Abstract
Land-based mitigation measures are needed to achieve climate targets. One option is the mitigation of currently high greenhouse gas (GHG) emissions of nutrient-rich drained peatland forest soils. Continuous cover forestry (CCF) has been proposed as a measure to manage this GHG emission source; however, its emission reduction potential and impact on timber production at regional and national scales have not been quantified. To quantify the potential emission reduction, we simulated four management scenarios for Finnish forests: (i) The replacement of clear-cutting by selection harvesting on nutrient-rich drained peatlands (CCF) and (ii) the current forest management regime (BAU), and both at two harvest levels, namely (i) the mean annual harvesting (2016–2018) and (ii) the maximum sustainable yield. The simulations were conducted at the stand scale with a forest simulator (MELA) coupled with a hydrological model (SpaFHy), soil C model (Yasso07) and empirical GHG exchange models. Simulations showed that the management scenario that avoided clear-cutting on nutrient-rich drained peatlands (i.e. CCF) produced approximately 1 Tg CO2 eq. higher carbon sinks annually compared with BAU at equal harvest level for Finland. This emission reduction can be attributed to the maintenance of a higher biomass sink and to the mitigation of soil emissions from nutrient-rich drained peatland sites.
Similar content being viewed by others
Introduction
Peatlands are some of the largest carbon reserves globally and store approximately 85% of the carbon (~ 550 Gt) of the Northern Hemisphere1. Approximately 12% of peatlands have been drained for a variety of land-use purposes. Peatland drainage induces aerobic decomposition, which is a significant source of CO2 into the atmosphere, particularly in Finland, where more than half of the peatland area (> 5 mill. ha) has been drained—mainly for forestry2. Avoiding peatland loss and degradation has an emission reduction potential of 813 Tg CO2 eq. in freshwater wetlands at the global level3, translating into annual peatland emission reduction by 8.35 Tg CO2 eq. in Finland (peat extent based on the International Mire Conservation Group Global Peatland Database4). According to the national greenhouse gas (GHG) inventory5, the CO2 emissions of drained peatland forest soils were 8.2 Tg in Finland for 2021, with an uncertainty of 83% (twice the relative standard error). Implementing measures that halt the degradation of peat soil and peatland vegetation would reduce these emissions. Global studies on the land-based mitigation of climate change3 report substantial uncertainties for their estimates, and therefore, national studies quantifying emission reduction potentials with bottom-up data and modelling are needed. National bottom-up studies are valuable also because drained peatlands are unevenly distributed across Finland, with most of them being in the western and northern regions6. Therefore, optimal land-based mitigation measures vary across regions, e.g., due to the extent of the drained peatland, their soil type, forest structure, regional wood demand and climatic conditions.
In Finland, forestry on peatlands has a considerable economic and ecological importance, because about a 20% of the productive forestry land i.e., about 4.1 mill ha is in drained peatlands7. The current practice in peatland forestry is largely based on rotation forestry, where tree stands are managed by thinning from below, the water table is kept at a low level by ditch network maintenance treatments, and after clear cut, the soil surface prepared by tillage such as mounding with furrow ploughing, and thereafter planted with seedlings.
On nutrient-rich drained peatland sites, continuous cover forestry (CCF) is a promising option as there are environmental and economic benefits, such as improved water quality and the avoidance of the costs of stand regeneration and ditch maintenance8,9. CCF on drained peatlands has been proposed as a measure to reduce soil GHG emissions through managing the water table level10,11,12. The rationale for using CCF on drained peatlands builds on the fact that a continuous and managed forest cover maintains the water table at the level which allows trees to grow, further ditch-network maintenance is not needed, and both nutrient loading to waterways and GHG emissions are reduced9. The reduction of soil GHG emissions with CCF on drained peatlands assumes that the mean water table level (WTL) is higher, reducing soil CO2 emissions compared with even-aged management. Site level studies and models show that peat hydraulic conductivity, ditch spacing, ditch depth and vegetation evapotranspiration affect WTL and the thickness of the fast decomposing aerobic peat layer13,14.
Nutrient-rich drained peatlands are typically occupied by Norway spruce-dominated forests, and they are the most productive site types of the boreal forests2. In Finland, most of these sites have been drained for forestry more than 50 years ago and are currently in the late stage of drainage succession, where the mire plant species have been replaced by upland forest vegetation consisting of mesic herb and forest moss species and where bilberry (Vaccinium myrtillus) is the most frequent dwarf shrub. The age structure of drained peatland forests is such, that a large fraction of those are currently at mature stage and approaching decision whether those will be clear-cut or managed with selection harvests.
Studies estimating the climate benefits of CCF management on drained peatlands are few and focused on stand level assessments15,16. Stand level climate change mitigation studies exclude potential leakage effects where e.g., reduced loggings in a study site are taking place elsewhere, and therefore, larger-scale simulation studies with alternative forest management strategies are needed. At a stand level, CCF harvesting typically removes between 50 and 70% of the merchantable timber, while clear-cut harvesting removes nearly 100% of the biomass, and in Southern Finland a mature nutrient-rich drained peatland sites may have 250–400 m3 in stem volume per ha. The application of CCF has been limited partly due to limited evidence based on tree growth and productivity compared with that of even-aged rotation forestry, especially with suppressed Norway spruce trees after selection harvests. On a peatland site in Canada, the release effect on the diameter growth of suppressed Black spruce (Picea mariana (Mill.)) trees took 12 years17. On a drained peatland site (Lettosuo) in southern Finland, it took less than 5 years for suppressed Norway spruces to recover their diameter growth after selection harvest; however, the carbon uptake by trees was increased immediately after the harvest treatment18.
