Forest biodiversity and structure modulate human health benefits and risks

Forest biodiversity and structure modulate human health benefits and risks | 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 Article Forest biodiversity and structure modulate human health benefits and risks Loic Gillerot, Dries Landuyt, Audrey Bourdin, Kevin Rozario, Taylor Shaw, and 21 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4669329/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 19 May, 2025 Read the published version in Nature Sustainability → Version 1 posted You are reading this latest preprint version Abstract Forest risks and benefits to human health are widely recognised. Yet, variation across forest types and their ecological characteristics driving health effects remain underexplored. Based on empirical data from an interdisciplinary European forest network, we developed a Bayesian Belief Network to quantify seven causal pathways relating different forest types to physical and mental health. Results show that forests always generate net health benefits regardless of their ecological characteristics. Forest canopy density and tree species diversity emerge as key drivers, but their effect size and directionality are strongly pathway-dependent. Changes in forest canopy density can generate trade-offs. For example, forests optimised for heat buffering and air pollution mitigation may compromise medicinal plant yield and enhance Lyme disease prevalence. Tree diversity effects were weaker but more consistently positive. Forest management should therefore account for such trade-offs to tailor forest biodiversity and functioning to local public health needs of priority. Biological sciences/Psychology Earth and environmental sciences/Ecology/Biodiversity Earth and environmental sciences/Ecology/Ecological epidemiology Earth and environmental sciences/Ecology/Forest ecology Dr.FOREST ecosystem service nature-based solution environmental health environmental psychology Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction A growing body of scientific literature hails contact with nature for its physical, mental and social wellbeing benefits 1 – 5 . With 74% of deaths worldwide caused by non-communicable diseases (e.g., cardiovascular and respiratory conditions) 6 and an increasing global burden of mental health disorders 7 , preventive nature-based solutions can counterbalance multiple public health issues. Increasing the surface area of publicly accessible greenspaces has wide-ranging benefits observable at the scale of cities including, for example, reduced sales of mood disorder and cardiovascular medication 8 , heat mortality mitigation 9 and, more generally, annual preventable death reductions by 0.2–5.5% 10 . Not all greenspaces are equal, and identifying greenspace characteristics that optimise health outcomes can help offsetting the limitation posed by space restrictions 1 . Firstly, the vegetation type will modulate health outcomes 11 , 12 . Reviews comparing greenspaces such as forests, grasslands and others suggest that forests are more strongly linked with a range of physical and mental benefits 1 , 12 . With their high level of biomass and tall canopies, trees and forests may create immersive environments that are more strongly decoupled from ambient conditions than low-stature vegetation. Forests thereby foster cooler 13 and subjectively quieter 3 environments with – in most circumstances – better air quality 14 that benefit physical and mental health. A global evidence assessment affirms the supportive role of forests in the physical, mental, social and spiritual health of humans throughout all life stages 3 . Compared to greenspace quantity, the quality of a given type of greenspace also affects human health but is much less studied 11 , 12 . Quantity is commonly assessed using well-defined proxies such as surface area of vegetated land-use classes or the Normalised-Difference Vegetation Index (NDVI) 9 , 10 , representing greenspaces as relatively homogenous entities 1 , 11 . In contrast, greenspace quality is often inconsistently quantified and can encompass aspects of accessibility, safety, amenities and biodiversity 11 , 12 , 15 . In health sciences, the latter is often proxied by indices that are merely indirectly related to biodiversity including, for instance, perceived bird, plant and butterfly species richness 16 or ‘naturalness’ 5 . These are poorly aligned with standards for rigorous ecological inventories established in the natural sciences 11 , 15 , 17 . Yet, thorough quantification of biodiversity holds great promise. Five decades of biodiversity-ecosystem functioning (BEF) research have demonstrated that ecologically-relevant characteristics such as species diversity, functional traits and structural complexity greatly influence the functions and services that ecosystems provide 18 – 20 – ultimately shaping health outcomes 17 . Until now, the integration of BEF into biodiversity-health research has been limited, accentuating the need for interdisciplinary approaches. The biodiversity-health relationship is increasingly recognised, including by the Convention on Biological Diversity (CBD) and the World Health Organisation (WHO) 4 , 19 , 21 . Still, the majority of nature-health studies to date overlook biodiversity effects and fail to unravel underlying mechanisms 11 , 12 . Reported greening effects on reductions in heat mortality 9 , antidepressant prescriptions 22 and overall mortality 10 are compelling and invaluable to illustrate the magnitude of nature’s disease-prevention potential at large scales. However, the many factors co-varying with biodiversity in broad-scale longitudinal datasets limit attribution of causal effects 11 , 12 . These scientific efforts should be complemented with experimental studies with proper control conditions to understand the full complexity of nature-health relationships and their inherent context-dependency 1 , 17 , 18 . Identifying mechanisms and context-dependencies may inform how and when management interventions can optimise health outcomes. For example, increasing tree canopy cover may reduce heat and air pollution levels in rural settings, but may aggravate these environmental stressors in cities depending on specific building and street design by locally trapping heat and particulate matter 23 , 24 . Interindividual differences can also alter health outcomes, with perceived rather than actual biodiversity generally better predicting mental wellbeing outcomes 16 , 25 . Subjectivity may also affect physical wellbeing, with forest environments reducing perceived heat 26 and perceived noise 27 more strongly than expected from objective measurements. Nature and forests may also cause harm by enhancing exposure to disease-bearing vectors 28 and potentially allergenic pollen 29 . Such ecosystem disservices are rarely considered in some disciplines 17 , 21 , though physical risks related to wildlife exposure are thoroughly covered in the field of tropical medicine 3 . At last, a handful of studies insinuate that nature-health outcomes may positively or negatively interact 26 , 30 . How health benefits can be optimised and risks minimised, leveraging synergies and avoiding trade-offs between pathways, is largely unknown. Here, we use a unique continental-scale interdisciplinary database gathered in situ in 164 European forest stands (Fig. 1 a) by a team composed of ecologists, environmental psychologists, neuroscientists and public health experts to examine seven distinct forest-health pathways: i) mental wellbeing via visual stimuli; ii) mental wellbeing via auditory stimuli; iii) thermal comfort; iv) polyphenol content of medicinal plants; v) mushrooms and nutrition, vi) air quality, and vii) ticks and Lyme disease. The latter represents an important ecosystem disservice in Europe. Forest ecological characteristics were assessed using well-established forest inventory methods, enabling the attribution of health outcomes to recognized ecological variables (Fig. 1 b). The diversity in assembled datasets was synthesised using a Bayesian Belief Network (BBN), combined with an adapted version of the Ecosystem Services Cascade Model 19 , 31 providing the mechanistic framework. This flexible statistical approach allowed for the modelling of complex interaction chains, the detection of synergies and trade-offs and the inclusion of crucial but unsampled variables. Results Effects of forest characteristics on functions, services and health outcomes The effect magnitude and direction of seven focal forest characteristics depended strongly on specific forest-health pathways and their considered stage along the Cascade Model (i.e. function, service or health outcome; Fig. 2 ). Canopy density was the most influential forest property overall, followed by stand density and tree diversity. Canopy height, proportion of deciduous trees and leaf size had relatively weak effects, and stem density was least influential. However, effect magnitude was highly dependent on the forest-health pathway. Increased tree diversity was a dominant driver for mental wellbeing via visual stimuli and mushroom pathways. Canopy density had a positive effect on the thermal comfort pathway yet a negative effect on medicinal plants. Increased canopy density led to increased risk of Lyme disease, mainly by strongly increasing relative humidity thereby supporting high tick population densities. Mental wellbeing via visual stimuli was almost not influenced by actual forest characteristics (i.e., in opposition to what is subjectively perceived). Changes in canopy density, represented by Leaf Area Index (LAI), were explained by mean leaf size (6.2% of variance explained), followed by stand density (3.1%), tree diversity (1.3%) and canopy height (0.7%). We highlight these results for canopy density as a central driver, but statistics on associations between other forest characteristics can be found in the Supplementary Information (p. 11). The more intermediate steps between two variables in the model, the less sensitive they were to each other. This dilution effect starts with the forest characteristics, proceeding via function to service and then to health outcome variables. However, this top-down approach largely conceals the predictive power of intermediate variables. This can be demonstrated with the mechanisms underlying mental wellbeing via visual stimuli. While tree diversity only