s41467-026-68668-x.pdf - Page 1

Article https://doi.org/10.1038/s41467-026-68668-x European forest carbon and biodiversity policies have a limited win-win potential Lorenzo Balducci 1 ,E l e n aH a e l e r2,Y o a nP a i l l e t3,E d o a r d oA l t e r i o4, Christian Ammer 5, Frédéric Archaux 6,S t e f f e nB o c h7, Christophe Bouget6, Gediminas Brazaitis 8, Francesco Chianucci9,A n d r e aC u t i n i9, Pallieter De Smedt 10, Inken Doerﬂer11,D a n i e lD v ořák12,M a r k u sF i s c h e r13, Paolo Giordani14, Marion Gosselin 6, Jacob Heilmann-Clausen15, Eva Holá 16, Jeňýk Hofmeister17,J a nH ošek18, Itziar García-Mijangos 19,P h i l i p p eJ a n s s e n3, Kirsten Jung 20, Sebastian Kepfer-Rojas 21, Nathalie Korboulewsky 6, Daniel Kozák17,T o m áš Kuras 22,T h i b a u l tL a c h a t23,J iříM a l íček24, Anders Mårell 6,M a r t i nM i k o l áš 25,J ö r gM ü l l e r26, Francesca Napoleone 1, Björn Nordén27, Péter Ódor 28,Z d e někP a l i c e24, Peter Schall 5, Tommaso Sitzia 4, Kastytis Šimkevičius8, Miroslav Svoboda 25,A l eš Tenčík18, Flóra Tinya 28, Giovanni Trentanovi 29, Mariana Ujhazyova30, Kris Vandekerkhove 31, Michael Wohlwend32,W o l f g a n gW .W e i s s e r33 & Sabina Burrascano 1 Climate change mitigation and biodiversity conservation are key forest func- tions, but how to pursue them jointly in timber-managed forests is still unclear. W eu s eaE u r o p e - w i d ed a t a s e to ff o r est multi-taxon diversity and stand structure to (i) evaluate the importance of aboveground carbon stocks in determining species richness of six taxonomic groups; (ii) assess relationships between species richness and carbon stocks; (iii) discuss the potential to jointly enhance carbon and biodiversity and policy implications. Carbon- diversity relationships are positive for several groups, but mostly when deadwood pools are considered. Forest policies should consider the complex relationship between different carbonpools and taxonomic groups. Environ- mental policies emphasizing carbon sequestration in aboveground living biomass may conﬂict with biodiversity conservation by promoting homo- geneous, fast-growing forests that fail to support species diversity of multiple groups. Sustainable forest management should acknowledge that deadwood carbon instead may translate into positive outcomes for both carbon storage and biodiversity conservation. European forests are expected to mitigate climate change through carbon storage, since the carbon stock associated with living biomass keeps increasing, i.e., from 8 to almost 12 Mt from 1990 to 2020 1. However, measures designed to increase forest carbon pools may conﬂict with timber production. As a matter of fact, while long-lasting wood products ensure that the sequestered carbon, i.e., about 40 MtCO2e/year2, remains out of the atmospheric cycle for extended periods3, a signiﬁcant portion of European forest wood is utilized for fuel or short-lived products, with approximately 24% of the EU ’s roundwood production in 2022 used as fuelwood. The remaining 76% Received: 29 July 2024 Accepted: 9 January 2026 Check for updates A full list of af ﬁliations appears at the end of the paper. e-mail: [email protected] Nature Communications| (2026) 17:1976 1 1234567890():,; 1234567890():,; was processed into industrial roundwood, including pulp and paper, which typically have short lifespans4. Regardless of the fate of woody biomass, removing carbon stocks from forest ecosystems hampers its support to forest biodiversity 5. In Europe, about 30% of forest area is referred to as a habitat of conservation value. Accordingly, biodiversity conservation is among the priorities of the European forest strategy 6. While forest area is increasing in Europe, about 80% of assessments of forest habitats report an unfavorable conservation status 7. The need to improve the condition of forest biodiversity is intrinsically associated with sustainable forest management practices that are expected to simultaneously support biodiversity conserva- tion, climate regulation 7, and timber production. One approach to reconcile these objectives is to spatially balance productive and environmental functions. In this view, the triad fra- mework was recently applied to forests to balance land sparing and land sharing approaches8 by encompassing: areas devoted to high- yield plantations9, areas managed extensively for multiple functions, and unmanaged