Continuous cover forest management decreases nutrient loads to water courses

Continuous cover forest management decreases nutrient loads to water courses | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Continuous cover forest management decreases nutrient loads to water courses Eppu Honkanen, Mika Nieminen, Tero Heinonen, Timo Pukkala This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5063189/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This study looked at the effects of even-aged forest management (BAU), continuous cover forestry (CCF), and any-aged forestry (AAF) on the loads of phosphorus and nitrogen to downstream water bodies when these loads were minimized by forest planning. The impact of aiming at minimal nutrient loads instead of maximal economic profit was inspected within each of the three silvicultural systems. The analyses were conducted with and without even-flow-cutting constraints. The data for the calculations was a random sample of forest stands, representing two municipalities in North Karelia, eastern Finland. The results showed that the transport of nitrogen and phosphorus from managed commercial forests to water courses can be substantially reduced by increased use of continuous cover forestry. Significant reductions were possible in both peatland forests and mineral sois. Nutrient loads could be reduced also in even-aged BAU management by decreasing the use of clear-felling, particularly in spruce mires. Nutrient loads increased with increasing harvests in all silvicultural systems. Significant reductions in nutrient loads can be achieved with a reasonably small loss in the profitability of forest management. Optimal use of continuous cover management makes it possible to simultaneously increase economic profitability and decrease nutrient loads. Forest planning nutrient load landscape level nitrogen phosphorus optimization Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Introduction Recent studies suggest that forestry is a significantly greater source of nitrogen and phosphorus to water courses than previously thought (Nieminen et al. 2017b, 2018c, 2022; Finér et al. 2021). In Finland, for example, it was estimated that forestry on peatlands accounts for only 1–3 percent of all human-induced nutrient loads (Finér et al. 2010), but recent studies suggest that this percentage may be as much as 15–20% (Nieminen et al. 2020a). Many studies have further indicated that nutrient concentrations in waters discharging from drained peatland forests are still increasing (Nieminen et al. 2017b, 2018c, 2022; Räike et al. 2019). A significant additional water quality problem in the waters discharging from forested catchments is their brownification due to enhanced loads of dissolved organic carbon and iron (Finstad et al. 2016; Škerlep et al. 2021; Williamson et al. 2021). Wide-spread brownification of waters has most often been attributed to recovery from acid deposition (Monteith et al. 2007; Erlandsson al. 2008) or climatic warming (Sarkkola et al. 2009; Laudon et al. 2012), but recent studies suggest that forests and forestry also contribute to increased organic carbon loads. Forest growth has increased particularly in the northern latitudes due to intensified silviculture and climate warming, resulting in increased carbon stocks in soils and forests, and consequently increased carbon losses from forests to surface waters. In peatland-dominated areas, such as large parts of Scandinavia, the Baltic states, and the British Isles, drainage of peatlands for forestry and management of peatland forests also contribute to organic carbon loads and water brownification (Nieminen et al. 2021; Härkönen et al. 2023). While the awareness of the effects of forestry on nutrient and carbon loads to watercourses has increased, it has also been acknowledged that current water protection measures may be inefficient in retaining nutrients and carbon (Nieminen et al. 2018; Härkönen et al. 2023). Most protection measures only retain suspended sediments and adhered nutrients and do not affect the loads of dissolved nutrients and dissolved carbon. The study by Nieminen et al. (2024) suggests that the only means to effectively decrease the loads of dissolved nutrients in forested catchments is to convey water to extensive wetland buffers. However, their study further suggests that it may only be feasible to use wetland buffers to reduce phosphorus loads as very large buffers are needed to significantly reduce nitrogen and organic carbon loads. Constructing wetland buffers by restoration of drained sites is also challenging in the sense that it initially increases rather than decreases nutrient and carbon loads (Koskinen et al. 2017; Nieminen et al. 2020). The problems related to the efficiency and operational use of water quality protection practices have raised a question of whether it would be more feasible to decrease forestry-induced release of dissolved carbon and nutrients rather than try to capture them from discharged water flow (Nieminen et al. 2018, 2024; Härkönen et al. 2023). It has been proposed that managing forests with continuous cover forestry (CCF) instead of the prevailing even-aged forestry management could be an effective means of decreasing forestry-induced nutrient and carbon loads (Nieminen et al. 2018a, 2024; Härkönen et al. 2023). CCF decreases nutrient loads particularly because it avoids large clear-cuts and eliminates the need for regular ditch cleanings in drained sites (Nieminen et al. 2018a). It may also maintain water tables higher than in dense even-aged peatland forests, which reduces aerobic nutrient mineralization (Ojanen and Minkkinen 2019) and the leaching of nutrients and carbon from deep peat layers (Nieminen et al. 2018a; Nieminen et al. 2022). A recent simulation study by Nieminen et al. (2023), which covered all privately owned forests of Finland, suggested that nutrient loads are lower from CCF than the prevailing even-aged forestry practices (referred to as BAU, i.e., business as usual). In their study, the silvicultural systems were implemented by following the cutting guidelines developed for them. However, as the cutting guidelines aim to maximize wood production or economic profit, simulations based on them do not show the performance of a silvicultural system when one target of forest management is to minimize nutrient loads. Previous analyses have also not covered so-called any-aged forestry (AAF) where management is not classified as either even-aged or continuous cover forestry but has features of both systems (Haight & Monserood 1990a, 1990b; Pukkala 2022). A problem with simulations based on silvicultural guidelines is also that they may result in different harvest levels in different silvicultural systems, making it difficult to separate the effect of the silvicultural system from that of the harvest level. As an alternative approach, optimization makes it possible to find efficient Pareto-optimal combinations of variables of interest. Pareto-optimality refers to management where a management objective is achieved with the smallest possible loss in another, conflicting objective. Moreover, comparisons can be conducted under different constraints. A common constraint in large-area analyses is the even flow of harvested timber. The current study used optimization to analyze the effect of timber harvests and profit maximization on nutrient loads to water bodies. Optimization allows fair comparisons of different silvicultural practices since the harvested volume can be constrained to be the same in all systems, and all systems can aim at the same management objectives. Another addition, as compared to earlier research, was the analysis of any-aged forest management. Since all cutting types of both even-aged rotation forestry and CCF are possible in AFF, we hypothesize that AFF is the most efficient silvicultural system when the aim is to maximize economic profit while concurrently minimizing forestry-induced nutrient loads. The study first analyzed the relationship between economic profit and nutrient loads in Pareto optimal forest management, separately in CCF and BAU. Second, the possibilities to reduce nutrient loads with fixed harvest removals were analyzed, separately for BAU, CCF, and AAF. This was done by minimizing nutrient loads instead of maximizing economic profit while keeping the periodical harvest volume constant. The study utilizes private forests from two municipalities, with Kitee representing a region with a large share of drained peatland forests and Liperi representing a region where their share is small. Building upon the results of Nieminen et al. (2023), our research hypothesis assumes that AAF and CCF provide more possibilities than BAU to decrease nutrient loads from managed forests to water systems. Materials and methods Calculation of nutrient loads This study used the same methods as Nieminen et al. (2023, 2024) in estimating the effects of CCF, AAF and BAU on the loads of nitrogen (N) and phosphorus (P) to receiving watercourses. We assume as in previous studies that forestry increases nutrient loads due to the long-term legacy effect of past peatland drainage operations (Nieminen et al. 2017b, 2018c, 2022; Räike et al. 2019), as well as the shorter-term effects of ditch network maintenance (DNM), forest fertilization, and harvesting. We further assume that the long-term legacy effect of drainage is controlled by nutrient mineralization (Nieminen et al. 2022), which, in turn, is controlled by site drainage conditions, that is, the soil water table level (Ojanen and Minkkinen 2019). As in the studies by Nieminen et al. (2023, 2024), the estimation of the legacy effect of drainage begins by predicting the water table level (WTL) during late summer conditions for all forest compartments that represent drained peatland forests (Fig. 1A). This is done by the empirical equation by Sarkkola et al. (2010), which predicts WTLs as a function of the growing stock volume (its evapotranspiration demand), monthly precipitation, ditch depth, and the location of the site (latitude). The legacy effect of drainage is then obtained as the relationship between WTL and N and P loads (Fig. 1B), as calculated by the SUSI peatland simulator (Laurén et al. 2021) for sites with differing fertility (Nieminen et al. 2023). The combined use of the two models shows the effect of growing stock volume on the legacy effect of drainage (Fig. 1C). The shorter-term effects of ditch network maintenance (DNM), fertilization, and clear-felling of mineral soil forests on N and P loads (kg ha -1 year -1 ) were estimated using the load coefficients by Nieminen et al. (2024). All these effects are temporary; the effects of clear-felling and DNM last for ten years, the annual loads peaking in the first year and decreasing gradually towards the tenth year. Fertilization-induced phosphorus loads in drained peatland forests last for five years (Finér et al. 2010; Nieminen et al. 2023) and fertilization-induced nitrogen loads in mineral soil forests for two years (Saura et al. 1995). The nutrient loads caused by the harvesting of peatland forests were calculated with the empirical models by Nieminen et al. (2023) for the relationship between harvested stem wood volume and harvest-induced nutrient loads. According to their models, small stem wood removals typical to BAU thinning treatments and CCF harvests (removed volume less than 100 m 3 ha -1 ) result in small nutrient loads, but nutrient loads increase when the volume of harvested stem wood approaches the volumes typical to clear-felling (Fig. 2). The excess loads induced by harvesting were assumed to last for six years on nutrient-rich peatland forests and four years on nutrient-poor peatland sites (Nieminen et al. 2023). Forest stand data Simulation and optimization were based on municipality-specific forest resource data from the two municipalities, Kitee (62°5'54.96''N, 30°8'15.00''E) and Liperi (62°31'59.88''N, 29°22'59.88''E). The data were obtained from the Finnish Forest Centre's open database, which focuses on private forests. For both municipalities, a 5% random sample was drawn from the forest resource data accounting for 4.4 % of all forests in Kitee and Liperi (Table 1). The average temperature in the region is about 4 °C, with -8°C in February, and 17 °C in July (https://en.ilmatieteenlaitos.fi/). The municipalities receive an average precipitation of approximately 600 mm, of which about 200 mm falls as snow (Jokinen et al. 2021). Table 1 Information about the stands selected for the sample. Kitee Liperi Number of stands 4068 2263 Total area of the sample, ha 4460.0 2379.0 Area of productive forest, ha 4380.3 2368.2 Share of peatland forest, % 21.9 12.6 Spruce mires, % 