increases up to 1e2 Cin change and increased N deposition. Briceno-Elizondo et al. summer and 2e3 C in winter during the next 50 years [4]. [10] suggested that northern boreal forests will have higher

Increased temperatures and elevated concentrations of CO2 production potential due to increased temperature. Kirilenko may stimulate forest production in the boreal coniferous and Sedjo [11] reported that climate change is likely to region, due to a longer growing season and more favourable increase the total standing forest biomass by 10e30% in the

* Corresponding author. Present address: Linnaeus University, 351 95 Va¨xjo¨ , Sweden. Tel.: þ46 470 70 8997; fax: þ46 470 76 85 40. E-mail address: [email protected] (L. Gustavsson). 0961-9534/$ e see front matter ª 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.biombioe.2011.08.005 biomass and bioenergy 35 (2011) 4340e4355 4341

next 100 years in northern Europe, if the current harvest level The aim of this study is to estimate the effects of climate is maintained. Eggers et al. [12] claimed that climate change change on future forest production in north-central Sweden, may increase forest productivity by 12e14% and carbon stock and to explore the potential feedback effects of increased by 23e31% in the next 50 years in Europe. Bergh et al. [6,13] forest product use on climate change mitigation. We use the estimated that the temperature increase and elevated process-based model BIOMASS [23] and the empirical forecast concentrations of CO2 could increase net primary production model HUGIN [24] to estimate the effects of climate change on by 20e40% in northern Europe, which may lead to a 24e37% forest production for the next 100 years. Using various harvest increase in forest growth over the next 100 years. Swedish scenarios, we calculate the potential to use the additional Forest Agency [14] projected a 25% increase in annual stem forest production to substitute for non-renewable materials wood production in Swedish forests due to the direct effects of and fuels. We use the Q model [25] to estimate the effects of climate change over the next 100 years. climate change and biomass removals on the forest soil The increased productivity and standing biomass in boreal carbon stock. We then quantify the overall carbon balance forests caused by climate change may in turn be used as based on the avoided fossil emissions due to biomass substi- a negative feedback on climate change due to greater tution and the changes in carbon stock in forest biomass, potential for mitigation activities. The increased biomass forest soils, and wood products. production can be used to substitute fossil fuels and carbon- intensive materials, leading to a net reduction of greenhouse gas emissions [15e18]. Biomass substitution affects energy 2. Methods and assumptions and carbon balances through several mechanisms [19,20], including changes in the carbon stock in living forest biomass 2.1. Study area and soil; the use of biomass as biofuel to replace fossil fuels; the lower fossil energy used to manufacture wood products The studied forest area is in Ja¨mtland and Va¨sternorrland compared with non-wood materials; the avoidance of counties in north-central Sweden (Fig. 1). Ja¨mtland and Va¨st- industrial process carbon emissions from e.g. cement ernorrland are characterised by a boreal climate, and range manufacture; the physical storage of carbon in wood prod- from 61 330 to 65 070 N latitude and from 12 090 to 19 180 E ucts; and the possible carbon sequestration in, and methane longitude. The elevation ranges from sea level to 1500 m alti- emissions from, wood residues deposited in landﬁlls. In tude. Annual precipitation ranges from 600 to 700 mm in the Sweden, the use of forest biomass-related products is eastern part to 1500 mm in the western part [26] and the range increasing. For example, wood fuel sales have increased by of growing season with mean temperature above þ5 Cis 30% over the past 7 years [21]. There is also growing interest approximately 120e160 days [27]. Average monthly tempera- in Sweden for using wood material in place of concrete and ture during the period 1961-1990 varied in winter between other construction materials [22]. þ1 C and 12 C and in summer between 10 C and 14 C.

Fig. 1 e The study area of Ja¨mtland and Va¨sternorrland counties, Sweden, (Data source: Department of Land Survey, Sweden). 4342 biomass and bioenergy 35 (2011) 4340e4355

These models give production forecasts for the next 100 years Climate scenarios for the forest land, quantities of harvested biomass of different types, and input of dead biomass to the forest soil. A biomass substitution model determines the carbon balance BIOMASS effect of using the harvested biomass in place of fossil fuels and non-wood materials. Estimates of soil carbon stock HUGIN changes are made based on litter input, biomass removals and climate scenarios. The resulting overall carbon balance incorporates carbon stock changes in tree biomass, forest soils Annual increment of forest and wood products, and avoided fossil carbon emissions due to biomass substitution. The time period of the study is from Available harvest Tree mortality and 2010 to 2109, broken down into 10-year time steps. retained residues We compare a Reference scenario that assumes an unchanged climate and a Climate change scenario that assumes a warmer climate and thus increased forest production. For Forest C-stock balance Biomass substitution Soil C-stock balance each of these scenarios we consider 2 biomass use options: Stem wood and Whole tree. The stem wood option considers only the use of stem wood including bark, while the whole tree C- Balance option considers the use of stem wood, bark, branches, foliage, Fig. 2 e Schematic structure of the analytical approach, tree tops, and recoverable stumps and coarse roots. showing the relation between scenario inputs, models, and outputs. 2.3. Climate change

The Intergovernmental Panel on Climate Change (IPCC) The forest land area of Ja¨mtland and Va¨sternorrland Special Report on Emission Scenarios (SRES) describes climate counties is about 3.5 and 1.9 million hectare, respectively. Of scenarios of how greenhouse gas emissions, and resulting this area, productive forests, which have productive capacity climate changes, might develop during the 21st century [2]. of at least 1 m3 ha-1y 1 and excludes protected areas, cover These scenarios are based on the main driving forces of about 2.6 and 1.7 million hectare, respectively [21]. The greenhouse gas emissions, e.g. population growth, socio- dominant tree species are Norway spruce (Picea abies (L.) economic development and technological changes, consid- Karst.), Scots pine (Pinus sylvestris L.) and silver birch (Betula ering their underlying uncertainties. The climate change spp.). The mean annual increment (MAI) of forests in the scenario used in this analysis is based on SRES B2, corre- counties has increased from 15.2 to 19.6 hm3 during the past sponding to moderate emissions of greenhouse gases leading

10 years, while the annual harvest of stem wood has main- to an atmospheric CO2 concentration of 572 ppm by the year tained between 11.5 and 13.5 hm3. In comparison, the annual 2085. Other scenarios such as SRES A2 assume higher levels of gross felling in Sweden as a whole has increased by 22% greenhouse gas emissions (Fig. 3a). Raupach et al. [28] showed during the past 10 years [21]. that actual emissions are greater than those assumed by SRES A2, though a concerted global effort to signiﬁcantly reduce 2.2. Modelling structure and scenarios greenhouse gas emission may reduce emissions to the level of SRES B2. Given this uncertainty of future climate conditions, Our analytical approach is shown schematically in Fig. 2. we discuss the implications of different climate scenarios in Climate change scenarios are input into forest growth models. Section 4.

e Fig. 3 Global CO2 concentration history and scenario projections (left) [3,30], and past temperature ﬂuctuation history and future temperature projections for north-central Sweden (right) [26]. biomass and bioenergy 35 (2011) 4340e4355 4343

Regional climate scenarios based on IPCC global scenarios induced production effect. An increase of SI by 4 m is assumed are made with the downscaled climate model RCA3 (Rossby to occur linearly during the 100-year projection. Centre’s regional atmospheric model) that uses a grid of approximately 50 km 50 km [26]. In the calculations, RCA3 2.4.2. HUGIN model uses global driving variables from the general circulation HUGIN is a model system for long-term forecasts of timber yield model ECHAM4/OPYC3 [29]. The model covers the period of and potential harvest level. The system is used for planning on 1961e2100, of which 1961-1990 is taken as reference. Climate a regional level and for strategic planning for large companies. projections for north-central Sweden are for an increase in The model uses sample plots from the Swedish national forest average regional annual temperature of more than 4 Cby inventory (NFI) to define initial conditions for the model, to 2100 for SRES B2, and by over 5 C for SRES A2 (Fig. 3b). The describe the forest conditions. A basic assumption is that the variability and extreme changes in the minimum temperature natural site productivity and the climate conditions will remain are larger than those for the average temperature, and the unchanged during the studied period (normally 100 years), expected temperature increase is higher in the eastern part though we modified the model to incorporate the effects of than in the western part of the region [26]. a changed climate (see Section 2.4.1). The growth simulators For soil carbon modelling, we use a shift in latitude as are constructed to be valid for all forest land in Sweden, for all a proxy for warmer mean temperatures expected in the Climate types of stands, and within a wide range of management change scenario. The increased average temperature of 4 C alternatives. Different methods are used for describing the corresponds to a shift of about 4 degrees of latitude towards the growth in young stands (stand establishment) and for the rest equator. Thus, in our soil carbon model we assume a latitude of the life-time of the stands (established stands). of 58e59N, instead of the actual latitude of 62e63N. 2.4.2.1. Stand establishment. In order to obtain suitable data 2.4. Forest production modelling from young stands with known stand history, about 800 stands were inventoried. The inventory used systematic We use two models, BIOMASS and HUGIN, to estimate the sampling with five circular plots with an area of 100 m2 in each effects of climate change on forest production. The BIOMASS stand. These data were used for the construction of growth model is used to incorporate the effects of climate change into functions for young stands [31]. The survey plots were re- the HUGIN model. measured after five growing seasons. The main objective was to improve the functions used for prediction of stand 2.4.1. BIOMASS model development. Furthermore, this data serves as a database for BIOMASS is a process-based growth model describing the the projections of new stands. The stands in the database processes of radiation absorption, canopy photosynthesis, were then adjusted to correspond to unthinned state at 3 m phenology, allocation of photosynthates among plant organs, height. litter-fall, and stand water balance. BIOMASS consists of a After a final harvest on a NFI plot a stand quality index at 3 m series of equations based on established theories of height is estimated, as a function of site conditions and plant-physiological processes and soil-water dynamics. For regeneration method. Depending on this index, to which a detailed description of the BIOMASS model, see McMurtrie a random component is added, one specific sample plot is et al. [23]. selected from the database. Information about individual trees Climate data from the SRES B2 scenario, with a spatial on the plot is then available for prediction of growth. In the resolution of 40 40 km, was used to run the BIOMASS model. stand establishment, the height is used as dependent variable Output data on net primary production (NPP) from the in the growth functions. To obtain diameter distributions and BIOMASS model was aggregated to mean values for the whole volume growth, special functions are used to transform height counties and used to determine the growth functions on to diameter, and diameter to volume. Fig. 4 shows a flow chart a county level in the HUGIN model [24]. Transient simulations of how simulation in young stands is carried out. for 1961e2100 were performed with BIOMASS for silver birch, Functions are used to estimate damages and mortality in Norway spruce and Scots pine, and output data on NPP were young forests [32]. These functions may be changed depend- summarized in four different periods: 1961-1990 (reference ing on what is known about different predators. When the climate), 2011-2040, 2041-2070 and 2071-2100 [6]. The mean average height of the trees on the plot has reached about 7 m, increase in NPP for Ja¨mtland county, for period 2071-2100 the stand is considered to be established and the method for compared with 1961e1990, was 28.9, 21.4 and 21.6% for growth forecast is changed. Norway spruce, Scots pine and silver birch, respectively [14]. For established seedlings the early development of the The corresponding numbers for Va¨sternorrland county were young stands to the pole stage (i.e. mean height of approxi- 21.6, 11.0 and 13.9%. mately 7e8 m) is projected by height-age curves for the main The effects of climate change on forest production are crop trees. Diameter, age and volume are estimated by static assumed to occur linearly, calculated for each tree species and relationships using height as the dependent variable. Mortality county such that the full increase in NPP for each species and and damage are predicted by a partly stochastic model. Addi- county is achieved after 100 years. The calculated effect is tional sub-models are also included in the growth simulator to used for established forest stands in the HUGIN system, both simulate the effects of pre-commercial thinning, intensive in old and pre-mature forests. For young forests, a calibration fertilization and use of genetically improved material. of site index (SI), a relative measure of site quality based on the Functions for mean height development in young stands height of the dominant tree, is made to find the climate were based on sample-tree data on height and total age from 4344 biomass and bioenergy 35 (2011) 4340e4355

(ii) Existing sapling stands, mean height < 7.0 m stand -level information Individual- tree information

(i) Existing bare land at the Frequency Expected Tree - initial state or Growth Regeneration Height , h list simulated Prediction success future regenerarion cuttings Database with young stands (Nyskog)

Fig. 4 e Flow chart of the simulation of the stand establishment and early growth.

future crop trees (1600 stem ha 1) in the HUGIN young stand gives the distribution for different regeneration methods in survey. In total about 15 000 trees were measured. Height is classes of SI and soil moisture. Whatever the regeneration expressed as a function of SI and total age. For spruce, SI is method, i.e., planting, sowing or natural regeneration, the

H100spruce. For pine, birch and aspen, SI is H100pine. species to be used in planting and number of plants is specified in the same way. The system will then randomly choose 2.4.2.2. Established stands. A review of growth and yield the regeneration method and if needed tree species, by using forecasts in established stands is given by Ha¨gglund [33]. The the distribution provided by the user. simulators for growth in established stands are based on NFI When pre-commercial thinning is performed, the number data, and use the basal area growth for each single tree as of stems to be left depends on SI and species, based on the dependent variable [34]. The growth period for the simulators recommendations in the Swedish Forest Act [36]. Thinning is is five years, and after each period a volume is obtained by possible to perform on plots with a mean height over 11 m and applying static form height functions. To get the net growth with a relative age less than 0.9. To rank thinning plots for each period, functions for mortality and ingrowth (trees a priority function (PT) is used: passing 5 cm in breast height) are used. The effect of thinning PT ¼ RBD þ a H100 þ b H þ c BL (1) is estimated by means of thinning response functions based dom on experimental data [35]. where H100 is SI in metres; Hdom is dominant height in metres; BL is a ratio of the basal area of broadleaf trees to the total 2.4.2.3. Treatments. The calculations of different treatments basal area and a, b and c are coefficients. This function is are based on single trees within sample plots, and the plots dependent on the relative basal area difference (RBD), SI and are selected according to priority rules. The priority rules are dominant height. The actual felling of trees, distributed based upon a standardised silvicultural management. A among species and diameters, can be defined by parameter- varied amount (35e50%, depending on harvest methods) of isation, dependent on SI and soil moisture. The RBD is calcu- harvested volume is randomly selected from the trees of NFI lated as: plots. The selected trees from the sample plots are treated for a specific growth period to simulate mismanagement and the basal area g RBD ¼ r (2) stand-plot problem. These selected trees are considered as gm gr a unit of biomass measurement. Although many trees are measured, they are not considered as a stand since they where gr is recommended basal area after thinning according represent different forest areas. to thinning schedule and gm is maximum basal area before The way to prescribe how treatments should be carried out thinning according to thinning schedule. in the system is very flexible. To make the use of the system easier, standard management programs are included, based 2.4.2.4. Level of fellings. The level of fellings is based on the upon what is considered as good management according to growth level in every period, and then calibrated depending present standards. Division into different treatment classes is on the state of the forest. The amount of the total fellings is made according to relative age, where relative age is defined calculated depending on age distribution and thinning rate, as the ratio of actual stand age and age for final felling. adjusted by the actual growing stock on thinning plots relative Depending on tree species, the age for final felling of each plot to expected growing stock on these plots. is determined from SI classes according to the Swedish Forest a b g d 3 G V A : þ A : A : : Act [36]. Total fellings ¼ G$ 0 5 $ 0 $ 1 0 $ 0 9 10 $ 0 7 0 9 G V R : þ R : R : : After a final felling, a young plot is chosen depending on 6 10 n 1 0 0 9 10 0 7 0 9 (3) site conditions and regeneration method (see Fig. 4). The user biomass and bioenergy 35 (2011) 4340e4355 4345

where G is total growth during latest ten year period, G0e5 is For pines, the height of a tree after t0 is given according to e e growth 0 5 years before actual year, G6e10 is growth 6 10 Elfving and Kiviste [40] and basal area growth for trees older years before actual year, V0 is standing volume, Vn is calcu- than t0 is taken from Elfving and Norgren [41]. For spruce, lated volume with actual age distribution and site conditions, height growth for northern Sweden (regions 1e12) is taken

A1:0þ ; A0:91:0 ; A0:70:9 are the areas over relative age 1.0, from Ha¨gglund [42] and for southern Sweden (regions 13e30)

0.9e1.0 and 0.7e0.9, respectively, R1:0þ ; R0:91:0 ; R0:70:9 are the from Ha¨gglund [43]. The height development is adjusted to desired areas over relative age 1.0, 0.9e1.0 and 0.7e0.9, give the right height at an age of 100 years. Basal area growth respectively, and a, b, g, d, and 3 are coefficients. under bark for spruce has been simplified from Eriksson [44] In the first two 10-year periods, the criteria for choosing and from which basal area over bark is calculated. Number plots for thinning and final felling is done by probability of trees per hectare is empirically determined for SI ranging functions, based on how forest owners have made their from 11 to 36 [44]. priority for harvest according to permanent plots of the NFI in the beginning of 2000. After the first two periods, the thinning 2.5.2. Biomass allocation decision is made as described above, and decision for final We assume that all trees are of equal size, and from their felling is made by volume increment percentage. The felling average diameter over bark we calculate the biomass allo- volume for each decade is presented in cubic meter over bark, cation to different tree components [38]. The tree compo- cubic meter under bark, and kilogram dry matter. Functions nents considered here are: needles, branches, tops, stems, used to determine tree biomass were developed by Marklund stumps, coarse roots >5 cm, coarse roots <5cm,andfine [37,38] and Petersson and Sta˚hl [39]. The total carbon stock in roots <2 mm. The ratio of mass of fine roots to mass of standing forest biomass is also calculated by HUGIN in both needles is set to 1.5. standing stem wood and in whole tree biomass. Pines are assumed to lose 25% of their needles and 5% of their branches per year [45]. The corresponding parameters 2.5. Soil carbon modelling for spruce are 10% and 2%, respectively [45]. These parameter values should vary with latitude but we do not have enough Soil carbon stock changes may be expected due to a warmer information to include this aspect. Fine root litter production climate causing increased soil respiration leading to is set to 1.5 times the needle litter fall [46]. Other types of litter decreased soil C stock, increased tree growth and litter fall are only formed when trees are felled (i.e., we neglect leading to increased soil C stock, and increased biomass mortality of single trees between harvest events). At harvest, removal leading to decreased soil C stock. 25% of the needles, branches, and tops of the felled trees, 50% We estimate soil carbon effects during the 100-year study of the stumps and coarse roots, and 100% of fine roots are left period using a linear extrapolation between the reference on the site to decompose. climate in 2010 and the 4 C warmer climate in 2110. The soil carbon analysis uses forest production data calculated inde- 2.5.3. Decomposition of tree litter and harvesting residues pendently from that described in Section 2.4. The manage- The decomposition of the different litter fractions is calcu- ment plan for pine stands ranges from two thinnings during lated using the Q model [25]. The equation g(Lat,t) is the a rotation period of 145 years for the least productive stands to fraction of initial carbon remaining in a needle or fine root three thinnings during a rotation period of 90 years in the cohort after time t: highest productive stands. Corresponding values for least productive spruce stands are two thinnings during a rotation gðLat; tÞ:¼ð1 þ aðP; LatÞ$tÞ z (6) period of 140 years and four thinnings during a rotation period of 70 years for the highest productive spruce stands. At thin- where a is a function of latitude (Lat) such that: nings, 14e35% of the basal area is removed depending on SI and stand age. At each thinning, the number of trees is b a Lat :¼ fc$b$h $u0ðLatÞ$q (7) decreased by the same fraction as the basal area. The thin- 11 0 nings are stem-only thinning. The parameter fC stands for carbon concentration and it is assumed to be 50% of organic matter. The parameter b refers h 2.5.1. Growth functions to abiotic factors and it is set to 7. The parameter 11 is a scale 2 -1 Development of basal area, BA (m ha ) from age 0 up to an parameter and it is set to 0.36. The parameter q0 describes age t0 (year) of the stand is assumed to follow an exponential initial quality and is set to 1. The parameter u0 is coupled to function: decomposer growth rate and depends on climate, which is correlated with latitude such that: rt BAðtÞ¼N0A0e ð Þ:¼½ : þ : $ð : : $ Þ 1 u0 Lat 0 0855 0 0157 50 6 0 768 Lat (8) where N0 is the initial number of trees ha , which varies depending on SI, A0 is the basal area of a tree at age 0 and is The remaining mass of woody components is expressed by 4 assumed to be 10 square meter, and t0 is the age when the the functions G1(t,tmax) and G2(t,tmax) where invasion rates of trees reach 10 m of height. The relative growth rate, r, is litter types are taken into account. G1 describes decomposition estimated from until all wood is infected by decomposers and G2 describes

decomposition after everything is infected, tmax is the time 1 BAðt Þ r ¼ ln 0 (5) taken for decomposers to completely invade these litter t0 N0A0 components. 4346 biomass and bioenergy 35 (2011) 4340e4355