areas accumulating old growth features 10.T h ed i f - ferent levels of management intensity are unevenly represented across European forests 11, and result in an uneven representation of the three carbon pools here considered. High-yield plantations result in wide temporal ﬂuctuations of standing live carbon depending on the har- vesting regime. Unmanaged conservation areas face a slow yet con- stant increase in all carbon pools with a greater proportion of lying and standing deadwood. Extensively managed forests maximize the standing live carbon pools with a relative temporal stability. These patterns in carbon pools, however, may have different outcomes in terms of biodiversity through direct and indirect mechanisms that are hard to disentangle. On the one hand, the amount of carbon stored in living trees may be associated with species and taxonomic groups related to long local ecological continuity and old- growth conditions 12,13, often dependent on woody substrates, i.e., epiphytic lichens and bryophytes. Not all species groups, however, beneﬁt from high carbon stocks: trade-offs could emerge for those taxonomic groups whose species diversity is linked to light resources, e.g., vascular plants 14– 16. On the other hand, the amounts of lying and standing deadwood, along with the carbon it stores, play a crucial role in sustaining specialized species, including saproxylic beetles and wood-decaying fungi 1,17,18. Overall, the impact of carbon stock on bio- diversity varies greatly depending on the carbon pool and the taxo- nomic group 15,19,20. Notwithstanding the relevance of both climate mitigation and biodiversity conservation in European forests, only few studies have explicitly tested the relationship between forest biodiversity and car- bon stocks 20– 22 or related functions20. In-depth knowledge of the trade- offs between biodiversity conservation and climate change mitigation in European forests is urgently needed since management and restoration strategies focused only on carbon sequestration could interfere with biodiversity conservation targets by promoting homo- geneous and fast-growing stands of tree species with high carbon sequestration rates 23– 25. Understorey vegetation, epiphytic lichens and birds are less diverse in plantation forests, either dominated by native25 or non-native tree species26,27. This is particularly relevant since one the main objectives of the 2030 EU forest strategy is to plant 3 billion trees by 2030 6.D e s p i t et h er e c e n ta w a r e n e s so nt h ei m p o r t a n c eo ff o r e s t s for jointly addressing biodiversity conservation and climate change mitigation, current forest policies, i.e., 2015 Paris agreement and the REDD+ initiative only partially incorporate this link 28. Here, we test the possibility of forests across Europe to simulta- neously sequester high amounts of carbon and host high levels of biodiversity for multiple taxonomic groups to inform trade-offs between competing forest functions. Our speciﬁc aims are to (i) assess the relative importance of dif- ferent aboveground carbon stocks in determining species richness of six taxonomic groups in European forests; (ii) identify the relationship between the species richness of these groups and carbon stocks; (iii) discuss the potential to jointly enhance carbon stocks and biodiversity and the related policy implications. We use the recently built European multi-taxonomic database from the “Bottoms-Up” platform 29 to derive the standardized species richness of six taxonomic groups, i.e., vascular plants; epiphytic and epixylic bryophytes and lichens, hereafter bryophytes and lichens; wood-inhabiting fungi, hereafter fungi; saproxylic beetles, hereafter beetles; and birds. We perform boosted regression trees (BRTs) to assess the response of their species richness to the carbon stocks in living aboveground biomass, standing and lying deadwood, also accounting for potential confounding effects. We expect carbon stocks to be highly relevant in driving the species richness of multiple taxonomic groups. However, given the distinct roles that different carbon pools play in supporting various components of forest biodiversity 1,30, we expect taxon-speci ﬁc responses that vary across carbon pools. Photosynthetic groups such as vascular plants, bryophytes, and lichens may decline with increasing aboveground carbon stocks 14– 16. However, this