explained 1.2% of the variance in perceived biodiversity, the latter was strongly explained by more intermediate variables: perceived density (6.7%) and perceived naturalness (5.4%). Furthermore, perceived biodiversity itself effectively determined restorativeness (13.9%), attention restoration (4.4%) and stress reduction (4.3%). For a complete understanding, sensitivities should thus be scrutinised case-by-case using the full model (Supplementary Fig. S1 ). Trade-offs and synergies in health outcomes Based on mutual information held among variables in function, service and health outcome stages, several synergies and trade-offs emerged. A quartet of pathways were consistently associated in function, service and health outcome stages: thermal comfort, medicinal plants, air quality and ticks and Lyme disease (Fig. 3 ). This means that forest characteristics that effectively reduced heat stress also improved air quality through reducing air particulate matter. Similarly, medicinal plant production and Lyme risk reduction also synergized. However, these two pairs of pathways were antagonistic to each other. Forests that provide a high level of medicinal plant production and regulation of disease (reduced risk in Lyme) had a reduced capacity for microclimate and air quality regulation. Some weaker synergies and trade-offs were found with respect to the three remaining pathways, where especially both mental wellbeing pathways were relatively disconnected from other health outcomes. Comparing overall forest effects to inter-forest variability When health outcomes were defined relative to non-forest conditions, forest presence generated substantial absolute health benefits – regardless of forest characteristics (Fig. 4 a). Interpretation was more challenging when explicit controls were absent (Fig. 4 b), requiring a case-by-case consideration. For example, mental wellbeing outcomes were compared before and after being exposed to visual or acoustic stimuli for the mental wellbeing pathways. Overall forest effects were compared to relative differences caused by changing canopy density and tree diversity. Some health outcomes appeared to be nearly exclusively determined by forests in general, with very small changes generated by their ecological characteristics. This is most notably the case for anxiety reduction (visual), negative affect reduction (auditory), positive affect increase (visual and auditory) and PM 10 risk mitigation. For cases where forest characteristics did matter, tree diversity was most influential for mushroom health benefits. Yet, canopy density was again the strongest driver of variation: most evidently in microclimate (thermal comfort) and disease (Lyme) regulation, followed by medicinal plants- benefits and PM 2.5 risk mitigation. Analogous analyses of contrasted stand types (young plantations vs . mature natural forests) were yielding even smaller differences than shown in Fig. 4 . While changes in stand type affected canopy height very strongly, the latter had generally little influence on other variables (Fig. 2 ). Discussion Drivers of forest functioning, services and health outcomes We here report on the first trans-continental study linking forest biodiversity, functioning and health outcomes, capitalising on interdisciplinary approaches. For example, recent heat mitigation research reported which forest characteristics determine surface or air temperature and relative humidity of forest microclimates 35 , but few have connected this to indices relevant to human perception and health outcomes 33 . In parallel, heat mitigation by ‘homogenous’ greenspaces is already thoroughly reported 9 , 36 , but ignores ecological intricacies. By using a BBN to merge multiple data sources, this gap was effectively bridged for multiple forest-health pathways. The two most influential and independent forest characteristics were forest canopy density and tree diversity. Canopies are key drivers of ecosystem functioning 18 , creating a physical, chemical and biological filter to ambient conditions 37 . Tree diversity is also a recognised determinant of ecosystem function, ecosystem service provision and non-tree biodiversity 18 , 38 . Beyond confirming these known consequences, we here show that they also shape human health impacts. Even though effect sizes are relatively small, it may make a substantial difference at the scale of populations at which greenspaces can prevent well over hundreds of deaths 9 , 10 . The process of disentangling mechanistic pathways via the Ecosystem Service Cascade Model also revealed dependencies at spatiotemporal, societal and individual levels. At the spatiotemporal level, the importance of forests to human health was defined by the surrounding environmental conditions, where locations with high particulate matter pollution and periods with high heat loads benefited proportionally more from forest tempering effects 13 . Likewise, seasonality determines production of mushrooms 39 and medicinal plants 40 , and the amount of songbird-generated biophony with potential mental benefits 41 . Societal context also plays a crucial role: local legislation and overall foraging culture are determinants for potential health benefits