Gðt; t Þ: restrictions to harvest both above-ground biomass and max stumps from the same harvest site. ¼ 2 $ 1 $ ð þ $ Þ1z t ð Þ 1 a t 1 tmax a$ 1 z tmax We assume that all recovered biomass for energy purposes h i 2 1 (biofuels) are used to replace either coal or fossil gas in þ $ $ 1 ð1 þ a$tÞ2 z t2 a2$ð1 zÞ$ð2 zÞ stationary plants with conversion efficiencies of 100% and 96% max 2 relative to the conversion efficiency of the respective fossil t þ 1 (9) fuel-fired plants [20]. Values of specific carbon emission from tmax fossil fuels as atomic carbon are assumed to be 30 kg GJ-1 for coal, 22 kg GJ-1 for oil, and 18 kg GJ-1 for natural gas, and G2ðt; tmaxÞ : include emissions during the entire fuel-cycle from the 2 1 2 1 ¼ $ ð1 þ a$tÞ1 zþ $ natural resource to the delivered energy service [49]. Energy $ð Þ 2 2$ð Þ$ð Þ tmax a 1 z tmax a 1 z 2 z used for recovery and transport of biofuels is assumed to be h i diesel fuel, calculated as a percentage of the heat energy $ ½ þ a$ð Þ2 zð a$ Þ2 z ð Þ 1 t tmax 1 t 10 content of the biofuel. We assume this percentage is 10% for stumps, 5% for slash and small diameter stem wood, and 1% The parameter combination z is defined as follows: for processing residues [50]. Emissions from forest management activities are based on Berg and Lindholm [51]. 1 e z ¼ 0 (11) Large diameter stem wood is assumed to be used for b$h $ 11 e0 producing wood construction material that substitutes in place of conventional reinforced concrete. The material The parameter e is decomposer growth efficiency or 0 substitution effects are based on a case study of a multi-story production-to-assimilation ratio and it is set to 0.25 [47]. apartment building constructed in Sweden using wood The calculations of remaining carbon are made by year for structural framing, compared to a functionally equivalent each litter component and each litter cohort separately such building constructed with a reinforced concrete frame [20]. that the total soil carbon consists of the sum of remaining Calculations take into account the differences between the litter cohorts of different ages and from different tree buildings due to the emission from fossil fuels used to components. There is no vertical partitioning of soil carbon in manufacture and transport the materials, the cement calci- our model and all soil carbon should therefore be included. nation and carbonation process emissions, and substitution of This assumes that we can neglect climatic variability with fossil fuel by biomass residues from wood processing and depth, as well as differences in the interaction between soil construction. The reference concrete building uses some carbon and the mineral soil matrix. wood materials, and the substitution benefit is calculated based on the reduction in net carbon emission per unit of 2.6. Biomass use modelling additional wood needed to make the wood-frame building. Large roundwood is converted into a mix of sawn lumber, In the stem wood biomass use option, we assume that we plywood, and particleboard, with a product/roundwood dry harvest and remove 95% of the felled stem wood from the weight ratio of 0.53. Some wood processing residue is used for forest. Large diameter pine and spruce stem wood is used for particleboard manufacture, some is used internally as bio- production of wood construction material, with the process- energy for e.g. kiln-drying lumber, while the remainder is ing residues used for bioenergy. Small diameter pine and available externally for use as biofuel. We assume a building spruce stem wood, and all deciduous stem wood, is used for life span of 100 years, and calculate the carbon stock stored in bioenergy. We assume a minimum diameter of 20 cm for large the wood building materials during that time. diameter stem wood. Stem wood is sorted entirely by diameter without consideration of quality aspects, thus we are likely to overestimate the quantity of stem wood used for 3. Results construction material. Small-diameter stem wood is generally used in Sweden for pulp and paper production. In this study 3.1. Forest biomass production and harvest we analyse the use of additional quantities of biomass produced due to climate change, but we have not analysed The annual forest biomass production and harvest under the whether there is demand for additional pulpwood in the pulp scenarios are presented in Table 1. The whole tree biomass industry. We therefore show the climate effects of using this production was significantly larger under the Climate change small diameter stem wood for bioenergy, acknowledging that scenario compared to the Reference scenario. The annual this may not be the most economically beneficial use of this biomass production increased by 1.84 Tg y-1 dry mass in the wood [48]. Reference scenario, and by 5.60 Tg y-1 dry mass in the Climate In the whole tree biomass use option, in addition to har- change scenario. The annual production increased during the vesting 95% of the felled stem wood, we also assume that 75% 100-year study period by 33% more in the Climate change of needles, branches and tops, and 50% of stumps and coarse scenario than in the Reference scenario. roots are harvested and used for bioenergy. According to the In the Reference scenario, the calculated average harvest of Swedish forest agency [14], current Swedish harvest guideline whole tree biomass over 100 years was 8.95 Tg y-1, while in the recommends 20% of logging residues should be left at the site Climate change scenario it was 10.18 Tg y-1. The harvest of stem and 15e25% of the stumps at clearfelling, and there is no wood biomass followed a similar trend. The annual harvest of L Table 1 e Average annual forest biomass production and harvest during each 10-year period for each scenario (Tg y 1 dry mass), cumulative totals for 100 years, and differences between the scenarios. 2010-2019 2020-2029 2030-2039 2040-2049 2050-2059 2060-2069 2070-2079 2080-2089 2090-2099 2100-2109 Cumulative total (Tg) ims n ieeg 5(01 4340 (2011) 35 bioenergy and biomass Reference scenario Biomass production Whole tree biomass 10.99 11.64 11.46 12.18 12.41 12.16 11.97 12.05 12.67 12.83 1204 Stem wood biomass 6.22 6.15 6.17 6.48 6.60 6.48 6.38 6.40 6.66 6.74 643 Biomass harvest Whole tree biomass 8.17 8.07 8.80 8.85 9.40 9.45 9.06 9.04 9.09 9.60 895 Stem wood biomass 5.00 4.91 5.36 5.40 5.75 5.85 5.63 5.63 5.63 5.97 551

Climate change scenario Biomass production Whole tree biomass 11.34 12.15 12.40 13.53 13.98 14.17 14.37 14.77 16.10 16.94 1397 Stem wood biomass 6.40 6.42 6.67 7.18 7.44 7.58 7.68 7.86 8.51 8.97 747 Biomass harvest Whole tree biomass 8.27 8.33 9.35 9.53 10.19 10.52 10.83 10.91 11.42 12.41 1018 Stem wood biomass 5.06 5.09 5.70 5.82 6.27 6.51 6.77 6.84 7.19 7.77 630

Difference between Climate change and Reference scenarios e

Biomass production 4355 Whole tree biomass 0.35 0.51 0.94 1.35 1.57 2.01 2.40 2.72 3.43 4.11 194 Stem wood biomass 0.18 0.27 0.50 0.70 0.84 1.10 1.30 1.46 1.85 2.23 104 Biomass harvest Whole tree biomass 0.10 0.26 0.55 0.68 0.79 1.07 1.77 1.87 2.33 2.81 122 Stem wood biomass 0.06 0.18 0.34 0.42 0.52 0.66 1.14 1.21 1.56 1.80 79 4347 4348 biomass and bioenergy 35 (2011) 4340e4355

14 14 Reference Climate 12 12 Stemwood Stumps Slash Bark Stemwood Stumps Slash Bark -1 10 10 8 8 6 6

Dry mass Tg4 y 4 2 2 0 0 2010-2019 2020-2029 2030-2039 2040-2049 2050-2059 2060-2069 2070-2079 2080-2089 2090-2100 2100-2109 2010-2019 2020-2029 2030-2039 2040-2049 2050-2059 2060-2069 2070-2079 2080-2089 2090-2100 2100-2109 Year

Fig. 5 e Average annual biomass harvest by tree component over time for Reference (left) and Climate change (right) scenarios L (Tg y 1 dry mass).

whole tree biomass in the Reference scenario increased about because of the low presence of birch in older forests today due 18% over the 100-year period, while in the Climate change to earlier silviculture practices when birch was unwanted. scenario it increased by 50% during the period. Moreover, HUGIN considers the past data and removes less Average annual harvest of forest biomass broken down by birch in thinning considering few trees in an area, thus birches tree component is shown in Fig. 5. Stem wood biomass remains uncut in each thinning which might increase the production averaged 6.30 Tg y-1 in the Climate change scenario, volume in the long run. Another explanation might be that the compared to an average of 5.51 Tg y-1 for Reference scenario. HUGIN model assumes deciduous trees live longer than they Stump biomass production increased by 47% for Climate actually do, which will overestimate the volumes in a long- change scenario, but only increased 14% for Reference scenario term perspective, especially in reserves and environmental over the 100-year period. The quantities of stump biomass care areas where the trees are not felled. may be overestimated, because Marklund’s revised function [40] used by the HUGIN model estimates deciduous tree stump 3.2. Carbon balance values based on spruce stumps. Pine harvest increases significantly particularly in Climate The average annual net carbon balance for each 10-year period, change scenario, while spruce harvest decreases in early years composed of carbon-stock increases in trees, soil and wood and then increases after 30 years of management (Fig. 6). For products, and avoided fossil and cement process emissions due deciduous species, the harvest is higher during the initial 50 to material and fuel substitution, is shown in Table 2. The net years but thereafter it decreases in Reference scenario. However, substitution values in the table are net values after deducting in Climate change scenario, deciduous species increase signifi- forest operations emissions from the substitution benefits. cantly. The higher increase for deciduous species might be Biomass substitution is the largest single contributor to the

14 Reference 14 Climate 12 12 Pine Spruce Deciduous Pine Spruce Deciduous 10 10 -1 8 8 6 6 4 4

Dry mass Tg2 y 2 0 0 2010-2019 2020-2029 2030-2039 2040-2049 2050-2059 2060-2069 2070-2079 2080-2089 2090-2100 2100-2109 2010-2019 2020-2029 2030-2039 2040-2049 2050-2059 2060-2069 2070-2079 2080-2089 2090-2100 2100-2109 Year

L Fig. 6 e Average annual biomass harvest by species over time for Reference (left) and Climate change (right) scenarios (Tg y 1 dry mass). biomass and bioenergy 35 (2011) 4340e4355 4349

carbon balance. The increased biomass production in the biomass substitution effects under Reference and Climate change Climate change scenario can be used to replace more fossil fuels scenarios. The results suggest that there will be a significant and non-wood products, resulting in substantially more carbon increase in forest production in the next 100 years in north- emission reduction than in the Reference scenario. Whole tree central Sweden due to climate change. Our forest modelling biomass usage results in greater substitution benefits because shows a 49% increase in total annual forest biomass produc- the slash and stump biomass is used to replace fossil fuels. tion for the Climate change scenario during the next 100 years, When coal instead of fossil gas is replaced, the substitution slightly higher than the estimate by Pussinen et al. [52] for the benefits increase by about 38% for stem wood usage and about whole of Europe. Compared to the increased biomass produc- 48% for whole tree usage. The carbon stock in forest biomass tion in our Reference scenario, the annual forest biomass continues to increase in both Reference and Climate change production for the Climate change scenario is 33% greater at the scenarios, but increases at a faster rate in the Climate change end of the 100-year period. scenario. Soil carbon stock increases in both scenarios, but The reasons why the production increases in the Reference increases more in the Climate change scenario, presumably due scenario are because large areas with low productive forests to the greater amounts of litter fall produced by the faster are replaced with new planted forests with improved genetic growing trees. Removing slash and stumps causes the soil material. This has been going on for several decades and will carbon to increase at a slower rate than when only stem wood is continue into the future. Some of these planted forests have removed. However, the carbon emission benefits of using the now reached their most productive phase and the amount will slash and stumps to replace fossil fuel are significantly greater increase further in the future. The total effect caused by than the net decrease in soil carbon accumulation. When coal is genetic improvement in the Reference scenario is about 7% at the reference fossil fuel, the increased substitution benefits are the end of the scenario. Other explanations for growth about 5.2 times greater than the soil carbon net decrease, and for increase are soil scarification during plantation, increased fossil gas are about 2.6 times greater. Carbon stock in wood pre-commercial thinning, and fertilization. Soil scarification is products is greater in the Climate change scenario than the done in 36 000 ha y-1 and pre-commercial thinning is carried Reference scenario because part of the additional biomass out in 31 000 ha y-1. Nitrogen fertilizer is applied in the young production is stored in additional wood products. stands of Norway spruce at an average rate of 20.9 kg ha-1 y-1 Fig. 7 shows cumulative avoided carbon emission over 100 throughout the study period. The area of fertilization at the years for the Reference and Climate change scenarios, assuming beginning is 9200 ha y-1 and reaches 10 000 ha y-1 at the end of whole tree biomass use and coal reference fuel. Biomass the study period. The total fertilized area at the end of study substitution is the largest single factor contributing to avoided period is 25% of the total production forest. C emission. Carbon stock increases in soil, tree biomass and Per hectare of productive forest, our results suggest that wood products each have roughly the same magnitude. Total annual whole tree biomass production should be 3.9 Mg ha-1y 1 avoided emissions during the 100-year period are 104 Tg C dry mass for Climate change scenario in 2110. This is smaller greater in the Climate change scenario than in the Reference than the forecast of 5e6Mgha-1y 1 by Bergh et al. [13] for scenario for whole tree biomass use with coal reference fuel, northern Swedish forests, but is higher than the projection of and 91 Tg C greater for stem wood biomass use with coal 2.4 Mg ha-1y 1 by Kellomaki et al. [53] for Finnish forests. The reference fuel. increased harvest level in Climate change scenario gives Table 3 shows the six factors contributing to the avoided increased possibility of forest product use, resulting in signifi- carbon emission due to biomass substitution. Stem wood use cant additional substitution benefits during the 100-year includes the reduced fossil energy and cement process emis- period. The modelled climate change also results in increased sions when wood construction materials are produced instead carbon stock in standing tree biomass, forest soil, and wood of reinforced concrete materials, and the biomass from wood products. Whole tree harvesting can increase usable biomass processing residues and small-diameter stem wood used to by about 58% compared to stem wood-only harvesting. replace fossil fuel. Whole tree use includes these factors plus However, stump biomass calculation for broadleaf trees in the slash and stumps used as biofuel to replace fossil fuel. Each of HUGIN model is based on spruce stumps, which overestimates these substitution factors is greater in the Climate change stump biomass for broadleaf trees. Nevertheless, the recovery scenario than the Reference scenario, because more biomass is of slash and stumps can significantly increase the amount of produced and used for substitution. biofuel per hectare and rotation period. Fig. 8 shows the differences in cumulative avoided carbon Carbon stock in living tree biomass continuously increases emissions between the Climate change scenario and the Refer- in both Reference and Climate change scenarios, though it ence scenario. The difference is greatest with whole tree har- increases more in the Climate change scenario. Our average vesting and when coal is the reference fossil fuel. The forest carbon stock increase is smaller than the estimate by difference in avoided carbon emission between whole tree use Pussinen et al. [52], who assumed the SRES A2 climate and stem wood use is roughly the same as the difference scenario and intensive forest harvesting in the whole of between coal and fossil gas reference fuel. Europe. In our analysis, increased harvest levels does not result in reduced carbon stock, as the forest production increase due to climate change is greater than the harvest 4. Discussion and conclusions increase. Although early studies reported that increasing tempera- We developed an integrated analytical approach to calculate ture does not necessarily increase the soil respiration and forest biomass production, harvest, carbon stock changes, and thus soil carbon decomposition [54], later studies have 4350

L Table 2 e Average annual carbon stock increases and avoided emissions due to biomass substitution during each 10-year period (Tg y 1 carbon). 2010-2019 2020-2029 2030-2039 2040-2049 2050-2059 2060-2069 2070-2079 2080-2089 2090-2099 2100-2109

Reference scenario Forest C stock change 0.38 0.76 0.22 0.55 0.34 0.20 0.35 0.40 0.67 0.45 Soil C stock change Stem wood biomass use 0.78 0.78 0.78 0.78 0.78 0.78 0.78 0.78 0.78 0.78 Whole tree biomass use 0.58 0.58 0.58 0.58 0.58 0.58 0.58 0.58 0.58 0.58 Wood product C stock change 0.77 0.72 0.62 0.63 0.65 0.67 0.72 0.74 0.74 0.76 Net biomass substitution: Coal

Stem wood biomass use 3.78 3.63 3.59 3.63 3.82 3.89 3.88 3.93 3.93 4.12 4340 (2011) 35 bioenergy and biomass Whole tree biomass use 4.75 4.61 4.63 4.67 4.91 4.96 4.90 4.94 4.96 5.20 Net biomass substitution: Fossil gas Stem wood biomass use 2.81 2.68 2.56 2.59 2.72 2.77 2.81 2.85 2.86 2.98 Whole tree biomass use 3.30 3.17 3.08 3.11 3.27 3.32 3.33 3.37 3.38 3.53 Total (Coal ref) Stem wood biomass use 5.70 5.90 5.22 5.59 5.59 5.54 5.73 5.85 6.12 6.11 Whole tree biomass use 6.48 6.67 6.05 6.42 6.48 6.42 6.55 6.67 6.95 6.99 Total (Fossil gas ref) Stem wood biomass use 4.73 4.94 4.18 4.55 4.49 4.43 4.66 4.78 5.05 4.97 Whole tree biomass use 5.03 5.24 4.51 4.87 4.84 4.77 4.97 5.09 5.37 5.32

Climate Scenario Forest C stock change 0.49 0.85 0.35 0.80 0.63 0.53 0.45 0.61 0.96 0.75 Soil C stock change Stem wood biomass use 0.78 0.79 0.79 0.80 0.81 0.82 0.82 0.83 0.84 0.84 Whole tree biomass use 0.58 0.58 0.59 0.60 0.61 0.62 0.63 0.64 0.64 0.65 Wood product C stock change 0.79 0.75 0.64 0.66 0.74 0.77 0.86 0.86 0.96 1.06 e

Net biomass substitution: Coal 4355 Stem wood biomass use 3.83 3.78 3.79 3.88 4.21 4.37 4.67 4.69 5.05 5.51 Whole tree biomass use 4.82 4.77 4.89 4.99 5.39 5.57 5.88 5.89 6.30 6.89 Net biomass substitution: Fossil gas Stem wood biomass use 2.85 2.79 2.69 2.76 3.01 3.13 3.38 3.38 3.68 4.03 Whole tree biomass use 3.35 3.29 3.25 3.32 3.61 3.73 4.00 4.00 4.32 4.73 Total (Coal ref) Stem wood biomass use 5.89 6.16 5.58 6.14 6.39 6.49 6.80 6.99 7.80 8.16 Whole tree biomass use 6.68 6.96 6.47 7.05 7.37 7.49 7.82 8.00 8.87 9.36 Total (Fossil gas ref) Stem wood biomass use 4.90 5.17 4.48 5.02 5.19 5.24 5.52 5.68 6.44 6.68 Whole tree biomass use 5.21 5.48 4.83 5.38 5.59 5.65 5.93 6.10 6.88 7.19 biomass and bioenergy 35 (2011) 4340e4355 4351

Fig. 7 e Cumulative avoided carbon emission (Tg carbon) for Reference (left) and Climate change (right) scenarios, assuming whole tree biomass use and coal reference fuel.

suggested that a warmer climate may decrease soil carbon as nutrients, and continued removal of needles may lead to long-

CO2 is released to the atmosphere [55]. Our results show soil term nutrient deﬁciency [63]. In practice, however, needles carbon stock increasing during the study period in both the often do not detach from the tree branches, even when felling Reference and Climate change scenarios. The carbon increase is residue is left in the forest to dry before harvesting. Thus, the slightly greater (60 kg ha-1 y-1 by 2010) in the Climate change quantity of needles that is harvested is uncertain and difﬁcult scenario than in the Reference scenario, suggesting that to control. Nevertheless, this factor will have minimal impact increasing temperature has a net positive effect on soil carbon on our conclusions. If 25% of needles are harvested instead of stock in boreal forests. Stro¨ mgren and Linder [7] also found 75%, the reduced amount of bioenergy used to substitute fossil that soil respiration was not increased by temperature as an fuels will change the total carbon balance by about 1%. At the effect of a shift in the respiration response to temperature. same time, however, soil carbon stock will increase faster if