9.2 6.7 Pine bogs, % 12.8 5.9 Herb-rich site or better, % 38.0 34.1 Mesic site, % 44.7 48.0 Sub-xeric site, % 13.6 16.1 Xeric site, % 3.7 1.4 Mean volume, m 3 ha -1 159.7 165.1 Pine 56.7 57.9 Spruce 62.0 68.9 Birch 40.5 37.9 The share of peatland forest was 21.9 % in Kitee and 12.6 % in Liperi. In both municipalities, most stands represented Norway spruce ( Picea abies (L.) Karst.) and Silver birch ( Betula pendula Roth) dominated herb-rich and mesic sites. The share of sub-xeric and xeric Scots pine ( Pinus sylvestris L.) dominated sites was 17.3% in Kitee and 17.5% in Liperi. Simulation of forest management alternatives The development of forest stands and their management were simulated for 50 years using the models developed by Pukkala et al. (2021). The 50 years were divided into five 10-year spans. Harvestings and other treatments were simulated in the middle of each 10 years. The models by Pukkala et al. (2021) are based on the measurements of the permanent sample plots (PSPs) of the 10 th and 11 th National Forest Inventory (NFI) of Finland. The models describe the diameter growth of trees, the probability of tree survival, and the ingrowth. The ingrowth models describe natural regeneration and predict the number of new trees per hectare that pass the 1.3-m height limit over five years (Pukkala et al., 2021). The predictive variables in the Pukkala et al. (2021) models do not include stand age, dominant height, or site index based on age and dominant height, thus allowing simulations where dominant trees are removed in harvestings. The site variables in the models are temperature sum (>+5 °C threshold), an indicator variable for peatland, and site fertility classified as mesotrophic, herb-rich, mesic, sub-xeric, xeric, and barren. The treatments of the management schedules in the prevailing even-aged forestry (BAU) were based on the silvicultural guidelines by Äijälä et al. (2014). The guidelines for cuttings are based on dominant tree species, dominant height, mean tree diameter, and the total basal area of the trees (m 3 ha -1 ). The prevailing type of thinning in Finnish forest management is thinning from below but there is a gradual shift towards more versatile thinning options. To reflect this evolution of forest management, we simulated BAU with thinning from below, uniform thinning, and thinning from above, with probabilities of 0.7, 0.1, and 0.2, respectively. In thinning from below, half of the removed basal area was obtained using the same harvest rate for all diameter classes. The other half was obtained by removing trees from the smallest diameter classes until the recommended post-thinning basal area was reached. In thinning from above, half of the removed basal area was taken by thinning all diameter classes with the same intensity, and the other half was taken by removing the largest trees. In uniform thinning the thinning rate was the same in all diameter classes. In CCF and AAF, the guidelines by Äijälä et al. (2014) for even-aged management were applied to the first commercial harvesting of a young forest. Later on, optimization-based guidelines developed for all-aged forest management were used (Pukkala 2022). In these guidelines, it is first checked whether a stand is mature for harvesting, or whether it would be optimal to let the stand continue to grow. If the optimal decision is to harvest, the second part of the guideline tells whether the cutting is partial cutting (thinning) or final felling. In the case of thinning, the third part of the guideline determines the optimal harvest rate for different diameter classes. The only difference in simulating CCF and AAF was that clearcuttings and seed tree cuttings were ruled out in CCF. A common type of cutting in all management systems was the removal of the upper story in two-storied stands. This option was selected if the lower story of small (non-merchantable) trees formed a dense enough stand and the volume of the removed upper story was at least 50 m 3 ha -1 . The allowed removal in other partial cuttings was 50–200 m³ ha -1 . If the removal was less than 50 m³ ha -1 , the treatment was postponed, and if the removal was more than 200 m³ ha -1 , the guidelines for remaining basal area (BAU) or thinning intensity (CCF and AAF) were adjusted. In the thinning treatments of CCF and AFF, additional requirements were set for the minimum post-cutting basal area: mesic or better sites 11-12 m² ha -1 , sub-xeric sites 9-10 m² ha -1 , and xeric sites 7-8 m² ha -1 , the lower values representing peatland forests. The harvest rates of the guidelines were adjusted if they resulted in lower basal areas than these limits. The regeneration method for even-aged forestry on mesic and better sites involved clear-cutting, soil preparation, and planting. In the sub-xeric site, the regeneration method was clear felling, site preparation, and sowing of Scots pine seeds. Natural regeneration via seed trees was used in xeric and poorer sites. In mesic and herb-rich sites, the planted tree species were randomized. In herb-rich sites, the probability of planting Norway spruce was 0.8 and that of planting Silver birch 0.2. On mesic sites, the probabilities of planting pine, spruce, and birch were 0.2, 0.7, and 0.1, respectively. The growth of all planted trees was set at 10% better and the growth of seeded trees at 5% better than the growth of natural trees to account for their breeding effect. Several alternative 50-year management schedules were simulated for each stand to provide a decision space for forest-level optimization. The recommendations of Äijalä et al. (2014) give the range for the basal area at thinning (m 2 ha -1 ) and mean tree diameter at final felling (cm). The lowest limits of these ranges were used as the earliest possible timing for thinning and final felling. Alternative schedules were simulated by postponing the cutting by one or more 10-year periods. The guidelines for AAF include the discount rate as one factor that affects the timing and intensity of cuttings. Alternative CCF and AAF cutting schedules were obtained by applying the recommendations with a 1, 3, or 5 % discount rate. A no-cutting schedule was also simulated for all stands in all three silvicultural systems. Nitrogen fertilizations were simulated in the BAU management on mineral soils for spruce-dominated mesic sites and pine-dominated sub-xeric sites. Fertilization was simulated with a probability of 0.5 when the mean diameter of the trees was 20–30 cm, the basal area was 20–30 m 2 ha -1 and at least 10 years had passed since the previous fertilization (Heinonen et al. 2018). In practice, these two stand types were fertilized once during the rotation period (Nieminen et al. 2023). The amount of added nitrogen was 150 kg ha -1 . The response of tree growth to fertilization was simulated based on the research by Kukkola and Saramäki (1983). In drained peatland forests, phosphorus+potassium fertilization was simulated with a probability of 0.3 when the site fertility was sub-xeric or better, the mean tree diameter was 5–30 cm, the basal area ranged from 10 to 40 m²ha -1 , and at least 50 years had passed since the previous fertilization (Nieminen et al. 2023). Since the main purpose of this fertilization is to maintain tree growth by avoiding nutrient deficits, it was assumed that fertilization did not improve growth compared to the model prediction without fertilization. Phosphorus+potassium fertilizations in drained peatland forests were simulated in all silvicultural systems. Additionally, ditch network maintenance (DNM) was simulated for drained peatlands by assuming a similar growth effect as in Heinonen et al. (2018). DNM was simulated with a probability of 0.1 when the mean tree diameter of the stand was 5–30 cm, the basal area was 2–25 m² ha -1 , and at least 30 years had passed since the previous DNM. However, DNM was not simulated if the stand volume was >150 m³ ha -1 , as such mature stands with high evapotranspiration capacity maintain sufficient drainage, even in the case of poorly functioning ditches (Sarkkola et al., 2010). DNM was also simulated after clearcutting in BAU and AAF. Optimizations The simulated annealing metaheuristic was used to search for the best combinations of the simulated treatment schedules (Reeves 1993; Bettinger et al. 2002). A utility function was maximized (Pukkala and Kangas 1993). The utility function included one to three objective variables, which were the net present value of timber production ( NPV ), nitrogen load ( N load ), and phosphorus load ( P load ). NPV was maximized while the nutrient loads were minimized. An additive utility function was assumed: U = w 1 u 1 ( NPV ) + w 2 u 2 ( N l oad ) + w 3 u 3 ( P load ) where U is utility, NPV of the net present value calculated with a 4% discount rate (€), N load is the 50-year nitrogen load from the forest to water bodies (kg ha -1 ) and P load is the respective phosphorus load (kg ha -1 ), w 1 , w 2 and w 3 are, respectively, the weights of NPV , N load , and P load , and u 1 , u 2 and u 3 are sub-utility functions for the tree management objectives. Since the sum of the weights ( w 1 , w 2 and w 3 ) was 1 and the sub-utilities ranged from 0 to 1, the total utility ( U ) also ranged from 0 to 1. The sub-utility functions were linear and scaled the range of the variation of NPV , N load , or P load between 0 and 1 so that the minimum possible NPV yielded a value of 0 and the maximum a value of 1. For the nutrient loads, the minimum possible load yielded sub-utility 1, and the highest possible load 0. When analyzing the trade-off between NPV and nutrient loads, the weight of NPV ( w 1 ) was first set equal to 1 ( w 2 and w 3 were zero), and the optimization problem was solved. The solution gives the maximum possible NPV. Then, a part of the weight was gradually transferred to either N load or P load , and the problem was solved again. Solving all these optimization problems produced the trade-off curves between NPV and N load , or between NPV and P load . This analysis was carried out separately for BAU and CCF. The total harvested volume of every solution was also recorded, which made it possible to determine the nutrient load at different harvest levels. The rest of the analyses involved an even-flow harvesting constraint. The 10-year harvested volume was found by trial and error. The volume was equal to the maximum 10-year removal that could be removed in each silvicultural system during the whole 50-year time span. It was checked that this harvest removal did not decrease the growing stock volume of the forest during the 50-year planning horizon. The 10-year target removal of industrial wood (saw logs and pulpwood) was 67.5 m 3 ha -1 in Kitee and 71.5 m 3 ha -1 in Liperi. However, some flexibility was allowed since a constant harvest level may not be economically optimal as it ignores the effect on forest structure on the optimal cutting level. Sawmills and paper mills usually purchase timber from areas substantially larger than a municipality, which means that the annual harvest volume in a single municipality may vary. Therefore, the 10-year harvest volume was allowed to deviate 10 % from the target removals. Equation 1 was maximized with this constraint so that the weight of NPV in Equation 1 was either 1, 0.5, or 0, and the rest of the total weight was divided equally to N load and P load . The purpose was to see how much nutrient loads can be decreased with an even-flow harvest requirement when nutrient loads are minimized concurrently with profit maximization, or when nutrient loads are decreased as the only management objective. This analysis was conducted separately for BAU, CCF, and AAF. Results The trade-off curves between NPV and nutrient loads, when NPV was maximized and nutrient load minimized, show a clear superiority of CCF over BAU (Fig. 3). The lowest loads were nearly the same in BAU and CCF. However, when NPVs were maximized, the loads were nearly two times higher in BAU. The nutrient loads were higher in Kitee where the share of peatland forests was almost two times higher than in Liperi. It was possible to decrease nutrient loads in both municipalities and all silvicultural systems when forest planning aimed at minimizing nutrient loads without considering its impact on economic profit (Fig. 4). Managing forests with either AAF or CCF reduced nutrient loads on average by around 50% compared to the BAU management. On average, 76% of the nutrient loads in Kitee and 67% in Liperi were due to forest management on peatlands. Minimizing nutrient loads instead of maximizing economic profit decreased the nutrient loads of BAU by 32% in Kitee and 41% in Liperi. In CCF, the reductions were 24% (Kitee) and 30% (Liperi), and in AAF, 