from non-timber forest products 42 . For example, while mushroom and berry collection is common in e.g. Poland, Germany and France 42 , it is legally not allowed in Flanders, Belgium. Mental wellbeing is also determined by societal realities: forests and dense vegetation can provoke fear of assault by criminals or wild animals 43 . This implies that our findings are best applicable to the world’s forests that are perceived as ‘safe’. At last, individual experiences may further shape health outcomes. Fear of crime varies with individual factors such as age, gender and culture 43 . Mental wellbeing was relatively insensitive to actual changes in forest characteristics in our analysis. This is partly explained by the pivotal role of perceived (visual and auditory) biodiversity, which can be based on cues like species richness, greenness, acoustic complexity or functional traits like colourfulness 25 – thereby recurrently deviating from actual biodiversity 21 . Half of variation in thermal comfort is determined by factors other than the physical microclimate, including physiological acclimatisation, biological sex, thermal preference and expectations 44 . Trade-offs bar all-round optimisation Health outcome interactions were most clear for heat mitigation, medicinal plants, air quality and Lyme disease risk, with canopy density as the common denominator. The latter enhances mitigation of heat stress 33 , 35 and particulate matter 34 , but reduces understory light availability 45 which impedes production of medicinal plants. Closed canopies also create ideal conditions for tick activity and survival by maintaining relative humidity levels that rarely fall below 80% 46 . Whereas canopy density is the most influential forest characteristic, corroborating a recent study on ecosystem services 47 , it can inadvertently lead to unwanted side-effects. This does not apply to tree diversity: though less influential, it always returned positive effects – except for a small increase in Lyme disease risk. Note that the effect of increasing tree diversity and forest complexity on Lyme risk in Europe is contested, with reports of both negative 48 and positive 49 outcomes. Some other, weaker, interactions were observed or theoretically expected. A separate analysis found enhanced levels of thermal and mental wellbeing to mutually reinforce each other, where forest environments incite this synergy 26 . This non-linear, subtle interaction did not clearly emerge here (Fig. 3 ). Other indirectly explored interactions include the increase in questing tick nymphs with higher levels of medicinal plants, because ticks use understory vegetation to reach vertebrate hosts 28 , 49 . At the same time, more lush understory vegetation affected perceived density, which was a driver for perceived biodiversity and thus potentially determines mental wellbeing. Greater bird diversity increases biophony, but also influences tick nymphal infection prevalence as birds are suitable hosts for multiple Borrelia strains 50 . Some interesting interactions could not be explored. High rainfall frequency and abundance may increase mushroom biomass 39 , but also improve the air filtering capacity of the forest by cleaning the leaves and rendering them more sticky 51 . Given that air pollution may worsen mental health 52 , filtering of particulate matter by forests may have additional mental health effects. Again, this emphasises the need to consider results in the light of local needs. In a theoretical region with low-to-absent Lyme prevalence and foraging regulations restricting medicinal plant harvesting, silvicultural interventions may aim for closed canopies to foster the synergy in heat and particulate matter improvements. Prioritisation of target health outcomes should follow the health impact, the number of people affected and forest effect magnitude given the local context. For instance, in 2022, air pollution and heat caused an estimated 300 000 53 and 60 000 54 deaths in Europe, respectively, which could be effectively mitigated by forests 9 . In contrast, despite widespread malnutrition in Europe 53 , harvesting mushroom and medicinal plants may only have marginal health impacts compared to industrial sources of food and medicine production 42 . Still, these may have great importance locally and have recreational co-benefits not addressed here 42 . Suboptimal forests are better than no forests Despite the large variability, our findings indicate that forests generally have net benefits to health, regardless of their biodiversity and structure (Fig. 4 ). Mental wellbeing effects were especially little influenced by actual measures because they essentially revolve around perceived forest characteristics, whereas predicted heat mortality strongly varied between forest types. Yet, even the least optimal forest stands led to strong heat reductions. Indeed, even young plantations have been found to reduce perceived temperatures by 10°C under hot conditions 33 . In fact, the whole variety of forest conditions was generally unequivocally favourable to human health, except for ticks which are especially abundant in forests compared to any other habitat 28 . Altering forest characteristics is thus less