trend may be counter- balanced by an increase in substrate availability for epiphytic groups 12,13. In contrast, saproxylic fungi and beetles may respond positively to deadwood carbon stocks1,17,18. More mobile taxa, such as birds, may show mixed responses, in ﬂuenced by landscape-level fac- tors rather than stand-level carbon stocks alone31. Results Carbon stocks importance for variations in species richness Carbon pools had a prominent role in explaining the variation in spe- cies richness, with differences across taxonomic groups (Fig. 1). The carbon stocks of lying deadwood showed the highest shares of explained deviances, ranging from 36% for fungi to 9% for birds. The carbon stocks of standing living trees were related to the species richness of birds and vascular plants (highest share of deviance, 22%, for birds, lowest, 5%, for lichens). Standing deadwood had the greatest effect (9%) on the species richness of beetles and the lowest (3%) on lichens. A high share of deviance was explained by the models for lichens (84%), beetles (82%), bryophytes (80%), fungi (77%) and birds (77%), while the predictive value was lower for vascular plants (55%) (Table1). 0 25 50 75 100 Beetles Birds Bryophytes Fungi Lichens Vascular Plants relative importance (%) C stock lying dead C stock stand alive C stock stand dead Forest category Elevation Protocol Silvicultural system Fig. 1 | Relative importance of the predictor variables on scaled species richness across taxonomic groups.Relative importance values indicate the total reduction in model deviance attributable to each variable, averaged across all trees and scaled to sum to 100%, as derived from boosted regression tree (BRT) models ﬁtted separately for six taxonomic groups (beetles, birds, bryophytes, fungi, lichens, and vascular plants). Each bar represents the proportion of explained deviance con- tributed by individual predictors. Source data are provided as a Source data ﬁle. Article https://doi.org/10.1038/s41467-026-68668-x Nature Communications| (2026) 17:1976 2 Model parameters and validation metrics are reported to document the consistency and reliability of the BRTs across taxonomic groups. It should be noted that the variable accounting for the effect of site and sampling protocol explained most of the variation in species richness, ranging from 49% to 14% of explained variance for birds and fungi respectively (Fig. 1). Elevation emerged as the second most inﬂuential factor, accounting for 39% of the explained deviance in bryophytes to 10% in lichens (Fig. 1). The silvicultural system showed the highest value of explained deviance (13%) for lichens and the lowest one (2%) for fungi. Forest category showed the lowest shares of explained deviances, ranging from 7% for beetles to 1% for lichens. Taxon-speciﬁc effects of carbon stocks on species richness Standing alive trees showed a positive effect mainly on fungi (Fig. 2), with an increase of the marginal effect on the logit of the scaled species richness of 0.2, starting at 1.5 (31.6 t C/ha); and a negative effect on vascular plants, with a decrease of 0.2, starting at 0 (1 t C/ha). Lying deadwood showed the strongest positive relationships with fungi and lichens (Fig.3), with an increase of the marginal effect on the logit of the scaled species richness of respectively 0.4 and 0.2. Inter- estingly, the scaled species richness of these two taxonomic groups showed a steep increase at different points in the gradient of carbon stocked in lying deadwood, i.e., while a value of −1( 0 . 1 t C / h a )w a s sufﬁcient to point out an increase in lichen species richness, an increase in fungi species richness was evident at −1.5 (0.03 t C/ha). In contrast, vascular plants displayed a negative relationship with lying deadwood, with a decrease in the marginal effect on the logit of the scaled species richness by 0.3, starting at −3( 0 . 0 0 1tC / h a ) . Standing deadwood had a positive effect on saproxylic beetles and wood-inhabiting fungi (Fig. 4), with an increase in the marginal effect on the logit of the scaled species richness by 0.3 and 0.1, starting at −1( 0 . 