30% (Kitee) and 38% (Liperi). As expected, minimizing nutrient loads instead of maximizing NPV decreased the profitability of forest management (Fig. 5). However, the decrease in NPV was small, implying small costs for load reduction. AAF and CCF resulted in about 15% higher NPV than BAU, indicating that switching from BAU to CCF or AAF makes it possible to concurrently improve the profitability of forest management and decrease nutrient loads. The trade-offs obtained from the optimizations with the even-flow harvesting constraint are visualized in Figure 6. CCF and AAF were again more efficient than BAU in controlling nutrient loads, while concurrently maximizing the profit of forest management. As expected, AAF was the most efficient silvicultural system. However, the margin to CCF was small. In the management systems where clear-felling was a possible management option (BAU and AAF), the clear-felled area decreased significantly when more weight was given to load minimization in forest planning (Fig. 7). The decrease was more than 50% in BAU when the nutrient load was minimized in forest planning instead of maximizing NPV. The area of clear-felling decreased particularly in spruce mires, where almost 80% of the area was clearcut when NPV was maximized, but less than 10% when forest planning minimized nutrient loads. In AAF, the clear-felled area was 13% (Liperi) or 16% (Kitee) when NPV was maximized but less than 2% when nutrient loads were minimized either as the only objective or together with NPV maximization. In the BAU management, the area of DNM decreased by 56–73% when the target was to minimize nutrient loads (Fig. 8). The fertilization area was not sensitive to the weights of NPV maximization and nutrient load minimization (Equation 1). In AAF, the area of DNM during 50 years was small, at a maximum of 3% of the total area of peatland forests. Discussion The effects of forestry on nutrient loads to water courses have previously been assessed regionally using simulation (Finér et al. 2021; Nieminen et al. 2023, 2024). This was the first Finnish study where the minimization of nutrient loads was one management objective in optimization-based forest planning. The results showed that, within each of the three inspected silvicultural systems, it is possible to significantly reduce nutrient loads with reasonable economic losses. If the starting point is even-aged forestry, it is possible to concurrently decrease nutrient loads and improve profitability by increasing the use of CCF-type management. By shifting from BAU to CCF or AAF, it is easier to decrease nutrient loads from mineral soils than drained peatlands where not only clear-cuts but also partial harvests increase nutrient loads (Nieminen et al. 2024b), and where the long-term legacy effect of drainage dominates nutrient loads. The legacy effect is not particularly sensitive to forest management methods (Nieminen et al. 2023, 2024). In mineral soils, clear-felling and nitrogen fertilization were assumed to be the only sources of nutrient loads (Finér et al. 2010). As both were executed in BAU simulations and neither in CCF simulations, it is obvious that CCF was a better option in mineral soil forests from the water quality management viewpoint. However, the general assumption that partial harvests do not affect nutrient loads from mineral soil forests (Finér et al. 2010; Nieminen et al. 2024) still needs to be verified empirically. While the legacy effect of drainage is not highly sensitive to forest management methods in the long term, it may induce significant differences in nutrient loads in the shorter term. This is because the nutrient load was assumed to be caused by aerobic peat mineralization, which varies considerably depending on the level of the water table in the peat (Ojanen and Minkkinen 2019). The legacy effect of drainage was therefore significantly lower from young stands with low evapotranspiration (EVT) capacity and consequent high water table, compared to mature stands with high EVT and low water table in peat (Sarkkola et al. 2010). Therefore, although clear-fellings in the BAU management increase nutrient loads in the short term, they may decrease them temporarily in the longer term due to the low stocking and EVT of young even-aged forests. Later on, when the volume of the even-aged stand increases, the BAU management may again lead to higher nutrient loads than CCF. Although the effect of water table on nutrient loads has not been assessed empirically, the results by Nieminen et al. (2022) indicate significantly smaller nutrient concentrations in waters discharging from mature peatland stands with low water tables than for young stands with high water tables (Sarkkola et al. 2010). This study utilized municipality-specific harvest removal constraints to provide realistic removals and management scenarios. Optimization with even-flow harvesting constraints resulted in the replacement of clear-fellings by less harmful partial cuttings when management optimization aimed at reducing nutrient loads. One reason for the decreased clear-felling area in peatland forests when load minimization was prioritized is that ditch maintenance treatment is executed after every clear-felling. DNM causes a peak in nutrient loads and increases forest management costs. There is uncertainty for instance in simulating the emergence of new understorey trees, which are important for the sustainability of CCF. There is considerable variation in forest regeneration under tree canopies, both within the stand and between them. The development of even-aged, artificially established stands on drained peatlands is also uncertain because the area of drained peatland forests that have been clear-felled and planted is small and the plantations are still young. The assumption made in this study was that planted peatland forests develop nearly the same way as planted mineral soil forests of the same site fertility category. The empirical survival models used in the simulation (Pukkala et al. 2021) predict slightly lower tree survival for peatland forests compared to similar mineral soil forests. The predicted diameter increment of pine is lower in peatland forests, but similar as in mineral soils in spruce and birch forests. Uncertainty in our simulations is also introduced by climate change, which may increase forest damages, such as windthrows and water stress during prolonged droughts. Spruce, in particular, is sensitive to changing climatic conditions. Windthrows may be reduced by shifting from BAU to CCF (Pukkala et al. 2016), but the probability of some other damages, such as root rot and bark beetle outbreaks, may increase particularly if spruce becomes the dominant tree species. In the BAU management, spruce can be easily replaced by deciduous trees by clear-felling spruce and planting deciduous species. In general, however, planted monocultures, especially those of spruce, are seen as the most vulnerable stand type against beetle attacks and other biotic and abiotic hazards (Messier et al. 2019). Maintaining and increasing the carbon stocks and biodiversity of forests has been increasingly used as a management objective of forest planning (e.g., Mönkkönen et al. 2014; Heinonen et al. 2017), to increase the overall responsibility and acceptability of commercial forestry (Pukkala 2021). However, especially in Finland with a large area of drained peatland forests (20% of land area), sustainable and responsible forest management should also consider water quality. This can be done by including nutrient load minimization as one target of forest planning in the models used earlier for carbon sequestration and biodiversity maintenance ( Díaz-Yáñez et al. 2019). Such coherent planning is timely as forestry is responsible for considerable biodiversity loss (Mönkkönen et al. 2022) and water quality deterioration (Nieminen et al. 2022, 2023) in areas with intensive forest management and large areas of drained peatland forests, such as Finland, Sweden, the Baltic States, and the British Isles. Conclusions The study showed that the transport of nitrogen and phosphorus from managed commercial forests to water courses can be substantially reduced by increased use of continuous cover forestry. Nutrient loads can be reduced also in even-aged BAU management if the area of clear-fellings is reduced, particularly in spruce mires. Nutrient loads increase with increasing harvested volume in all silvicultural systems. Significant reductions in nutrient loads can be achieved with reasonably small economic losses. Optimal use of continuous cover management makes it possible to simultaneously increase economic profitability and decrease nutrient loads. Declarations Author Contribution All authors contributed to the study conception and design. T.P. prepared the software. T.P. and E.H. and did the analysis. T.P., M.N. and T.H. wrote the main manuscript text. All authors reviewed the manuscript. Funding No funding was received for conducting this study. Conflicts of interest/Competing interests No conflict of interests Availability of data and material Not applicable Code availability Not applicable References Bettinger P, Graetz D, Boston K, Sessions J, Chung W (2002) Eight heuristic planning techniques applied to three increasingly difficult wildlife planning problems. Silva Fennica 36:561–584 Díaz-Yáñez O, Pukkala T, Packalen P, Peltola H (2019) Multifunctional comparison of different management strategies in boreal forests. Forestry 93(1):84–95. https://doi.org/10.1093/forestry/cpz053 Finér L, Mattsson T, Joensuu S, Koivusalo H, Laurén A, Makkonen T, Nieminen M, Tattari S, Ahti E, Kortelainen P, Koskiaho J, Leinonen A, Nevalainen R, Piirainen S, Saarelainen J, Sarkkola S, Vuollekoski M (2010) Metsäisten valuma-alueiden vesistökuormituksen laskenta. Suomen ympäristö 10/2010. http://hdl.handle.net/10138/37973 Haight RG, Monserud RA (1990a) Optimizing any-aged management of mixed-species stands: II. Effects of decision criteria. Forest Science 36(1):125–144 Haight RG, Monserud RA (1990b) Optimizing any-aged management of mixed-species stands. I. Performance of a coordinate-search process. Canadian Journal of Forest Research 20(1):15–25 Heinonen T, Pukkala T, Mehtätalo L, Asikainen A, Kangas J, Peltola H (2017) Scenario analyses for the effects of harvesting intensity on development of forest resources, timber supply, carbon balance and biodiversity of Finnish forestry. Forest Policy and Economics 80:80–98. https://doi.org/10.1016/j.forpol.2017.03.011 Heinonen T, Pukkala T, Asikainen A, Peltola H (2018) Scenario analyses on the effects of fertilization, improved regeneration material, and ditch network maintenance on timber production of Finnish forests. Eur J Forest Res 137:93–107. https://doi.org/10.1007/s10342-017-1093-9 Hökkä H, Stenberg L, Laurén A (2020) Modelling depth of drainage ditches in forested peatlands of Finland. Balt For 26:220–228. https://doi.org/10.46490/BF453 Jokinen P, Pirinen P, Kaukoranta J-P, Kangas A, Alenius P, Eriksson P, Johansson M, Wilkman S (2021) Tilastoja Suomen ilmastosta ja merestä 1991–2020 (Climatological and oceanographic statistics of Finland 1991–2020). Raportteja (reports) 2021:8. Ilmatieteen laitos (Finnish Meteorological Institute), Helsinki, Finland. https://doi.org/10.35614/isbn.9789523361485 Kukkola M, Saramäki J (1983) Growth response in repeatedly fertilized pine and spruce stands on mineral soils. Comm Inst For Fenn 114:1–55. http://urn.fi/URN:ISBN:951-40-0622-4 Laurén A, Palviainen M, Launiainen S, Leppä K, Stenberg L, Urzainki I, Nieminen M, Laiho R, Hökkä H (2021) Drainage and stand growth response in peatland forests – description, testing, and application of mechanistic peatland simulator SUSI. Forests 12:293. https://doi.org/10.3390/f12030293 Mattsson T, Ahtiainen M, Kenttämies K, Haapanen M (2006) Avohakkuun ja ojituksen pitkäaikaisvaikutukset valuma-alueen ravinne- ja kiintoainehuuhtoumiin. Julkaisussa: Kenttämies K, Mattsson T (toim) Metsätalouden vesistökuormitus. MESUVE-projektin loppuraportti. Suomen ympäristö 816:73–81. http://hdl.handle.net/10138/40492 Messier C, Bauhus J, Doyon F, Maure F, Sousa-Silva R, Nolet P, Mina M, Aquilé N, Fortin M-J, Puettmann K (2019) The functional complex network approach to foster forest resilience to global changes. For Ecosyst 6:21. https://doi.org/10.1186/s40663-019-0166-2 Mönkkönen M, Juutinen A, Mazziotta A, Miettinen K, Podkopaev D, Reunanen P, Tikkanen O, Kouki J (2014) Spatially dynamic forest management to sustain biodiversity and economic returns. J Environ Manage 134:80–89. https://doi.org/10.1016/j.jenvman.2013.12.021 Mönkkönen M, Aakala T, Blattert C, Burgas D, Duflot R, Eyvindson K, Kouki J, Laaksonen T, Punttila P (2022) More wood but less biodiversity