Extent and costs of forest-based climate change mitigation in Germany: accounting for substitution

Abstract

The objective of this study is to quantify the contribution of the German forestry to climate change mitigation and to calculate the associated costs at the national level. For that, the forest and harvested wood products carbon pools are considered as well as energy and material substitution. We compare five different scenarios, each referring to an alternative level of timber harvests (due to changing rotation lengths or setting aside of forest areas). The study shows that enhancing the use of wood products at the expense of other materials, such as steel, plastic or concrete, can have a substantial impact on the overall carbon balance.

Keywords:

• Germany
• forestry
• carbon sequestration
• substitution effects
• scenario analysis

Introduction

Forests play an important role in climate change mitigation. They influence the local and regional climate by moderating temperature and humidity, and sequester atmospheric carbon through photosynthesis and tree growth (forest carbon pool). When wood is used, for example in boards, furniture or packaging material, the carbon remains stored in the wood products (harvested wood products [HWP] carbon pool). In addition, wood use can also replace fossil fuels such as gas, oil or coal in two ways. On the one hand, wood can be burnt directly for fuel (energy substitution). On the other hand, the production and disposal of wood products usually require less energy than competing products made of materials such as plastic, metal or concrete (material substitution). Both types of substitution effects have been quantified and applied in previous studies [1–7].

While the potential of forests to reduce net greenhouse gas emissions is widely recognized, its quantification on a national scale is quite challenging. Recently, an investigation by Bösch et al. [8] assessed the CO₂ effects and the costs of alternative forest management measures for Germany. However, that study quantified only effects on the forest and HWP carbon pools. This was done in accordance with climate reporting rules where substitution effects are implicitly recorded in the industry and energy sectors instead of being recognized as a contribution of the forestry sector. However, when discussing the impacts of alternative forest management measures on the CO₂ balance, it can be illustrative to include substitution effects in the calculation. Otherwise, the quantitative impacts of forest management measures on climate change mitigation would only be partly captured, and the resulting abatement costs (i.e. Euros per ton of CO₂) might be misinterpreted.

The purpose of this paper is to quantify the contribution of German forestry to climate protection in a comparison of the scenarios used in Bösch et al. [8], integrating all aspects of carbon storage as well as material and energy substitution effects. It is important to note that to date there are no universally accepted standards for calculating the substitution potential of wood which is just supplied by forests and not already appointed to a specified use. Therefore, we use average displacement factors from the literature. Moreover, we calculate the associated costs when substitution effects are included. For this, we use the approach devised by Bösch et al. [8] that accounts for possible impacts on the values of market goods as well as non-market goods and services of forestry as an aggregate cost measure at the national level.

Material and methods

Scenarios

We analyze the CO₂ effects of different management regimes in German forestry by comparing five scenarios, each referring to an alternative level of timber harvests between 2014 and 2048 (see Figure 1). For the analysis, we compare the base scenario and each of the other four scenarios in turn. The objective of this scenario analysis is not a forecast of the future, but to show the wide range of potential developments under different management objectives [see e.g. 9–11]. Originally, Scenarios 0, I and II were constructed within the so-called WEHAM framework (‘Forest Development and Timber Resource Modeling’) and applied to data from the second National Forest Inventory (2002) and the Inventory Study 2008 (IS 08; see [12–14]). Scenario 0, the ‘base scenario’, simulates management as intended by forest managers for the next few decades, taking market conditions and legal requirements into account.

Figure 1. Overview of timber harvests under the forest management scenarios (in million m³ under bark per year) (Source: Bösch et al. [8]).

Increasing/decreasing the rotation length of tree stands usually leads to a larger/smaller carbon stock in forests [15–17]. Scenarios I and II reflect the outcomes of varying rotation periods. In Scenario I, the final cutting of the trees is postponed by approximately 20 years compared to the base scenario (technically, the diameter threshold for harvesting is increased by 10 cm). Scenario II delineates the potential roundwood availability which will result when growing stocks per hectare are gradually reduced from the stocks of 2002 to the level of 1987 (i.e. the base year of the first German National Forest Inventory). Accordingly, average rotation lengths are shortened by about 17 years [18].

Scenario III reflects the goal of the National Biodiversity Strategy [19], namely to reach a 5% share of set-aside forest area. In Scenario IV, the existing harvest restrictions due to, for example, the preservation of nesting and hollow trees are supplemented with the assumption of a 10% reduction in timber harvests, systematically distributed across all age classes and tree species. These harvest reductions cover a reduction of the productive area to 90% plus additional harvest restrictions for preserving hollow and nesting trees. Thus, the reduction in harvest potential is larger than just doubling the reductions of Scenario III.

Modeling approach

Modeling carbon pools

The modeling of both the forest and the HWP carbon pools is taken from Bösch et al. [8]. The assessment of the CO₂ effects in the forest according to the five management scenarios is based on an inventory study that was conducted to estimate carbon stocks in German forests at the beginning of the first commitment period of the Kyoto Protocol in 2008 (IS 08; see [14]). The aim of this inventory was to assess carbon stocks in living biomass and dead wood. The volumes taken from the forest inventory were used to estimate timber harvest potentials and carbon stocks in wood (see also Bösch et al. [8] for more details).