1tC / h a ) . Discussion Carbon stocks played a prominent role in explaining the variation in species richness. Deadwood is a fundamental component of the forest carbon pool (7% of growing stock in Europe) and has a key role in nutrient cycles1. It affects several taxonomic groups that are crucial in the detritus food web 32 and provides habitat to many saproxylic insects33, lichens, fungi34, and bryophytes35. Standing living trees also contributed to variation in species richness, with the largest relative importance observed for birds and vascular plants, though their inﬂuence was generally smaller than that of deadwood. By inﬂuencing forest structure 30, microclimate, and resource availability 36, living biomass remains an important carbon pool. Retaining trees in the forest, either as living biomass or deadwood, supports biodiversity across multiple taxonomic groups and enhances long-term carbon storage 37, underscoring the potential impacts of timber harvesting on forest ecosystem functions. The relationship between aboveground living carbon stock and species richness across taxonomic groups was generally weakly positive 22, with some exceptions, most notably vascular plants, which declined above ~1 t C/ha. This decline likely re ﬂects reduced light availability at the forest ﬂoor14, where increases in shade-tolerant species only partially compensate for the decrease of species related to open areas 14. In old-growth forests, both groups may coexist within ﬁne-scale mosaics of successional phases 38, but such conditions are underrepresented in Europe and in our dataset. For most other taxo- nomic groups, the absence of strong positive responses suggests that the accumulation of carbon in living biomass and species richness are often driven by different ecological processes. Although increases in living biomass are typically linked with an increase in stand age 39 and may favor species linked to long ecological continuity and habitat structures associated with late-successional conditions 13; a substantial share of forest biodiversity, including species adapted to disturbance or with high dispersal ability, relies on structural heterogeneity or periodic disturbance rather than continuous biomass accumulation 13. These species can represent an important component of forest diversity, thereby weakening any simple positive link between above- ground living carbon and total species richness. Moreover, in the European context, high aboveground living carbon stocks are not necessarily linked to long ecological continuity, as anthropogenic disturbances strongly drive changes in forest carbon stocks 40.F o r e s t s managed for timber production purposes can reach high carbon sto- rage rates due to the silvicultural promotion of biomass accumulation processes (e.g., fast-growing species in even-aged forests). However, such stands often lack the structural complexity and ecological con- tinuity required by many taxa 24,25. This decoupling underscores that aboveground living carbon is a poor surrogate for forest biodiversity. High carbon stocks in living biomass do not necessarily coincide with high multi-taxon diversity. We found signi ﬁcant associations between deadwood carbon stocks and species richness of several taxonomic groups 17,18. Carbon in lying deadwood had strong positive links with fungi and lichens spe- cies richness in European forests. This can be explained by lying deadwood’s role as both substrate and resource for these two taxo- nomic groups. Interestingly, their scaled species richness showed a steep increase at different values of lying deadwood carbon, with fungi responding at slightly smaller carbon stocks than epiphytic lichens. Fungal diversity is known to increase with lying deadwood carbon stock 41,42, with low amounts of lying deadwood potentially supporting a large number of species43. Lichen diversity, on the other hand, may respond only to greater carbon stocks due to the light conditions associated with the occurrence of large amounts of deadwood, i.e., large gap maker. On the other hand, we found a strong negative rela- tionship between carbon in lying deadwood and vascular plant species richness, up to mid to high levels of carbon stocks. An increase in the carbon stock associated with lyi ng deadwood is often linked to a higher stand age 44, with a potential reduced light availability at the forest ﬂoor14,15. However, at mid to high levels of carbon stocks this trend is not evident, likely counterbalanced by the occurrence of gap makers in forests with high amounts of deadwood 45.B r y o p h y t e sd i d not show a positive response to increasing carbon stock in lying deadwood, even if previous works have