To develop climate policies and incentives for land-based climate change mitigation, countries should be able to quantify the technical and feasible potential of different climate change mitigation measures. Currently, the national level scenario studies focusing on land-based mitigation strategies in forests are few and mostly consider the impact of harvest levels or climate change on forest carbon sinks19,20. In a market-driven economy, reducing harvests as a mitigation measure can be applied on state lands, but on private lands, this is more challenging without additional policy instruments. Therefore, there is an urgent need for alternative forest management strategies that allow timber provisioning while having simultaneous environmental benefits. Previous scenario studies have shown the impacts of harvesting on forest carbon sinks19,20,21, but there are a few studies that quantify the potential impact of a management regime change (from rotation forestry to CCF) to national-level climate benefits.
To explore the potential of land-based mitigation strategies with forests, we simulated different scenarios for Finnish forests with the MELA forest simulator for 2016–2050, the SpaFHy-peat hydrological model, the Yasso07 soil carbon model and GHG exchange models (see “Methods”). We quantified the GHG exchange of tree biomass and soils for Finnish forests (both mineral soils and drained peatlands) under current (BAU) management regime and with the compulsory avoidance of clear-cut on nutrient-rich drained peatland forests (CCF). We assessed the impacts of these scenarios following two harvest levels, actual harvests (AH, which was mean annual harvesting level of the years 2016–2018), and the maximum sustained yield (MSY), which was substantially larger (7–17 mill m3 per year) harvest level compared with AH. The period of 2016–2018 for AH logging level is also a basis for the land-use sector targets for EU member states for 2030 (EU regulation 2018/841). The MSY logging level is the basis of Finnish forest policy by providing level of loggings that are considered feasible without compromising future logging possibilities. This allows policy-makers and the forest industry to consider timber industry development. The MSY logging level integrates current conservation statuses, however ignores possible future policy shifts, such as shifts required to meet climate targets or coming requirements of EU restoration act.
Specifically, we focused on the emissions and sink estimation of drained peatlands by incorporating the latest research findings on CO2, N2O and CH4 dynamics and the impact of hydrology to emissions into our simulations. In addition to BAU and CCF, we simulated a CCF-regressed (CCF-reg) scenario to study the sensitivity of our results on the growth level of suppressed Norway spruces after selection harvest. The compulsory avoidance of clear-cuts (CCF scenario) was based on studies that have shown a reduction in soil GHG emissions from drained fertile peatlands after the water table level had been increased11,12.
Our aims were as follows: (i) to quantify the forest GHG exchange differences between the current management regime (BAU), targeted peatland CCF, and increased harvest levels in Finland, (ii) to quantify CO2, N2O and CH4 exchange estimates for drained peatland forest soils and (iii) to identify the regions in Finland that have climate change mitigation potential with their drained peatlands.
Results
Our results indicate that by avoiding clear-cutting through the use of selection harvesting (CCF scenario) on nutrient-rich drained peatlands, it is possible to obtain climate benefits, i.e., higher vegetation carbon sinks and lower soil GHG emissions, compared with business as usual (BAU) forest management (Fig. 1). With actual harvesting (AH) levels (Table 1), the CCF scenario produced 0.7–1.3 Tg CO2 eq. year−1 higher carbon sink compared with BAU throughout the simulation period (Table 2). This difference between the CCF and the BAU scenario was mainly due to larger areas of clear-cuts in the BAU scenario and emissions related to those, as contrary to our expectations water table levels were equal between BAU and CCF or deeper for the CCF (see Supplementary Fig. S3). Our simulations for the first 10-year period showed that under the CCF scenario, the annual clear-cut area in Finland was 61,000 ha on nutrient-poor peatlands (168,000 ha mineral soils), whereas under the BAU scenario, it was 75,000 ha on all peatlands (157,000 ha in mineral soils). The total amount of harvested wood from drained nutrient-rich peatland forests was 6.7 mill. m3 under CCF and 9.4 mill. m3 under BAU for the first 10-year period, due to leakage of harvesting from peatlands to mineral soils in the CCF scenario. With the actual harvesting levels (AH), the total annual harvesting area for commercial timber was 466,300 ha for the BAU scenario and 481,700 ha for the CCF scenario in the first 10-year period.
With the maximum sustained yield (MSY), the total annual harvesting amount was approximately 3 mill. m3 lower under the CCF management scenario compared with the BAU scenario (Table 1). This difference was due to fact that clear-cuts were not allowed on the nutrient-rich drained peatlands under the CCF scenario. For the period of 2022–2027, the Finnish forests were emission sources, according to the MSY harvesting level, with BAU management resulting in emissions that were 4.2 Tg CO2 year−1 eq. higher than those under the CCF scenario (Table 2).