consequential than the mere presence of forests, which has two positive implications. First, studies considering forests as homogenous entities certainly ignore plenty of variation, yet their findings may be roughly applicable to a wide range of forest conditions. Second, increasingly prevalent reforestation efforts, despite being mostly constituted of young plantation forests 55 , may already yield substantial health benefits to people. Strengths and weaknesses define avenues for future research While our modelling effort brings about insights on largely uncharted scientific terrain, limitations require consideration before extrapolating our conclusions. From the technical side, BBNs were elemental to linking forest characteristics with their health effects owing to the capacity to model complex causal pathways with heterogeneous data sources. Yet, they have inherent weaknesses. First and foremost, the mandatory discretisation of continuous variables is disadvantageous for environmental modelling, where such variables are omnipresent. This leads to significant information loss that can partly be compensated for by increasing the number of states – though this ramps up the number of required observations to populate Conditional Probability Tables 56 . We compensated for this by using expert opinion as priors, which improves information content for state combinations that are not covered by empirical data 31 . Despite these measures, a clear effect size dilution was visible especially for long causal chains. While partly due to discretisation, this incidentally also reflects real-life processes: as the myriad of unmeasured confounders and aforementioned dependencies modulate nature-health relationships, small final effects are expected. Forest-health benefits and their modulation by forest characteristics should therefore not be exaggerated. From the conceptual side, a major strength and singularity is the direct integration of multiple forest-health pathways. Yet, our selection of forest-health effects is not comprehensive. Others include, for example, reduction of noise stress, modulation of pollen allergy prevalence and fostering of spiritual wellbeing 3 . These may create additional synergies and trade-offs that we have not covered. Next, our analysis focuses on local scales (plot- or stand level) but some pathways require consideration of landscape-scale dynamics. Most notably, highly mobile vertebrates such as ungulates and birds are likely influenced by characteristics of whole forest complexes. At last, outcomes should be cautiously interpreted in light of assumptions listed in the Supplementary Information. In sum, pathways assume that users interact intimately with the forest environment: our ‘virtual guinea pig’ is consistently exposed to forest microclimate, atmosphere, soundscape and visual aspects and regularly consumes non-timber forest products. These assumptions may be naive in many cases, especially for urbanites, and resonates with the challenge of quantifying nature exposure, accounting for the frequency, duration and type of nature contact 57 . In short, future research may turn towards following avenues: i) directly comparing multiple forest-health pathways and their interactions, ii) considering mechanistic processes via frameworks like the Cascade model and coupling statistical approaches like Structural Equation Models or BBNs, iii) incorporating relevant undesirable health effects, and iv) investigating how biodiversity-health relationships apply to (semi-)urban environments. Management recommendations and conclusions Despite the case-specificity and complex pathway interactions, general patterns stand out. Firstly, any European forest is likely to have beneficial effects regardless of how it is managed, certainly compared to urban environments 5 . Any type of forest is also likely to be more beneficial than low-stature vegetation for heat and particulate matter reduction, mushroom production and mental wellbeing, though forests host more ticks 28 . Yet, forest managers and policy makers have several degrees of freedom to enhance health effects and mitigate risks, with potentially considerable impact at the public health level. Our findings suggest that targeting canopy density (leaf area index) and stand density (basal area) are the most effective tools to modulate physical health outcomes in particular. Management strategies aiming at densification such as continuous cover forestry may enhance heat buffering and air filtering capacities, but may lead to an increased risk of Lyme and reduced medicinal plant yield. This clear trade-off requires careful consideration of local priorities regarding these four health outcomes. Canopy cover changes will also affect forest functioning itself, with potential cascading effects. For example, thinning may lead to increased water availability for certain tree species under drought 58 , but rapid canopy opening may also perturb understory biodiversity 35 . In contrast, increasing tree diversity represents a relatively safe, no-regret intervention with net advantages 1 and important co-benefits for biodiversity conservation 4 , 59 . In that sense, this may make it a compelling all-round management