The role of warmer climate and elevated CO2 concentration on fewer needles are harvested, making the net effect on total soil C stock is still under debate [56e62]. Nevertheless, we carbon balance even less signiﬁcant. The residual removal expect that the uncertainty in soil carbon stock change will after thinning and clear felling may reduce forest growth by have very little effect on our conclusions. In the Climate change 10e15% and the reduction tends to be higher for Norway spruce scenario, our estimate of soil carbon stock change is about compared with Scots pine [64e66]. The growth reduction is due 12% and 8% of the total carbon balance with coal reference to temporary reduction because of loss in nitrogen primarily in fuel, with stem wood and whole tree biomass use, respectively needles [65]. This reduction is likely to occur after extended (Table 2). Thus, if our estimate of soil carbon stock change is periods of residue removal when there is no nutrient over- or underestimated by 20%, the total carbon balance will compensation provided. The nutrient can be compensated by vary by only 2e3%. Hence, we expect that the uncertainty in N-fertilization [66]. There appears to be no reduced growth of soil carbon stock change will have a minor impact on our trees due to stump harvest [66]. results and will not change our conclusions. Carbon stock in wood products increases more in the We have assumed that 75% of needles from felled trees are Climate change scenario compared to Reference scenario, harvested and used as bioenergy. Tree needles are rich in because of the greater quantities of stem wood converted into

L Table 3 e Average carbon emission avoided (Tg y 1 carbon) due to the six factors comprising biomass substitution, with reference fossil fuel of coal or fossil gas. Coal fuel Fossil gas fuel

Reference Climate change Reference Climate change

Material substitution: Production energy 0.96 1.10 0.73 0.84 Material substitution: Cement emissions 0.99 1.14 0.99 1.14 Biofuel: Wood processing residues 0.92 1.06 0.53 0.61 Biofuel: Small stem wood 1.00 1.13 0.56 0.63 Biofuel: Slash 0.59 0.66 0.33 0.37 Biofuel: Stumps 0.54 0.61 0.29 0.33 4352 biomass and bioenergy 35 (2011) 4340e4355

120 Cumulative differences: Climate and Reference scenario fossil fuel in another region [68]. Gustavsson et al. [69] showed that woody biofuels can be economically transported 100 Whole tree, coal fuel internationally, and the GHG reduction per unit of biofuel 80 Whole tree, fossil gas depends more on the fossil fuel replaced than on the transport Stemwood, coal fuel distance. By exporting biomaterials and biofuels to be used in 60 Stemwood, fossil gas applications that result in high greenhouse gas emission reductions per unit of biomass, the total emission reduction 40 from the available supply of biomass could by increased. 20 There are significant uncertainties in our analysis. The Avoided carbon emissions Tg emissions carbon Avoided degree of temperature change in the future is not known with 0 certainty, and will depend on several factors including economic growth, population, and implemented climate mitigation activities. We have used the SRES B2 scenario of 2100-2109 2000-2009 2010-2019 2020-2029 2030-2039 2040-2049 2050-2059 2060-2069 2070-2079 2080-2089 2090-2099 Ye a r temperature increase requiring significant mitigation efforts. Regional temperatures may rise by more than we modelled, e Fig. 8 Difference in cumulative avoided carbon emission and the forest biomass production would further increase. (Tg carbon) between the Climate change scenario and the Furthermore, the HUGIN model overestimates biomass in Reference scenario. deciduous stumps and in mature deciduous trees that are not felled. Climate change may result in natural disturbances, for example, wind damages, fire risks, pathogens, and insect outbreaks, which may affect forest growth and mortality long-lived wood building materials. We assume a 100-year life [70e75]. Some of these potential disturbances do not have span of the buildings produced with the forest biomass, thus direct effect on growth, such as forest fires and storm-felling, the carbon stock in building materials increases continuously however it can damage timber or make more difficult to during the 100-year study period because no wood products harvest. Other disturbances like pathogens and insects can are yet removed from service. At the end of this period, have major effect on growth following the disturbances in tree however, buildings will begin to be demolished and the stems and in the leaf area. A Swedish governmental survey annual carbon stock change in wood products will be the [76] pointed out these uncertainties and risks as possible difference between new wood products entering service and threats to future forestry in Sweden. In Swedish forestry, old wood products leaving service. Post-use wood products however, because of intensive forest management practices can be a significant source of bioenergy [17], which would and easy access to the forest, it is possible to take precautions appear in our carbon balance calculations if the time period to reduce potential disturbances. Although these uncer- were longer. tainties may alter the exact values that we have calculated, Using forest biomass as raw material for construction our general conclusions appear to be robust. We expect future materials and bioenergy provides permanent and cumulative climate change to increase both biomass production and emissions reductions by avoiding fossil emissions. Storing substitution potential in forests in north-central Sweden. increasing amounts of carbon in living biomass is not a viable climate change mitigation strategy in the long-term, as forests will eventually reach a dynamic equilibrium where carbon Acknowledgement uptake is balanced by carbon release [50]. Moreover, biomass fractions such as slash and stumps that are left in the forest We gratefully acknowledge the support of European after fellings will also decompose over time and release most of Commission: Objective II, European Structural Funds, County their stored carbon into the atmosphere. Instead of forest Administrative Board of Ja¨mtland, Sveaskog AB, SCA Forest biomass decaying naturally in the forest, while simultaneously Products, Norrskog/SA˚ TAB, and Ja¨mtkraft AB. This analysis using fossil fuels and materials to provide services to society, was done partly in collaboration with the project Future the forest biomass could provide those services resulting in Forests at the Faculty of Forestry at SLU, Umea˚. We also thank less net CO emission to the atmosphere [20]. The time 2 the anonymous reviewers of this paper. dynamics of forest residue oxidation vary. Forest residues left to decompose naturally in the forest slowly release CO2 into the atmosphere over a time scale of decades, while residues references removed from the forest and used as biofuel release CO2 when burned. This can result in varied radiative forcing, the signifi- cance of which depends on the time horizon under consider- [1] Keeling RF, Piper SC, Bollenbacher AF, Walker JS. ation [67]. This effect is more pronounced for slower-decaying Atmospheric CO2 records from sites in the SIO air sampling biomass such as stumps. network. Oak Ridge National Laboratory, US Department of Energy, Oak Ridge, USA. In: Trends: a compendium of data on The avoided fossil CO2 emissions are greater when coal is global change. Tennessee, USA: Carbon Dioxide Information substituted, due to its greater carbon intensity. International Analysis Center [Accessed: March 2009]. Available from: trade in bioenergy is increasing rapidly, made feasible http://cdiac.ornl.gov/trends/co2/sio-mlo.html; 2009. by the development of efficient long-distance transport [2] Nakicenovic N, Swart R. Emissions scenarios, a special report methods allowing biofuel produced in one region to replace of working group III of the Intergovernmental Panel on biomass and bioenergy 35 (2011) 4340e4355 4353

Climate Change (IPCC). Cambridge, United Kingdom: [21] Statistical yearbook of forestry 2008. Jo¨ nko¨ ping: Swedish Cambridge University Press; 2000. Forest Agency [Skogsstyrelsen] [Accessed: August 2009]. [3] Houghton JT, Ding Y, Griggs DJ, Noguer M, PJvd Linden, Dai X, Available from: http://www.svo.se/episerver4/default.aspx? et al. Climate change 2001: the scientific basis. Contributions id¼38515; 2008. of working group I to the third assessment report of the [22] Departementsserien [Ministry Series]. Mer tra¨ i byggandet: Intergovernmental Panel on Climate Change (IPCC). underlag fo¨r en nationell strategi att fra¨mja anva¨ndning av Cambridge, United Kingdom: Cambridge University Press; tra¨ i byggandet (More wood in construction: basis for 2001. a national strategy to promote the use of wood in [4] Carter TR, Jylha¨ K, Perrels A, Fronzek S, Kankaanpaä¨ S. construction). Stockholm: Na¨ringsdepartementet [Ministry FINADAPT scenarios for the 21st century: alternative futures of Enterprise, Energy, and Communication]; 2004. for considering adaptation to climate change in Finland. [23] McMurtrie RE, Rook DA, Kelliher FM. Modelling the yield of Helsinki: Finnish Environment Institute; 2005. 42 pp. Pinus radiata on a site limited by water and nitrogen. For Ecol Mimeograph no.: 332. Manag 1990;30:381e413. [5] Bergh J, Freeman M, Sigurdsson B, Kelloma¨ki S, Laitinen K, [24] Lundstro¨ mA,So¨derberg U. Outline of the HUGIN system for Niinisto¨ S, et al. Modelling the short-term effects of climate longterm forecasts of timber yields and possible cut. change on the productivity of selected tree species in Nordic proceedings no. 5. In: Païvinen R, Roihuvuo L, Siitonen M, countries. For Ecol Manag 2003;183:327e40. editors. Large-scale forestry scenario models: experiences [6] Bergh J, Nilsson U, Kjartansson B, Karlsson M. Impact of and requirements. Joensuu: European Forest Institute; 1996. climate change on the productivity of silver birch, Norway p. 63e77. spruce and Scots pine stands in Sweden with economic [25] Hyvo¨ nen R, A˚ gren GI. Decomposer invasion rate, implications for timber production. Ecol Bull 2010;53:185e95. decomposer growth rate, and substrate chemical quality: [7] Stro¨mgren M, Linder S. Effects of nutrition and soil warming how they influence soil organic matter turnover. Can J For on stemwood production in a boreal Norway spruce stand. Res 2001;31:1594e601. Glob Change Biol 2002;8:1194e204. [26] SMHI. Climatic data. Norrko¨ ping: Swedish Meteorological [8] Rasmussen C, Southard RJ, Horwath WR. Mineral control of and Hydrological Institute [Accessed: August 2009]. Available organic carbon mineralization in a range of temperate from: http://www.smhi.se/klimatdata; 2009. conifer forest soils. Glob Change Biol 2006;12:834e47. [27] Sveriges National Atlas [Swedish National Atlas]. Klimat, [9] Pussinen A, Karjalainen T, Ma¨kipaä¨ R, Valsta L, Kelloma¨ki S. Sjo¨ ar och Vattendrag (Climate, Lakes and Rivers). Ho¨ gana¨s, Forest carbon sequestration and harvests in Scots pine stand Sweden: Bra Bo¨ cker; 1995. under different climate and nitrogen deposition scenarios. [28] Raupach MR, Marland G, Ciais P, Le Que´re´ C, Canadell JG, For Ecol Manag 2002;158:103e15. Klepper G, et al. Global and regional drivers of accelerating e [10] Bricen˜ o-Elizondo E, Garcia-Gonzalo J, Peltola H, Matala J, CO2 emissions. Proc Natl Acad Sci USA 2007;104:10288 93. Kelloma¨ki S. Sensitivity of growth of Scots pine, Norway [29] Kjellstro¨ mE,Ba¨rring L, Gollvik S, Hansson U, Jones C, spruce and silver birch to climate change and forest Samuelsson P, et al. A 140-year simulation of European management in boreal conditions. For Ecol Manag 2006;232: climate with the new version of the Rossby Centre regional 152e67. atmospheric climate model (RCA3). SMHI Reports in [11] Kirilenko AP, Sedjo RA. Climate change impacts on forestry. Meteorology and Climatology. Norrko¨ ping: Swedish Proc Natl Acad Sci USA; 2007:19697e702. Meteorological and Hydrological Institute; 2006. 54 pp. Report [12] Eggers J, Lindner M, Zudin S, Zaehle S, Liski J. Impact of no.:108. changing wood demand, climate and land use on European [30] Tans D, NOAA/ESRL, Keeling R. Trends in atmospheric forest resources and carbon stocks during the 21st century. carbon dioxide. Boulder: Earth System Research Laboratory; Glob Change Biol 2008;14:2288e303. 2009 [Accessed: July 2009]. Available from: http://www.esrl. [13] Bergh J, Linder S, Bergstro¨ m J. Potential production of noaa.gov/gmd/ccgg/trends/ and http://scrippsco2.ucsd.edu. Norway spruce in Sweden. For Ecol Manag 2005;204:1e10. [31] Elfving B. HUGINs ungskogstaxering 1976-1982 (HUGIN’s [14] Swedish Forest Agency [Skogsstyrelsen]. Skogliga young forest assessments 1976-1982). Umea˚: Swedish konsekvensanalyser (Forestry impact analysis) SKA-VB 08. University of Agricultural Sciences, Department of Forest Jo¨ nko¨ ping: Skogsstyrelsen; 2008. 157 pp. Rapport.: 25. Survey; 1982. 87 pp. (In Swedish). [15] Schlamadinger B, Apps M, Bohlin F, Gustavsson L, [32] Na¨slund B-A˚ . Simulation of damage and mortality in young Jungmeier G, Marland G, et al. Towards a standard stands and associated stand development effects. Umea˚: methodology for greenhouse gas balances of bioenergy Swedish University of Agricultural Sciences, Department of systems in comparison with fossil energy systems. Biomass Silviculture; 1986. 147 pp. (In Swedish with English Bioenerg 1997;13(6):359e75. summary). [16] Gustavsson L, Madlener R, Hoen HF, Jungmeier G, [33] Ha¨gglund B. Forecasting growth and yield in established Karjalainen T, Klo¨ hn S, et al. The role of wood material for forests. An outline and analysis of the outcome of greenhouse gas mitigation. Mitig Adapt Strat Glob Change a subprogram within the HUGIN project, 1981 analysis of the 2006;11:1097e127. outcome of a subprogram within the HUGIN project. Umea˚: [17] Gustavsson L, Sathre R. Variability in energy and carbon Swedish University of Agricultural Sciences, Department of dioxide balances of wood and concrete building materials. Forest Survey; 1981. 145 pp. Report no.: 31. Build Environ 2006;41:940e51. [34] So¨ derberg U. Funktioner fo¨ r skogliga produktionsprognoser- [18] Sathre R, O’Connor J. Meta-analysis of greenhouse gas Tillva¨xt och formho¨ jd fo¨ r enskilda tra¨d av inhemska tra¨dslag displacement factors of wood product substitution. Environ i Sverige (Functions for forecasting of timber yields. Section Sci Policy 2010;13:104e14. of Forest Mensuration and Management). Umea˚: Swedish [19] Lippke B, Wilson J, Perez-Garcia J, Boyer J, Meil J. CORRIM: University of Agricultural Sciences; 1986. 251 pp. Report no.: life-cycle environmental performance of renewable building 14. (In Swedish). materials. For Prod J 2004;54:8e19. [35] Jonsson B. The thinning response of Scots pine (Pinus [20] Gustavsson L, Pingoud K, Sathre R. Carbon dioxide balance of sylvestris) in Northern Sweden. Umea˚: Skogsho¨gskolan, Inst wood substitution: comparing concrete- and wood-framed fo¨ r skogsproduktion [College of Forestry, Department of buildings. Mitig Adapt Strat Glob Change 2006;11:667e91. forest]; 1974. 41 pp. 4354 biomass and bioenergy 35 (2011) 4340e4355

[36] Skogsstyrelsen [Swedish Forest Agency]. Skogsstyrelsens [57] Lindroth A, Klemedtsson L, Grelle A, Weslien P, Langvall O. fo¨ reskrifter och allma¨nna ra˚d till skogsva˚rdslagen (Swedish Measurement of net ecosystem exchange, productivity and Forest Agency’s regulations and general advice to the respiration in three spruce forests in Sweden shows Forestry Act). Jo¨ nko¨ ping: Swedish Forest Agency; 1993. unexpectedly large soil carbon losses. Biogeochemistry 2008; [37] Marklund L. Biomass functions for Norway spruce (Picea 89:43e60. abies) in Sweden. Umea˚: Swedish University of Agricultural [58] Luo Y. Terrestrial carbonecycle feedback to climate Sciences, Department of Forest Survey; 1987. 127 pp. warming. Ann Rev Ecol Evol Syst 2007;38:683e712. [38] Marklund LG. Biomass functions for pine, spruce and birch in [59] Davidson EA, Janssens IA. Temperature sensitivity of soil Sweden. Report 45: Department of Forest Survey, Swedish carbon decomposition and feedbacks to climate change. University of Agricultural Sciences; 1988, p. 73 (In Swedish Nature 2006;440:165e73. with English summary). [60] Friedlingstein P, Cox P, Betts R, Bopp L, von Bloh W, Brovkin V, [39] Petersson H, Sta˚hl G. Functions for belowground biomass of et al. Climateecarbon cycle feedback analysis: results from Pinus sylvestris, Picea abies, Betula pendula and Betula pubescens the C4MIP model intercomparison. J Clim 2006;19:3337e53. in Sweden. Scand J For Res 2006;21:84e93. [61] Knorr W, Prentice IC, House JI, Holland EA. Long-term [40] Elfving B, Kiviste A. Construction of site index equations for sensitivity of soil carbon turnover to warming. Nature 2005; Pinus sylvestris L. using permanent plot data in Sweden. For 433:298e301. Ecol Manag 1997;98:125e34. [62] Kirschbaum MUF. Will changes in soil organic carbon act as [41] Elfving B, Norgren O. Volume yield superiority of lodgepole a positive or negative feedback on global warming? pine compared to Scots pine in Sweden. In: Lindgren D, Biogeochemistry 2000;48:21e51. editor. Pinus contorta-from untamed forest to domesticated [63] Skogsstyrelsen [Swedish Forest Agency]. 20. Recom- crop. Umea˚: Department of Forest Genetics and Plant mendations for the extraction of forest fuel and Physiology, Swedish University of Agricultural Sciences; compensation fertilizing, 2011 forest fuel and compensation 1993. p. 69e80. fertilizing, vol. 3. Jonko¨ ping: National Board of Forestry [42] Hagglund B. Site index curves for Norway spruce in northern Message [Accessed: May 2011]. Available from: http://shop. Sweden. Stockholm: Department of Forest Yield Research, skogsstyrelsen.se/sv/meddelanden/recommendations-for- Royal College of Forestry; 1972. the-extraction-of-forest-fuel-and-compensati.html; 2002. [43] Hagglund B. Site index curves for Norway spruce in southern [64] Egnell G, Valinger E. Survival, growth, and growth Sweden. Stockholm: Department of Forest Yield Research, allocation of planted Scots pine trees after different levels Royal College of Forestry; 1973. of biomass removal in clear-felling. For Ecol Manag 2003; [44] Eriksson H. Yield of Norway spruce in Sweden. Stockholm: 177:65e74. Department of Forest Yield Research, Royal College of [65] Egnell G. Is the productivity decline in Norway spruce following Forestry; 1976. whole-tree harvesting in the ﬁnal felling in boreal Sweden [45] A˚ gren GI. Nitrogen productivity of some conifers. Can J For permanent or temporary? For Ecol Manag 2011;261:148e53. Res 1983;13:494e500. [66] Egnell G, Westling O, Dahlberg A, Bergh J, Rytter L. [46] Majdi H. Root sampling methods - applications and Miljo¨ effekter av skogsbra¨nsleuttag och aska˚terfo¨ring i limitations of the minirhizotron technique. Plant Soil 1996; Sverige e Ensyntes av Energimyndighetens 185:255e8. forskningsprogram 1997 till 2004 (Environmental effects of [47] A˚ gren GI, Bosatta E. Theoretical ecosystem ecology: logging residue removals and ash recycling in Sweden e A understanding element cycles. Cambridge, New York: synthesis of the Swedish Energy Agency’s research program Cambridge University Press; 1998. 1997 to 2004). Eskilstuna, Sweden: Swedish Energy Agency [48] Sathre R, Gustavsson L. Process-based analysis of added value Publication Service. 211 Available from: http://webbshop.cm. in forest product industries. For Policy Econ 2009;11:65e75. se/System/TemplateView.aspx?p¼Energimyndigheten&vi [49] Gustavsson L, Bo¨ rjesson P, Johansson B, Svenningsson P. ew¼all&cat¼/Rapporter&id¼eb21987e62fc49fe8a6b437ac

Reducing CO2 emissions by substituting biomass for fossil 7f9f940; 2006. pp.[In Swedish with English summary]. fuels. Energy 1995;20:1097e113. [67] Sathre R, Gustavsson L. Time-dependent climate benefits of [50] Eriksson E, Gillespie AR, Gustavsson L, Langvall O, Olsson M, using forest residues to substitute fossil fuels. Biomass Sathre R, et al. Integrated carbon analysis of forest Bioenerg 2011;35:2506e16. management practices and wood substitution. Can J For Res [68] Junginger M, Bolkesjø T, Bradley D, Dolzan P, Faaij A, 2007;37:671e81. Heinimo¨ J, et al. Developments in international bioenergy [51] Berg S, Lindholm E-L. Energy use and environmental impacts trade. Biomass Bioenerg 2008;32:717e29. e of forest operations in Sweden. J Clean Prod 2005;13:33 42. [69] Gustavsson L, Eriksson L, Sathre R. Costs and CO2 benefits of [52] Pussinen A, Nabuurs GJ, Wieggers HJJ, Reinds GJ, recovering, refining and transporting logging residues for Wamelink GWW, Kros J, et al. Modelling long-term impacts fossil fuel replacement. Appl Energ 2011;88:192e7. of environmental change on mid- and high-latitude [70] Battles J, Robards T, Das A, Waring K, Gilless J, Biging G, et al. European forests and options for adaptive forest Climate change impacts on forest growth and tree mortality: management. For Ecol Manag 2009;258:1806e13. a data-driven modeling study in the mixed-conifer forest of the [53] Kelloma¨ki S, Peltola H, Nuutinen T, Korhonen KT, Sierra Nevada, California. Climatic Change 2008;87:193e213. Strandman H. Sensitivity of managed boreal forests in [71] Netherer S, Schopf A. Potential effects of climate change on Finland to climate change, with implications for adaptive insect herbivores in European forests-General aspects and management. Phil Trans R Soc Lond B 2008;363:2339e49. the pine processionary moth as specific example. For Ecol [54] Lloyd J, Taylor JA. On the temperature dependence of soil Manag 2010;259:831e8. respiration. Funct Ecol 1994;8:315e23. [72] Kurz WA, Dymond CC, Stinson G, Rampley GJ, Neilson ET, [55] Jones RJA, Hiederer R, Rusco E, Montanarella L. Estimating Carroll AL, et al. Mountain pine beetle and forest carbon organic carbon in the soils of Europe for policy support. Eur J feedback to climate change. Nature 2008;452:987e90. Soil Sci 2005;56:655e71. [73] Lindner M, Maroschek M, Netherer S, Kremer A, Barbati A, [56] Bond-Lamberty B, Thomson A. Temperature-associated Garcia-Gonzalo J, et al. Climate change impacts, adaptive increases in the global soil respiration record. Nature 2010; capacity, and vulnerability of European forest ecosystems. 464:579e82. For Ecol Manag 2010;259:698e709. biomass and bioenergy 35 (2011) 4340e4355 4355

[74] Blennow K, Andersson M, Sallna¨s O, Olofsson E. Climate [76] SOU [Government Ofﬁcial Report]. Sverige info¨ r change and the probability of wind damage in two Swedish klimatfo¨ ra¨ndringarna - hot och mo¨ jligheter. Slutbeta¨nkande forests. For Ecol Manag 2010;259:818e30. av Klimat- och sa˚rbarhetsutredningen (Sweden facing [75] Schelhaas M-J, Nabuurs G-J, Schuck A. Natural disturbances climate change - threats and opportunities. Final Report of in the European forests in the 19th and 20th centuries. Glob the Climate and Vulnerability). Stockholm: Statens offentliga Change Biol 2003;9:1620e33. utredningar; 2007. Report no.:60.