in forests in Finland: a historical evaluation. Memoranda Societatis pro Fauna et Flora Fennica 98(Supplement 2):1–11. https://journal.fi/msff/article/view/120306 Nieminen M, Hökkä H, Laiho R, Juutinen A, Ahtikoski A, Pearson M, Kojola S, Sarkkola S, Launiainen S, Valkonen S, Penttilä T, Lohila A, Saarinen M, Haahti K, Mäkipää R, Miettinen J, Ollikainen M (2018) Could continuous cover forestry be an economically and environmentally feasible management option on drained boreal peatlands? Forest Ecol Manage 424:78–84. https://doi.org/10.1016/j.foreco.2018.04.046 Nieminen M, Sarkkola S, Haahti K, Sallantaus S, Koskinen M, Ojanen P (2020) Metsäojitettujen soiden typpi- ja fosforikuormitus. Suo 71:1–13 Nieminen M, Hasselquist EM, Mosquera V, Ukonmaanaho L, Sallantaus T, Sarkkola S (2022) Post-drainage stand growth and peat mineralization impair water quality from forested peatlands. J Environ Qual 51:1211–1221. https://doi.org/10.1002/jeq2.20412 Nieminen M, Pukkala T, Stenberg L, Sarkkola S, Vihonen A, Valkeapää A (2023) Jatkuvan kasvatuksen ja tasaikäismetsätalouden vaikutus metsäisten valuma-alueiden vesistökuormitukseen Suomessa. Metsätieteen aikakauskirja 2023-22001. Tutkimusartikkeli, 18 s. https://doi.org/10.14214/ma.22001 Nieminen M, Stenberg L, Leppä K, Lohila A, Minkkinen K, Ojanen P, Korkiakoski M, Penttilä T, Sarkkola S (2024) Effect of partial harvesting on loads of dissolved organic carbon and nutrients from drained boreal pine mires. Boreal Environment Research 29:65–76 Ojanen P, Minkkinen K (2019) The dependence of net soil CO2 emissions on water table depth in boreal peatlands drained for forestry. Mires Peat 24:27. https://doi.org/10.19189/MaP.2019.OMB.StA.1751 Pukkala T (2021) Measuring the social performance of forest management. Journal of Forestry Research 32:1803–1818. https://doi.org/10.1007/s11676-021-01321-z Pukkala T (2022) Improved guidelines for any-aged forestry. Journal on Forestry Research 33:1443–1457. https://doi.org/10.1007/s11676-022-01473-6 Pukkala T, Kangas J (1993) A heuristic optimization method for forest planning and decision making. Scandinavian Journal of Forest Research 8:560–570 Pukkala T, Laiho O, Lähde E (2016) Continuous cover forestry decreases wind damage. Forest Ecology and Management 372:120–127. https://doi.org/10.1016/j.foreco.2016.04.014 Pukkala T, Vauhkonen J, Korhonen KT, Packalen T (2021) Self-learning growth simulator for modeling forest stand dynamics in changing conditions. Forestry 94:333–346. https://doi.org/10.1093/forestry/cpab008 Reeves CR (Ed.) (1993) Modern heuristic techniques for combinatorial problems. Blackwell Scientific Publications, 320 p. Sarkkola S, Hökkä H, Koivusalo H, Nieminen M, Ahti E, Päivänen J, Laine J (2010) Role of tree stand evapotranspiration in maintaining satisfactory drainage conditions in drained peatlands. Can J For Res 40:1485–1496. https://doi.org/10.1139/X10-084 Sikström U, Hökkä H (2016) Interactions between soil water conditions and forest stands in boreal forests with implications for ditch network maintenance. Silva Fenn 50:1416. https://doi.org/10.14214/sf.1416 Äijälä O, Koistinen A, Sved J, Vanhatalo K, Väisänen P (Eds.) (2014) Hyvän metsänhoidon suositukset – metsänhoito. Metsätalouden kehittämiskeskus Tapion julkaisuja. ISBN 978-952-6612-32-4. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5063189","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":491177976,"identity":"8d89f70f-c760-45a7-a95a-5984df69d25c","order_by":0,"name":"Eppu Honkanen","email":"","orcid":"","institution":"University of Eastern Finland","correspondingAuthor":false,"prefix":"","firstName":"Eppu","middleName":"","lastName":"Honkanen","suffix":""},{"id":491177977,"identity":"9e48bb5f-f355-4c93-a0e3-e9fc39b84692","order_by":1,"name":"Mika Nieminen","email":"","orcid":"","institution":"Natural Resources Institute Finland","correspondingAuthor":false,"prefix":"","firstName":"Mika","middleName":"","lastName":"Nieminen","suffix":""},{"id":491177978,"identity":"d6f4c49b-81ce-45ea-8f6e-829b1a117b17","order_by":2,"name":"Tero Heinonen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA20lEQVRIiWNgGAWjYLACxgYGHgirgoGBvYHxwQF8qnlQtZwBihxgNiBKC9S6NogWvG6yZ+99+IBxh42MOXvvwceF8w7L8TAwM+K3hee4sQHjmTQey55zycYztx02BmphwK9FIo1NgrHtMI/BjRwzad5thxP3M/AfwK9F/hlIy3+QFvPfvHMO1/cQtoUNpOUA2BZm3obDCYQddiaN2SCxLZnH4MwZY2meY+mGPcwEtLC3H2N88LHNzt7geI/hZ54aa3ke9mbmD/i0gEECCo+ZoPpRMApGwSgYBYQAAM3CQI4BST7aAAAAAElFTkSuQmCC","orcid":"","institution":"University of Eastern Finland","correspondingAuthor":true,"prefix":"","firstName":"Tero","middleName":"","lastName":"Heinonen","suffix":""},{"id":491177979,"identity":"d12ca834-1e64-44ad-9a24-8fa8161f4a0a","order_by":3,"name":"Timo Pukkala","email":"","orcid":"","institution":"University of Eastern Finland","correspondingAuthor":false,"prefix":"","firstName":"Timo","middleName":"","lastName":"Pukkala","suffix":""}],"badges":[],"createdAt":"2024-09-10 08:52:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5063189/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5063189/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":93193136,"identity":"b3a9517f-b1e1-413a-8092-ed3271495238","added_by":"auto","created_at":"2025-10-10 05:05:04","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":92452,"visible":true,"origin":"","legend":"\u003cp\u003eThe effect of growing stock volume on the depth of the water table when latitude is 65 degrees and ditch depth is 60 cm (A) and the effect of water table depth on the long-term nutrient load (legacy effect) from drained peatland forests. The lowest diagram (C) shows, based on A and B, the relationship between growing stock volume and the legacy effect of drainage.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-5063189/v1/ea4be45e38ffadaf428284ee.png"},{"id":93193084,"identity":"218a83ef-0bbe-4736-9886-06025ae1e43d","added_by":"auto","created_at":"2025-10-10 04:57:04","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":10403,"visible":true,"origin":"","legend":"\u003cp\u003eEffect of harvested stem volume on the annual loads of nitrogen and phosphorus from harvested peatland forests.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-5063189/v1/616dbcf87a12fe18bc11e29c.png"},{"id":93193637,"identity":"fef6f484-5092-4d54-bab8-a95f32013e0d","added_by":"auto","created_at":"2025-10-10 05:13:04","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":50045,"visible":true,"origin":"","legend":"\u003cp\u003eTop: Trade-off curves between net present value and nitrogen and phosphorus loads when maximizing net present value and minimizing loads. Bottom: Respective harvest removals. Black = Kitee, red = Liperi, solid line = BAU, dashed line = CCF.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-5063189/v1/b834df03033a310f2380151f.png"},{"id":93193134,"identity":"7cecb7d0-18e4-4545-9d72-3d17bf0d18e6","added_by":"auto","created_at":"2025-10-10 05:05:04","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":71275,"visible":true,"origin":"","legend":"\u003cp\u003eThe effect of management objective on nutrient loads with an even-flow harvesting constraint. The weight of maximizing NPV and minimizing loads (Equation 1) is 1 and 0 (maxNPV), 0.5 and 0.5 (50-50), or 0 and 1 (min Load), respectively.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-5063189/v1/9712281ea1ef2c5deb66a671.png"},{"id":93193085,"identity":"5754d8a1-c977-49a5-962e-7ea29585b19f","added_by":"auto","created_at":"2025-10-10 04:57:04","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":38582,"visible":true,"origin":"","legend":"\u003cp\u003eThe effect of management objective on the average net present value of the forest. The weight of maximizing NPV (Equation 1) is 1 (max NPV), 0.5 (50-50), or 0 (min Load), and the corresponding total weight of minimizing nitrogen and phosphorus load is 0, 0.5, or 1.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-5063189/v1/83cea352e8a043fe7b8adc74.png"},{"id":93193088,"identity":"38952a10-214a-4c2e-a808-412d1978ce67","added_by":"auto","created_at":"2025-10-10 04:57:04","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":65548,"visible":true,"origin":"","legend":"\u003cp\u003eThe trade-off between economic profit and nutrient load in different silvicultural systems with an even-flow harvest constraint. The ten-year removal is\u0026nbsp;\u0026nbsp; 67.5±6.75 m\u003csup\u003e3 \u003c/sup\u003eha\u003csup\u003e-1\u003c/sup\u003e in Kitee and 71.5±7.15 m\u003csup\u003e3 \u003c/sup\u003eha\u003csup\u003e-1\u003c/sup\u003e in Liperi.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-5063189/v1/18246e2badf43bab98dba7c9.png"},{"id":93193089,"identity":"7b070ccd-c148-4589-b6f3-efd9bfd30cba","added_by":"auto","created_at":"2025-10-10 04:57:04","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":31991,"visible":true,"origin":"","legend":"\u003cp\u003eClear-felled areas during 50 years in Kitee and Liperi in the BAU management when the management objective was to maximize net present value (max NPV), minimize nutrient loads (min Load), or simultaneously maximize NPV and minimize nutrient loads (50-50).\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-5063189/v1/38ffc27dd68174bfc5af8551.png"},{"id":93193090,"identity":"39b80735-c6f9-4b0d-84e5-bf600c24940d","added_by":"auto","created_at":"2025-10-10 04:57:04","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":17724,"visible":true,"origin":"","legend":"\u003cp\u003eDitch maintenance area during 50 years, expressed as the percentage of the total area of peatland forests. Max NPV: net present value was maximized; 50-50: net present value was maximized and nutrient loads were minimized; min Load: nutrient loads were minimized.\u003c/p\u003e","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-5063189/v1/16a5077c7661c9cf84388961.png"},{"id":94472140,"identity":"16ec406d-03e7-48a3-9f56-91167485f019","added_by":"auto","created_at":"2025-10-27 15:40:52","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":888252,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5063189/v1/838b7829-5b4c-4fb6-9a30-4b64f3f1c5ae.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Continuous cover forest management decreases nutrient loads to water courses","fulltext":[{"header":"Introduction","content":"\u003cp\u003eRecent studies suggest that forestry is a significantly greater source of nitrogen and phosphorus to water courses than previously thought (Nieminen et al. 2017b, 2018c, 2022; Fin\u0026eacute;r et al. 2021). In Finland, for example, it was estimated that forestry on peatlands accounts for only 1\u0026ndash;3 percent of all human-induced nutrient loads (Fin\u0026eacute;r et al. 2010), but recent studies suggest that this percentage may be as much as 15\u0026ndash;20% (Nieminen et al. 2020a). Many studies have further indicated that nutrient concentrations in waters discharging from drained peatland forests are still increasing (Nieminen et al. 2017b, 2018c, 2022; R\u0026auml;ike et al. 2019).\u003c/p\u003e\n\u003cp\u003eA significant additional water quality problem in the waters discharging from forested catchments is their brownification due to enhanced loads of dissolved organic carbon and iron (Finstad et al. 2016; \u0026Scaron;kerlep et al. 2021; Williamson et al. 2021). Wide-spread brownification of waters has most often been attributed to recovery from acid deposition (Monteith et al. 2007; Erlandsson al. 2008) or climatic warming (Sarkkola et al. 2009; Laudon et al. 2012), but recent studies suggest that forests and forestry also contribute to increased organic carbon loads. Forest growth has increased particularly in the northern latitudes due to intensified silviculture and climate warming, resulting in increased carbon stocks in soils and forests, and consequently increased carbon losses from forests to surface waters. In peatland-dominated areas, such as large parts of Scandinavia, the Baltic states, and the British Isles, drainage of peatlands for forestry and management of peatland forests also contribute to organic carbon loads and water brownification (Nieminen et al. 2021; H\u0026auml;rk\u0026ouml;nen et al. 2023).