strategy in many situations with a win-win for biodiversity and human health, especially if local public health priorities are difficult to identify. Diverse forests are especially appealing given their higher resilience to multiple global change drivers and insect pests 60 , which may secure provision of health benefits in the longer term 59 . In sum, despite rather low direct health impacts, increasing forest tree diversity emerges as a robust strategy for promoting human health whereas altering forest density can be used to maximise specific benefits according to local contexts. Materials and methods Study sites and design We deployed an international network of ten forest sites along a West-East gradient in Central Europe (Fig. 1 a), covering oceanic to sub-continental climates. Of these ten sites, four comprised mature forests, while six comprised young plantations within tree diversity experiments. The mature forest stands reflected natural conditions that varied in tree diversity levels. These included the Białowieża (PL), Hainich (DE) and MIXLOR (FR) sites 61 , and stands of the TREEWEB network 62 . All were designed to obtain an even distribution of species per species richness level, and to minimise variation in environmental factors such as soil properties and topography that could covary with species composition. The young plantations 63 were aged nine to 17 years at the time of sampling. These were all located within a small perimeter using a uniform planting design that standardised the density and spatial arrangement of different species, where only tree diversity varied with negligible covariation of other environmental factors. These included B-TREE (AT), BIOTREE (DE), three FORBIO sites (BE) and ORPHEE (FR). Both site types were specifically designed to study BEF relationships, enabling the quantification of causal effects of forest ecological characteristics on human health. Selection of forest stands revolved around tree species diversity as one of the major determinants supporting overall forest biodiversity and functioning 18 , 38 . At each site, an equal number of monospecific and mixed forest plots was selected. Mixed plots were always composed of a randomised selection of species present as monospecific stands. All mixed plots contained three target tree species, which was the diversity level that all sites had in common. Non-target species were often present in mature natural stands but were never dominant, having established after plot demarcation 61 . Our selection covers a total of 164 permanent forest plots and 18 target tree species. Participating researchers were encouraged to cover as many of these plots as possible to generate directly comparable forest-health data for identical plots (“all measurements on all plots” strategy). However, due to COVID-related fieldwork impediments, most researchers failed to cover the full selection (see Supplementary Information for number of plots per pathway). The same set of forest ecological characteristics was sampled over all plots (Fig. 1 b). Central herein was tree diversity, expressed as a Shannon index. Unlike species richness, Shannon Diversity accounts for species’ relative abundances, here based on species-specific basal area shares 64 . A second characteristic receiving special attention in our analyses is canopy density, proxied by the Leaf Area Index via indirect sampling (using hemispherical photographs). Details on remaining characteristics are found in the Supplementary Information. Forest-health pathways This meta-analysis synthesises the published and unpublished works of six separate doctoral studies that gathered empirical data and insights for seven forest-health pathways, covering both mental and physical outcomes that can be considered services or disservices. Pathways are listed and shortly described in Table 1 and described in more detail in the Supplementary Information. Table 1 | Overview of submodels and their constituting studies . Submodels represent the seven forest-health pathways which have all been covered by multiple studies using interdisciplinary approaches. We briefly discuss the setup of studies of which data populated the final model, and some key findings. Included references refer to readily published studies done within the overarching project. The number of sites refers to the network shown in Fig. 1a and the symbols refer to forest-health pathways as shown in Fig. 1c. Outcome Researchers Setup Key findings A1 KR, RRYO, MRM & ABON In situ experiment involving 223 participants subjected to non-forest baseline conditions and different peri-urban forests with contrasting levels of tree species richness 32 . Complemented with a lab-based EEG experiment. Biophysical structure has weak effects, with only tree diversity and understory productivity having limited influence on short-term mental health and wellbeing. Mainly subjectively perceived forest characteristics, and especially perceived biodiversity, fostered positive mental wellbeing effects 25 . We found little congruence to actual biodiversity. A2 TS, SM, KR, AGCR, RRYO, ABON & MSL Monitoring of bird song diversity and abundance using audiologgers at eight sites. Complemented with a lab-based