Paper II

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 5 ( 2 0 1 2 ) 1 0 6 – 1 2 4

Available online at www.sciencedirect.com

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

Potential effects of intensive forestry on biomass production

and total carbon balance in north-central Sweden

a, a,c a,b a,c

Bishnu Chandra Poudel *, Roger Sathre , Johan Bergh , Leif Gustavsson ,

d e

Anders Lundstro¨m , Riitta Hyvo¨nen

Ecotechnology and Environmental Science, Mid Sweden University, O¨ stersund, Sweden

Southern Swedish Forest Research Centre, Swedish University of Agricultural Sciences, Alnarp, Sweden

Linnaeus University, Va¨ xjo¨, Sweden

Department of Forest Resource Management, Swedish University of Agricultural Sciences, Umea˚ , Sweden

Department of Ecology, Swedish University of Agricultural Sciences, Uppsala, Sweden

a b s t r a c t

a r t i c l e i n f o

We quantify the potential effects of intensive forest management activities on forest

Published on line 20 October 2011

production in north-central Sweden over the next 100 years, and calculate the potential

climate change mitigation feedback effect due to the resulting increased carbon stock and

Keywords:

increased use of forest products. We analyze and compare four different forest management

Climate change

scenarios (Reference, Environment, Production, and Maximum), all of which include the expected

Forest biomass

effects of climate change based on SRES B2 scenario. Forest management practices are

Wood substitution

intensiﬁed in Production scenario, and further intensiﬁed in Maximum scenario. Four differ-

Forest management

ent models, BIOMASS, HUGIN, Q-model, and Substitution model, were used to quantify net

Carbon emission reduction

primary production, forest production and harvest potential, soil carbon, and biomass

substitution of fossil fuels and non-wood materials, respectively. After integrating the

models, our results show that intensive forestry may increase forest production by up to

26% and annual harvest by up to 19%, compared to the Reference scenario. The greatest single

effect on the carbon balance is from using increased biomass production to substitute for

fossil fuels and energy intensive materials. Carbon stocks in living tree biomass, forest soil

and wood products also increase. In total, a net carbon emission reduction of up to 132 Tg

(for Maximum scenario) is possible during the next 100 years due to intensive forest

management in two Swedish counties, Ja¨mtland and Va¨sternorrland.

(IPCC, 2007b; Keeling et al., 2009). This temperature increase is

1. Introduction

expected to continue during the 21st century (IPCC, 2007b) and

to be more pronounced at higher latitudes (Houghton et al.,

The observed increase of atmospheric greenhouse gas (GHG) 2001). In northern Europe the mean temperature may increase

concentration is mainly due to emissions from fossil fuel 1–2 8C in summer and 2–3 8C in winter during the next 50 years

combustion, land-use changes and industrial processes (Carter et al., 2005). Hence, the boreal coniferous forest regions

(Matthews and Caldeira, 2008; Quadrelli and Peterson, 2007). in northern Europe will be especially affected.

Increased GHG concentration has led to an increased mean The increased temperature and CO2 concentration is

surface temperature on the earth during the 20th century expected to increase forest growth. Net primary production

* Corresponding author. Tel.: +46 63 165535, fax: +46 63 165500.

E-mail addresses: [email protected], [email protected] (B.C. Poudel).

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 5 ( 2 0 1 2 ) 1 0 6 – 1 2 4 107

of Swedish boreal forests might increase by 20–40% over the carbon emissions such as in cement manufacturing, (iii) the

next 100 years (Poudel et al., 2011). Several studies suggest that storage of carbon in wood products, and (iv) the use of wood

larger harvest levels will be available in future because of by-products and post-use wood products to replace fossil

climate change (Bricen˜ o-Elizondo et al., 2006; Eggers et al., fuels. Biomass substitution is recognized as an important part

2008; Kirilenko and Sedjo, 2007; Poudel et al., 2011; Pussinen of a long-term strategy to use forest resources for climate

et al., 2002). Poudel et al. (2011) projected that climate change change mitigation (IPCC, 2007a).

based on the Intergovernmental Panel on Climate Change The aim of this study is to assess the potential effects of

(IPCC) B2 scenario (Nakicenovic and Swart, 2000) will increase intensive forest management activities on forest production in

the forest biomass production by 33% in the next 100 years in north-central Sweden, and the potential uses of the increased

north-central Sweden. forest production to mitigate climate change. Using various

Besides the effects of climate change, forest production scenarios for forest production and biomass utilization, we (i)

may vary with different management methods. Boreal forests estimate the increased biomass availability and the potential

are generally characterized by low productivity because of to use the additional forest production to substitute for non-

low nitrogen availability (Tamm, 1991), which is indirectly an renewable materials and fuels, (ii) estimate the effects of

effect of soil temperatures that slow mineralization and biomass removals on the carbon stock in living tree biomass

decomposition rates of soil organic matter (Kirschbaum, and forest soils, (iii) quantify the emissions due to forest

2000; Vanhala et al., 2008). Bergh et al. (2005) showed that if operations and emissions (CO2, N2O, and CH4) associated with

nutrient availability is non-limiting, forest production can fertilizer production and use, and (iv) quantify the overall

increase by 300% in northern Sweden and by 80% in southern carbon balance of each scenario based on the changes in

Sweden compared to current production. In addition, selec- carbon stock in forest biomass, forest soils and wood products,

tion of species according to soil moisture, texture, and and the avoided fossil emissions due to biomass substitution.

structure might further increase forest productivity. Further-

more, management practices such as regeneration methods,

thinning, and ﬁnal harvesting may also increase forest 2. Study area and scenarios

productivity.

Increased forest biomass production can be used to 2.1. Study area

mitigate climate change by using the biomass to substitute

fossil fuels (Schlamadinger et al., 1997) as well as carbon- The study area (Fig. 1) encompasses the counties of

intensive materials such as concrete (Gustavsson et al., 2006a; Ja¨ mtland and Va¨ sternorrland in north-central Sweden, from

0 0 0 0

Poudel et al., 2011). Biomass substitution can affect energy and 61833 to 65807 N latitude and from 12809 to 19818 E

carbon balances through several mechanisms (Sathre and longitude. The elevation ranges from sea level to 1500 m

O’Connor, 2010). These may include (i) the lower primary altitude. Annual precipitation ranges from 600 to 700 mm in

energy used to manufacture wood products compared with the eastern part to 1500 mm in the western part (SMHI, 2009)

alternative materials, (ii) the avoidance of industrial process and the range of growing season (mean temperature above

Fig. 1 – The study area of Ja¨mtland and Va¨sternorrland counties, Sweden.

Data source: Department of Land Survey, Sweden.

108 e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 5 ( 2 0 1 2 ) 1 0 6 – 1 2 4

Table 1 – Overview of scenarios analyzed in this study.

Scenario name Forest management goals Biomass use

Reference Reference (current management) Stemwood or whole tree

Environment Fulﬁll environmental goals Stemwood or whole tree

Production Increase biomass production Stemwood or whole tree

Maximum Maximize biomass production Stemwood or whole tree

+5 8C) is approximately 120–160 days per year (Sveriges- production rate. The four scenarios are employed to estimate

Nationalatlas, 1995). Average monthly temperature during the changes in forest biomass production caused by explicitly

1961–1990 varied in January between +1 8C and 12 8C, and different forest management practices. The Reference scenario

in July between 10 8C and 14 8C. describes the development of the forest with silvicultural

The forest land area of the counties of Ja¨mtland and techniques present in Swedish forestry today and with the

Va¨sternorrland is about 3.5 million ha and 1.9 million ha, current environmental policy, where speciﬁed environmental

respectively. Of this area, productive forests excluding goals will be fulﬁlled (Naturva˚rdsverket and Skogsstyrelsen,

protected areas cover about 2.6 million ha and 1.7 million ha, 2005).

respectively (Skogsstyrelsen, 2010). The dominant tree species The Environment scenario describes the development of the

are Norway spruce (Picea abies (L.) Karst.), Scots pine (Pinus forest with increased environmental ambitions. In this

sylvestris L.) and birch (Betula spp.). Scots pine with site index scenario 8% of total forest land is set aside for protected

(H100) T20–T22 covers 71% and 83% of the site index reserves compared to 4% in the Reference scenario, and 14% of

distribution of forest land in Ja¨mtland and Va¨sternorrland, total forest land is set aside for special environmental care

respectively, while Norway spruce with site index (H100) G18– compared to 8% in the Reference scenario. These ambitions

G22 covers 57% and 69% of the site index distribution of forest are on a higher level than proposed for 2020 by the Swedish

land in Ja¨mtland and Va¨sternorrland. Age class distribution of Environmental Agency (Naturva˚rdsverket, 2008).

the forest stands shows that 12% of forest land has trees of In the Production scenario, different silvicultural actions are

age class 21–30 years, and 12% has trees of age class 41–60 taken to increase forest production. These include improved

years. Over-mature trees that are older than 100 years cover genetic material in seedling plantations, soil scariﬁcation,

30% of forest land, of which a third has trees older than 140 selection of tree species suitable for particular site conditions,

years (Skogsstyrelsen, 2010). Standing biomass of the produc- and shorter time between clear-cut and planting. Increased

tive forest area is 526 million m (2005–2009 average), which pre-commercial thinning and increased area of traditional

corresponds to a carbon content of 105 Tg C. In the past 12 fertilization is included. Balanced nutrient supply in young

years the annual stem increment of forests in Ja¨mtland and stands of Norway spruce and former agricultural land forest is

Va¨sternorrland has increased from 15 to 20 million m per included to intensify the production. These investments in

year. In terms of carbon, the annual increment has increased forestry are high but economically realistic (Skogsstyrelsen,

from 3 Tg C to 4 Tg C in the span of 12 years. The annual 2008). The environmental ambition is the same as in the

harvest of stemwood has varied between 11.5 and 13.5 mil- Reference scenario.

lion m per year, which in terms of carbon is 2.3 Tg C and 3.7 Tg In the Maximum scenario, additional silvicultural actions

C. The annual gross felling in the whole of Sweden has are taken to maximize the forest production. Replacement of

increased by 22% during the past 10 years (Skogsstyrelsen, Scots pine with Lodgepole pine (Pinus contorta), and balanced

2010). nutrient supply (fertilization) in young stands of Lodgepole

pine and Norway spruce on all suitable forest land, are

2.2. Climate and management scenarios included. The environmental ambition is the same as in the

Reference scenario.

The regional climate scenarios were produced by downscaling The average age for ﬁnal felling is 113 years for Reference,

global climatic models to Sweden and surrounding areas 110 years for Environment and Production scenario, and 106

(Kjellstro¨ m et al., 2006; SMHI, 2009), based on the SRES B2 years for Maximum scenario. The details of the silvicultural

scenario (Nakicenovic and Swart, 2000). The climate modeling operations and management practices including fertilization

covers the period of 1961–2100, including the reference climate for all scenarios are given in Table 2.

period of 1961–1990 (for detail see Poudel et al., 2011). The SRES Traditional fertilization includes application of 150 kg of N

B2 climate change scenario corresponds to moderate emis- ha at least once in 10 years. The balanced fertilization

sions of greenhouse gases leading to an atmospheric CO2 regime includes both N and nitrogen–phosphorus–potassium

concentration of 621 ppm by the year 2100 (Levy et al., 2004). (NPK) fertilizers applied as needed according to foliar analysis

Four forest management scenarios are used to represent every second year in young stands and every tenth year in

different intensities of silvicultural activities and environ- mature stands. Balanced fertilization is discussed in detail in

mental protection (Table 1). Anticipated climate change is Sathre et al. (2010).

included in all four scenarios. Alteration of management For each of the scenarios Reference, Environment, Production,

practice can affect the forest carbon stock and the forest and Maximum, we consider ‘‘stemwood’’ and ‘‘whole tree’’

biomass harvest and use options. In the stemwood option we

Non-productive forests are deﬁned as forests with production assume that only stemwood (tree stem from base to diameter

capacity less than 1 m per ha and year. of 20 cm over bark) is harvested, while in the whole tree option

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 5 ( 2 0 1 2 ) 1 0 6 – 1 2 4 109

Table 2 – Forest management practices and fertilization in different scenarios (units are in 1000 ha year ).

Description of activities Management practices in 1000 ha year in each scenario

Reference Environment Production Maximum

Final felling 42 39 45 46

Pre-commercial thinning 31 29 40 42

Thinning 64 63 63 62

Traditional fertilization 10 9 48 46

Balanced fertilization 0 0 3 14

Plantation, Scots pine 16 15 18 0

Plantation, Norway spruce 12 11 15 19

Plantation, Lodgepole pine 1 1 8 22

Natural regeneration 13 12 4 5

Soil scariﬁcation 36 33 43 44

we assume that in addition to stemwood, the thinning fuels and non-wood materials. The overall carbon balance

residues, the harvest slash (branches, foliage, and tree tops), incorporates carbon stock changes in tree biomass, forest soils

and recoverable stumps and coarse roots are removed for use and wood products, and avoided fossil carbon emissions due

as bioenergy. to biomass substitution of non-wood materials and fossil

fuels. The time period of the study is from 2010 to 2109, broken

down into 10-year time steps.

3. Component models and model based

analyses 3.2. BIOMASS model

3.1. Schematic framework of modeling BIOMASS is a process-based growth model describing the

processes of radiation absorption, canopy photosynthesis,

The general structure of our analytical approach is shown phenology, allocation of photosynthates among plant organs,

schematically in Fig. 2, which is described in detail by Poudel litter-fall, and stand water balance. It consists of a series of

et al. (2011). The different forest production scenarios are equations which are based on established theories of plant–

simulated over the next 100 years. The process-based model physiological processes and soil–water dynamics (McMurtrie

BIOMASS (McMurtrie et al., 1990) is used to estimate biomass et al., 1990). Transient simulations were performed with

production, the empirical forecast model HUGIN (Lundstro¨m climate data from the IPCC SRES B2 scenario to get estimates of

and So¨derberg, 1996) is used to estimate yield and harvest net primary production (NPP) for 1961–2100 (Poudel et al.,

level, and the Q model (Hyvo¨nen and A˚ gren, 2001) is used to 2011). These data are then used to determine the growth

estimate soil carbon stock changes based on litter input and functions on a county level in the HUGIN model. Simulated

biomass removals. A biomass substitution model determines mean increase in NPP for Ja¨mtland County for the period 2071–

the net avoided carbon emissions due to substitution of fossil 2100 compared with 1961–1990 was 28.9%, 21.4% and 21.6% for

Norway spruce, Scots pine and birch, respectively. The

corresponding numbers for Va¨sternorrland County were

21.6%, 11.0% and 13.9%.

Linear effect of climate change is assumed for each tree

species and county, such that the full increase in NPP occurs

after 100 years. The differences in leaf area index is used to

estimate the growth effects in dry mass values of foliage, stem,

branches and roots corresponding to the amount of absorbed

photosynthetically active radiation and respiring biomass.

These calculated effects are used for established forest stands

in the HUGIN system, both in mature and pre-mature forests.

For young forests, a calibration of site index is made to ﬁnd the

climate induced production effect. An increase of site index by

4 m is assumed to occur linearly during the 100-year projection.

3.3. HUGIN model

HUGIN is a regional-level strategic planning model that

forecasts timber yield and possible harvest level (Lundstro¨ m

and So¨ derberg, 1996). It uses sample plots from the Swedish

National Forest Inventory (NFI) to deﬁne the model baseline.