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhile the awareness of the effects of forestry on nutrient and carbon loads to watercourses has increased, it has also been acknowledged that current water protection measures may be inefficient in retaining nutrients and carbon (Nieminen et al. 2018; H\u0026auml;rk\u0026ouml;nen et al. 2023). Most protection measures only retain suspended sediments and adhered nutrients and do not affect the loads of dissolved nutrients and dissolved carbon. The study by Nieminen et al. (2024) suggests that the only means to effectively decrease the loads of dissolved nutrients in forested catchments is to convey water to extensive wetland buffers. However, their study further suggests that it may only be feasible to use wetland buffers to reduce phosphorus loads as very large buffers are needed to significantly reduce nitrogen and organic carbon loads. Constructing wetland buffers by restoration of drained sites is also challenging in the sense that it initially increases rather than decreases nutrient and carbon loads (Koskinen et al. 2017; Nieminen et al. 2020). \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe problems related to the efficiency and operational use of water quality protection practices have raised a question of whether it would be more feasible to decrease forestry-induced release of dissolved carbon and nutrients rather than try to capture them from discharged water flow (Nieminen et al. 2018, 2024; H\u0026auml;rk\u0026ouml;nen et al. 2023). It has been proposed that managing forests with continuous cover forestry (CCF) instead of the prevailing even-aged forestry management could be an effective means of decreasing forestry-induced nutrient and carbon loads (Nieminen et al. 2018a, 2024; H\u0026auml;rk\u0026ouml;nen et al. 2023). CCF decreases nutrient loads particularly because it avoids large clear-cuts and eliminates the need for regular ditch cleanings in drained sites (Nieminen et al. 2018a). It may also maintain water tables higher than in dense even-aged peatland forests, which reduces aerobic nutrient mineralization (Ojanen and Minkkinen 2019) and the leaching of nutrients and carbon from deep peat layers (Nieminen et al. 2018a; Nieminen et al. 2022).\u003c/p\u003e\n\u003cp\u003eA recent simulation study by Nieminen et al. (2023), which covered all privately owned forests of Finland, suggested that nutrient loads are lower from CCF than the prevailing even-aged forestry practices (referred to as BAU, i.e., business as usual). In their study, the silvicultural systems were implemented by following the cutting guidelines developed for them. However, as the cutting guidelines aim to maximize wood production or economic profit, simulations based on them do not show the performance of a silvicultural system when one target of forest management is to minimize nutrient loads. Previous analyses have also not covered so-called any-aged forestry (AAF) where management is not classified as either even-aged or continuous cover forestry but has features of both systems (Haight \u0026amp; Monserood 1990a, 1990b; Pukkala 2022).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eA problem with simulations based on silvicultural guidelines is also that they may result in different harvest levels in different silvicultural systems, making it difficult to separate the effect of the silvicultural system from that of the harvest level. As an alternative approach, optimization makes it possible to find efficient Pareto-optimal combinations of variables of interest. Pareto-optimality refers to management where a management objective is achieved with the smallest possible loss in another, conflicting objective. Moreover, comparisons can be conducted under different constraints. A common constraint in large-area analyses is the even flow of harvested timber.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe current study used optimization to analyze the effect of timber harvests and profit maximization on nutrient loads to water bodies. Optimization allows fair comparisons of different silvicultural practices since the harvested volume can be constrained to be the same in all systems, and all systems can aim at the same management objectives. Another addition, as compared to earlier research, was the analysis of any-aged forest management. Since all cutting types of both even-aged rotation forestry and CCF are possible in AFF, we hypothesize that \u0026nbsp;AFF is the most efficient silvicultural system when the aim is to maximize economic profit while concurrently minimizing forestry-induced nutrient loads.\u003c/p\u003e\n\u003cp\u003eThe study first analyzed the relationship between economic profit and nutrient loads in Pareto optimal forest management, separately in CCF and BAU. Second, the possibilities to reduce nutrient loads with fixed harvest removals were analyzed, separately for BAU, CCF, and AAF. This was done by minimizing nutrient loads instead of maximizing economic profit while keeping the periodical harvest volume constant.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe study utilizes private forests from two municipalities, with Kitee representing a region with a large share of drained peatland forests and Liperi representing a region where their share is small. Building upon the results of Nieminen et al. \u0026nbsp;(2023), our research hypothesis assumes that AAF and CCF provide more possibilities than BAU to decrease nutrient loads from managed forests to water systems.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cp\u003e\u003cstrong\u003eCalculation of nutrient loads\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study used the same methods as Nieminen et al. (2023, 2024) in estimating the effects of CCF, AAF and BAU on the loads of nitrogen (N) and phosphorus (P) to receiving watercourses. We assume as in previous studies that forestry increases nutrient loads due to the long-term legacy effect of past peatland drainage operations (Nieminen et al. 2017b, 2018c, 2022; R\u0026auml;ike et al. 2019), as well as the shorter-term effects of ditch network maintenance (DNM), forest fertilization, and harvesting. We further assume that the long-term legacy effect of drainage is controlled by nutrient mineralization (Nieminen et al. 2022), which, in turn, is controlled by site drainage conditions, that is, the soil water table level (Ojanen and Minkkinen 2019).\u003c/p\u003e\n\u003cp\u003eAs in the studies by Nieminen et al. (2023, 2024), the estimation of the legacy effect of drainage begins by predicting the water table level (WTL) during late summer conditions for all forest compartments that represent drained peatland forests (Fig. 1A). This is done by the empirical equation by Sarkkola et al. (2010), which predicts WTLs as a function of the growing stock volume (its evapotranspiration demand), monthly precipitation, ditch depth, and the location of the site (latitude). The legacy effect of drainage is then obtained as the relationship between WTL and N and P loads (Fig. 1B), as calculated by the SUSI peatland simulator (Laur\u0026eacute;n et al. 2021) for sites with differing fertility (Nieminen et al. 2023). The combined use of the two models shows the effect of growing stock volume on the legacy effect of drainage (Fig. 1C).\u003c/p\u003e\n\u003cp\u003eThe shorter-term effects of ditch network maintenance (DNM), fertilization, and clear-felling of mineral soil forests on N and P loads (kg ha\u003csup\u003e-1\u003c/sup\u003e year\u003csup\u003e-1\u003c/sup\u003e) were estimated using the load coefficients by Nieminen et al. (2024). All these effects are temporary; the effects of clear-felling and DNM last for ten years, the annual loads peaking in the first year and decreasing gradually towards the tenth year. Fertilization-induced phosphorus loads in drained peatland forests last for five years (Fin\u0026eacute;r et al. 2010; Nieminen et al. 2023) and fertilization-induced nitrogen loads in mineral soil forests for two years (Saura et al. 1995).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe nutrient loads caused by the harvesting of peatland forests were calculated with the empirical models by Nieminen et al. (2023) for the relationship between harvested stem wood volume and harvest-induced nutrient loads. According to their models, small stem wood removals typical to BAU thinning treatments and CCF harvests (removed volume less than 100 m\u003csup\u003e3\u003c/sup\u003eha\u003csup\u003e-1\u003c/sup\u003e) result in small nutrient loads, but nutrient loads increase when the volume of harvested stem wood approaches the volumes typical to clear-felling (Fig. 2). The excess loads induced by harvesting were assumed to last for six years on nutrient-rich peatland forests and four years on nutrient-poor peatland sites (Nieminen et al. 2023).\u003c/p\u003e\n\u003ch2\u003eForest stand data\u003c/h2\u003e\n\u003cp\u003eSimulation and optimization were based on municipality-specific forest resource data from the two municipalities, Kitee (62\u0026deg;5\u0026apos;54.96\u0026apos;\u0026apos;N, 30\u0026deg;8\u0026apos;15.00\u0026apos;\u0026apos;E) and Liperi (62\u0026deg;31\u0026apos;59.88\u0026apos;\u0026apos;N, 29\u0026deg;22\u0026apos;59.88\u0026apos;\u0026apos;E). The data were obtained from the Finnish Forest Centre\u0026apos;s open database, which focuses on private forests. For both municipalities, a 5% random sample was drawn from the forest resource data accounting for 4.4 % of all forests in Kitee and Liperi (Table 1). The average temperature in the region is about 4 \u0026deg;C, with -8\u0026deg;C in February, and 17 \u0026deg;C in July (https://en.ilmatieteenlaitos.fi/). The municipalities receive an average precipitation of approximately 600 mm, of which about 200 mm falls as snow (Jokinen et al. 2021).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 1\u003c/strong\u003e Information about the stands selected for the sample.\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"468\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 222px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003eKitee\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003eLiperi\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 222px;\"\u003e\n \u003cp\u003eNumber of stands\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e4068\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e2263\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 222px;\"\u003e\n \u003cp\u003eTotal area of the sample, ha\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e4460.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e2379.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 222px;\"\u003e\n \u003cp\u003eArea of productive forest, ha\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e4380.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e2368.2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 222px;\"\u003e\n \u003cp\u003eShare of peatland forest, %\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e21.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e12.6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 222px;\"\u003e\n \u003cul\u003e\n \u003cli\u003eSpruce mires, %\u003c/li\u003e\n \u003c/ul\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e9.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e6.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 222px;\"\u003e\n \u003cul\u003e\n \u003cli\u003ePine bogs, %\u003c/li\u003e\n \u003c/ul\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e12.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e5.9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 222px;\"\u003e\n \u003cp\u003eHerb-rich site or better, %\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e38.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e34.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 222px;\"\u003e\n \u003cp\u003eMesic site, %\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e44.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e48.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 222px;\"\u003e\n \u003cp\u003eSub-xeric site, %\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e13.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e16.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 222px;\"\u003e\n \u003cp\u003eXeric site, %\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e3.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e1.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 222px;\"\u003e\n \u003cp\u003eMean volume, m\u003csup\u003e3\u003c/sup\u003eha\u003csup\u003e-1\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e159.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e165.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 222px;\"\u003e\n \u003cul\u003e\n \u003cli\u003ePine\u003c/li\u003e\n \u003c/ul\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e56.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e57.9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 222px;\"\u003e\n \u003cul\u003e\n \u003cli\u003eSpruce\u003c/li\u003e\n \u003c/ul\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e62.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e68.9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 222px;\"\u003e\n \u003cul\u003e\n \u003cli\u003eBirch\u003c/li\u003e\n \u003c/ul\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e40.