experiment using similar psychological questionnaires to submodel A1. Forest characteristics, especially stand density, effectively explained variation in bird song diversity. Yet, further effects were strongly tempered since mental wellbeing is essentially driven by subjectively perceived acoustic diversity, analogous to submodel A1. B LG, PDF, DL, BM & KV A thermal comfort index (Physiologically Equivalent Temperature) was measured for 14 months in eight rural sites 33 and in an urban setting 65 . How this was subjectively perceived by people was tested in a complementary experiment 26 . Cooling effects were very strongly explained by forest characteristics, with especially variables related to forest density and height being impactful, followed by multiple composition-related variables. Cooling capacity increased with increasing ambient heat stress. This was effectively experienced as such by participants. Mental wellbeing (A1) was found to positively interact with thermal comfort in forests 26 . C1 KS & BJ Inventory of medicinal plants occurring over eight sites. The biologically active substances, polyphenols, were measured in a selection of species, and extrapolated at forest plot scales. Polyphenol content is mainly determined by medicinal plant productivity rather than plant diversity, and canopy density crucially determined productivity by altering the light regime. Soil fertility had little effect. C2 KS & BJ Continuous and exhaustive sampling of mushrooms at 20 Białowieża plots over two years 39 . This was complemented by a single sampling campaign at five other sites. Because many macroscopic mushrooms are mycorrhizal and thus associate to specific tree species, tree diversity was a more influential variable than canopy cover. Health benefits are mostly driven by mushroom productivity and only little by mushroom diversity. D MS, DH, BM & DG Detailed case-study at B-TREE with quantification of particulate matter deposition for different size classes (<100, <10, < 2.5µm) 34 . Meta-analysis on tree characteristics driving deposition rates 51 . Complemented with direct LAI measures at eight sites and PM monitoring at four of them. Variation in PM pollution mitigation by forests was mainly influenced by the canopy closure and its roughness, whereas mitigation becomes stronger as ambient pollution levels increase – analogous to submodel B. E ABOU, TV & HJ Detailed case-study at ORPHEE where ticks were sampled and screened for various Borrelia strains 49 . A second sampling campaign covered several other sites. A meta-analysis 28 on forest effects on ticks further defined submodel conceptualisation. Complemented with data from the Forester 48 project, to quantify links between forest characteristics, mouse- and deer abundance and tick-related variables (DON, NIP and DIN). Forest characteristics were hypothesised risks posed by ticks and Lyme via three pathways: microclimate humidity (mediated by canopy density), understory vegetation and the vertebrate host community. Humidity levels and deer abundance strongly predicted tick densities (DON), whereas mouse abundance and understory productivity had only modest effects. The choice for Bayesian Belief Networks A BBN consists of a Directed Acyclic Graph, composed of a set of nodes (i.e. variables) connected via causal links. If node B is the causal agent of node A, node A is the child node and node B is a parent node. Each node is partitioned into a limited set of states that the node can reach. The probability of a node being in a particular state is based on so-called Conditional Probability Tables (CPT) which store probability distributions for all state combinations of parent nodes. BBNs do not generate p-values. Instead, the certainty of a given state (outcome) is related to the narrowness of the probability distribution. See Jensen & Nielsen (2007) 66 for a standard technical reference or Cain (2001) 56 for an applied introduction. We chose BBN as an analytical method as they provide a very flexible approach to integrate data from heterogeneous sources including expert knowledge, quantitative literature findings and empirical research 31 . This was essential given that empirical data from its constituting studies – despite interdisciplinary methodologies – sometimes failed to cover the whole mechanistic pathway from forest until health outcome. Other advantages include the capacity to handle small datasets and missing data, the highly visual and interactive model building process, the fast processing times, and the very intuitive way of dealing with uncertainty 31 , 67 . BBNs also have disadvantages: feedback loops and cyclic processes cannot be handled, complex models with long interaction chains lead to ‘dilution’ of effect sizes, random effects to account for e.g. spatial autocorrelation cannot be directly implemented, and generating child node CPTs becomes exponentially more cumbersome with each additional parent node 31 , 67 . Another limitation is the absent direct implementation of hierarchical data structures (e.g. random factors). Most importantly, continuous nodes need to be discretised into a limited number of states, leading to information loss 67 . Model building steps BBNs were built