In the present analysis, natural site productivity and the

climate conditions in the model were modiﬁed to incorpo-

Fig. 2 – Schematic structure of the analytical approach. rate the effects of a changed climate (see Section 3.2). The

110 e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 5 ( 2 0 1 2 ) 1 0 6 – 1 2 4

growth simulators are constructed to be valid for all forest stemwood as bioenergy, acknowledging that this may not be

stands within a wide range of management alternatives in the most economically beneﬁcial use of the wood (Sathre and

Sweden. Separate methods are used to describe the growth Gustavsson, 2009).

in young stands (until the average height of the trees on a In the whole tree biomass use option, in addition to

plot has reached about 7 m) and for the rest of the life-time harvesting 95% of stemwood from thinning and ﬁnal fellings

of established stands. When pre-commercial thinning is we assume that 75% of needles, branches and tops, and 50% of

performed, the number of stems to be left depends on site stumps and coarse roots are harvested and used for bioenergy.

index and species, based on the recommendations in the We assume that all recovered biofuels are used to replace

Swedish Forest Act (Skogsstyrelsen, 1993). The level of felling either coal or fossil gas in stationary plants with relative

is based on the growth level in every period, and then conversion efﬁciencies of 100% and 96%, respectively (Gus-

calibrated depending on the state of the forest. The amount tavsson et al., 2006b). Values of speciﬁc C emission from the

of the total felling is calculated depending on age distribu- combustion of fossil fuels are assumed to be 30 kg C GJ for

1 1

tion and thinning rate, adjusted by the actual growing stock coal, 22 kg C GJ for oil, and 18 kg C GJ for fossil gas, and

on thinning plots relative to expected growing stock on these include emissions during the entire fuel-cycle from the

plots. The HUGIN model is described in more detail by Poudel natural resource to the delivered energy service (Gustavsson

et al. (2011). et al., 1995). Energy used for recovery and transport of biofuels

is assumed to be diesel fuel, calculated as a percentage of the

3.4. Soil carbon modeling heat energy content of the biofuel. We assume this percentage

is 10% for stumps, 5% for slash and small diameter stemwood,

Changes in soil carbon stock may be expected due to a warmer and 1% for processing residues (Eriksson et al., 2007).

climate causing increased soil respiration leading to decreased Large diameter stemwood is assumed to be used for

soil C stock, increased tree growth and litter fall leading to producing wood construction material that substitutes in place

increased soil C stock, and increased biomass removal leading of conventional reinforced concrete. The material substitution

to decreased soil C stock. We estimate climate-related soil effects are based on a case study of a multi-story apartment

carbon effects using a linear extrapolation between the building constructed in Sweden using wood structural framing,

reference climate in 2010 and the 4 8C warmer climate in compared to a functionally equivalent building constructed

2109 (A˚ gren and Wetterstedt, 2007). Forest production data for with a reinforced concrete frame (Gustavsson et al., 2006b). The

the soil carbon analysis is calculated independently from that substitution beneﬁts take into account the differences between

described in Section 3.3, and is detailed by Poudel et al. (2011). the buildings due to the emission from fossil fuels used to

The decomposition of the different litter fractions is calculated manufacture and transport the materials, the avoided cement

using the Q model where invasion rates of litter types are calcination and carbonation process emissions from concrete

taken into account (Bosatta and A˚ gren, 1996; Hyvo¨nen and substituted by wood material, and substitution of fossil fuel by

A˚ gren, 2001). The calculations of remaining carbon are made residues from wood processing and construction. The reference

by year for each litter component and each litter cohort concrete building uses some wood materials, and the substitu-

separately such that the total soil carbon consists of the sum of tion beneﬁt is calculated based on the reduction in net carbon

remaining litter cohorts of different ages and from different emission per unit of additional wood needed to make the wood-

tree components. frame building. Large roundwood is converted into a mix of

sawn lumber, plywood, and particleboard, with a product/

3.5. Biomass substitution modeling roundwood dry weight ratio of 0.53. Some wood processing

residue is used for particleboard manufacture, while the

The biomass recovered from ﬁnal felling and thinnings is remainder is used as bioenergy. Bark and wood-based con-

assumed to be used in different ways, depending on the struction site waste is assumed to be recovered and used as

biomass usage option. In the stemwood biomass use option, bioenergy. We assume a building lifespan of 100 years, and

we assume that 95% of stemwood from ﬁnal fellings is calculate the carbon stock stored in the wood building materials

harvested from the forest. Large diameter pine and spruce during that time.

stemwood is used for production of wood construction

material, with the processing residues used as bioenergy. 3.6. Emissions from forest operations and fertilization

Small diameter pine and spruce stemwood, and all deciduous

stemwood, is used as bioenergy. We assume a minimum Emissions from fossil fuels used for forest management

diameter of 20 cm for large diameter stemwood, which is activities are based on Berg and Lindholm (2005), and include

standard size to convert roundwood to sawnwood. Stemwood stand establishment, thinnings, ﬁnal harvest of roundwood,

is sorted entirely by diameter without consideration of quality and forwarding and transport of roundwood to mills. Energy

aspects, thus we are likely to overestimate the quantity of use data for northern Sweden are used.

stemwood used for construction material. Small-diameter GHG emissions associated with the production and use of

stemwood is generally used in Sweden for pulp and paper fertilizers include CO2, N2O and CH4. In our calculations we use

production. In this study we analyze the use of additional a global warming potential (GWP) of 298 for N2O and 25 for CH4,

quantities of biomass produced due to climate change, but we relative to the radiative forcing of CO2 over a 100-year time

have not analyzed whether there is demand for additional horizon (IPCC, 2007b). Primary energy use and GHG emissions

pulpwood in the pulp industry. We therefore show the climate for the production of Skog-CAN fertilizer (27-0-0) and Opti-

mitigation effects of using this additional small diameter Crop fertilizer (24-4-5) are based on Davis and Haglund (1999).

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 5 ( 2 0 1 2 ) 1 0 6 – 1 2 4 111

The amount of fossil fuel used for fertilizer application by Reference scenario. Stump biomass harvest in the Environment,

helicopter is based on Mead and Pimentel (2006), correspond- Production, and Maximum scenarios increase by 40%, 58%, and

ing to 0.022 kg CO2 emission per kg fertilizer. The use of N 66%, respectively, compared to a 47% increase for Reference

fertilizer may lead to emission of N2O from the soil, due to the scenario over the 100-year period.

microbiological processes of nitriﬁcation and denitriﬁcation. Biomass growth of different tree species is shown in Fig. 5.

Nordin et al. (2009) state that between 0.5% and 1% of the N in Spruce growth decreases in the beginning, then increases in

the applied fertilizer can be expected to be released as N2O, the end in all scenarios. The Production scenario has an earlier

and MacDonald et al. (1997) found that 1% of the N deposited recovery in growth. Growth decrease in the beginning may be

on upland spruce forests in Scotland was emitted as N2O. In a result of increased harvest level compared to the Reference

this study we assume that 1% of the N in the applied fertilizers and Environment scenarios. Rapid increase of spruce in the

is released as N2O. We do not consider the potential for N later period in the Maximum scenario is due to improved

addition to inhibit the oxidation of CH4 by soil microorgan- genetic plantation material and fertilization effect. Pine

isms, which is expected to be negligible in comparison to soil growth increases in all scenarios. Lodgepole pine planted in

carbon stock changes due to fertilization (Nordin et al., 2009). place of Scots pine gives continuous increase in pine growth.

Higher growth is observed for deciduous species for the initial

70 years but thereafter decreases in all scenarios. A clear

4. Results distinction in biomass growth between species is observed in

the Maximum scenario; spruce initially decreases and then

4.1. Change in forest biomass and carbon stock increases prominently, pine increases linearly, and deciduous

decreases after the middle of the study period. In Reference and

The forest biomass production and harvest for the four Environment scenarios there is an increase in the ﬁrst 70–80

scenarios for both whole tree and stemwood use options are years and then stabilizes and slightly decreases. In Production

shown in Table 3. Biomass production and harvest are highest and Maximum the increase is shorter for about 60 years and

for the Maximum scenario, followed by the Production, Reference, thereafter a clear decrease occurs. The reason for these

and Environment scenarios. Compared to the Reference scenario, decreases is differences in silviculture, where Production and

cumulative whole tree production over the 100-year period Maximum scenarios promote conifers more than in Reference

increased by 225 Tg dry mass in the Maximum scenario, and Environment. More conifers in regenerations and geneti-

increased by 105 Tg dry mass in the Production scenario, but cally strong material for the planted species and more

decreased by 37 Tg dry mass in the Environment scenario. frequent and effective pre-commercial thinning are used,

Compared to the increase in annual biomass production in the which leads to reduced growth of deciduous trees. This may be

Reference scenario over the 100 years from 2010 to 2109, the explained as the outcome of intensive silviculture and

annual production increase in the Maximum scenario is 26% fertilization to promote the growth.

greater for whole tree biomass and 25% greater for stemwood Total standing biomass in the year 2010 and the year 2109 for

biomass. In the Environment scenario, however, the annual each scenario are shown in Fig. 6. All scenarios have larger

production increase is 9% less for whole tree and 7% less for standing biomass at the end of the study period compared to the

stemwood biomass, compared to the Reference scenario. The standing biomass today. Maximum scenario has the largest

higher production in the Maximum scenario is caused by standing biomass, followed by the Environment scenario, which

species change, fertilization, and other intensive silvicultural is followed by the Production and Reference scenarios. Greater

actions. The decrease in the Environment scenario is because of standing biomass in Environment scenario may be explained as a

increased set aside area and probable mortalities. result of increased set aside land and lower harvest intensity.

Differences in biomass production and harvest for Environ- Standing biomass increases by 31%, 44%, 37% and 56% for

ment, Production and Maximum scenarios are compared to the Reference, Environment, Production and Maximum scenarios,

Reference scenario (Fig. 3). The difference between Maximum and respectively over the 100-year period.

Reference scenarios steadily increases during the whole study Average annual carbon stock changes in tree biomass are

period, while the difference between Environment and Reference shown in Table 4. The average tree biomass carbon stock

scenarios is negative and also increases steadily during the increases continually during the 100-year period, but not

study period. The difference between Production and Reference linearly. The amount of stock increase suddenly reduces after

scenarios increases during the initial decades and then 20 years of management before rising again in the next 10

decreases. The Maximum scenario gives higher production years. The sudden decrease might be because of increased

and harvest than all other scenarios. harvest in that period.

Average annual harvests for 10-year periods for different

products are shown in Fig. 4. The whole tree biomass harvest 4.2. Change in soil carbon stock

in the Reference scenario increases about 50% over the 100-

year period. The Production and Maximum scenarios allow 12% Soil carbon stock changes are shown in Table 5. Soil carbon

and 19% more biomass harvest, respectively, compared to the stock increases for all scenarios during the 100-year study

Reference scenario for the same period whereas the Environ- period. The soil carbon stock is calculated based on input from

ment scenario has a 5% decrease in harvest. Average stem- litter fall from standing trees and harvesting/thinning resi-

wood biomass harvest for Environment, Production, and dues. A fraction of standing biomass of needles, ﬁne roots and

1 1

Maximum scenarios are 5.84 Tg year , 6.69 Tg year , and branches is lost in litter fall each year. At clear-fellings and

1 1

6.87 Tg year , respectively, compared to 6.30 Tg year for thinnings a fraction of stems (5% of standing biomass) is left at 112

S Table 3 – Average annual forest biomass production and harvest during each 10-year period for each scenario (Tg year 1 dry mass) for Ja¨mtland and Va¨sternorrland. 2010–2019 2020–2029 2030–2039 2040–2049 2050–2059 2060–2069 2070–2079 2080–2089 2090–2099 2100–2109 Cumulative total (Tg) Reference scenario Biomass production e

Whole tree biomass 11.3 12.1 12.4 13.5 14.0 14.2 14.4 14.8 16.1 16.9 1397 n

Stemwood biomass 6.4 6.4 6.7 7.2 7.4 7.6 7.7 7.9 8.5 9.0 747 v

Biomass harvest o

Whole tree biomass 8.3 8.3 9.4 9.5 10.2 10.5 10.8 10.9 11.4 12.4 1018 n

Stemwood biomass 5.1 5.1 5.7 5.8 6.3 6.5 6.8 6.8 7.2 7.8 630 e

Environment scenario a

Biomass production s

Whole tree biomass 11.3 12.3 12.5 13.5 13.7 13.7 13.9 14.3 15.2 15.8 1361 c

Stemwood biomass 6.4 6.5 6.7 7.2 7.3 7.4 7.5 7.7 8.1 8.5 732 e

Biomass harvest c

Whole tree biomass 7.7 7.8 8.8 9.0 9.5 9.7 9.7 9.9 11.0 11.1 942 &

Stemwood biomass 4.7 4.8 5.4 5.5 5.9 6.1 6.1 6.2 6.9 7.0 584 p

Production scenario i

Biomass production y

Whole tree biomass 11.5 12.8 13.7 15.1 15.7 15.6 15.6 16.0 16.8 17.6 1502 5

(

Stemwood biomass 6.5 6.7 7.3 7.9 8.3 8.3 8.3 8.5 8.9 9.4 801 2

Biomass harvest 1

Whole tree biomass 8.2 8.4 9.6 10.1 10.8 11.6 11.8 11.9 12.5 13.3 1082 2

)

Stemwood biomass 5.0 5.1 5.9 6.1 6.6 7.2 7.4 7.5 7.8 8.3 669 1

Maximum scenario –

Biomass production 2

4 Whole tree biomass 11.5 12.8 13.9 15.9 16.5 16.7 17.3 18.1 19.3 20.2 1622 Stemwood biomass 6.5 6.7 7.3 8.3 8.7 8.9 9.2 9.6 10.2 10.7 861 Biomass harvest Whole tree biomass 8.3 8.6 9.9 10.3 10.8 11.7 12.1 12.4 12.9 14.1 1112 Stemwood biomass 5.1 5.2 6.0 6.3 6.7 7.2 7.6 7.8 8.1 8.8 687

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 5 ( 2 0 1 2 ) 1 0 6 – 1 2 4 113

Max imum minus Reference Production minus Reference Environment minus Reference

Differe nces in biomass production: whole tree Differences in biomass production: stem wood 4 4 3.5 3.5

-1 3 3 2.5 2.5 2 2 1.5 1.5 1 1 0.5 0.5

0 0 Dry mass Tgyear -0.5 -0. 5

-1 -1 -1.5 -1. 5

2010-2019 2040-2049 2080-2089 2020-2029 2030-2039 2050-2059 2060-2069 2070-2079 2090-2099 2100-2109 2010-2019 2020-2029 2030-2039 2040-2049 2050-2059 2060-2069 2090-2099 2070-2079 2080-2089 2100-2109

Differences in biomass harvest: whole tree Differe nces in biomass harvest: stem wood 4 4 3.5 3.5

-1 3 3 2.5 2.5 2 2 1.5 1.5 1 1 0.5 0.5

0 0

Dry mass Tgyear -0.5 -0. 5

-1 -1 -1.5 -1. 5

2010-2019 2020-2029 2030-2039 2060-2069 2070-2079 2080-2089 2090-2099 2040-2049 2050-2059 2100-2109 2020-2029 2040-2049 2050-2059 2060-2069 2070-2079 2080-2089 2090-2099 2100-2109 2010-2019 2030-2039

Ye a r Ye a r

Fig. 3 – Differences of average annual forest biomass production and harvest during each 10-year period between the

Environment, Production and Maximum scenarios and the Reference scenario for Ja¨mtland and Va¨sternorrland.

the site, and at thinnings the stumps and coarse roots are left year 2010. The total soil carbon stocks for the three scenarios

at the site contributing to the soil carbon stock. Soil carbon in the end of year 2109 are 16.4 Tg C, 18.2 Tg C, and 20.3 Tg C,

derived from previous stands and management operations respectively.

was not included in these calculations because of lack of Soil carbon stock increases more with stemwood biomass

information on land use history. This means that we are able use than with whole tree biomass use. However, the rate of

only to compare the effects of different harvesting intensities increase is large during the ﬁrst 10-year period and then slows

from the present stands on soil carbon dynamics. Total soil down for all scenarios over the rest of the study period.

carbon stock, excluding available old soil carbon, for the Although the biomass left in the forest is assisting carbon

Reference scenario with whole tree harvest option is calculated build up, eventually the soil carbon stock approaches

12.2 Tg C for the year 2010 and 16.3 Tg C for the year 2109. Total equilibrium for each management scenario, so that thereafter

soil carbon values for Environment, Production and Maximum only small amounts of soil carbon are added to the stock.

scenario with whole tree harvest option are all 12.2 Tg C for the Whole tree biomass harvest continues to add lower amounts

Table 4 – Changes in average annual tree biomass carbon stock during each 10-year period (Tg C year ) for Ja¨mtland and

Va¨sternorrland.

2010–2019 2020–2029 2030–2039 2040–2049 2050–2059 2060–2069 2070–2079 2080–2089 2090–2099 2100–2109

Reference 0.49 0.85 0.35 0.80 0.63 0.53 0.45 0.61 0.96 0.75

scenario

Environment 0.81 1.25 0.75 1.08 0.90 0.81 0.93 0.97 0.76 0.98

scenario

Production 0.61 1.08 0.79 1.22 1.07 0.59 0.47 0.59 0.65 0.53

scenario

Maximum 0.54 1.01 0.73 1.52 1.44 1.08 1.07 1.34 1.58 1.31 scenario

114 e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 5 ( 2 0 1 2 ) 1 0 6 – 1 2 4

Fig. 4 – Average annual biomass harvest by product for each 10-year period for the different scenarios for Ja¨mtland and

Va¨sternorrland.

of soil carbon each year compared to stemwood biomass 554 Tg C, 515 Tg C, 599 Tg C and 618 Tg C, respectively, over 100

harvest. years. When stemwood biomass is used for substitution, the

cumulative avoided carbon emission with coal reference fossil

4.3. Change in carbon stock in wood products fuel for Reference, Environment, Production and Maximum scenarios

are 438 Tg C, 407 Tg C, 474 Tg C and 489 Tg C, respectively, over

Wood product carbon stock changes are shown in Table 6. The 100 years. The substitution beneﬁts are lower when fossil gas is

changes are based on the differences between new wood the reference fossil fuel because of its lower carbon intensity.

products entering service and old wood products that are Substitution beneﬁts are lower for the Environment scenario

demolished. The carbon stock in wood products increases in because of the lower amount of biomass harvest compared to

all scenarios throughout the study period. The Production and the Reference scenario.

Maximum scenarios have signiﬁcantly increased rate of wood

product carbon stock over 100-years while Reference and 4.5. Total carbon balance

Environment scenarios have slower rates, which is due to the

smaller harvest in these scenarios. The total annual 10-year average net carbon balance for each

scenario, based on tree biomass carbon stock changes, soil

4.4. Reduced emissions due to biomass substitution carbon stock changes, wood product carbon stock changes,

and avoided carbon emissions due to material and fuel

The avoided fossil and cement reaction carbon emissions due to substitution are shown in Table 8. Biomass substitution,

using forest biomass to substitute for fossil fuels and energy- particularly when replacing coal fuel, is the largest contributor

intensive non-wood products are shown in Table 7. The net to the carbon balance. Whole tree biomass use results in lower

substitution values are net values after deducting emissions soil carbon stock increase (Table 5) and higher avoided

from forest management and biomass logistics. The increased emissions due to substitution (Table 7) compared to stemwood

biomass production in the Production and Maximum scenarios biomass use. In Reference scenario, when coal is the reference

has greater substitution potential, resulting in a substantially fossil fuel, the increased substitution beneﬁt of whole tree use

larger carbon emission reduction than in the Reference scenario. is about 4 times greater than the change in soil carbon increase

Total cumulative avoided carbon emission resulting from at the beginning of study period, and reaches up to 40 times

whole tree biomass substitution with coal reference fossil fuel greater at the end of the 100-year period. The Production and

for Reference, Environment, Production and Maximum scenarios are Maximum scenario have greater substitution beneﬁts com-

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 5 ( 2 0 1 2 ) 1 0 6 – 1 2 4 115

35 35 Refere nce Environment 30 30 -1 25 25

20 20

15 15

10 10

Dry mass Tg year

Pine Spruce Deciduous Pine Spruce Deciduous 5 5 2020-2029 2040-2049 2060-2069 2080-2089 2090-2099 2000-2009 2010-2019 2030-2039 2070-2079 2100-2109 2050-2059 2000-2009 2070-2079 2090-2099 2100-2109 2010-2019 2020-2029 2030-2039 2040-2049 2050-2059 2060-2069 2080-2089

35 35 Production Max imum 30

-1 30 25 year 25 20 20 15 15 10 Dry mass Tg

5 10

Pine Spruce Deciduous Pine Spruce Deciduous 0 5

2000-2009 2010-2019 2020-2029 2030-2039 2040-2049 2070-2079 2080-2089 2090-2099 2100-2109 2050-2059 2060-2069 2000-2009 2010-2019 2020-2029 2030-2039 2060-2069 2100-2109 2040-2049 2050-2059 2070-2079 2080-2089 2090-2099

Ye a r Ye a r

Fig. 5 – Average annual biomass growth by species for each 10-year period for the different scenarios for Ja¨mtland and

Va¨sternorrland.

pared to the Reference scenario. All scenarios have greater with coal reference fuel. However, Environment scenario has

avoided carbon emissions in the whole tree use option 16 Tg C less avoided carbon emission than in the Reference

compared to the stemwood only use option. scenario for whole tree biomass use with coal reference fuel.

Cumulative avoided net carbon emission during the 100- The lower net avoided carbon in the Environment scenario is

year study period is shown in Fig. 7. Biomass substitution is the because of the reduced amount of biomass harvest and

largest single contributor to the avoided carbon emissions. substitution, which outweighs the increased carbon stock in

Total avoided emissions in the Production and Maximum standing biomass.

scenario are 68 Tg C and 132 Tg C greater than in the Reference The differences in cumulative avoided carbon emissions

scenario during the 100-year period for whole tree biomass use between the Reference scenario and Environment, Production and

Maximum scenarios are shown in Fig. 8. The difference is

largest between the Maximum and Reference scenario with

whole tree biomass use and with coal as reference fossil fuel.

Coal reference fuel gives greater avoided carbon emissions

difference than natural gas reference fuel for the Production

and Maximum scenarios, but lower avoided carbon emissions

difference for the Environment scenario because the biomass

harvest and substitution is lower in the Environment scenario

than in the Reference scenario.

5. Discussion and conclusions

Our result shows that intensive forest management can

signiﬁcantly increase biomass production in north-central

Sweden during the next 100 years. The increase in annual

Fig. 6 – Standing forest biomass at the beginning of the whole tree biomass production over the 100 years from 2010 to

study period (2010) and at the end of study period (2109) in 2109 is 4% and 26% greater in the Production and Maximum

the different scenarios for Ja¨mtland and Va¨sternorrland. scenarios, respectively, compared to the increase in the 116

S Table 5 – Changes in average annual soil carbon stock during each 10-year period (Tg C year 1 ) for Ja¨mtland and Va¨sternorrland.