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 123px;\"\u003e\n \u003cp\u003e37.9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eThe share of peatland forest was 21.9 % in Kitee and 12.6 % in Liperi. In both municipalities, most stands represented Norway spruce (\u003cem\u003ePicea abies\u003c/em\u003e (L.) Karst.) and Silver birch (\u003cem\u003eBetula pendula\u003c/em\u003e Roth) dominated herb-rich and mesic sites. The share of sub-xeric and xeric Scots pine (\u003cem\u003ePinus sylvestris\u003c/em\u003e L.) dominated sites was 17.3% in Kitee and 17.5% in Liperi.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eSimulation of forest management alternatives\u003c/h2\u003e\n\u003cp\u003eThe development of forest stands and their management were simulated for 50 years using the models developed by Pukkala et al. (2021). The 50 years were divided into five 10-year spans. Harvestings and other treatments were simulated in the middle of each 10 years. The models by Pukkala et al. (2021) are based on the measurements of the permanent sample plots (PSPs) of the 10\u003csup\u003eth\u003c/sup\u003e and 11\u003csup\u003eth\u003c/sup\u003e National Forest Inventory (NFI) of Finland. The models describe the diameter growth of trees, the probability of tree survival, and the ingrowth. The ingrowth models describe natural regeneration and predict the number of new trees per hectare that pass the 1.3-m height limit over five years (Pukkala et al., 2021).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe predictive variables in the Pukkala et al. (2021) models do not include stand age, dominant height, or site index based on age and dominant height, thus allowing simulations where dominant trees are removed in harvestings. The site variables in the models are temperature sum (\u0026gt;+5 \u0026deg;C threshold), an indicator variable for peatland, and site fertility classified as mesotrophic, herb-rich, mesic, sub-xeric, xeric, and barren.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe treatments of the management schedules in the prevailing even-aged forestry (BAU) were based on the silvicultural guidelines by \u0026Auml;ij\u0026auml;l\u0026auml; et al. (2014). The guidelines for cuttings are based on dominant tree species, dominant height, mean tree diameter, and the total basal area of the trees (m\u003csup\u003e3\u003c/sup\u003eha\u003csup\u003e-1\u003c/sup\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe prevailing type of thinning in Finnish forest management is thinning from below but there is a gradual shift towards more versatile thinning options. To reflect this evolution of forest management, we simulated BAU with thinning from below, uniform thinning, and thinning from above, with probabilities of 0.7, 0.1, and 0.2, respectively. In thinning from below, half of the removed basal area was obtained using the same harvest rate for all diameter classes. The other half was obtained by removing trees from the smallest diameter classes until the recommended post-thinning basal area was reached. In thinning from above, half of the removed basal area was taken by thinning all diameter classes with the same intensity, and the other half was taken by removing the largest trees. In uniform thinning the thinning rate was the same in all diameter classes.\u003c/p\u003e\n\u003cp\u003eIn CCF and AAF, the guidelines by \u0026Auml;ij\u0026auml;l\u0026auml; et al. (2014) for even-aged management were applied to the first commercial harvesting of a young forest. Later on, optimization-based guidelines developed for all-aged forest management were used (Pukkala 2022). In these guidelines, it is first checked whether a stand is mature for harvesting, or whether it would be optimal to let the stand continue to grow. If the optimal decision is to harvest, the second part of the guideline tells whether the cutting is partial cutting (thinning) or final felling. In the case of thinning, the third part of the guideline determines the optimal harvest rate for different diameter classes. The only difference in simulating CCF and AAF was that clearcuttings and seed tree cuttings were ruled out in CCF.\u003c/p\u003e\n\u003cp\u003eA common type of cutting in all management systems was the removal of the upper story in two-storied stands. This option was selected if the lower story of small (non-merchantable) trees formed a dense enough stand and the volume of the removed upper story was at least 50 m\u003csup\u003e3\u0026nbsp;\u003c/sup\u003eha\u003csup\u003e-1\u003c/sup\u003e. The allowed removal in other partial cuttings was 50\u0026ndash;200 m\u0026sup3; ha\u003csup\u003e-1\u003c/sup\u003e. If the removal was less than 50 m\u0026sup3; ha\u003csup\u003e-1\u003c/sup\u003e, the treatment was postponed, and if the removal was more than 200 m\u0026sup3; ha\u003csup\u003e-1\u003c/sup\u003e, the guidelines for remaining basal area (BAU) or thinning intensity (CCF and AAF) were adjusted. In the thinning treatments of CCF and AFF, additional requirements were set for the minimum post-cutting basal area: mesic or better sites 11-12 m\u0026sup2; ha\u003csup\u003e-1\u003c/sup\u003e, sub-xeric sites 9-10 m\u0026sup2; ha\u003csup\u003e-1\u003c/sup\u003e, and xeric sites 7-8 m\u0026sup2; ha\u003csup\u003e-1\u003c/sup\u003e, the lower values representing peatland forests. The harvest rates of the guidelines were adjusted if they resulted in lower basal areas than these limits.\u003c/p\u003e\n\u003cp\u003eThe regeneration method for even-aged forestry on mesic and better sites involved clear-cutting, soil preparation, and planting. In the sub-xeric site, the regeneration method was clear felling, site preparation, and sowing of Scots pine seeds. Natural regeneration via seed trees was used in xeric and poorer sites. In mesic and herb-rich sites, the planted tree species were randomized. In herb-rich sites, the probability of planting Norway spruce was 0.8 and that of planting Silver birch \u0026nbsp;0.2. On mesic sites, the probabilities of planting pine, spruce, and birch were 0.2, 0.7, and 0.1, respectively. The growth of all planted trees was set at 10% better and the growth of seeded trees at 5% better than the growth of natural trees to account for their breeding effect.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSeveral alternative 50-year management schedules were simulated for each stand to provide a decision space for forest-level optimization. The recommendations of \u0026Auml;ijal\u0026auml; et al. (2014) give the range for the basal area at thinning (m\u003csup\u003e2\u0026nbsp;\u003c/sup\u003eha\u003csup\u003e-1\u003c/sup\u003e) and mean tree diameter at final felling (cm). The lowest limits of these ranges were used as the earliest possible timing for thinning and final felling. Alternative schedules were simulated by postponing the cutting by one or more 10-year periods. The guidelines for AAF include the discount rate as one factor that affects the timing and intensity of cuttings. Alternative CCF and AAF cutting schedules were obtained by applying the recommendations with a 1, 3, or 5 % discount rate. A no-cutting schedule was also simulated for all stands in all three silvicultural systems.\u003c/p\u003e\n\u003cp\u003eNitrogen fertilizations were simulated in the BAU management on mineral soils for spruce-dominated mesic sites and pine-dominated sub-xeric sites. Fertilization was simulated with a probability of 0.5 when the mean diameter of the trees was 20\u0026ndash;30 cm, the basal area was 20\u0026ndash;30 m\u003csup\u003e2\u0026nbsp;\u003c/sup\u003eha\u003csup\u003e-1\u003c/sup\u003e and at least 10 years had passed since the previous fertilization (Heinonen et al. 2018). In practice, these two stand types were fertilized once during the rotation period (Nieminen et al. 2023). The amount of added nitrogen was 150 kg ha\u003csup\u003e-1\u003c/sup\u003e. The response of tree growth to fertilization was simulated based on the research by Kukkola and Saram\u0026auml;ki (1983).\u003c/p\u003e\n\u003cp\u003eIn drained peatland forests, phosphorus+potassium fertilization was simulated with a probability of 0.3 when the site fertility was sub-xeric or better, the mean tree diameter was 5\u0026ndash;30 cm, the basal area ranged from 10 to 40 m\u0026sup2;ha\u003csup\u003e-1\u003c/sup\u003e, and at least 50 years had passed since the previous fertilization (Nieminen et al. 2023). Since the main purpose of this fertilization is to maintain tree growth by avoiding nutrient deficits, it was assumed that fertilization did not improve growth compared to the model prediction without fertilization. Phosphorus+potassium fertilizations in drained peatland forests were simulated in all silvicultural systems.\u003c/p\u003e\n\u003cp\u003eAdditionally, ditch network maintenance (DNM) was simulated for drained peatlands by assuming a similar growth effect as in Heinonen et al. (2018). DNM was simulated with a probability of 0.1 when the mean tree diameter of the stand was 5\u0026ndash;30 cm, the basal area was 2\u0026ndash;25 m\u0026sup2; ha\u003csup\u003e-1\u003c/sup\u003e, and at least 30 years had passed since the previous DNM. However, DNM was not simulated if the stand volume was \u0026gt;150 m\u0026sup3; ha\u003csup\u003e-1\u003c/sup\u003e, as such mature stands with high evapotranspiration capacity maintain sufficient drainage, even in the case of poorly functioning ditches (Sarkkola et al., 2010). DNM was also simulated after clearcutting in BAU and AAF.\u003c/p\u003e\n\u003ch2 id=\"_Toc158629512\"\u003eOptimizations\u003c/h2\u003e\n\u003cp\u003eThe simulated annealing metaheuristic was used to search for the best combinations of the simulated treatment schedules (Reeves 1993; Bettinger et al. 2002). A utility function was maximized (Pukkala and Kangas 1993). The utility function included one to three objective variables, which were the net present value of timber production (\u003cem\u003eNPV\u003c/em\u003e), nitrogen load (\u003cem\u003eN\u003c/em\u003e\u003csub\u003eload\u003c/sub\u003e), and phosphorus load (\u003cem\u003eP\u003c/em\u003e\u003csub\u003eload\u003c/sub\u003e). NPV was maximized while the nutrient loads were minimized. An additive utility function was assumed:\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eU\u003c/em\u003e = \u003cem\u003ew\u003c/em\u003e\u003csub\u003e1\u003c/sub\u003e\u003cem\u003eu\u003c/em\u003e\u003csub\u003e1\u003c/sub\u003e(\u003cem\u003eNPV\u003c/em\u003e) + \u003cem\u003ew\u003c/em\u003e\u003csub\u003e2\u003c/sub\u003e\u003cem\u003eu\u003c/em\u003e\u003csub\u003e2\u003c/sub\u003e(\u003cem\u003eN\u003csub\u003el\u003c/sub\u003e\u003c/em\u003e\u003csub\u003eoad\u003c/sub\u003e) + \u003cem\u003ew\u003c/em\u003e\u003csub\u003e3\u003c/sub\u003e\u003cem\u003eu\u003c/em\u003e\u003csub\u003e3\u003c/sub\u003e(\u003cem\u003eP\u003c/em\u003e\u003csub\u003eload\u003c/sub\u003e)\u003c/p\u003e\n\u003cp\u003ewhere \u003cem\u003eU\u003c/em\u003e is utility, \u003cem\u003eNPV\u003c/em\u003e of the net present value calculated with a 4% discount rate (\u0026euro;), \u003cem\u003eN\u003c/em\u003e\u003csub\u003eload\u003c/sub\u003e is the 50-year nitrogen load from the forest to water bodies (kg ha\u003csup\u003e-1\u003c/sup\u003e) and \u003cem\u003eP\u003c/em\u003e\u003csub\u003eload\u003c/sub\u003e is the respective phosphorus load (kg ha\u003csup\u003e-1\u003c/sup\u003e), \u003cem\u003ew\u003c/em\u003e\u003csub\u003e1\u003c/sub\u003e, \u003cem\u003ew\u003c/em\u003e\u003csub\u003e2\u003c/sub\u003e and \u003cem\u003ew\u003c/em\u003e\u003csub\u003e3\u003c/sub\u003e are, respectively, the weights of \u003cem\u003eNPV\u003c/em\u003e, \u003cem\u003eN\u003c/em\u003e\u003csub\u003eload\u003c/sub\u003e, and \u003cem\u003eP\u003c/em\u003e\u003csub\u003eload\u003c/sub\u003e, and \u003cem\u003eu\u003c/em\u003e\u003csub\u003e1\u003c/sub\u003e, \u003cem\u003eu\u003c/em\u003e\u003csub\u003e2\u003c/sub\u003e and \u003cem\u003eu\u003c/em\u003e\u003csub\u003e3\u003c/sub\u003e are sub-utility functions for the tree management objectives. Since the sum of the weights (\u003cem\u003ew\u003c/em\u003e\u003csub\u003e1\u003c/sub\u003e, \u003cem\u003ew\u003c/em\u003e\u003csub\u003e2\u003c/sub\u003e and \u003cem\u003ew\u003c/em\u003e\u003csub\u003e3\u003c/sub\u003e) was 1 and the sub-utilities ranged from 0 to 1, the total utility (\u003cem\u003eU\u003c/em\u003e) also ranged from 0 to 1.