According to our results, the relative climate impact per change in harvests (ΔTg CO2 eq. per Δ harvested mill. m3) were smaller when converting from BAU (AH) to CCF (MSY) management on nutrient-rich peatlands compared with conversion from BAU (AH) to BAU (MSY) (Fig. 2). Increasing harvests from the current management (BAU) and AH logging levels to MSY harvesting level but with CCF management increased less CO2 emissions than conversion from BAU (AH) to BAU (MSY). Reducing harvesting at the MSY harvesting level by switching from BAU (80.5 mill. m3) to CCF (78.3 mill. m3) produced 2–3.5 Tg CO2 eq. climate benefits per 1 mill. m3, whereas that ratio was 1.3–2 Tg CO2 eq. per 1 mill. m3 when the harvesting level changed between the AH level (73 mill. m3) and the MSY harvesting level (80.5 mill. m3) under the BAU management (Fig. 2). Most of the climate benefits due to reducing harvests with the transition from BAU (MSY) to CCF (MSY) were obtained by the increased tree biomass sink. Note that these estimates exclude carbon sinks with harvested forest product pool and potential substitution benefits.
At actual harvest levels, the net emissions from drained peat soils were 14.5 Tg CO2 eq. year−1 for 2022–2035 in the BAU scenario and approximately 13.5 Tg CO2 eq. year−1for the CCF scenario (Table 3). The soil emissions reductions decreased in periods close to 2040s (Figs. 1, 3). The largest individual component of these emissions was the CO2 emission from the peat layer, being approximately 12 Tg CO2 year−1 before 2035 for the BAU scenario. The second largest component was the N2O emission with approximately 4.1 Tg CO2 eq. year−1 before 2035 for the BAU scenario (Table 3). Both CO2 and N2O emissions were higher than those reported earlier as we also accounted for the additional emissions from the clear-cut area22,23,24.
Most of the regions showed reduced emissions for the CCF scenario compared with the BAU scenario with total ecosystem GHG exchange for the period of 2022–2035 (Fig. 4) and for 2022–2050, with identical harvesting levels (actual harvestings, Fig. 4 and Supplementary Table S9). These emission reductions can be attributed to the regions that have significant areas of nutrient-rich drained peatlands and areas that have high harvesting possibilities during these periods25 on these sites (Supplementary Tables S6–S9), such as Central Finland, Lapland, Northern Ostrobothnia and North Karelia (see also Supplementary Material for more details).
Changing the management regime from BAU to CCF resulted in a leakage effect, where more harvestings targeted the mineral soils in the CCF scenario compared to the BAU scenario. The magnitude of the leakage for the mineral soil carbon stock change was between − 0.3 and − 0.9 Mt CO2 over different simulation periods with AH levels (Table 4). It should be noted that in spite of the leakage, the CCF scenario resulted in 0.2–1.3 Tg CO2 eq. year−1 higher carbon sinks for Finnish forests compared with the BAU scenario.
Based on the results of the sensitivity analysis (CCF-reg scenario), which assumed a reduced growth of suppressed Norway spruce trees after selection harvests (− 25% over next 5 years) reduced the carbon sink in Finland compared with BAU and CCF scenario (Supplementary Table S10 in the Supplementary Material). According to the CCF-reg scenario ran with the AH levels, the total carbon sink of Finnish forests was 1–3 Tg CO2 year−1 eq. lower than that under the BAU scenario.
Discussion
GHG exchange of scenarios
We found that the CCF scenario provides approximately 1 Tg CO2 eq. higher carbon sinks for Finnish forests before 2035 compared with the BAU scenario, which can be attributed to the avoidance of clear-cuttings, resulting in reduced soil emissions and stronger carbon sinks in tree biomass. The role of clear-cut emissions is significant, especially as the interplay between reduced ditch depth and increased LAI resulted, that there were more deeper water table levels with CCF compared to BAU with drained peatlands. Avoiding clear-cuttings under the CCF scenario resulted in an increase in the total annual harvesting area of commercial timber by 15,000 ha, with a more pronounced harvesting of mineral soil areas. This leakage to mineral soils can be seen in the results, where the sink of mineral soils was reduced by 0.3–0.9 Tg CO2 eq. from 2027 onwards under the CCF scenario compared with that under the BAU scenario with AH. Despite this reduction of the sink due to leakage however, the CCF scenario produced higher overall carbon sinks for Finland than the BAU scenario. This result shows that there are opportunities to maintain timber production and simultaneously reduce GHG emissions with climate-smart forest management planning.
Relative climate impacts of different scenarios with varying harvest amounts
The relative climate impact (the change in GHG sink relative to the change in the harvesting amount) doubled when reducing harvestings by converting from BAU (MSY) to CCF (MSY) compared with reducing harvestings while retaining BAU management [i.e. from BAU (MSY) to BAU (AH)]. Also, it was found that increasing harvests from the AH level to MSY level, emissions increased less per 1 mill m3 if the management was converted to CCF instead staying with current forest management (BAU). Most of these climate benefits were due to stronger tree biomass sink. The soil-related climate benefits were larger in the case of a management regime change from BAU to CCF compared with changes in the harvesting levels with current management (BAU) regime due to fact that conversions from BAU to CCF reduces the amount of stands with recent clear-cuts and high soil related emissions.
Our results also show that these relative climate impacts, measured as a ratio between the change in GHG sink relative to the change in the harvesting amount is not a constant and varies over time; its dynamics differ between conversions from current management regime (BAU) to CCF compared with maintained management regime (BAU), but changed harvests.