highlighted decaying wood as a crucial substrate for the diversity of bryophytes in general within forest ecosystems 35,46. Nevertheless, the relatively low number of Table 1 | Statistics of the selected boosted regression trees (BRTs) Learning rate Tree complexity Bag fraction Explained deviance Cross-validated mean correlation coef ﬁcient Self statistics Beetles 0.005 5 0.75 0.82 0.78 0.91 Birds 0.005 5 0.75 0.77 0.81 0.88 Bryophytes 0.005 5 0.75 0.80 0.76 0.90 Fungi 0.005 5 0.75 0.77 0.77 0.88 Lichens 0.005 5 0.75 0.84 0.86 0.92 Vascular plants 0.005 5 0.75 0.55 0.61 0.75 Article https://doi.org/10.1038/s41467-026-68668-x Nature Communications| (2026) 17:1976 3 bryophyte species in European forest ecosystems47,48 may have ham- pered the detection of a consistent increasing trend in species richness across our database. Deadwood diversity, stand openness and large lying deadwood volume are largely recognized as the main drivers of saproxylic beetle species richness in temperate forests 49. However, we were not able to detect a substantial positive response of this group to an increasing carbon stock in lying deadwood. This unexpected out- come may be associated with the fact that the vast majority of sam- pling units in the dataset showed a relatively low amount of lying deadwood volume, mostly composed of relatively ﬁne deadwood fragments (see Supplementary Fig. 2), which may not be able to sup- port the larvae of a high share of saproxylic beetle species. Interest- ingly, we found different results for standing deadwood, with a strong positive response of saproxylic beetles and a weak positive response for fungi, likely in relation to the generally larger sizes of standing deadwood, which, however, is less easily colonized by fungi 49.T h e o - retically, an increasing amount of standing deadwood carbon is asso- ciated with an increasing amount of surface available for epixylic groups, i.e., bryophytes and lichens. For these groups, however, an increase of available surface does not necessarily mean an increase in s p e c i e sr i c h n e s s ,w h i c hi sr a t h e ri nﬂuenced by tree species composition 50, stand-scale ecological continuity 51, local species dis- persal capacity 52, microclimatic conditions 53 or seasonal rainfall distribution54. Increasing aboveground carbon stock in forests does not neces- sarily lead to a signiﬁcant increase in species richness across different taxonomic groups. Species richness responses vary considerably depending on the carbon pool an d taxonomic group. While above- ground living carbon stock showed weak or negative responses for most taxonomic groups, deadwood carbon stocks, including both lying and standing deadwood, demonstrated positive relationships with saproxylic, epiphytic, and epixylic groups. European and global policies focused on climate change mitigation may not jointly achieve the biodiversity conservation targets they emphasize 6 if they continue to prioritize carbon stocks in aboveground living biomass, following carbon accounting initiatives (e.g., REDD+)28. Although recent EU fra- meworks, such as the nature restoration law indicators 55,i n t e g r a t e other carbon pools (e.g., lying and standing deadwood, soil carbon), these components remain underrepresented in practical imple- mentation and policy targets. Pursuing the target of carbon and bio- diversity rich forests should account for the complex pathways that may link these two objectives, e.g., by promoting biodiversity-friendly measures in highly productive forest stands through integrated closer- to-nature forest management 56. Similarly, measuring living biomass without accounting for successional pathways, human and natural disturbance regimes, landscape context and species composition, may not be used as a proxy of multi-taxon forest biodiversity, whereas it is commonly used as such 28,57. This is particularly relevant in view of the current proposal for a European Union forest monitoring law and the LULUCF regulation 58, whose implementation continues to prioritize carbon stocks in living biomass59. On the other hand, our work highlights the importance of dead- wood carbon pools for both climate change mitigation and biodi- versity conservation targets. Over the past decades, the extraction of wood from European forests has been increasing, with growing pres- sures on logging residuals 60,61, as required by the need to