2010–2019 2020–2029 2030–2039 2040–2049 2050–2059 2060–2069 2070–2079 2080–2089 2090–2099 2100–2109 e

Reference scenario i

Stemwood 0.36 0.12 0.07 0.05 0.06 0.05 0.04 0.06 0.11 0.10 o

biomass use m

Whole tree 0.14 0.03 0.01 0.01 0.02 0.02 0.02 0.04 0.07 0.06 e

biomass use t

Environment scenario s

Stemwood 0.34 0.11 0.07 0.05 0.05 0.04 0.04 0.08 0.10 0.08 i

biomass use n

Whole tree 0.13 0.04 0.02 0.02 0.02 0.02 0.03 0.04 0.06 0.05 e

biomass use &

Production scenario o

Stemwood 0.37 0.13 0.08 0.07 0.08 0.07 0.05 0.07 0.11 0.09 c

biomass use 1

Whole tree 0.14 0.04 0.03 0.04 0.06 0.06 0.05 0.05 0.07 0.06 5

(

biomass use 2

Maximum scenario 2

) Stemwood 0.38 0.14 0.08 0.06 0.08 0.07 0.06 0.09 0.13 0.11 1

biomass use 6

–

Whole tree 0.14 0.04 0.03 0.06 0.07 0.08 0.08 0.09 0.11 0.11 1

biomass use 4

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 5 ( 2 0 1 2 ) 1 0 6 – 1 2 4 117

Table 6 – Changes in average annual wood product carbon stock (Tg C year ) during each 10-year period for Ja¨mtland and

Va¨sternorrland.

2010–2019 2020–2029 2030–2039 2040–2049 2050–2059 2060–2069 2070–2079 2080–2089 2090–2099 2100–2109

Reference 0.79 0.75 0.64 0.66 0.74 0.77 0.86 0.86 0.96 1.06

scenario

Environment 0.73 0.71 0.60 0.62 0.70 0.72 0.82 0.86 0.86 0.96

scenario

Production 0.78 0.76 0.67 0.70 0.81 0.88 1.01 1.03 1.13 1.24

scenario

Maximum 0.79 0.76 0.69 0.75 0.83 0.88 1.05 1.10 1.18 1.35 scenario

Table 7 – Average annual avoided carbon emission (Tg C year ) due to forest biomass substitution during each 10-year

period for Ja¨mtland and Va¨sternorrland. 2010– 2020– 2030– 2040– 2050– 2060– 2070– 2080– 2090– 2100– 2019 2029 2039 2049 2059 2069 2079 2089 2099 2109

Reference scenario

Net biomass substitution: coal

Stemwood 3.8 3.8 3.8 3.9 4.2 4.4 4.7 4.7 5.0 5.5

biomass use

Whole tree 4.8 4.8 4.9 5.0 5.4 5.6 5.9 5.9 6.3 6.9

biomass use

Net biomass substitution: fossil gas

Stemwood 2.8 2.8 2.7 2.8 3.0 3.1 3.4 3.4 3.7 4.0

biomass use

Whole tree 3.4 3.3 3.2 3.3 3.6 3.7 4.0 4.0 4.3 4.7

biomass use

Environment scenario

Net biomass substitution: coal

Stemwood 3.5 3.5 3.6 3.7 4.0 4.1 4.3 4.4 4.7 5.0

biomass use

Whole tree 4.5 4.5 4.6 4.7 5.0 5.2 5.4 5.5 5.9 6.2

biomass use

Net biomass substitution: fossil gas

Stemwood 2.6 2.6 2.5 2.6 2.8 2.9 3.1 3.2 3.4 3.6

biomass use

Whole tree 3.1 3.1 3.1 3.1 3.4 3.5 3.7 3.8 4.0 4.3

biomass use

Production scenario

Net biomass substitution: coal

Stemwood 3.8 3.8 3.9 4.1 4.5 4.9 5.2 5.3 5.7 6.1

biomass use

Whole tree 4.8 4.8 5.1 5.3 5.8 6.2 6.6 6.6 7.1 7.6

biomass use

Net biomass substitution: fossil gas

Stemwood 2.8 2.8 2.8 2.9 3.3 3.5 3.8 3.9 4.2 4.5

biomass use

Whole tree 3.3 3.3 3.4 3.5 3.9 4.2 4.5 4.6 4.9 5.3

biomass use

Maximum scenario

Net biomass substitution: coal

Stemwood 3.9 3.8 4.0 4.3 4.6 4.9 5.4 5.6 5.9 6.6

biomass use

Whole tree 4.9 4.9 5.2 5.5 5.9 6.2 6.8 7.0 7.4 8.2

biomass use

Net biomass substitution: fossil gas

Stemwood 2.9 2.8 2.9 3.0 3.3 3.5 4.0 4.1 4.3 4.9

biomass use

Whole tree 3.4 3.3 3.5 3.7 4.0 4.2 4.6 4.8 5.1 5.7

biomass use 118

S Table 8 – Total annual average carbon stock increases and avoided fossil emissions (Tg C year 1 ) during each 10-year period for Ja¨mtland and Va¨sternorrland. 2010–2019 2020–2029 2030–2039 2040–2049 2050–2059 2060–2069 2070–2079 2080–2089 2090–2099 2100–2109 Reference scenario Total (coal ref) e

Stemwood biomass use 5.5 5.5 4.8 5.4 5.6 5.7 6.0 6.2 7.1 7.4 n

Whole tree biomass use 6.2 6.4 5.9 6.5 6.8 6.9 7.2 7.4 8.3 8.8 i

Total (Fossil gas ref) o

Stemwood biomass use 4.5 4.5 3.7 4.3 4.4 4.5 4.7 4.9 5.7 5.9 n

Whole tree biomass use 5.0 5.0 4.3 4.8 5.0 5.1 5.4 5.5 6.3 6.6 e

Environment scenario a

Total (coal ref) s

Stemwood biomass use 5.4 5.6 5.0 5.4 5.6 5.6 6.1 6.3 6.4 7.0 c

Whole tree biomass use 6.1 6.5 6.0 6.4 6.7 6.7 7.1 7.4 7.6 8.2 n

Total (fossil gas ref) c

Stemwood biomass use 4.5 4.7 3.9 4.4 4.5 4.5 4.9 5.2 5.1 5.6 &

Whole tree biomass use 5.0 5.2 4.5 4.9 5.0 5.0 5.5 5.7 5.7 6.3 p

Production scenario c

Total (coal ref) 1

Stemwood biomass use 5.6 5.8 5.5 6.1 6.5 6.4 6.8 7.0 7.6 8.0 5

(

Whole tree biomass use 6.3 6.7 6.6 7.3 7.7 7.7 8.1 8.3 8.9 9.5 2

Total (fossil gas ref) 1

Stemwood biomass use 4.6 4.8 4.4 4.9 5.2 5.1 5.4 5.6 6.1 6.4 )

Whole tree biomass use 5.1 5.3 4.9 5.5 5.9 5.7 6.0 6.3 6.8 7.2 0

Maximum scenario –

Total (coal ref) 2

4 Stemwood biomass use 5.6 5.7 5.5 6.6 7.0 6.9 7.6 8.1 8.8 9.3 Whole tree biomass use 6.3 6.7 6.7 7.8 8.2 8.3 9.0 9.5 10.2 10.9 Total (fossil gas ref) Stemwood biomass use 4.6 4.7 4.4 5.4 5.7 5.6 6.1 6.6 7.2 7.6 Whole tree biomass use 5.1 5.2 5.0 6.0 6.3 6.2 6.8 7.3 8.0 8.5

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 5 ( 2 0 1 2 ) 1 0 6 – 1 2 4 119

Fig. 7 – Cumulative avoided carbon emission (Tg C) for each 10-year period for the different scenarios, assuming whole tree

biomass use and coal as reference fuel.

Reference scenario. These increases are the result of intensive material. In addition, increased temperature and atmospheric

practices that includes fertilization, species changes from CO2 fertilization is another major contributor to forest

comparatively slow growing Scots pine to fast growing production (Nadelhoffer et al., 1999), and all scenarios include

Lodgepole pine, application of balanced nutrient supply in the effect of temperature increase in forest growth. The

young forest stands and use of genetically improved planting Environment scenario has 9% lower increase in annual biomass

140 140 Whole tree biomass use Env minus Ref Coal fuel Env minus Ref Coal fuel Stem wood biomass use 120 120 Env minus Ref Fossil gas Env minus Ref Fossil gas 100 Prod minus Ref Coal fuel 100 Prod minus Ref Coal fuel 80 Prod minus Ref Fossil gas 80 Prod minus Ref Fossil gas Max minus Ref Coal fuel Max minus Ref Coal fuel 60 60 Max minus Ref Fossil gas Max minus Ref Fossil gas 40 40

20 20 Avoided C emissions (Tg) 0 0

-20 -20

2030-2039 2070-2079 2030-2039 2070-2079 2000-2009 2010-2019 2020-2029 2040-2049 2050-2059 2060-2069 2080-2089 2090-2099 2100-2109 2000-2009 2010-2019 2020-2029 2040-2049 2050-2059 2060-2069 2080-2089 2090-2099 2100-2109

Year Year

Fig. 8 – Differences in cumulative avoided carbon emission (Tg C) between the Environment, Production and Maximum

scenarios and the Reference scenario for each 10-year period for Ja¨mtland and Va¨sternorrland.

120 e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 5 ( 2 0 1 2 ) 1 0 6 – 1 2 4

production compared to the Reference scenario. The standing assumed that the buildings produced with the forest biomass

biomass in Environment scenario is likely to be overestimated will have a 100-year life span, thus the carbon stock in building

because the HUGIN model imperfectly models mortality in materials increases continuously during the 100-year study

overmature trees, thus the difference may actually be greater period because no wood products were yet assumed to be

than 9%. removed from service. At the end of this period, however,

Our estimates of increased biomass production are buildings will begin to be demolished and the annual carbon

greater than the estimates of Skogsstyrelsen (2008), Pussinen stock change in wood products will be the difference between

et al. (2009), and Poudel et al. (2011). The reason is that those new wood products entering service and old wood products

studies only considered the effects of climate change on leaving service. Post-use wood products can be a signiﬁcant

forest growth, while our results also include changes in source of bioenergy (Dodoo et al., 2009; Gustavsson and Sathre,

forest management regimes. The 49% increase in biomass 2006), which would also appear in our carbon balance

production in Reference scenario even without changes in calculations if the time period were longer.

management practices is likely due to higher quality In this analysis we have not considered the economic

planting material, inclusion of marginal lands in the forested feasibility of the modeled management practices, and have

area, and that forests that were planted in the past have not included demand analysis of additional forest products.

better production capacity, in addition to the effects of This work does not include any economical analysis to allow

1 1

climate change. The average increases of 3.0 Mg ha year the recommendation of management practices in terms of

1 1

and 2.8 Mg ha year in dry mass production for Maximum economic gain. Furthermore, the adoption of new forest

and Production scenario for whole tree biomass were smaller management practices is complex and has not been studied

1 1

than the estimate of 5–6 Mg ha year by Bergh et al. (2005) here. However, there is a trend towards increased bioenergy

for northern Swedish forests. This is because Bergh et al. use in Sweden (Energimyndigheten, 2010) and a strong

(2005) showed the potential production on a stand level, interest in increasing wood use in Swedish construction

while the present analysis is on a landscape level. The (Na¨ringsdepartementet, 2004). The potential contributions of

Environment scenario results in an average increase of forest resources to energy security and climate change

1 1

2.5 Mg ha year dry mass, which is less than the average mitigation suggest that future demand for forest products

1 1

increase of 2.6 Mg ha year in the Reference scenario. will rise. Furthermore, the use of bioenergy is increasing

The whole tree harvest can contribute 38% more cumulative globally, and international trade in bioenergy is made feasible

usable biomass harvest compared to harvesting only stem- by efﬁcient long-distance transport methods (Junginger et al.,

wood, in all scenarios. The whole tree biomass calculation for 2008). Gustavsson et al. (2011) calculated that woody biofuels

deciduous species might be slightly overestimated, since stump can be economically transported internationally, and the GHG

biomass calculation for deciduous species in the HUGIN model balance depends largely on the fossil fuel replaced by the

is based on spruce stumps biomass (Petersson and Sta˚hl, 2006; bioenergy, and much less on the transport distance. The

Poudel et al., 2011). Nonetheless, the harvest of slash and avoided fossil CO2 emissions for coal substitution are greater

stumps can signiﬁcantly increase the amount of bioenergy, as compared to fossil gas substitution, due to coal’s greater

well as help to prepare the regeneration area. It has been carbon intensity. By exporting biomass to be used in

suggested that logging residues removal could affect the forest applications that result in high CO2 emissions reductions

soil nutrients (Smolander et al., 2010) and could in some cases per unit of biomass, the total emissions reduction from the

reduce growth to a minor extent (Jacobson et al., 2000). In our available supply of biomass could by increased. Similarly,

calculations these potential minor growth reductions are not wood products can be exported to substitute for non-wood

considered. Our results show that the soil carbon content products in other countries. For example, the total number of

continues to increase if whole tree biomass is removed, but at a new buildings built per year in Sweden is small in relation to

slower rate than if only stemwood is removed. the total quantities of biomass potentially available. If the

The rate of carbon stock increase in living tree biomass in export potential were ignored, the additional biomass would

the Reference scenario ﬂuctuates during the study period, and be used for other uses with lower efﬁciency of emission

in the Production and Environment scenarios the rate of increase reduction. However, if additional biomass is exported and

declines after the middle of the study period. The ﬂuctuation used instead of non-wood buildings in other countries, the

and decline of increase may be due to harvest intensity in the higher emission reduction per unit of biomass could be gained

previous years. Ma¨kipa¨a¨ (2006) suggests that this phenomenon by a larger share of the biomass, thus resulting in a greater

may be due to forest ﬁres, wind damages and harvesting. Since overall emission reduction globally.

forest ﬁre impact is negligible in Sweden and wind damages There are signiﬁcant uncertainties in our analysis. The

are included in the NFI, a reasonable argument would be the degree of temperature change in the future is not known with

increased harvesting intensity. Overall, Reference and Produc- certainty, and will depend on many factors including

tion scenarios have lower rates of carbon stock increase, while economic growth, population, and climate mitigation activi-

Environment and Maximum scenarios have higher rates. ties that are implemented. We have used the SRES B2 scenario

Carbon stock in wood products increases moderately in the of temperature increase requiring signiﬁcant mitigation

Reference (35%) and Environment scenarios (32%), and more efforts. Regional temperatures may rise by more than we

signiﬁcantly in the Production (59%) and Maximum (71%) modeled, and the forest biomass production may further

scenarios. The reason for the differences in carbon stock for increase. A number of factors that could not be taken into

different scenarios is the greater quantities of stemwood account might give different results in NPP estimations. Earlier

converted into long-lived wood building materials. We have onset of bud-burst in spring, as an effect of elevated

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 5 ( 2 0 1 2 ) 1 0 6 – 1 2 4 121

temperature, might increase the risk of frost injury (Cannell, indeﬁnitely, because the biomass stocks will eventually

1985; Cannell and Smith, 1986; Ha¨nninen, 1991; Kramer, 1994). become ‘‘saturated’’ at an equilibrium level, and may release

The increased risk of frost damage, especially in plantations their stored carbon due to natural disturbances. Instead, a

where consecutive years of frost injury might cause plant more viable long-term strategy is to avoid CO2 emissions from

mortality (Krasowski et al., 1995), and its effects on NPP is not fossil fuels and energy-intensive materials, while forest

included. In a soil-warming experiment in northern Sweden landscapes maintain a dynamic equilibrium in the process

the stemwood production after six seasons was 100% greater of cyclical carbon uptake and release. Furthermore, the

for the heated plots compared with unheated (Stro¨ mgren and biomass fractions that are typically left in the forest also

Linder, 2002). The increased demand of nutrients for increased decompose over time and release their stored carbon into the

biomass production must be met by increased mineralization, atmosphere. Instead of allowing forest biomass to decay

nutrient availability, and uptake by roots; otherwise the naturally, while at the same time using fossil fuels and carbon

growth response will stagnate at a lower level (Bonan and intensive materials to provide services to the society, using

Cleve, 1991; Houghton et al., 1998; McMurtrie et al., 2001; forest biomass to provide those services will result in less net

Melillo et al., 1993). In Reference and Environment scenarios climate impact (Sathre and Gustavsson, 2011). Although

traditional fertilization is assumed to supply additional forecasts of future events are inherently uncertain, the

nutrients, whereas in the other two scenarios balanced general conclusions of this study appear to be robust.

fertilization is provided to assist soil nutrient dynamics. The Intensive forest management and utilization of forest

changes in the amount of fertilization and the availability of biomass can play an important role in transitioning to a

water resources during growth periods might bring slightly carbon-neutral economy.

different values of NPP. Increased mineralization and nutrient

availability as an effect of increased soil temperature have

been found in earlier studies (Cleve et al., 1990; Jarvis and Acknowledgements

Linder, 2000; Lu¨ ckewille and Wright, 1997; Peterjohn et al.,

1994). A study in Oregon in the northwest USA that involved We gratefully acknowledge the support of the European

stump harvest found that the amount of soil N and C is likely Union, County Administrative Board of Ja¨mtland, Sveaskog

to reduce (Zabowski et al., 2008). In the present study, the AB, SCA Forest Products, Norrskog/SA˚ TAB, and Ja¨mtkraft AB.

assumed 50% stump harvest after ﬁnal felling may decrease This analysis was partly done in collaboration with the project

the amount of soil N and C, but remains uncertain. ‘‘Future Forests’’ at the Faculty of Forestry of the Swedish

Climate change may result in wind damages, ﬁre risk, University of Agricultural Sciences, Umea˚. We also thank the

pathogens, and insect outbreaks, which may affect forest anonymous reviewers of this paper.

growth and mortality (Battles et al., 2008; Blennow et al., 2010;

Kurz et al., 2008; Lindner et al., 2010; Netherer and Schopf,

r e f e r e n c e s

2010). Such risks have not been considered in this study, which

could be substantial in a Swedish context. A Swedish

governmental survey (SOU, 2007) pointed out these uncer-

A˚ gren, G.I., Wetterstedt, J.A˚ .M., 2007. What determines the

tainties and risks as possible threats to future forestry in

temperature response of soil organic matter decomposition?

Sweden. These new threats do not directly cause growth

Soil Biol. Biochem. 39, 1794–1798.

reduction, since in most cases they only affect vitality and Battles, J., Robards, T., Das, A., Waring, K., Gilless, J., Biging, G.,

wood quality in mature stands which can be harvested if et al., 2008. Climate change impacts on forest growth and

needed. Some problems can also be avoided by changed tree mortality: a data-driven modeling study in the mixed-

conifer forest of the Sierra Nevada, California. Climatic

silviculture methods and preventative measures against

Change 87, 193–213.

major outbreaks of insects and pathogens.

Berg, S., Lindholm, E.-L., 2005. Energy use and environmental

As responses of climatic change on forest productivity are

impacts of forest operations in Sweden. J. Clean Prod. 13,

of great interest, models are often used as tools to predict 33–42.

responses of vegetation to environmental change. A number Bergh, J., Linder, S., Bergstro¨ m, J., 2005. Potential production of

of models are available at patch and regional scale (Landsberg Norway spruce in Sweden. For. Ecol. Manage. 204, 1–10.

et al., 2003) and studies of climatic change on forest Blennow, K., Andersson, M., Sallna¨s, O., Olofsson, E., 2010.

Climate change and the probability of wind damage in two

productivity on a regional scale has been done earlier (Coops

Swedish forests. For. Ecol. Manage. 259, 818–830.

and Waring, 2001; Koca et al., 2006). Many of the models,

Bonan, B.G., Cleve, K.v., 1991. Soil temperature, nitrogen

however, are not suitable for predicting forest productivity in

mineralisation, and carbon source–sink relationships in

boreal and cold-temperate environments, since they do not

boreal forests. Can. J. For. Res. 2, 2.

include essential low-temperature effects on physiological Bosatta, E., A˚ gren, G.I., 1996. Theoretical analyses of carbon and

processes. This work presents estimates of NPP in boreal nutrient dynamics in soil proﬁles. Soil Biol. Biochem. 28,

forests, where low-temperature effects have been included 1523–1531.

Bricen˜ o-Elizondo, E., Garcia-Gonzalo, J., Peltola, H., Kelloma¨ki,

and the response of elevated temperature is predicted in a

S., 2006. Carbon stocks and timber yield in two boreal forest

more robust way.

ecosystems under current and changing climatic conditions

This study documents the cumulative carbon emission

subjected to varying management regimes. Environ. Sci.

reduction made possible by using forest biomass to substitute Policy 9, 237–252.

for fossil fuels and non-wood materials. Accumulating carbon Cannell, M.G.J., 1985. Analysis of risks of frost damage to forest

in living biomass will not help to mitigate climate change trees in Britain. In: Tigerstedt, P.M.A., Puttonen, P., Koski, V.