The varying relative climate impacts with time suggests that climate benefits will increase over time for CCF compared with BAU. Earlier studies have shown that these ratios vary from 1.2 to 1.8 (depending on the models used), when the unit was the change in the total forest sink in CO2 compared with that change in the harvested stem biomass as CO2 (excluding emissions from drained peatland soils)26. According to a review27, the ratio between forest carbon stock change and change in the harvesting varied over time and across regions. Increasing harvestings by 1 unit of C would decrease the forest carbon sink by 1.6 units of C on average27 (with high variability and a standard deviation of 0.8–1.1, based on the statistical cut-off). This is supported by our results regarding the case, where staying with current management, where the ratio between changes in tree biomass sink and in harvests varied between 1.2 and 1.6. The ratio varied between 1.8 and 2.7 with management change from the current management to CCF; the unit is Δ Tg CO2 eq. per Δ harvested mill. m3. These values convert to 1.45–1.94 (staying with current management) and to 2.18–3.27 (from current management to CCF), respectively, unit being as change in carbon sink relative to change in harvested stem biomass in carbon, assuming a biomass expansion factor of 0.45 Mg m−3 and a carbon content of 50%28.
Our results show that for tree biomass, the relative climate impact, as a ratio between changes in sink and harvesting amount reported in this study is much larger than those reported earlier27 when harvesting reduction is a result of management practice change (current management to CCF). Likely, the reason for this relates to the fact that the conversion from current management to CCF results in a larger area managed with lower intensity by selection harvests and in a smaller total clear-cut area, leading to a smaller area of young seedling stands with negligible volume growth and marginal tree biomass carbon accumulation.
The impact of clear-cut emissions
The inclusion of additional clear-cut emissions of CO2 and N2O based on22,23,24 resulted in higher GHG emissions from drained peatland soils at national level than those reported under the GHG inventory6. Our estimates for CO2 emissions as a result of peat and litter decomposition varied between 9.2 and 10.2 Tg CO2, depending on the period, whereas the reported emissions for Finland in the year 2021 were 8.2 Tg CO25. This difference can be mainly attributed to the emissions from clear-cut areas as the annual CO2 emissions during the first 5 years after clear-cut on nutrient-rich drained peatlands vary between 8 and 31 Mg CO2 per ha, whereas the total annual clear-cut area on drained peatlands reached 30,000 ha per year in the BAU scenario. Also, the contribution of the N2O emissions to the total GHG exchange of drained peatlands was larger in our study compared with that of the GHG inventory29, where N2O emission estimates are solely based on data from tree-covered study sites, excluding those that have been recently clear-cut30.
Regional results on the GHG exchange
Our regional results on the climate benefits of converting from BAU to CCF in nutrient-rich peatlands can be roughly divided into three categories: (i) regions with no significant differences, (ii) regions where benefits were found with tree biomass and (iii) regions where benefits were found with tree biomass and organic soils. Typically, regions where climate benefits were not found were those without significant areas of nutrient-rich drained peatland sites, whereas the regions where benefits were observed, especially regarding tree biomass, were those with high recent harvesting levels (e.g., in 2016–2018) (actual harvestings larger than the maximum sustained yield). Results of this work could be utilised when disseminating results, planning incentives, streamlining exiting policy instruments and when prioritising regions for conversion from current management to CCF of nutrient-rich drained peatlands.
Future development needs for the scenario modelling of CCF
Our modelling framework has been built on various published models that have been here linked together. However, a clear drawback with our approach is that the forest simulator MELA lacks the feedback from soil conditions (beyond static forest site fertility type), weather variability (beyond annual temperature sum and site location) and understorey vegetation competition to growth, which means that, e.g., dynamic nutrient availability, elevated peatland water table and droughts on mineral soils are not reflected in the simulated tree increment. Overcoming these shortcomings should be a future research priority. For tree increment we used empirical growth models of MELA for peatlands to describe the development of the stands after selection harvestings. To evaluate the potential of reduced growth from selection harvestings and the impact of growth models, we conducted a sensitivity analysis (CCF-reg scenario) where we downscaled the growth of suppressed Norway spruce trees by 25% for 5 years after selection harvesting. This resulted in the loss of climate benefits that were shown by the CCF scenario compared by the BAU scenario. The assumed 25% growth reduction for 5 post-harvest years is highly uncertain and mimics the potential release effect of diameter growth, which typically occurs as a delay after management treatment; however, it is likely an overestimation of the reduction in carbon uptake after selection harvest18.