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe sub-utility functions were linear and scaled the range of the variation of \u003cem\u003eNPV\u003c/em\u003e, \u003cem\u003eN\u003c/em\u003e\u003csub\u003eload\u003c/sub\u003e, or \u003cem\u003eP\u003c/em\u003e\u003csub\u003eload\u003c/sub\u003e between 0 and 1 so that the minimum possible NPV yielded a value of 0 \u0026nbsp;and the maximum a value of \u0026nbsp;1. For the nutrient loads, the minimum possible load yielded sub-utility 1, and the highest possible load 0. \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhen analyzing the trade-off between NPV and nutrient loads, the weight of NPV (\u003cem\u003ew\u003c/em\u003e\u003csub\u003e1\u003c/sub\u003e) was first set equal to 1 (\u003cem\u003ew\u003c/em\u003e\u003csub\u003e2\u003c/sub\u003e and \u003cem\u003ew\u003c/em\u003e\u003csub\u003e3\u003c/sub\u003e were zero), and the optimization problem was solved. The solution gives the maximum possible NPV. Then, a part of the weight was gradually transferred to either \u003cem\u003eN\u003c/em\u003e\u003csub\u003eload\u003c/sub\u003e or \u003cem\u003eP\u003c/em\u003e\u003csub\u003eload\u003c/sub\u003e, and the problem was solved again. Solving all these optimization problems produced the trade-off curves between NPV and \u003cem\u003eN\u003c/em\u003e\u003csub\u003eload\u003c/sub\u003e, or between NPV and \u003cem\u003eP\u003c/em\u003e\u003csub\u003eload\u003c/sub\u003e. This analysis was carried out separately for BAU and CCF. The total harvested volume of every solution was also recorded, which made it possible to determine the nutrient load at different harvest levels.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe rest of the analyses involved an even-flow harvesting constraint. The 10-year harvested volume was found by trial and error. The volume was equal to the maximum 10-year removal that could be removed in each silvicultural system during the whole 50-year time span. It was checked that this harvest removal did not decrease the growing stock volume of the forest during the 50-year planning horizon.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe 10-year target removal of industrial wood (saw logs and pulpwood) was 67.5 m\u003csup\u003e3\u0026nbsp;\u003c/sup\u003eha\u003csup\u003e-1\u003c/sup\u003e in Kitee and 71.5 m\u003csup\u003e3\u0026nbsp;\u003c/sup\u003eha\u003csup\u003e-1\u003c/sup\u003e in Liperi. However, some flexibility was allowed since a constant harvest level may not be economically optimal as it ignores the effect on forest structure on the optimal cutting level. Sawmills and paper mills usually purchase timber from areas substantially larger than a municipality, which means that the annual harvest volume in a single municipality may vary. Therefore, the 10-year harvest volume was allowed to deviate 10 % from the target removals.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eEquation 1 was maximized with this constraint so that the weight of NPV in Equation 1 was either 1, 0.5, or 0, and the rest of the total weight was divided equally to \u0026nbsp;\u003cem\u003eN\u003c/em\u003e\u003csub\u003eload\u003c/sub\u003e and \u003cem\u003eP\u003c/em\u003e\u003csub\u003eload\u003c/sub\u003e. The purpose was to see how much nutrient loads can be decreased with an even-flow harvest requirement when nutrient loads are minimized concurrently with profit maximization, or when nutrient loads are decreased as the only management objective. This analysis was conducted separately for BAU, CCF, and AAF.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eThe trade-off curves between NPV and nutrient loads, when NPV was maximized and nutrient load minimized, show a clear superiority of CCF over BAU (Fig. 3). The lowest loads were nearly the same in BAU and CCF. However, when NPVs were maximized, the loads were nearly two times higher in BAU. The nutrient loads were higher in Kitee where the share of peatland forests was almost two times higher than in Liperi.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIt was possible to decrease nutrient loads in both municipalities and all silvicultural systems when forest planning aimed at minimizing nutrient loads without considering its impact on economic profit (Fig. 4). Managing forests with either AAF or CCF reduced nutrient loads on average by around 50% compared to the BAU management. On average, 76% of the nutrient loads in Kitee and 67% in Liperi were due to forest management on peatlands. Minimizing nutrient loads instead of maximizing economic profit decreased the nutrient loads of BAU by 32% in Kitee and 41% in Liperi. In CCF, the reductions were 24% (Kitee) and 30% (Liperi), and in AAF, \u0026nbsp;30% (Kitee) and 38% (Liperi).\u003c/p\u003e\n\u003cp\u003eAs expected, minimizing nutrient loads instead of maximizing NPV decreased the profitability of forest management (Fig. 5). However, the decrease in NPV was small, implying small costs for load reduction. AAF and CCF resulted in about 15% higher NPV than BAU, indicating that switching from BAU to CCF or AAF makes it possible to concurrently improve the profitability of forest management and decrease nutrient loads.\u003c/p\u003e\n\u003cp\u003eThe trade-offs obtained from the optimizations with the even-flow harvesting constraint are visualized in Figure 6. CCF and AAF were again more efficient than BAU in controlling nutrient loads, while concurrently maximizing the profit of forest management. As expected, AAF was the most efficient silvicultural system. However, the margin to CCF was small.\u003c/p\u003e\n\u003cp\u003eIn the management systems where clear-felling was a possible management option (BAU and AAF), the clear-felled area decreased significantly when more weight was given to load minimization in forest planning (Fig. 7). The decrease was more than 50% in BAU when the nutrient load was minimized in forest planning instead of maximizing NPV. The area of clear-felling decreased particularly in spruce mires, where almost 80% of the area was clearcut when NPV was maximized, but less than 10% when forest planning minimized nutrient loads. In AAF, the clear-felled area was 13% (Liperi) or 16% (Kitee) when NPV was maximized but less than 2% when nutrient loads were minimized either as the only objective or together with NPV maximization.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn the BAU management, the area of DNM decreased by 56\u0026ndash;73% when the target was to minimize nutrient loads (Fig. 8). The fertilization area was not sensitive to the weights of NPV maximization and nutrient load minimization (Equation 1). In AAF, the area of DNM during 50 years was small, at a maximum of 3% of the total area of peatland forests.\u003c/p\u003e"},{"header":"Discussion ","content":"\u003cp\u003eThe effects of forestry on nutrient loads to water courses have previously been assessed regionally using simulation (Fin\u0026eacute;r et al. 2021; Nieminen et al. 2023, 2024). This was the first Finnish study where the minimization of nutrient loads was one management objective in optimization-based forest planning. The results showed that, within each of the three inspected silvicultural systems, it is possible to significantly reduce nutrient loads with reasonable economic losses. If the starting point is even-aged forestry, it is possible to concurrently decrease nutrient loads and improve profitability by increasing the use of CCF-type management. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eBy shifting from BAU to CCF or AAF, it is easier to decrease nutrient loads from mineral soils than drained peatlands where not only clear-cuts but also partial harvests increase nutrient loads (Nieminen et al. 2024b), and where the long-term legacy effect of drainage dominates nutrient loads. The legacy effect is not particularly sensitive to forest management methods (Nieminen et al. 2023, 2024).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn mineral soils, clear-felling and nitrogen fertilization were assumed to be the only sources of nutrient loads (Fin\u0026eacute;r et al. 2010). As both were executed in BAU simulations and neither in \u0026nbsp; CCF simulations, it is obvious that CCF was a better option in mineral soil forests from the water quality management viewpoint. However, the general assumption that partial harvests do not affect nutrient loads from mineral soil forests (Fin\u0026eacute;r et al. 2010; Nieminen et al. 2024) still needs to be verified empirically. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhile the legacy effect of drainage is not highly sensitive to forest management methods in the long term, it may induce significant differences in nutrient loads in the shorter term. This is because the nutrient load was assumed to be caused by aerobic peat mineralization, which varies considerably depending on the level of the water table in the peat (Ojanen and Minkkinen 2019). The legacy effect of drainage was therefore significantly lower \u0026nbsp;from young stands with low evapotranspiration (EVT) capacity and \u0026nbsp;consequent high water table, compared to mature stands with high EVT and low water table in peat (Sarkkola et al. 2010). Therefore, although clear-fellings in the BAU management increase nutrient loads in the short term, they may decrease them temporarily in the longer term due to the low stocking and EVT of young even-aged forests. Later on, when the volume of the even-aged stand increases, the BAU management may again lead to higher nutrient loads than CCF. Although the effect of water table on nutrient loads has not been assessed empirically, the results by Nieminen et al. (2022) indicate significantly smaller nutrient concentrations in waters discharging from mature peatland stands with low water tables than for young stands with high water tables (Sarkkola et al. 2010). \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis study utilized municipality-specific harvest removal constraints to provide realistic removals and management scenarios. Optimization with even-flow harvesting constraints resulted in the replacement of clear-fellings by less\u0026nbsp;harmful partial cuttings when management optimization aimed at reducing nutrient loads. One reason for the decreased clear-felling area in peatland forests when load minimization was prioritized is that ditch maintenance treatment is executed after every clear-felling. DNM causes a peak in nutrient loads and increases forest management costs.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThere is uncertainty for instance in simulating the emergence of new understorey trees, which are important for the sustainability of CCF. There is considerable variation in forest regeneration under tree canopies, both within the stand and between them. The development of even-aged, artificially established stands on drained peatlands is also uncertain because the area of drained peatland forests that have been clear-felled and planted is small and the plantations are still young. The assumption made in this study was that planted peatland forests develop nearly the same way as planted mineral soil forests of the same site fertility category. The empirical survival models used in the simulation (Pukkala et al. 2021) predict slightly lower tree survival for peatland forests compared to similar mineral soil forests. The predicted diameter increment of pine is lower in peatland forests, but similar as in \u0026nbsp;mineral soils in spruce and birch forests.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eUncertainty in our simulations is also introduced by climate change, which may increase forest damages, such as windthrows and water stress during prolonged droughts. Spruce, in particular, is sensitive to changing climatic conditions. Windthrows may be reduced by shifting from BAU to CCF (Pukkala et al. 2016), but the probability of some other damages, such as root rot and bark beetle outbreaks, may increase particularly if spruce becomes the dominant tree species. In the BAU management, spruce can be easily replaced by deciduous trees by clear-felling spruce and planting deciduous species. In general, however, planted monocultures, especially those of spruce, are seen as \u0026nbsp;the most vulnerable stand type against beetle attacks and other biotic and abiotic hazards (Messier et al. 2019).\u003c/p\u003e\n\u003cp\u003eMaintaining and increasing the carbon stocks and biodiversity of forests has been increasingly used as a management objective of forest planning (e.g., M\u0026ouml;nkk\u0026ouml;nen et al. 2014; Heinonen et al. 2017), to increase the overall responsibility and acceptability of commercial forestry (Pukkala 2021). However, especially in Finland with \u0026nbsp;a large area \u0026nbsp;of drained peatland forests (20% of land area), sustainable and responsible forest management should also consider water quality. This can be done by including nutrient load minimization as one target of forest planning in the models used earlier for carbon sequestration and biodiversity maintenance ( D\u0026iacute;az-Y\u0026aacute;\u0026ntilde;ez et al. 2019). Such coherent planning is timely as forestry is responsible for considerable biodiversity loss (M\u0026ouml;nkk\u0026ouml;nen et al. 2022) and water quality deterioration (Nieminen et al. 2022, 2023) in areas with intensive forest management and large areas of drained peatland forests, such as Finland, Sweden, the Baltic States, and the British Isles.