The calculation of the HWP contribution to the carbon balance in the five management scenarios is taken from Bösch et al. [8] and is based on the method used for deriving future emissions and removals from the HWP pool for the Forest Management Reference Level [20]. The calculation is implemented in the model WoodCarbonMonitor on the basis of Tier 2 methodology applied for reporting HWP in the second commitment period of the Kyoto Protocol [21,22].

Assessing material and energy substitution

Material and energy substitution are analyzed using average displacement factors. A displacement factor is determined as follows: C emissions associated with wood products (GHG_wood) are compared with C emissions attributed to non-wood alternatives (GHG_non-wood) with the same functionality on the basis of life-cycle assessments and then related to the mass of C in the wood product (WU_wood), as either an end-use product or wood used for energy [1,4,11]. Formally, a displacement factor (DF) can be expressed as: (1) $$DF=\frac{{GHG}_{non-\mathit{\text{wood}}}-{GHG}_{\mathit{\text{wood}}}}{{WU}_{\mathit{\text{wood}}}}$$(1)

A higher displacement factor means that more C emissions are avoided per unit of wood used (a negative displacement factor would indicate that emissions are higher when using the wood products). The displacement factors are given in units of tC of emission reduction per tC in the wood product.

For this study, we use an average displacement factor for material substitution of 1.5 tC/tC as estimated by Knauf et al. [4], who derived individual product displacement factors for 16 key product areas and weighted them on the basis of a material flow analysis for Germany [23,24].¹ For fossil fuel substitution, we employ an energy displacement factor of 0.67 tC/tC as estimated by Rüter [20], who derived this displacement factor from the difference in emissions between light fuel oil and wood. Both displacement factors are applied to the carbon content of the timber harvests allocated to material and energetic use according to Bösch et al. [25].² We assume that both displacement factors remain constant in the whole investigation period until 2048 [7].³

Cost calculation

Various studies have estimated the costs of forest management measures in order to address climate change mitigation. However, most previous studies focused on costs at the forest enterprise or forestry sector level rather than the national level [26–29,31,32]. These so-called bottom-up studies do not generally account for how one sector will adjust to changes in another [32]. For this study, we use the novel approach developed by Bösch et al. [8] that accounts for possible impacts on the values of market goods as well as non-market goods and services of forestry as an aggregate cost measure at the national level.

When full substitution is assumed, costs related to the values of market goods can be expected to be zero on balance. This is because substitution always implies a shift of economic activities to other sectors. For instance, decreasing timber harvests are likely to reduce wood-based construction, provided that the reduction in the supply of roundwood is not compensated by increasing imports [33].⁴ On the other hand, decreasing timber harvests are rather unlikely to reduce activities in the construction industry in general. As a consequence, CO₂ emissions would increase due to the usually higher energy needs for the production and disposal of alternative building materials (e.g. concrete). However, because the activities move from one sector (in this example, wood-based construction) to another sector of the national economy (construction based on alternative materials), it can be argued that the economic losses (i.e. lower value added) of the one sector would roughly be offset by the economic gains of the other sector (i.e. higher value added). This implies the absence of notable additional costs from an overall economic perspective.⁵ Similarly, technical equipment and labor forces that are no longer of use in forestry, for instance due to harvest reductions, are assumed to be used in other sectors of the economy (e.g. agriculture) at no additional cost from a national point of view.

In order to assess non-market values that are affected by forest management measures, we use the results of a choice experiment (CE) that dealt with the impacts of land-use changes in Germany on the monetary values of different ecosystem services of forests and other land uses [34]. In the CE, respondents were asked to choose their most preferred future development of the forests in a radius of 15 km around their homes (out of three alternatives, one of which was the status quo). The alternatives were each described by different attributes, including the harvest age of the forest stands (which is varied in Scenarios I and II) and the percentage of forest area set aside without further forest management (varied in Scenarios III and IV). A more detailed description of the CE can be found in Bösch et al. [8], Weller and Elsasser [35] and Fick and Gömann [34].

Results

Figures 2–5 exhibit CO₂ effects under Scenarios I–IV in relation to the base scenario. When only looking at the forest carbon pool, any decrease in roundwood supply increases the amount of carbon stored in the forest (Scenarios I, III and IV) compared to the base scenario, whereas shorter rotation lengths result in a decrease in forest carbon stocks (Scenario II), relative to the base scenario. The HWP carbon pool is affected in the opposite direction (albeit to a lesser degree): Any decrease in timber harvests reduces the amount of carbon stored in the wood products (Scenarios I, III and IV), while a shorter rotation length leads to an increase in the HWP carbon pool (Scenario II).

Figure 2. Relative emissions in Scenario I (in million tons CO₂ per year; negative values indicate removals from the atmosphere).

Figure 3. Relative emissions in Scenario II (in million tons CO₂ per year; negative values indicate removals from the atmosphere.