Uncertainty related to GHG exchange of the scenarios
According to the GHG inventory of Finland, the total uncertainty for the GHG exchange of forests land was 10.6 Tg CO2 (83%)29, which can be considered as a minimum estimate for the uncertainty with model-based values presented here for Finland. According to a Monte Carlo simulation study, standard deviation of forest carbon sink in Finland was 27% for 1990–200431 and it was found that soil model initialisation was the most influential parameter for total uncertainty. For mineral soils, in the GHG inventory it was reported that the propagated uncertainty (uncertainty of sampling, biomass model, litter rates and soil model parameters) of the soil carbon sink was 26–32% depending on the region in Finland32. For drained peatlands soil emissions uncertainties are relatively high in the GHG inventory, but still less than actual estimate showing that drained peat soils are an emissions source in Finland6. Our findings are based on comparisons between the BAU and CCF scenarios, and we can assume that the uncertainties of the estimates of these different scenarios are highly correlated20; consequently, the uncertainty in the difference between the scenarios is smaller. The major uncertainty regarding our results in addition to growth models of suppressed trees (see CCF-reg sensitivity analysis) is based on the estimation of CO2 and N2O emissions that originate after clear-cutting from drained peatland sites. To illustrate the impact of the chosen assumptions, we used a linear decrease of soil CO2 emissions for 8 years, instead of 10 years, showing that the difference between BAU and CCF would narrow down by 0.3 Tg CO2 per year if additional emissions after clear-cutting would reduce faster than assumed here (Supplementary Fig. S2 in the Supplementary Material). In contrast, using a 50% lower emission factor for N2O for nutrient-rich drained peatlands after clear-cutting would imply an approximately 0.6 Tg CO2 eq. smaller difference between BAU and CCF (Supplementary Fig. S1 in the Supplementary Material). This underlines the need for additional studies of GHG fluxes, especially N2O, after clear-cutting. Simultaneously, these extremely high CO2 and N2O emissions from clear-cut areas highlight the fact that even-aged forestry is questionable on these nutrient-rich drained peatlands and that alternative management strategies should be tested. We also highlight the need for a mechanistic model with mass-balance for drained peatland ecosystems, allowing the estimation of CO2, CH4 and N2O exchange.
There is an urgent need for land-based climate change mitigation strategies, and our results show that CCF forestry on nutrient-rich drained peatlands has potential for emission reduction without compromising wood production despite the leakage of harvestings to mineral soils.
Methods
To estimate the future development of GHG exchange of Finnish forests, we followed the methods and system boundaries of the Finnish GHG inventory for forest land29 with a few exceptions. Our simulations include productive forests, where growth is more than 1 m3 ha−1, equivalent to the national forest definition, whereas in the GHG inventory, they apply a slightly broader forest definition by the FAO. The total area of Finnish forests (productive and poorly productive forests) used in this study was 22.8 mill. ha, according to the National Forest Inventory (NFI) (2014–2018), and the area of nutrient-rich drained peatlands was approximately 2 mill. ha. The regions with the largest coverage with drained peatlands were North Ostrobothnia, Lapland, Kainuu, and North Karelia (see Supplementary Tables S7 and S8).
Our simulations assume constant land-use, and therefore, all potential future land-use changes were excluded. The main methodological difference between our work and the GHG inventory is that here GHG exchange of drained peatland forest soils was estimated with the water table depth-driven models and with models that consider increased soil emissions after clear-cut harvesting. In addition, these estimations were conducted separately for each calculation unit based on the NFI data (stand scale) and then upscaled to regional scale, whereas in the GHG inventory, the carbon stock change and GHG exchange estimates were determined at the larger regional scale, using regional mean values. Here, CH4 emissions and CO2 sinks of undrained peatland soils were excluded, whereas drained peatland soils were included, as in the GHG inventory (Fig. 5).
The simulated projections were initiated for the year 2016 using the National Forest Inventory data (NFI12)7, representing forests of Finland and measured in 2014–2018 for southern Finland and in 2012–2013 for northern Finland. The total number of calculation units on forests was 57,720, and they were used with the MELA forestry model (of which 12,246 were in drained peatland forests). These calculation units were classified into three categories according to restrictions concerning wood production: (1) no restrictions for wood production, (2) restrictions for wood production and (3) no wood production allowed. This classification defines the management operations that are allowed for each calculation unit.
MELA modelling framework and scenarios
The future development of forest resources in Finland was estimated using the MELA forestry model for 2016–2050, version MELA201633,34. With MELA, we simulated a large number of management schedules for each calculation unit throughout the calculation period, based on the individual tree growth and development models35, forest management guidelines36 and estimates of the economics using costs for management actions and prices for harvested wood37. Different scenarios for the forests were obtained from the simulated management schedules with the linear optimisation package33,38 under the chosen objectives and constraints37. Development of forests resources was conditional on the harvesting levels that changed the wood use and tree increment, according to the simulated harvestings. The results of the MELA modelling were used to determine the GHG exchange of forests growing on mineral soils and peatlands.
In this study, we had two scenarios for the harvesting level: (i) maximum sustained yield and (ii) actual harvesting (MELA 2021). The maximum sustained yield scenario considers the long-term wood production and the economics. The harvested removal was defined by maximising the net present value of the 4% discount rate subject to non-declining periodic total roundwood and energy wood removals, saw log removals and net income (MELA 2021). With the maximum sustained scenario, the harvesting volumes ranged between 76 and 90 mill. m3 annually. The actual harvesting scenario was estimated by using annual harvests equal the mean level of harvests from 2016 to 2018, being 73.6 mill. m3.
In addition to the two scenarios for harvest level, we had two different scenarios for the applied forest management regime. The business as usual (BAU) scenario mimics rotation forestry with thinning and clear-cutting, which is the most common regime in Finnish forestry, and in this scenario, CCF management was not allowed. The alternative management scenario (CCF) is based on the BAU scenarios but with the additional constraint that the nutrient-rich drained peatland forests will only be managed without clear-cuts, i.e., they were transferred to continuous cover forestry when MELA conducted harvesting operations and when the minimum basal area thresholds were met. Nutrient-rich drained peatlands were defined as sites being of the Myrtillus type or herb-rich type sites, representing the most productive peatland sites for wood production, excluding Scots pine-dominated Myrtillus sites39.