122 e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 5 ( 2 0 1 2 ) 1 0 6 – 1 2 4

(Eds.), Crop Physiology of Forest Trees. Helsinki University IPCC, 2007a. Climate Change 2007—Mitigation of Climate

Press, Helsinki, pp. 153–166. Change. Working group III contribution to the fourth

Cannell, M.G.R., Smith, R.I., 1986. Climatic warming, spring assessment report of the Intergovernmental Panel on

budburst and forest damage on trees. J. Appl. Ecol. 23, Climate Change (IPCC) Cambridge University Press,

177–191. Cambridge, United Kingdom, p. 862.

Carter, T.R., Jylha¨, K., Perrels, A., Fronzek, S., Kankaanpa¨a¨, S., IPCC, 2007b. In: Solomon, S., Qin, D., Manning, M. (Eds.), Climate

2005. FINADAPT Scenarios for the 21st Century: Alternative Change 2007—The Physical Science Basis. Working Group I

Futures for Considering Adaptation to Climate Change in Contribution to the Fourth Assessment Report of the

Finland. Finnish Environment Institute, Helsinki, p. 42. Intergovernmental Panel on Climate Change (IPCC).

Cleve, K.V., Oechel, W.C., Hom, J.L., 1990. Response of black Cambridge University Press, Cambridge, United Kingdom, p.

spruce (Piceamariana) ecosystems to soil temperature 1009.

modiﬁcation in interior Alaska. Can. J. For. Res. 20, Jacobson, S., Kukkola, M., Ma¨lko¨ nen, E., Tveite, B., 2000. Impact

1530–1535. of whole-tree harvesting and compensatory fertilization on

Coops, N.C., Waring, R.H., 2001. Assessing forest growth across growth of coniferous thinning stands. For. Ecol. Manage. 129,

southwestern Oregon under a range of current and future 41–51.

global change scenarios using a process model, 3-PG. Global Jarvis, P.G., Linder, S., 2000. Constraints to growth of boreal

Change Biol. 7, 15–29. forests. Nature 405, 904–905.

Davis, J., Haglund, C., 1999. Life Cycle Inventory (LCI) of Fertiliser Junginger, M., Bolkesjø, T., Bradley, D., Dolzan, P., Faaij, A.,

Production—Fertiliser Products Used in Sweden and Heinimo¨, J., et al., 2008. Developments in international

Western Europe. The Swedish Institute for Food and bioenergy trade. Biomass Bioenergy 32, 717–729.

Biotechnology (SIK), Go¨ teborg (SIK Report no. 654). Keeling, R.F., Piper, S.C., Bollenbacher, A.F., Walker, J.S., 2009.

Dodoo, A., Gustavsson, L., Sathre, R., 2009. Carbon implications Atmospheric CO2 records from sites in the SIO air sampling

of end-of-life management of building materials. Resour. network. In: Trends: A Compendium of Data on Global

Conserv. Recycl. 53, 276–286. Change, Carbon Dioxide Information Analysis Center, Oak

Eggers, J., Lindner, M., Zudin, S., Zaehle, S., Liski, J., 2008. Impact Ridge National Laboratory, U.S. Department of Energy,

of changing wood demand, climate and land use on OakRidge, TN, USA Available from: http://cdiac.ornl.gov/

European forest resources and carbon stocks during the 21st trends/co2/sio-mlo.html (accessed: March 2009).

century. Global Change Biol. 14, 2288–2303. Kirilenko, A.P., Sedjo, R.A., 2007. Climate change impacts on

Energimyndigheten, 2010. Energy in Sweden 2010. Swedish forestry. Proc. Natl. Acad. Sci. U.S.A. 104, 19697–19702.

Energy Agency, Eskilstuna, p. 142. Kirschbaum, M.U.F., 2000. Will changes in soil organic carbon

Eriksson, E., Gillespie, A.R., Gustavsson, L., Langvall, O., Olsson, act as a positive or negative feedback on global warming?

M., Sathre, R., et al., 2007. Integrated carbon analysis of Biogeochemistry 48, 21–51.

forest management practices and wood substitution. Can. J. Kjellstro¨m, E., Ba¨rring, L., Gollvik, S., Hansson, U., Jones, C.,

For. Res. 37, 671–681. Samuelsson, P., et al., 2006. A 140-year simulation of

Gustavsson, L., Bo¨ rjesson, P., Johansson, B., Svenningsson, P., European climate with the new version of the Rossby Centre

1995. Reducing CO2 emissions by substituting biomass for regional atmospheric climate model (RCA3), SMHI Reports in

fossil fuels. Energy 20, 1097–1113. Meteorology and Climatology, No. 108. SMHI, Norrko¨ ping,

Gustavsson, L., Eriksson, L., Sathre, R., 2011. Costs and CO2 Sverige, p. 54.

beneﬁts of recovering, reﬁning and transporting logging Koca, D., Smith, B., Sykes, M., 2006. Modelling regional climate

residues for fossil fuel replacement. Appl. Energy 88, change effects on potential natural ecosystems in Sweden.

192–197. Climatic Change 78, 381–406.

Gustavsson, L., Madlener, R., Hoen, H.F., Jungmeier, G., Kramer, K., 1994. A modelling analysis of the effects of climatic

Karjalainen, T., KlO¨ hn, S., et al., 2006a. The role of wood warming on the probability of spring frost damage to tree

material for greenhouse gas mitigation. Mitig. Adapt. Strat. species in The Netherlands and Germany. Plant Cell Environ.

Glob. Change 11, 1097–1127. 17, 367–377.

Gustavsson, L., Pingoud, K., Sathre, R., 2006b. Carbon dioxide Krasowski, M.J., Letchford, T., Caputa, A., Bergerud, W.A., 1995.

balance of wood substitution: comparing concrete- and Desiccation of white spruce seedlings planted in the

wood-framed buildings. Mitig. Adapt. Strat. Glob. Change 11, southern boreal forest of British Columbia. Water Air Soil

667–691. Pollut. 82, 133–146.

Gustavsson, L., Sathre, R., 2006. Variability in energy and carbon Kurz, W.A., Dymond, C.C., Stinson, G., Rampley, G.J., Neilson,

dioxide balances of wood and concrete building materials. E.T., Carroll, A.L., et al., 2008. Mountain pine beetle and

Build. Environ. 41, 940–951. forest carbon feedback to climate change. Nature 452,

Ha¨ nninen, H., 1991. Does climatic warming increase the risk 987–990.

of frost damage in northern trees? Plant Cell Environ. 14, Landsberg, J.J., Waring, R.H., Coops, N.C., 2003. Performance of

449–454. the forest productivity model 3-PG applied to a wide range of

Houghton, J.T., Ding, Y., Griggs, D.J., Noguer, M., Linden, P.J.v.d., forest types. For. Ecol. Manage. 172, 199–214.

Dai, X., et al., 2001. The Scientiﬁc Basis: Contribution of Levy, P.E., Cannell, M.G.R., Friend, A.D., 2004. Modelling the

Working Group I to the Third Assessment Report of the impact of future changes in climate, CO2 concentration and

Intergovernmental Panel on Climate Change (IPCC). land use on natural ecosystems and the terrestrial carbon

Cambridge University Press, Cambridge, United Kingdom, p. sink. Global Environ. Chang. 14, 21–30.

881. Lindner, M., Maroschek, M., Netherer, S., Kremer, A., Barbati, A.,

Houghton, R.A., Davidsson, E.A., Woodwell, G.M., 1998. Missing Garcia-Gonzalo, J., et al., 2010. Climate change impacts,

sinks, feedbacks, and understanding the role of terrestrial adaptive capacity, and vulnerability of European forest

ecosystems in the global carbon balance. Global ecosystems. For. Ecol. Manage. 259, 698–709.

Biogeochem. Cycles 12, 25–34. Lu¨ ckewille, A., Wright, R.F., 1997. Experimentally increased soil

Hyvo¨ nen, R., A˚ gren, G.I., 2001. Decomposer invasion rate, temperature causes release of nitrogen at a boreal forest

decomposer growth rate, and substrate chemical quality: catchment in southern Norway. Global Change Biol. 3, 13–21.

how they inﬂuence soil organic matter turnover. Can. J. For. Lundstro¨ m, A., So¨ derberg, U., 1996. Outline of the Hugin system

Res. 31, 1594–1601. for longterm forecasts of timber yields and possible cut.

e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 5 ( 2 0 1 2 ) 1 0 6 – 1 2 4 123

Large-scale forestry scenario models: experiences and experimentally elevated soil temperatures. Ecol. Appl. 4,

requirements. EFI Proc. 63–77. 617–625.

MacDonald, J.A., Skiba, U., Sheppard, L.J., Ball, B., Roberts, J.D., Petersson, H., Sta˚hl, G., 2006. Functions for below-ground

Smith, K.A., et al., 1997. The effect of nitrogen deposition biomass of Pinus sylvestris, Picea abies, Betula pendula and

and seasonal variability on methane oxidation and nitrous Betula pubescens in Sweden. Scand. J. Forest Res. 21, 84–93.

oxide emission rates in an upland spruce plantation and Poudel, B.C., Sathre, R., Gustavsson, L., Bergh, J., Lundstro¨ m, A.,

moorland. Atmos. Environ. 31, 3693–3706. Hyvo¨ nen, R., 2011. Effects of climate change on biomass

Ma¨kipa¨a¨, R., 2006. Integrated method to estimate the carbon production and substitution in north-central Sweden.

budget of forest—nation-wide estimates obtained by Biomass Bioenerg 35, 4340–4355.

combining forest inventory data with biomass expansion Pussinen, A., Karjalainen, T., Ma¨kipa¨a¨, R., Valsta, L., Kelloma¨ki,

factors, biomass turnover rates and dynamic soil C model. S., 2002. Forest carbon sequestration and harvests in Scots

In: Pilli, R., Anfodillo, T., Dalla, E. (Eds.), Valle stima del pine stand under different climate and nitrogen deposition

carbonio in foresta: metodologie ed aspetti normativi. Corso scenarios. For. Ecol. Manage. 158, 103–115.

di Cultura in Ecologia, Atti del 428 corso. Universita` di Pussinen, A., Nabuurs, G.J., Wieggers, H.J.J., Reinds, G.J.,

Padova, pp. 35–53. Wamelink, G.W.W., Kros, J., et al., 2009. Modelling long-term

Matthews, H.D., Caldeira, K., 2008. Stabilizing climate requires impacts of environmental change on mid- and high-latitude

near-zero emissions. Geophys. Res. Lett. 35, L04705. European forests and options for adaptive forest

McMurtrie, R.E., Medlyn, B.E., Dewer, R.C., 2001. Increased management. For. Ecol. Manage. 258, 1806–1813.

understanding of nutrient immobilisation in soil organic Quadrelli, R., Peterson, S., 2007. The energy-climate challenge:

matter is critical for predicting the carbon sink strength of recent trends in CO2 emissions from fuel combustion.

forest ecosystem over the next 100 years. Tree Physiol. 21, Energy Policy 35, 5938–5952.

831–839. Sathre, R., Gustavsson, L., 2009. Using wood products to mitigate

McMurtrie, R.E., Rook, D.A., Kelliher, F.M., 1990. Modelling the climate change: external costs and structural change. Appl.

yield of Pinus radiata on a site limited by water and nitrogen. Energy 86, 251–257.

For. Ecol. Manage. 30, 381–413. Sathre, R., Gustavsson, L., 2011. Time-dependent climate

Mead, D.J., Pimentel, D., 2006. Use of energy analyses in beneﬁts of using forest residues to substitute fossil fuels.

silvicultural decision-making. Biomass Bioenergy 30, Biomass Bioenergy 35, 2506–2516.

357–362. Sathre, R., Gustavsson, L., Bergh, J., 2010. Primary energy and

Melillo, J.M., McGuire, A.D., Kicklighter, D.W., Moore, B., greenhouse gas implications of increasing biomass

Vorosmarty, C.L., Schloss, A.L., 1993. Global climate production through forest fertilization. Biomass Bioenergy

change and terrestrial net primary production. Nature 363, 34, 572–581.

234–240. Sathre, R., O‘Connor, J., 2010. Meta-analysis of greenhouse gas

Nadelhoffer, K.J., Emmett, B.A., Gundersen, P., Kjønaas, O.J., displacement factors of wood product substitution. Environ.

Koopmans, C.J., Schleppi, P., et al., 1999. Nitrogen deposition Sci. Policy 13, 104–114.

makes a minor contribution to carbon sequestration in Schlamadinger, B., Apps, M., Bohlin, F., Gustavsson, L.,

temperate forests. Nature 398, 145–148. Jungmeier, G., Marland, G., et al., 1997. Towards a standard

Nakicenovic, N., Swart, R., 2000. Emissions Scenarios Special methodology for greenhouse gas balances of bioenergy

Report of Working Group III of the Intergovernmental Panel systems in comparison with fossil energy systems. Biomass

on Climate Change (IPCC). Cambridge University Press, Bioenergy 13, 359–375.

Cambridge, United Kingdom, p. 612. Skogsstyrelsen, 1993. Swedish Forestry Act (Skogsstyrelsens

Na¨ringsdepartementet, 2004. Mer tra¨ i byggandet: underlag fo¨ r fo¨ reskrifter och allma¨nna ra˚d till skogsva˚rdslagen).

en nationell strategi att fra¨mja anva¨ndning av tra¨ i Skogssterelsen, Jo¨nko¨ ping.

byggandet. (More Wood in Construction: Basis for a National Skogsstyrelsen, 2008. Skogliga konsekvensanalyser—SKA-VB 08

Strategy to Promote the Use of Wood in [Forest impact analyses—SKA-VB 08], Swedish Forest

Construction).Ministry of Enterprise, Energy, and Agency Report 25/2008. Skogsstyrelsen, Jo¨ nko¨ ping, Sweden,

Communication[[nl]]Na¨ringsdepartementet, Stockholm, p. p. 146.

191. Skogsstyrelsen, 2010. Forestry Statistics. Swedish Forest

Naturva˚rdsverket, 2008. Miljo¨ ma˚len–nu a¨r det bra˚ttom! Agency. Skogsstyrelsen, Jo¨ nko¨ ping Available from: http://

Miljo¨ ma˚lsra˚dets utva¨rdering av Sveriges miljo¨ma˚l (The www.svo.se/episerver4/default.aspx?id=38515 (accessed:

Environmental Objectives—Now it’s in a Hurry! September 2011).

Environmental Objectives Council’s assessment of Sweden’s SMHI, 2009. Climatic Data. Swedish Meteorological and

Environmental Objectives).Swedish Environmental Hydrological Institute Available from: http://www.smhi.se/

Protection Agency, Stockholm, p. 356. cmp/jsp/polopoly.jsp?d=8785&l=sv (accessed: August 2009).

Naturva˚rdsverket, Skogsstyrelsen, 2005. Nationell strategi fo¨ r Smolander, A., Kitunen, V., Tamminen, P., Kukkola, M., 2010.

formellt skydd av skog. (National Strategy for Formal Removal of logging residue in Norway spruce thinning

Protection of Forests).Environmental Protection Agency and stands: long-term changes in organic layer properties. Soil

Swedish Board of Forestry, Stockholm, Jo¨ nko¨ ping, p. 95. Biol. Biochem. 42, 1222–1228.

Netherer, S., Schopf, A., 2010. Potential effects of climate change SOU, 2007. Sverige info¨ r klimatfo¨ ra¨ndringarna—hot och

on insect herbivores in European forests—general aspects mo¨jligheter. Slutbeta¨nkande av Klimatoch

and the pine processionary moth as speciﬁc example. For. sa˚rbarhetsutredningen. (Sweden facing climate change—

Ecol. Manage. 259, 831–838. threats and opportunities. Final Report of Klimatoch

Nordin, A., Lundmark, T., Grip, H., Nilsson, M., Ericson, L., 2009. Vulnerability).In: Statens offentliga utredningar, Ministry of

Miljo¨ analys av behovsanpassad go¨ dsling pa˚ skogmark Environment, Stockholm, p. 35.

[Environmental analysis of demand-fertilization on forest Stro¨ mgren, M., Linder, S., 2002. Effects of nutrition and soil

land], Final Report to the Energy Agency P 30639-1. Swedish warming on stem wood production in a boreal Norway

University of Agricultural Sciences Report, Umea˚ (in spruce stand. Global Change Biol. 8, 1194–1204.

Swedish). Sveriges-Nationalatlas, 1995. Sveriges Nationalatlas: Klimat,

Peterjohn, W.T., Melillo, J.M., Steudler, S.A., Newkirk, K.M., 1994. Sjo¨ ar och Vattendrag. (Swedish National Atlas: Climate,

Responses of trace gas ﬂuxes and N availability to Lakes and Rivers]).Bra Bo¨cker, Ho¨ gana¨s, Sweden.

124 e n v i r o n m e n t a l s c i e n c e & p o l i c y 1 5 ( 2 0 1 2 ) 1 0 6 – 1 2 4

Tamm, C.O., 1991. Nitrogen in Terrestrial Ecosystems: Johan Bergh is a research leader at the Swedish University of

Questions of Productivity, Vegetational Changes, and Agricultural Sciences, Southern Swedish Forest Research Center

Ecosystem Stability. Springer-Verlag, Berlin/New York. in Alnarp, Sweden. His research focuses on how different environ-

Vanhala, P., Karhu, K., Tuomi, M., Bjo¨ rklo¨ f, K., Fritze, H., Liski, J., mental factors control forest production in boreal and cold-temper-

2008. Temperature sensitivity of soil organic matter ate forest ecosystems and implications of forest management,

decomposition in southern and northern areas of the boreal climate change mitigation and forest policy.

forest zone. Soil Biol. Biochem. 40, 1758–1764.

Zabowski, D., Chambreau, D., Rotramel, N., Thies, W.G., 2008. Leif Gustavsson is a professor in building technology at Linnaeus

Long-term effects of stump removal to control root rot on University and in Ecotechnology at Mid Sweden University in

forest soil bulk density, soil carbon and nitrogen content. Sweden. His research focuses on system analyses linked to sus-

For. Ecol. Manage. 255, 720–727. tainable development, especially in the ﬁelds of bioenergy, build-

ing construction, energy efﬁciency, forestry, and the interaction

Bishnu Chandra Poudel is a PhD candidate with the Ecotechnology between these ﬁelds.

research group in the Department of Engineering and Sustainable

Development at Mid Sweden University in O¨ stersund, Sweden. His Anders Lundstro¨ mis a research leader at the Department of Forest

research focuses on carbon balance implications of forest man- Resource Management at the Swedish University of Agricultural

agement and products utilization to contribute to climate change Sciences in Umea˚, Sweden. He works with production of long-term

mitigation. forest forecasts using NFI-data, and is also involved with the devel-

opment of systems for long-term forecasts for regional scenarios of

Roger Sathre is a researcher linked to Linnaeus University and the forest utilization in Sweden.

Ecotechnology research group at Mid Sweden University in O¨ ster-

sund, Sweden. His research focuses on the energy and climate Riitta Hyvo¨ nen is an associate professor in systems ecology at the

implications of forest management and forest product use, and Department of Ecology at the Swedish University of Agricultural

the potential contributions of forestry to sustainable development Sciences in Uppsala, Sweden. Her research focuses on modeling

and climate change mitigation. the cycling of carbon and other elements in forest ecosystems.

Paper III

World Renewable Energy Congress 2011 – Sweden Climate Change Issues (CC) 8-11 May 2011, Linköping, Sweden Climate change mitigation through increased biomass production and substitution: A case study in north-central Sweden Bishnu Chandra Poudel1,*, Roger Sathre1, Leif Gustavsson1,2, Johan Bergh1,3

1Ecotechnology, Mid Sweden University, Östersund, Sweden 2Linnaeus University, Växjö, Sweden 3Southern Swedish Forest Research Centre, Swedish University of Agricultural Sciences, Alnarp, Sweden * Corresponding author. Tel: +4663165535, E-mail: [email protected]

Abstract: In this study, we perform an integrated analysis to calculate the potential increases in forest biomass production and substitution as an effect of climate change and intensive management. We use the BIOMASS model to simulate change in Net Primary Production due to climate change. Then we estimate the development of forest biomass growth and harvest by using the HUGIN model, the change in soil carbon stock by the use of the Q-model, and the biomass substitution benefits by the use of an energy and material substitution model. Our results show that an average regional temperature rise of 4 °C could increase annual whole tree forest biomass production by 32% and harvest by 29% over the next 100 years. Intensive forest management including climate effect could increase whole tree biomass production by 58% and harvest by 47%. A total net reduction in carbon emissions of up to 89 Tg C and 182 Tg C over 100 years is possible due to climate change effect only and due to climate change plus intensive forestry, respectively. The carbon stock in standing biomass, forest soils and wood products all increase, but the carbon stock changes are less significant than the substitution benefits.