achieve cli- mate neutrality by substituting fossil-based materials62.A d d i t i o n a l l y , the limited assessment of deadwood forest carbon stock 63,c o n - strained by the scarcity of broad-scale data64 might have contributed to an underestimation of both the losses in deadwood carbon pools in European forests, and of the potential impacts of deadwood- dependent biodiversity 65. It is crucial for policymakers to consider the multifaceted nature of these ecological interactions. Innovative approaches and case studies that address these dual objectives are required to inform and re ﬁne policy frameworks towards compre- hensive and effective forest management strategies. Opportunities and limitations Local environmental conditions and sampling protocols in ﬂuenced the variations in species richness for several taxonomic groups. The -0.1 0.0 0.1 0.2 0.3 0123 C stock stand alive scaled species richness Beetles -0.1 0.0 0.1 0.2 0.3 0123 C stock stand alive scaled species richness Birds -0.1 0.0 0.1 0.2 0.3 0123 C stock stand alive scaled species richness Bryophytes -0.1 0.0 0.1 0.2 0.3 0123 C stock stand alive scaled species richness Fungi -0.1 0.0 0.1 0.2 0.3 0123 C stock stand alive scaled species richness Lichens -0.1 0.0 0.1 0.2 0.3 0123 C stock stand alive scaled species richness Vascular Plants Fig. 2 | Partial dependence plots of the marginal effect of standing alive C stock on scaled species richness of studied taxonomic groups.Solid lines represent ﬁtted functions from boosted regression-tree models, while dashed lines indicate smoothed trends summarizing overall patterns. Black stripes on the top of each graph represent the distribution of the data. Source data are provided as a Source data ﬁle. Article https://doi.org/10.1038/s41467-026-68668-x Nature Communications| (2026) 17:1976 4 discrepancies among sampling methods signiﬁcantly affect biodi- versity data, whose comparability across studies would greatly beneﬁt from harmonization and standardization processes66. This need for protocol standardization is particularly urgent for the effectiveness of forest monitoring in Europe59. Beyond methodological issues, local environmental conditions play a crucial role in shaping species diver- sity across taxonomic groups. Context-related predictors explained a high share of the variation in species richness we observed, thus emphasizing the need to further investigate the effects of local con- ditions on the carbon-biodiversity relationship. -0.1 0.0 0.1 0.2 0.3 -4 -2 0 2 C stock stand dead scaled species richness Beetles -0.1 0.0 0.1 0.2 0.3 -4 -2 0 2 C stock stand dead scaled species richness Birds -0.1 0.0 0.1 0.2 0.3 -4 -2 0 2 C stock stand dead scaled species richness Bryophytes -0.1 0.0 0.1 0.2 0.3 -4 -2 0 2 C stock stand dead scaled species richness Fungi -0.1 0.0 0.1 0.2 0.3 -4 -2 0 2 C stock stand dead scaled species richness Lichens -0.1 0.0 0.1 0.2 0.3 -4 -2 0 2 C stock stand dead scaled species richness Vascular Plants Fig. 4 | Partial dependence plots of the marginal effect of standing deadwood C stock on scaled species richness of studied taxonomic groups. Solid lines represent ﬁtted functions from boosted regression-tree models, while dashed lines indicate smoothed trends summarizing overall patterns. Black stripes on the top of each graph represent the distribution of the data. Source data are provided as a Source data ﬁle. -0.1 0.0 0.1 0.2 0.3 -2 -1 0 1 2 C stock lying dead scaled species richness Beetles -0.1 0.0 0.1 0.2 0.3 -2 -1 0 1 2 C stock lying dead scaled species richness Birds -0.1 0.0 0.1 0.2 0.3 -2 -1 0 1 2 C stock lying dead scaled species richness Bryophytes -0.1 0.0 0.1 0.2 0.3 -4 -2 0 2 C stock lying dead scaled species richness Fungi -0.1 0.0 0.1 0.2 0.3 -2 -1 0 1 2 C stock lying dead scaled species richness Lichens -0.1 0.0 0.1 0.2 0.3 -4 -2 0 2 C stock lying dead scaled species richness Vascular Plants Fig. 3 | Partial dependence plots of the marginal effect of lying deadwood C stock on scaled species richness of studied taxonomic groups. Solid lines represent ﬁtted functions from boosted regression-tree models, while dashed lines indicate smoothed trends summarizing overall patterns. Black stripes on the top of each graph represent the distribution of the data. Source data are provided as a Source data ﬁle. Article https://doi.org/10.1038/s41467-026-68668-x Nature Communications| (2026) 17:1976 5

Page 1 of 2