To study the sensitivity of the tree growth to the results, a scenario with decreased growth of previously suppressed Norway spruce after selection harvesting was tested, termed as CFF-regressed (CCF-reg) scenario. In this scenario, it was assumed that other species than Norway spruce grew according to the default model estimates. The CCF-reg scenario is a modification of the CCF scenario by reducing the growth of suppressed Norway spruce by 25% during the first 5-year period after the selection harvest. The reduction of 25% is an expert judgement but an conservative estimate as, most likely, the reduction in growth is not greater than 25%, as described for the Lettosuo drained peatland area18.
The time step for MELA simulations was 10 years for tree biomass, whereas that for soil carbon and GHG exchange was 1 year. The simulation results were presented at minimum with 5-year periods, where linear interpolation between the MELA simulations for tree biomass was assumed, and then merged with the annual soil GHG exchange data.
Implementation of selection harvests in the MELA system
Selection harvest practices were implemented into MELA by modifying thinning routines and applying selection harvesting with the CCF (and CCF-reg) scenario on those calculation units where clear-cuts were not allowed. In the selection harvest, the trees were removed also from the dominant canopy layers compared with normal thinning, where harvested trees were selected from the lower canopy layers, mainly targeting smaller trees.
Selection harvests were allowed once the stand basal area for the plot had achieved a threshold of 22 m2 per ha. The stand would be harvested until the basal area was taken down to 12 or 15 m2 per ha at minimum in southern and northern Finland, respectively, in Norway spruce-dominated stands. In Scots pine-dominated stands, the post-harvest basal area was at minimum 14 or 17 m2 per ha in southern and northern Finland, respectively. For sites dominated by deciduous tree species, selection harvest was allowed when the basal area was more than 16–21 m2, depending on the region and soil fertility. The basal area of deciduous stands after harvesting varied between 10 and 15 m2, depending on the region and the site’s fertility. These harvesting guidelines were applied according to the dominant species, site fertility and region, see thinning rules in Table 5 and Supplementary Fig. S2 in the Supplementary Material. Clear-cuttings (for BAU all lands and in CCF, excluding nutrient-rich drained peatlands) were allowed when stand age or mean stand diameter fulfilled the criteria presented in the silvicultural guidelines40,41. Stand age and mean stand diameter criteria for stand maturity vary according to the region, site type and tree species.
Estimation of tree biomass carbon stock change for all lands
The forest simulator, MELA estimates tree growth, waste wood (tops of stems and lost logs), natural mortality and harvesting removals for each scenario, by regions and tree species, for 10-year periods (MELA 2021). To estimate the annual carbon sink of trees, stem volume estimates were converted to biomass and carbon. The conversion from stem volume to biomass was conducted using group-specific biomass expansion factors (BEFs) for tree species, which have been developed for the GHG inventory of Finland and vary according to the region and type of biomass (i.e., growth, natural mortality and harvesting), see Appendix 142. These BEFs were estimated from the sample trees of the NFI; for increment BEFs, increment cores and height increment measurements were used for estimating tree biomass and volume (present and 5 years ago), allowing BEF derivation for the increment. For natural mortality and harvesting BEFs, permanent sample data were used, and BEFs were estimated for trees that were either harvested or had died between consecutive inventories. For biomass estimation, empirical models43,44 were used, and the model version was chosen according to the availability of the model predictors.
Estimation of mineral soil carbon exchange
The mineral soil carbon stock change was estimated by using the Yasso07 model45. The principles of generating litter inputs and the use of the environmental forcing data and the applied parameters followed those of the GHG inventory of Finland29. The main difference between the GHG inventory of Finland and this study was the fact that here, simulations were conducted for each calculation unit separately, not based on regional averages as in the GHG inventory. For details of the application of the Yasso07 model to mineral soils, see the Supplementary Material.
Estimation of GHG exchange of drained peat soils
To estimate the GHG exchange from drained peatland forests, we applied a stepwise modelling approach. For each calculation unit, we organised data on future scenarios by the MELA model as input to the hydrological SpaFHy-Peat model10,13,46 to calculate the daily water table levels. Resulting ditch depths and leaf area indices (LAI) drove water table levels. With the BAU management ditches were deeper, but LAI values were lower, resulting water table levels that were deeper for CCF (see Supplementary Fig. S3 in the Supplementary Material). Thereafter, the annual GHG exchange was estimated based on the mean growing season (mean of May–October) water table levels and site type, coupled with empirical soil GHG exchange models11,12,30. We also incorporated a temporary increase in CO2 and N2O emissions following clear-cut actions and the CH4 emissions from ditches. Further, the decomposition and CO2 exchange of harvest residues and natural mortality on the oxic layer in drained peatlands were quantified employing the Yasso07 model. Thereafter, the emissions were summed to the regional level by using the emissions and sinks per area and the area represented by each calculation unit. To convert other GHGs to CO2 equivalents, global warming potential values of 25 and 298 for CH4 and N2O, respectively, were applied, based on the IPCC Assessment Report 447. For details on the GHG exchange estimation in drained peatland soils, see the Supplementary Material.
Data availability
Plot-level national forest inventory data (NFI11 and NFI12) are available through opendata.luke.fi portal. For tree-level data, please get in touch with the national inventory team of the Natural Resources Institute Finland.