Figure 4. Relative emissions in Scenario III (in million tons CO₂ per year; negative values indicate removals from the atmosphere).

Figure 5. Relative emissions in Scenario IV (in million tons CO₂ per year; negative values indicate removals from the atmosphere).

Substitution effects turn out to have substantial impacts on the emission balance. In most periods, they are on the same order of magnitude as the effects on the forest carbon pool – however, with the opposite sign. Thus, the net sink effect of Scenarios I to IV (as compared to the base scenario) is dominated by the development of the forest carbon pool and the substitution effects: Only Scenario I brings some significant CO₂ removals compared to the base scenario (at least in the period 2019–2038), whereas all other scenarios result in higher relative emissions over the whole period. For this reason, we refrain from calculating cost-effectiveness ratios (i.e. a scenario’s cost per ton of CO₂) in this paper.

Figure 6 shows total costs for Scenarios I–IV in relation to the base scenario. In Scenario I, benefits rather than costs occur (on average about 500–800 million Euros per year), since postponing the final cutting of the trees is considered positive by the German population. The opposite is true in Scenario II: The shorter rotation periods result in considerable losses in the value of public goods, starting from 437 million Euros in the period 2014–2018 and increasing to nearly 4 billion Euros per year between 2044 and 2048. Scenario III does not come with significant costs, as the differences from the base scenario regarding the amount of unused forest area are negligible, and the age development is the same as in the base scenario. In Scenario IV, total costs rise to 324 million Euros per year compared to the base scenario, as there is a constant 10% share of unused forest area.

Figure 6. Relative costs of the forest management scenarios (in million Euros per year; negative values indicate benefits).

Discussion and conclusions

In this study, we compared and analyzed the effects of different forest management options on the climate change mitigation potential of forestry in Germany and calculated the associated costs. Our study differs from the recently published assessment by Bösch et al. [8] in that our carbon account here considers not only the forest and the HWP carbon pools, but also substitution effects (i.e. energy and material substitution). Even though the quantification of substitution effects suffers from information deficits and hence relies on several assumptions, our results demonstrate that substitution effects can have a substantial impact on the overall carbon balance.

The comparison of the scenarios shows that different mitigation strategies can produce very different CO₂ effects. Our study reveals that enhancing the use of wood products at the expense of other materials, such as steel, plastic, or concrete, can considerably decrease relative CO₂ emissions (as in Scenario I). However, some forest management scenarios that had turned out to be favorable in the investigation by Bösch et al. [8], when only considering forest and HWP carbon pools (e.g. the protection scenarios III and IV), now result in relative CO₂ emissions rather than removals when considering forests, HWP and substitution effects jointly.

In this study, the costs of the mitigation measures were calculated as the sum of possible impacts on the values of market goods as well as non-market goods and services of forestry at the national level. It is important to note that the high level of cost aggregation may conceal politically relevant distribution aspects. This is because forestry measures mainly affect the values of the market goods produced by forest enterprises and downstream forest-based industries. However, benefits of climate change mitigation accrue to society as a whole. This means that policies choosing the least costly option from an aggregate point of view (i.e. Scenario I) may lead to compensation demands from those who are negatively affected by this option.

There are other mitigation measures in forestry besides the ones assessed in this study (e.g. changes in timing and/or intensity of thinning, changes in tree species, application of fertilizer to increase productivity or afforestation of agricultural land). The framework for cost calculation used in this study is perfectly suited to assess the costs of those measures, as long as the supply of roundwood from forests and hence the amount of roundwood available to the economy is changing. However, measures that do not change the amount of wood removals cannot be assessed with this methodology. Moreover, it is important to keep in mind that implementing mitigation measures may also entail transaction costs, such as information-gathering and administration costs, which can be substantial. However, they are hard to quantify [32]. Consequently, we did not account for transaction costs in this study.

Our results emphasize the importance of forests and wood products in mitigating climate change. For policies aimed at mitigating climate change, an increased use of wood and wood products can play an important role. Our study shows that substitution effects are very relevant when assessing the contribution of managed forests to the reduction of atmospheric CO₂. Our approach for Germany could also be transferred to other regions of the world. However, the applied substitution factors are only valid for countries with a comparable share of fossil-based energy supply.

Disclosure statement

The authors declare no conflict of interest.

Notes

1 This displacement factor is in line with the average of the displacement factors for material substitution derived in the studies by Rüter et al. [23] and Hafner et al. [24].

2 We assume that 1 m³ of wood contains 250 kg C, which corresponds to 917.5 kg CO₂.

3 This assumption is not unusual [see e.g. 7]. However, it is quite strong, as it implies that the energy mix of Germany will remain constant over time. Even though this is not very realistic, any alternative specification of the development of the energy mix and thus of the respective displacement factors would be equally speculative.

4 This assumption can be justified by the fact that Germany’s net imports of wood and wood-based products as a percentage of total consumption have been rather small in recent years, in spite of increasing timber prices [33].

5 Of course, this holds only if the activities remain in the national economy. If the economic activities moved abroad, there would be substantial losses in value added which would be costs from an aggregate point of view.