Keywords: Forest biomass, Carbon, Bioenergy, Construction material, Intensive forestry

1. Introduction Increasing atmospheric concentration of greenhouse gases (GHGs) increase earth’s surface temperature [1]. The Intergovernmental Panel on Climate Change (IPCC) [2] affirms that the temperature increase will be greater in the higher latitudes. The regional climate model by the Swedish Meteorological and Hydrological Institute (SMHI) [3] based on IPCC B2 scenario projects a 4 °C average temperature increase in north-central Sweden in the next 100 years.

The increasing temperature has significant impact on physical systems. It provides a longer growing season for the trees and favorable conditions for photosynthesis that stimulates Net Primary Production (NPP) [4]. Thus, an increased temperature is expected to produce more biomass in the boreal forests [5]. Although the boreal forest is considered as less productive due to low soil temperature and low nitrogen availability [6], production can further be increased by adding nutrients, by species change, and by changes in management practices [7, 8]. As a result, there will likely be larger harvest levels available in future.

The increased forest biomass can be used as a substitute for fossil fuels and carbon intensive materials to reduce carbon emissions through several mechanisms [9]. Using biomass to substitute for fossil fuels directly avoids fossil carbon emissions, except to the extent that fossil fuels are used to operate the biomass system [10]. Using biomass to substitute for carbon-intensive materials may reduce carbon emissions by lowering fossil energy use during the manufacture of products, by avoiding industrial processes emissions, by increasing carbon stocks in wood materials, by using biomass residues to replace fossil fuels, and possibly by carbon sequestration or emissions from wood products deposited in landfills [11, 12].

This paper describes an integrated assessment of the potential forest biomass increase as an effect of climate change and intensive forestry practices in north-central Sweden, and potential climate change mitigation with increased biomass use. Using five scenarios for forest production and two scenarios for biomass utilization, we estimate the increased biomass 1

World Renewable Energy Congress 2011 – Sweden Climate Change Issues (CC) 8-11 May 2011, Linköping, Sweden production, its harvest level, and carbon benefits from the substitution of non-wood materials and fossil fuels. We also estimate the carbon stock changes in standing biomass, soil, and wood products, and finally quantify the overall carbon balance for each scenario.

2. Methodology 2.1. Study area This study considers Jämtland and Västernorrland, two counties in north-central Sweden. The area lies from 61° 33’ to 65° 07’ N latitude and from 12° 09’ to 19° 18’ E longitude. The forest land area is about 3.5 and 1.9 million ha, respectively. Of this area, productive forests1 excluding protected areas cover about 2.6 million ha and 1.7 million ha, respectively [13].

2.2. Climate change IPCC Special Report on Emission Scenarios (SRES) [14] describes potential GHG emissions pathways during the 21st century based on the main driving forces of GHG emissions, considering their underlying uncertainties. Our regional climate scenarios are based on the IPCC B2 global scenario, corresponding to moderate emissions of GHGs leading to an atmospheric CO2 concentration of 572 ppm by the year 2085, with model RCA3 (The Rossby Centre’s regional Atmospheric model) that uses a grid of approximately 50 km x 50 km. The RCA3 model uses global driving variables from the general circulation model ECHAM4/OPYC3 [15] covering the period of 1961-2100. Climate projections for north- central Sweden are for an increase in average annual temperature of 4 °C by 2100 [16].

2.3. Scenarios We compare forest biomass production in a Reference scenario that assumes the current forest management practices without any climate change effect, and four scenarios that include climate change effect: Current, Environment, Production, and Maximum, (Table 1).

Table 1. Overview of scenarios analyzed in this study. Scenario Climate Forest management goals Biomass use Reference No change Reference (current management) Stem wood or whole tree Current Change Current management Stem wood or whole tree Environment Change Fulfill environmental goals Stem wood or whole tree Production Change Increase biomass production Stem wood or whole tree Maximum Change Maximize biomass production Stem wood or whole tree

The Current scenario assumes a continuation of current forest management practices. The Environment scenario assumes the fulfillment of additional environmental goals, e.g. 8% of total forest land is set aside for reserves, and 14% of total forest land is given special environmental care. The Production scenario uses additional silvicultural practices, e.g. improved genetic material plantation, soil scarification, selection of tree species, increased traditional fertilization and a balanced nutrient supply [8]. The Maximum scenario uses silvicultural practices to maximize the production by replacing of Scots pine with lodge-pole pine (Pinus contorta) and by adding balanced nutrient supply in young stands of lodge-pole pine and Norway spruce. The effect of a lower and higher precipitation is considered in the

1 Non-productive forests are defined as forests with production capacity less than 1 m3 per ha and year.

World Renewable Energy Congress 2011 – Sweden Climate Change Issues (CC) 8-11 May 2011, Linköping, Sweden scenarios and had no significant effect on production in northern Sweden [4]. The scenarios are described in more detail in [17].

2.4. Forest production modeling The process-based growth model BIOMASS describes the processes of radiation absorption, canopy photosynthesis, phenology, allocation of photosynthates among plant organs, litterfall, and stand water balance in a tree [18]. BIOMASS uses Swedish National Forest Inventory (NFI) database and current temperature for Reference scenario and increased temperature based on RCA3 climate model (see Section 2.2) for the estimation of NPP. Values of daily precipitation are used in NPP simulations. The parameterization is valid for mesic soils, and the effects of water deficits would be more pronounced on drier sites [7]. Water deficits seem to be at a low level in simulations, since extremely dry summers didn’t occur in the data set, which could lead to substantial losses in production. Output data on NPP from the BIOMASS are used to determine the growth functions in the HUGIN model [5, 17] an empirical model for long-term forecasts of timber yield and harvest level [19]. It uses sample plots from NFI, defining initial conditions for the model in order to describe the forest conditions during the development of young stands, their establishment, and the sivicultural treatments. The growth simulators in HUGIN are valid for all forest land in Sweden, for all types of stands, with a wide range of management alternatives [19].

2.5. Soil carbon modeling The Q-model describes changes in soil carbon stock based on litter inputs, soil temperature, and the decomposition of litter fractions [20]. It uses old carbon in the soil, litter inputs from standing trees, thinnings and harvests, and soil carbon from fine roots. The decomposition in response to temperature is calculated as in Ågren et al. [21] and Bosatta and Ågren [20]. For more details see [5].

2.6. Biomass use modeling The harvested biomass in thinnings and final fellings is assumed to substitute non-wood materials and fossil fuels. In the “stem wood” option we assume that 95% of coniferous stem wood (>20 cm diameter) is harvested and used as construction material to replace reinforced concrete [5, 11, 17]. The small stems and processing residues of coniferous trees, and all deciduous stem wood, is used as biofuel. In the “whole tree” option, in addition to stem wood use, 75% of branches and tops, 25% of needles, and 50% of stumps are harvested and used as biofuel [5, 11, 17]. Biofuel is assumed to replace coal in stationary plants.

2.7. Forest operations and fertilization We account for emissions from fossil fuels used for forest management activities including stand establishment, thinnings, final harvest of round wood, and forwarding and transport of round wood to mills [22]. GHG emissions associated with production and use of fertilizers include CO2, N2O and CH4 and are converted to CO2 equivalent using global warming potentials (GWP) over a 100-year time horizon [1]. Primary energy use and GHG emissions for the production of Skog-CAN fertilizer (27-0-0) and Opti-Crop fertilizer (24-4-5) are based on Davis and Haglund [23]. The amount of fossil fuel used for fertilizer application by helicopter is based on Mead and Pimentel [24].

World Renewable Energy Congress 2011 – Sweden Climate Change Issues (CC) 8-11 May 2011, Linköping, Sweden 3. Results and discussions 3.1. Forest biomass production and harvest Figure 1 shows the annual forest biomass production and harvest in all scenarios at the end of the study period. The annual whole tree biomass production in the Reference scenario is 12.8 Tg year-1 at the end of the study period compared to 11.0 Tg year-1 in the beginning of the study, a 17% increase during the 100-year period. The corresponding increases for the Current, Environment, Production and Maximum scenarios are 49%, 40%, 53%, and 75%, respectively. Thus biomass production increases in the Current, Environment, Production and Maximum scenarios are respectively 32%, 23%, 36% and 58% greater compared to the Reference scenario increase. The whole tree harvest in the Current scenario increases by 31% more and in the Maximum scenario by 50% more than the Reference scenario. The increase in stem wood biomass production for the Maximum scenario is 57% greater and the harvest is 54% greater compared to the Reference scenario.

Whole tree production Whole tree harvest 20 Stem wood production Stem wood harvest -1 15

Dry biomass Tg year Tg biomass Dry 5

0 Reference 2110 Current 2110 Environment 2110 Production 2110 Maximum 2110 Fig. 1. Annual forest biomass production and harvest (Tg year-1) in all scenarios at the end of the study period in Jämtland and Västernorrland.

Figure 2 shows the cumulative biomass harvest for all products in different scenarios during the 100 year. Stem wood is the largest fraction, while stumps are the second largest. The stump values might be slightly overestimated because of the Marklund’s revised function [25] that estimates stump biomass based on spruce stumps which likely overestimates the deciduous tree stumps [5].

1200 Bark Needle Slash Stump Stem wood 1000

800

600

400

200 Cumulative harvest harvest (Tg) Cumulative

0 Reference Current Environment Production Maximum Fig. 2. Cumulative biomass harvest (Tg dry matter) for all products in the different scenarios during 100-years in Jämtland and Västersorrland.

The increased production in the Current scenario is caused by climate change, while in the other scenarios are the results of climate change plus intensive forestry practices such as species change, fertilization, and intensive silvicultural action [5]. 4

World Renewable Energy Congress 2011 – Sweden Climate Change Issues (CC) 8-11 May 2011, Linköping, Sweden 3.2. Carbon stock in standing biomass, forest soils, and wood products Annual carbon stocks in living tree biomass, soil and harvested wood products with whole tree harvest increase in all scenarios (Fig. 3). Maximum scenario has the largest carbon stock in forest biomass, followed in descending order by Environment, Production, Current and Reference scenarios. Maximum scenario has the largest soil carbon stock, followed by the Reference and Production scenarios. Caron stock increases slowly in the Environment and Current scenario after 30 years compared to the other scenarios (Fig. 3). Wood product carbon stock is also largest in Maximum scenario, followed by the Production and Current scenarios, while Environment and Reference have smaller. The larger carbon stock is due to larger volume of stem wood harvested and used as building materials, however, after the life span of buildings, the carbon stock may stabilize due to the demolition of old buildings [17].

Reference scenario Current scenario Environment scenario Production scenario Maximum scenario 340 20 120 Forest biomass C stock Forest soil C stock Wood product C stock 320 18 100 300 80 280 16 60 260 14 40 240 12 20

Carbon stock (Tg C) 220

200 10 0 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100 2110 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100 2110 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100 2110 Fig. 3. Annual carbon stocks (Tg C) in living forest biomass, forest soil, and wood products in the different scenarios during 100-years in Jämtland and Västernorrland. Note differences in scale.

3.3. Biomass substitution Biomass substitution is largest in Maximum scenario (Table 2). The cumulative avoided

Table 2. Average annual avoided carbon emissions due to forest biomass use (Tg C y-1) during each 10-year period for Jämtland and Västernorrland. 2010- 2020- 2030- 2040- 2050- 2060- 2070- 2080- 2090- 2100- 2019 2029 2039 2049 2059 2069 2079 2089 2099 2109 Reference Scenario Stem wood 3.8 3.7 3.6 3.7 3.9 3.9 3.9 4.0 4.0 4.2 Whole tree 4.9 4.7 4.8 4.8 5.1 5.1 5.0 5.1 5.1 5.4 Current Scenario Stem wood 3.9 3.8 3.8 3.9 4.3 4.4 4.7 4.7 5.1 5.6 Whole tree 5.0 4.9 5.0 5.1 5.6 5.7 6.1 6.1 6.5 7.1 Environment Scenario Stem wood 3.6 3.6 3.6 3.7 4.0 4.1 4.3 4.5 4.8 5.0 Whole tree 4.6 4.6 4.7 4.9 5.2 5.3 5.5 5.7 6.1 6.4 Production Scenario Stem wood 3.9 3.8 4.0 4.2 4.6 5.0 5.3 5.4 5.8 6.2 Whole tree 4.9 5.0 5.2 5.5 6.0 6.4 6.7 6.8 7.3 7.8 Maximum Scenario Stem wood 3.9 3.9 4.1 4.3 4.7 5.0 5.5 5.7 6.0 6.6 Whole tree 5.0 5.0 5.4 5.6 6.0 6.4 7.0 7.2 7.6 8.4

World Renewable Energy Congress 2011 – Sweden Climate Change Issues (CC) 8-11 May 2011, Linköping, Sweden carbon emission due to use of stem wood and whole tree biomass for Maximum scenario is 497 and 635 Tg C, respectively. As Maximum and Production scenarios have larger harvests, they will give larger substitution benefits. When compared to the Reference scenario, Maximum and Production scenarios give 136 Tg C and 116 Tg C greater benefits, while Environment and Current scenario have comparatively smaller substitution benefits.

3.4. Carbon balance Cumulative avoided carbon emission over 100 years for all scenarios with the use of whole tree biomass is shown in Figure 4. Biomass substitution is the largest contributor to the carbon balance. The avoided carbon emission due to biomass substitution with recovered harvest slash and stumps is greater than the reduced increase in soil carbon stock in the forest due to harvest of whole tree biomass. The total avoided emissions in the Maximum and Production scenarios are 182 Tg C and 117 Tg C greater than in the Reference scenario over the 100-year period for whole tree biomass use. The corresponding numbers for the Environment and Current scenarios are 31 and 48 Tg C, respectively.

Soil C-stock increa se Forest C-stock increase 900 Product C-stock increase Biomass substitution 800 700 600 500 400 300 200

Avoided C emissions (Tg C) (Tg C emissions Avoided 100 0 Reference Current Environment Production Maximum Fig. 4.Cumulative avoided carbon emissions (Tg C) over 100- years for all scenarios with whole tree biomass use for Jämtland and Västernorrland.

4. Conclusions Climate change may substantially increase the production of biomass in boreal forests. The production can be further increased by intensive forestry practices. The increased forest biomass, if used to substitute for fossil fuels and carbon intensive materials, can avoid carbon emissions, thus contributing to climate change mitigation. This is a negative feedback on climate change. The substitution potential depends on the amount and type of forest biomass harvest. Thus, larger biomass harvests create greater substitution potential.

In this study, significant uncertainties can be observed. Net increase of greenhouse gases, driving the temperature change, depends on several factors as the development of the global population, economic growth, and on mitigation actions to be taken. The results may vary somewhat with differences in the temperature and other corresponding variables. Regional temperature increases may differ from our assumptions. Fire risk, pathogens and insect damage which may cause tree mortalities are not accounted in our study. The HUGIN model overestimates biomass in deciduous living mature trees and stumps. HUGIN does not have a function to calculate self thinning in living deciduous trees, which may lead to an overestimation of biomass in Environment scenario. Nevertheless, although these

World Renewable Energy Congress 2011 – Sweden Climate Change Issues (CC) 8-11 May 2011, Linköping, Sweden uncertainties may alter the exact values that we have calculated, it appears that our general conclusions are robust.

Acknowledgement We gratefully acknowledge the support of European Union, County Administrative Board of Jämtland, Sveaskog AB, SCA Forest Products, Norrskog/SÅTAB, Jämtkraft AB. We also thank two anonymous reviewers for their comments.

References [1] IPCC, Climate Change: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, ed. S. Solomon, et al. 2007: Cambridge University Press, United Kingdom and New York, USA. [2] IPCC, The Scientific Basis: Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change. 2001, New York , USA: Cambridge University Press. [3] SMHI. Climatic data, Swedish Meteorological and Hydrological Institute. 2009 [cited 2009 2, August]; Available from: http://www.smhi.se/cmp/jsp/polopoly.jsp?d=8785&l=sv. [4] Bergh, J., U. Nilsson, B. Kjartansson, and M. Karlsson, Impact of climate change on the productivity of Silver birch, Norway spruce and Scots pine stands in Sweden with economic implications for timber production., Eco.Bulletins, 53(15), 2010: pp. 185-195. [5] Poudel, B.C., R. Sathre, L. Gustavsson, J. Bergh, A. Lundström, and R. Hyvönen, Effects of climate change on biomass production and substitution in north-central Sweden, Manuscript, 2010. [6] Tamm, C.O., Nitrogen in terrestrial ecosystems, Ecological Studies, (81) 1991: pp. 1- 115. [7] Bergh, J., U. Nilsson, B. Kjartansson, and M. Karlsson, Impact of climate change on the productivity of Silver birch, Norway spruce and Scots pine stands in Sweden with economic implications for timber production., Ecological Bulletins, 53(15), 2010: pp. 185-195. [8] Skogsstyrelsen, Skogliga konsekvensanalyser – SKA-VB 08 in Swedish Forest Agency Rapport. 2008, Skogsstyrelsen, Sweden. [9] Schlamadinger, B., M. Apps, F. Bohlin, L. Gustavsson, G. Jungmeier, G. Marland, K. Pingoud, and I. Savolainen, Towards a standard methodology for greenhouse gas balances of bioenergy systems in comparison with fossil energy systems, Biomass and Bioenergy, 13(6), 1997: pp. 359-375. [10] Hall, D.O. and J.I. House, Trees and biomass energy: Carbon storage and/or fossil fuel substitution?, Biomass and Bioenergy, 6(1-2), 1994: pp. 11-30. [11] Gustavsson, L., K. Pingoud, and R. Sathre, Carbon dioxide balance of wood substitution: comparing concrete- and wood-framed buildings, Mitigation and Adaptation Strategies for Global Change, 11(3), 2006: pp. 667-691.

World Renewable Energy Congress 2011 – Sweden Climate Change Issues (CC) 8-11 May 2011, Linköping, Sweden

[12] IPCC, Climate Change: Mitigation of Climate Change. Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. 2007: Cambridge University Press, United Kingdom and New York, USA. [13] Skogsstyrelsen. Forestry Statistics, Swedish Forest Agency. 2009 [cited 2009 2, August]; Available from: http://www.svo.se/episerver4/default.aspx?id=38515 [14] IPCC, Special Report on Emissions Scenarios, in Emission Scenarios. Special Report of Working Group Ш of the Intergovernmental Panel on Climate Change. 2000, Cambridge University Press, UK. [15] Kjellström, E., L. Bärring, S. Gollvik, U. Hansson, C. Jones, P. Samuelsson, M. Rummukainen, A. Ullerstig, Willén, and K. Wyser, A 140-year simulation of European climate with the new version of the Rossby Centre regional atmospheric climate model (RCA3), in SMHI Reports in Meteorology and Climatology, No. 108. 2006, SMHI: Norrköping, Sverige. pp. 54. [16] SMHI. Climatic data, Swedish Meteorological and Hydrological Institute. 2009 [cited 2009, August, 2]; Available from: http://www.smhi.se/cmp/jsp/polopoly.jsp?d=8785&l=sv. [17] Poudel, B.C., R. Sathre, L. Gustavsson, J. Bergh, A. Lundström, and R. Hyvönen, Potential effects of intensive forestry on biomass production and substitution in north- central Sweden, Manuscript, 2010. [18] McMurtrie, R.E., D.A. Rook, and F.M. Kelliher, Modelling the yield of Pinus radiata on a site limited by water and nitrogen, Forest Ecology and Management, 30(1-4), 1990: pp. 381-413. [19] Lundström, A. and U. Söderberg. Outline of the Hugin system for longterm forecasts of timber yields and possible cut. In Large-Scale Forestry Scenario Models: experiences and requirements. 1996: EFI proceeding.pp. 63-77 [20] Bosatta, E. and G.I. Ågren, Theoretical analyses of carbon and nutrient dynamics in soil profiles, Soil Biology and Biochemistry, 28(10-11), 1996: pp. 1523-1531. [21] Ågren, G., R. Hyvönen, and T. Nilsson, Are Swedish forest soils sinks or sources for CO2—model analyses based on forest inventory data, Biogeochemistry, 82(3), 2007: pp. 217-227. [22] Berg, S. and E.-L. Lindholm, Energy use and environmental impacts of forest operations in Sweden, Journal of Cleaner Production, 13(1), 2005: pp. 33-42. [23] Davis, J. and C. Haglund, Life cycle inventory (LCI) of fertiliser production - Fertiliser products used in Sweden and Western Europe, in SIK Report No. 654, SIK. 1999, The Swedish Institute for Food and Biotechnology Göteborg, Sweden. [24] Mead, D.J. and D. Pimentel, Use of energy analyses in silvicultural decision-making, Biomass and Bioenergy, 30(4), 2006: pp. 357-362. [25] Petersson, H. and G. Ståhl, Functions for belowground biomass of Pinus sylvestris, Picea abies, Betula pendula and Betula pubescens in Sweden, Scandinavian Journal of Forest Research, (21) 2006: pp. 84-93.