Code availability
The SpaFHy-Peat soil hydrological model is available through Github: https://github.com/LukeEcomod/SpaFHy-Peat. The Yasso07 soil carbon model is available through Github: https://github.com/YASSOmodel. For the MELA forest simulator, please get in touch with the model developing group of the Natural Resources Institute Finland.
References
Hugelius, G. et al. Large stocks of peatland carbon and nitrogen are vulnerable to permafrost thaw. Proc. Natl. Acad. Sci. 117, 20438–20446 (2020).
Päivänen, J. & Hånell, B. Peatland Ecology and Forestry—A Sound Approach (Helsingin yliopiston metsätieteiden laitos, 2012).
Griscom, B. W. et al. Natural climate solutions. Proc. Natl. Acad. Sci. 114, 11645–11650 (2017).
Mires and Peatlands of Europe (Schweizerbart Science Publishers, 2017).
Statistics Finland. Greenhouse gas emissions in Finland 1990 to 2021. National Inventory Report under the UNFCCC and the Kyoto Protocol. 537 (2023).
Alm, J. et al. A new method for estimating carbon dioxide emissions from drained peatland forest soils for the greenhouse gas inventory of Finland. EGUsphere. https://doi.org/10.5194/egusphere-2022-1424 (2022). Accessed 12 September 2023 (preprint).
Korhonen, K. T. et al. Forests of Finland 2014–2018 and their development 1921–2018. Silva Fenn. 55(5), 10662 (2021).
Juutinen, A. et al. Profitability of continuous-cover forestry in Norway spruce dominated peatland forest and the role of water table. Can. J. For. Res. 51, 859–870 (2021).
Nieminen, M. et al. Could continuous cover forestry be an economically and environmentally feasible management option on drained boreal peatlands?. For. Ecol. Manag. 424, 78–84 (2018).
Leppä, K. et al. Selection cuttings as a tool to control water table level in boreal drained peatland forests. Front. Earth Sci. 8, 576510 (2020).
Minkkinen, K., Ojanen, P., Koskinen, M. & Penttilä, T. Nitrous oxide emissions of undrained, forestry-drained, and rewetted boreal peatlands. For. Ecol. Manag. 478, 118494 (2020).
Ojanen, P. & Minkkinen, K. The dependence of net soil CO2 emissions on water table depth in boreal peatlands drained for forestry. Mires Peat https://doi.org/10.19189/MaP.2019.OMB.StA.1751 (2019).
Launiainen, S., Guan, M., Salmivaara, A. & Kieloaho, A.-J. Modeling boreal forest evapotranspiration and water balance at stand and catchment scales: A spatial approach. Hydrol. Earth Syst. Sci. 23, 3457–3480. https://doi.org/10.5194/hess-23-3457-2019 (2019).
Leppä, K. et al. Vegetation controls of water and energy balance of a drained peatland forest: Responses to alternative harvesting practices. Agric. For. Meteorol. 295, 108198 (2020).
Juutinen, A. et al. Profitability of continuous cover forestry in Norway spruce-dominated peatland forest and the role of water table. Can. J. For. Res. https://doi.org/10.1139/cjfr-2020-0305 (2020).
Shanin, V. et al. Simulation modelling of greenhouse gas balance in continuous-cover forestry of Norway spruce stands on nutrient-rich drained peatlands. For. Ecol. Manag. 496, 119479 (2021).
Groot, A. & Hökkä, H. Persistence of suppression effects on peatland black spruce advance regeneration after overstory removal. Can. J. For. Res. 30, 753–760 (2000).
Lehtonen, A. et al. Fast recovery of suppressed Norway spruce trees after selection harvesting on a drained peatland forest site. For. Ecol. Manag. 530, 120759 (2023).
Sievänen, R. et al. Carbon stock changes of forest land in Finland under different levels of wood use and climate change. Ann. For. Sci. 71, 255–265 (2014).
Soimakallio, S., Kalliokoski, T., Lehtonen, A. & Salminen, O. On the trade-offs and synergies between forest carbon sequestration and substitution. Mitig. Adapt. Strateg. Glob. Change 26, 4 (2021).
Kallio, A. M. I., Salminen, O. & Sievänen, R. Sequester or substitute—Consequences of increased production of wood based energy on the carbon balance in Finland. J. For. Econ. 19, 402–415 (2013).
Korkiakoski, M. The Short-Term Effect of Partial Harvesting and Clearcutting on Greenhouse Gas Fluxes and Evapotranspiration in a Nutrient-Rich Peatland Forest (Finnish Meteorological Institute, 2020). https://doi.org/10.35614/isbn.9789523361300.
Korkiakoski, M. et al. Partial cutting of a boreal nutrient-rich peatland forest causes radically less short-term on-site CO2 emissions than clear-cutting. Agric. For. Meteorol. 332, 109361 (2023).
Mäkiranta, P., Riutta, T., Penttilä, T. & Minkkinen, K. Dynamics of net ecosystem CO2 exchange and heterotrophic soil respiration following clearfelling in a drained peatland forest. Agric. For. Meteorol. 150, 1585–1596 (2010).
Hirvelä, H., Härkönen, K., Salminen, O., Eyvindson, K. & Lempinen, R. Suomen metsien hakkuumahdollisuudet vuosina 2016–2045 valtakunnan metsien 12. inventointiin perustuen. 62 (2023).
Kalliokoski, T. et al. Skenaarioanalyysi metsien kehityastä kuvaanvien mallien ennustieden yhtäläisyyksistä ja eriosta. (2019).