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eThe study showed that the transport of nitrogen and phosphorus from managed commercial forests to water courses can be substantially reduced by increased use of continuous cover forestry. Nutrient loads can be reduced also in even-aged BAU management if the area of clear-fellings is reduced, particularly \u0026nbsp;in spruce mires. Nutrient loads increase with increasing harvested volume in all silvicultural systems. Significant reductions in nutrient loads can be achieved with reasonably small economic losses. Optimal use of continuous cover management makes it possible to simultaneously increase economic profitability and decrease nutrient loads.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eAll authors contributed to the study conception and design. T.P. prepared the software. T.P. and E.H. and did the analysis. T.P., M.N. and T.H. wrote the main manuscript text. All authors reviewed the manuscript.\u003c/p\u003e\u003cp\u003eFunding\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNo funding was received for conducting this study.\u003c/p\u003e\n\u003cp\u003eConflicts of interest/Competing interests\u003c/p\u003e\n\u003cp\u003eNo conflict of interests\u003c/p\u003e\n\u003cp\u003eAvailability of data and material\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003eCode availability\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBettinger P, Graetz D, Boston K, Sessions J, Chung W (2002) Eight heuristic planning techniques applied to three increasingly difficult wildlife planning problems. Silva Fennica 36:561\u0026ndash;584\u003c/li\u003e\n\u003cli\u003eD\u0026iacute;az-Y\u0026aacute;\u0026ntilde;ez O, Pukkala T, Packalen P, Peltola H (2019) Multifunctional comparison of different management strategies in boreal forests. Forestry 93(1):84\u0026ndash;95. https://doi.org/10.1093/forestry/cpz053\u003c/li\u003e\n\u003cli\u003eFin\u0026eacute;r L, Mattsson T, Joensuu S, Koivusalo H, Laur\u0026eacute;n A, Makkonen T, Nieminen M, Tattari S, Ahti E, Kortelainen P, Koskiaho J, Leinonen A, Nevalainen R, Piirainen S, Saarelainen J, Sarkkola S, Vuollekoski M (2010) Mets\u0026auml;isten valuma-alueiden vesist\u0026ouml;kuormituksen laskenta. Suomen ymp\u0026auml;rist\u0026ouml; 10/2010. http://hdl.handle.net/10138/37973\u003c/li\u003e\n\u003cli\u003eHaight RG, Monserud RA (1990a) Optimizing any-aged management of mixed-species stands: II. Effects of decision criteria. Forest Science 36(1):125\u0026ndash;144\u003c/li\u003e\n\u003cli\u003eHaight RG, Monserud RA (1990b) Optimizing any-aged management of mixed-species stands. I. Performance of a coordinate-search process. Canadian Journal of Forest Research 20(1):15\u0026ndash;25\u003c/li\u003e\n\u003cli\u003eHeinonen T, Pukkala T, Meht\u0026auml;talo L, Asikainen A, Kangas J, Peltola H (2017) Scenario analyses for the effects of harvesting intensity on development of forest resources, timber supply, carbon balance and biodiversity of Finnish forestry. Forest Policy and Economics 80:80\u0026ndash;98. https://doi.org/10.1016/j.forpol.2017.03.011\u003c/li\u003e\n\u003cli\u003eHeinonen T, Pukkala T, Asikainen A, Peltola H (2018) Scenario analyses on the effects of fertilization, improved regeneration material, and ditch network maintenance on timber production of Finnish forests. Eur J Forest Res 137:93\u0026ndash;107. https://doi.org/10.1007/s10342-017-1093-9\u003c/li\u003e\n\u003cli\u003eH\u0026ouml;kk\u0026auml; H, Stenberg L, Laur\u0026eacute;n A (2020) Modelling depth of drainage ditches in forested peatlands of Finland. Balt For 26:220\u0026ndash;228. https://doi.org/10.46490/BF453\u003c/li\u003e\n\u003cli\u003eJokinen P, Pirinen P, Kaukoranta J-P, Kangas A, Alenius P, Eriksson P, Johansson M, Wilkman S (2021) Tilastoja Suomen ilmastosta ja merest\u0026auml; 1991\u0026ndash;2020 (Climatological and oceanographic statistics of Finland 1991\u0026ndash;2020). Raportteja (reports) 2021:8. Ilmatieteen laitos (Finnish Meteorological Institute), Helsinki, Finland. https://doi.org/10.35614/isbn.9789523361485\u003c/li\u003e\n\u003cli\u003eKukkola M, Saram\u0026auml;ki J (1983) Growth response in repeatedly fertilized pine and spruce stands on mineral soils. Comm Inst For Fenn 114:1\u0026ndash;55. http://urn.fi/URN:ISBN:951-40-0622-4\u003c/li\u003e\n\u003cli\u003eLaur\u0026eacute;n A, Palviainen M, Launiainen S, Lepp\u0026auml; K, Stenberg L, Urzainki I, Nieminen M, Laiho R, H\u0026ouml;kk\u0026auml; H (2021) Drainage and stand growth response in peatland forests \u0026ndash; description, testing, and application of mechanistic peatland simulator SUSI. Forests 12:293. https://doi.org/10.3390/f12030293\u003c/li\u003e\n\u003cli\u003eMattsson T, Ahtiainen M, Kentt\u0026auml;mies K, Haapanen M (2006) Avohakkuun ja ojituksen pitk\u0026auml;aikaisvaikutukset valuma-alueen ravinne- ja kiintoainehuuhtoumiin. Julkaisussa: Kentt\u0026auml;mies K, Mattsson T (toim) Mets\u0026auml;talouden vesist\u0026ouml;kuormitus. MESUVE-projektin loppuraportti. Suomen ymp\u0026auml;rist\u0026ouml; 816:73\u0026ndash;81. http://hdl.handle.net/10138/40492\u003c/li\u003e\n\u003cli\u003eMessier C, Bauhus J, Doyon F, Maure F, Sousa-Silva R, Nolet P, Mina M, Aquil\u0026eacute; N, Fortin M-J, Puettmann K (2019) The functional complex network approach to foster forest resilience to global changes. For Ecosyst 6:21. https://doi.org/10.1186/s40663-019-0166-2\u003c/li\u003e\n\u003cli\u003eM\u0026ouml;nkk\u0026ouml;nen M, Juutinen A, Mazziotta A, Miettinen K, Podkopaev D, Reunanen P, Tikkanen O, Kouki J (2014) Spatially dynamic forest management to sustain biodiversity and economic returns. J Environ Manage 134:80\u0026ndash;89. https://doi.org/10.1016/j.jenvman.2013.12.021\u003c/li\u003e\n\u003cli\u003eM\u0026ouml;nkk\u0026ouml;nen M, Aakala T, Blattert C, Burgas D, Duflot R, Eyvindson K, Kouki J, Laaksonen T, Punttila P (2022) More wood but less biodiversity in forests in Finland: a historical evaluation. Memoranda Societatis pro Fauna et Flora Fennica 98(Supplement 2):1\u0026ndash;11. https://journal.fi/msff/article/view/120306\u003c/li\u003e\n\u003cli\u003eNieminen M, H\u0026ouml;kk\u0026auml; H, Laiho R, Juutinen A, Ahtikoski A, Pearson M, Kojola S, Sarkkola S, Launiainen S, Valkonen S, Penttil\u0026auml; T, Lohila A, Saarinen M, Haahti K, M\u0026auml;kip\u0026auml;\u0026auml; R, Miettinen J, Ollikainen M (2018) Could continuous cover forestry be an economically and environmentally feasible management option on drained boreal peatlands? Forest Ecol Manage 424:78\u0026ndash;84. https://doi.org/10.1016/j.foreco.2018.04.046\u003c/li\u003e\n\u003cli\u003eNieminen M, Sarkkola S, Haahti K, Sallantaus S, Koskinen M, Ojanen P (2020) Mets\u0026auml;ojitettujen soiden typpi- ja fosforikuormitus. Suo 71:1\u0026ndash;13\u003c/li\u003e\n\u003cli\u003eNieminen M, Hasselquist EM, Mosquera V, Ukonmaanaho L, Sallantaus T, Sarkkola S (2022) Post-drainage stand growth and peat mineralization impair water quality from forested peatlands. J Environ Qual 51:1211\u0026ndash;1221. https://doi.org/10.1002/jeq2.20412\u003c/li\u003e\n\u003cli\u003eNieminen M, Pukkala T, Stenberg L, Sarkkola S, Vihonen A, Valkeap\u0026auml;\u0026auml; A (2023) Jatkuvan kasvatuksen ja tasaik\u0026auml;ismets\u0026auml;talouden vaikutus mets\u0026auml;isten valuma-alueiden vesist\u0026ouml;kuormitukseen Suomessa. Mets\u0026auml;tieteen aikakauskirja 2023-22001. Tutkimusartikkeli, 18 s. https://doi.org/10.14214/ma.22001\u003c/li\u003e\n\u003cli\u003eNieminen M, Stenberg L, Lepp\u0026auml; K, Lohila A, Minkkinen K, Ojanen P, Korkiakoski M, Penttil\u0026auml; T, Sarkkola S (2024) Effect of partial harvesting on loads of dissolved organic carbon and nutrients from drained boreal pine mires. Boreal Environment Research 29:65\u0026ndash;76\u003c/li\u003e\n\u003cli\u003eOjanen P, Minkkinen K (2019) The dependence of net soil CO2 emissions on water table depth in boreal peatlands drained for forestry. Mires Peat 24:27. https://doi.org/10.19189/MaP.2019.OMB.StA.1751\u003c/li\u003e\n\u003cli\u003ePukkala T (2021) Measuring the social performance of forest management. Journal of Forestry Research 32:1803\u0026ndash;1818. https://doi.org/10.1007/s11676-021-01321-z\u003c/li\u003e\n\u003cli\u003ePukkala T (2022) Improved guidelines for any-aged forestry. Journal on Forestry Research 33:1443\u0026ndash;1457. https://doi.org/10.1007/s11676-022-01473-6\u003c/li\u003e\n\u003cli\u003ePukkala T, Kangas J (1993) A heuristic optimization method for forest planning and decision making. Scandinavian Journal of Forest Research 8:560\u0026ndash;570\u003c/li\u003e\n\u003cli\u003ePukkala T, Laiho O, L\u0026auml;hde E (2016) Continuous cover forestry decreases wind damage. Forest Ecology and Management 372:120\u0026ndash;127. https://doi.org/10.1016/j.foreco.2016.04.014\u003c/li\u003e\n\u003cli\u003ePukkala T, Vauhkonen J, Korhonen KT, Packalen T (2021) Self-learning growth simulator for modeling forest stand dynamics in changing conditions. Forestry 94:333\u0026ndash;346. https://doi.org/10.1093/forestry/cpab008\u003c/li\u003e\n\u003cli\u003eReeves CR (Ed.) (1993) Modern heuristic techniques for combinatorial problems. Blackwell Scientific Publications, 320 p.\u003c/li\u003e\n\u003cli\u003eSarkkola S, H\u0026ouml;kk\u0026auml; H, Koivusalo H, Nieminen M, Ahti E, P\u0026auml;iv\u0026auml;nen J, Laine J (2010) Role of tree stand evapotranspiration in maintaining satisfactory drainage conditions in drained peatlands. Can J For Res 40:1485\u0026ndash;1496. https://doi.org/10.1139/X10-084\u003c/li\u003e\n\u003cli\u003eSikstr\u0026ouml;m U, H\u0026ouml;kk\u0026auml; H (2016) Interactions between soil water conditions and forest stands in boreal forests with implications for ditch network maintenance. Silva Fenn 50:1416. https://doi.org/10.14214/sf.1416\u003c/li\u003e\n\u003cli\u003e\u0026Auml;ij\u0026auml;l\u0026auml; O, Koistinen A, Sved J, Vanhatalo K, V\u0026auml;is\u0026auml;nen P (Eds.) (2014) Hyv\u0026auml;n mets\u0026auml;nhoidon suositukset \u0026ndash; mets\u0026auml;nhoito. Mets\u0026auml;talouden kehitt\u0026auml;miskeskus Tapion julkaisuja. ISBN 978-952-6612-32-4.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Forest planning, nutrient load, landscape level, nitrogen, phosphorus, optimization","lastPublishedDoi":"10.21203/rs.3.rs-5063189/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5063189/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis study looked at the effects of even-aged forest management (BAU), continuous cover forestry (CCF), and any-aged forestry (AAF) on the loads of phosphorus and nitrogen to downstream water bodies when these loads were minimized by forest planning. The impact of aiming at minimal nutrient loads instead of maximal economic profit was inspected within each of the three silvicultural systems. The analyses were conducted with and without even-flow-cutting constraints. The data for the calculations was a random sample of forest stands, representing two municipalities in North Karelia, eastern Finland. The results showed that the transport of nitrogen and phosphorus from managed commercial forests to water courses can be substantially reduced by increased use of continuous cover forestry. Significant reductions were possible in both peatland forests and mineral sois. Nutrient loads could be reduced also in even-aged BAU management by decreasing the use of clear-felling, particularly in spruce mires. Nutrient loads increased with increasing harvests in all silvicultural systems. Significant reductions in nutrient loads can be achieved with a reasonably small loss in the profitability of forest management. Optimal use of continuous cover management makes it possible to simultaneously increase economic profitability and decrease nutrient loads.\u003c/p\u003e","manuscriptTitle":"Continuous cover forest management decreases nutrient loads to water courses","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-10 04:57:00","doi":"10.21203/rs.3.rs-5063189/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"b66cffef-8578-4bee-8411-a34cc79b4669","owner":[],"postedDate":"October 10th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-10-27T14:20:05+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-10 04:57:00","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5063189","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5063189","identity":"rs-5063189","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}