Soimakallio, S. et al. Closing an open balance: The impact of increased tree harvest on forest carbon. GCB Bioenergy 14, 989–1000 (2022).
Lehtonen, A., Mäkipää, R., Heikkinen, J., Sievänen, R. & Liski, J. Biomass expansion factors (BEF) for Scots pine, Norway spruce and birch according to stand age for boreal forests. For. Ecol. Manag. 188, 211–224 (2004).
Statistics Finland. Greenhouse gas emissions in Finland 1990 to 2020. National Inventory Report under the UNFCCC and the Kyoto Protocol. 581 (2022).
Ojanen, P., Minkkinen, K., Alm, J. & Penttilä, T. Soil–atmosphere CO2, CH4 and N2O fluxes in boreal forestry-drained peatlands. For. Ecol. Manag. 260, 411–421 (2010).
Peltoniemi, M., Palosuo, T., Monni, S. & Mäkipää, R. Factors affecting the uncertainty of sinks and stocks of carbon in Finnish forests soils and vegetation. For. Ecol. Manag. 232, 75–85 (2006).
Lehtonen, A. & Heikkinen, J. Uncertainty of upland soil carbon sink estimate for Finland. Can. J. For. Res. 46, 310–322 (2016).
Hirvelä, H., Härkönen, K., Lempinen, R., Salminen, O. et al. MELA2016: Reference Manual (2017).
Kilpeläinen, H., Salminen, O., Härkönen, K., Miina, J. & Kurttila, M. Integrating bilberry yields into regional long-term forest scenario analyses. Scand. J. For. Res. 33, 378–386 (2018).
Hynynen, J. et al. Models for predicting stand development in MELA System. 1–116. https://jukuri.luke.fi/handle/10024/521469 (2002). Accessed 12 September 2023.
Äijälä, O., Koistinen, A., Sved, J., Vanhatalo, K. & Väisänen, P., Koistinen, A., Sved, J., Vanhatalo, K. & Väisänen, P. Metsänhoidon suositukset [The Good Practice Guidance for Forestry in Finland], vol. 2019 (Tapio Oy, 2014).
Luke. MELA Tulospalvelu. VMI12 (mittausvuodet 2014–2018). (2021).
Lappi, J. JLP: A Linear Programming Package for Management Planning. 1–134 (1992).
Vasander, H. & Laine, J. Site type classification on drained peatlands. In Finland—Fenland (eds. Riitta, K., Leila, K., Sakari, S.) 146–151 (Finnish Peatland Society Maahenki, 2008).
Äijälä, O., Koistinen, A., Sved, J., Vanhatalo, K. & Väisänen, P. Metsänhoidon suositukset. Tapion julkaisuja. 253 (2019).
Koistinen, A. et al. Metsänhoidon suositukset energiapuun korjuuseen, työopas. Tapion julkaisuja. Julkaistu Verkossa 2, 2016 (2016).
Lehtonen, A. et al. Maankäyttösektorin ilmastotoimenpiteet: Arvio päästövähennysmahdollisuuksista. (2021).
Repola, J. Biomass equations for Scots pine and Norway spruce in Finland. Silva Fenn. 43, 625–647 (2009).
Repola, J. Biomass equations for birch in Finland. Silva Fenn. 42, 605–624 (2008).
Tuomi, M., Rasinmäki, J., Repo, A., Vanhala, P. & Liski, J. Soil carbon model Yasso07 graphical user interface. Environ. Model. Softw. 26, 1358–1362 (2011).
Eyvindson, K. et al. Trade-offs between greenhouse gas mitigation and economic objectives with drained peatlands in Finnish landscapes. Can. J. For. Res. 53, 444–454 (2023).
IPCC. Climate Change 2007: The Physical Science Basis. (2007).
Acknowledgements
We thank the field staff of the Natural Resources Institute Finland for contributing to this study by collecting National Forest Inventory data used in this study and for developing and maintaining the MELA forest simulation and optimisation system. This work was supported by Academy of Finland funding for the SOMPA project “Novel soil management practices—key for sustainable bioeconomy and climate change mitigation”, decisions nr. 312912 and 336570, and by funding for the BiBiFe project “Biogeochemical and biophysical feedbacks from forest harvesting to climate change” (nr. 325680). This study received funding from the UNITE flagship of the Academy of Finland (decision 337655). Also, the Ministry of Agriculture and Forestry of Finland funded this work under the “Catch the Carbon program” (Project Tuima, VN/28648/2020-MMM-3).
Author information
Authors and Affiliations
Contributions
A.L., M.P., R.M. and K.H. planned and designed the study. A.L. led the writing of the manuscript and conducted the numerical analysis. K.H. and O.S. contributed to the analysis by running the MELA model. K.L. and S.L. contributed by supporting the SpaFHy-peat implementation. K.E. performed SpaFHy-peat and Yasso07 simulations. A.S. estimated ditch spacing for Finland by the management unit, while M.R. helped with Finnish NFI data. All authors contributed to the writing of the manuscript and gave final approval for its publication.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Lehtonen, A., Eyvindson, K., Härkönen, K. et al. Potential of continuous cover forestry on drained peatlands to increase the carbon sink in Finland. Sci Rep 13, 15510 (2023). https://doi.org/10.1038/s41598-023-42315-7
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-023-42315-7
- Springer Nature Limited