and adapted as individual research projects advanced. Model building was partitioned in three phases centred around three annual meetings (2021–2023) during which protocols were presented and discussed prior to application. Protocols were adapted from existing guidelines 56 , 68 . Phases overlapped in time due to continuously updating insights as fieldwork and data analysis progressed in parallel (Supplementary Information p.5). BBNs were developed in Netica (v7.11) 69 . Phase I The main goal of phase I was to create a Directed Acyclic Graph for each pathway/submodel. The framework we imposed was the Ecosystem Services Cascade Model by Haines-Young & Potschin 19 , previously adapted for BBNs 31 and here applied to health outcomes. This forced partners to disentangle their forest-health pathway into a mechanistic conceptual model with four stages. The first stage groups nodes defining the biophysical structure of the ecosystem (e.g. tree diversity, canopy height). This defines ecosystem functioning (e.g. heat buffering, particulate matter retention), which influences (dis-)service provisioning (e.g. medicinal plant productivity, attention restoration), which will ultimately influence the variables defining the health outcome (e.g. Lyme disease risk, positive affect). Partners avoided feedback loops and more than three parent nodes for the same child node whenever possible 68 . They prioritized nodes being empirically sampled and respected relationship causality: “ Draw arrows from node A to node B if changes in A will lead to changes in B, but not the opposite ”. We acknowledge the more recent Nature’s Contributions to People framework 70 , but use ecosystem services to comply with the Cascade model. Phase II Phase II aimed at identifying interactions between submodels and to define node’s state thresholds. Forest characteristics that were shared in multiple submodels were jointly defined and constitute the overarching model structure shown in Fig. 1 c. This represents current knowledge in forest ecological interactions. Next, we explored interactions between submodels to increase model cohesion and the potential to unveil trade-offs and synergies. At last, nodes were partitioning according to specific state thresholds. The following guidelines were provided for continuous variables: i) define the node’s unit and potential range, ii) define the minimum number of possible states that still makes logical sense (maximum five) 68 , iii) define threshold values based on published guidelines (e.g., WHO air quality guidelines), recognized classifications (e.g., heat stress categories) or logical reasoning (e.g., increasing vs . decreasing anxiety) 71 , and iv) detailed reasoning, define state names whenever possible, and list key references. When obvious threshold values were lacking, data percentiles were used. When a lot of empirical data was available and a child node had few parents, the state number limitation could be relaxed because the minimum data amount is alleviated. Phase III The last phase quantified and joined submodels. The four major classes that can be combined in BBNs are raw empirical data, quantitative literature results, logical equations or expert opinion 31 , 56 . Each has advantages and disadvantages. Empirical raw data are very easy to incorporate, especially when available for multiple nodes simultaneously, but the underlying counting algorithm requires sufficient data points for robust probability estimation. Quantitative literature findings can include formulae with regression coefficients, or estimates with error distributions. When available, priority was given to robust meta-analyses. They were most often used to quantify health outcomes, given that multiple of our studies only gathered empirical data until the service stage. Their advantage is the relatively easy implementation in Netica, which integrates numerous error distribution functions, but finding published results with the same predictor set becomes more challenging as the number of parent nodes increases. Logical equations usually link parent nodes via basic mathematical relationships. For example, the density of infected ticks (DIN) equals the density of questing ticks (DON) times the nymphal infection prevalence (NIP). They do not cause an effect dilution and are easy to incorporate, with few disadvantages. At last, expert opinion data are used when quantitative data are unavailable or not robust enough, which also represents their main advantage. We used a direct expert elicitation style based on recommendations by Kuhnert and colleagues 72 , assuming that our expert teams were made sufficiently familiar with probability theory through annual meetings and individual discussions. Here, experts had to estimate probability distributions for each combination of parent node states, thereby directly generating CPTs. The disadvantage is that this becomes a very cognitively-demanding endeavour as the number of parent nodes increases, which exponentially augments possible state combinations. To support partners in identifying data classes, a custom-made decision tree was presented (Supplementary Information p. 8). The decision tree aims to circumvent aforementioned weaknesses or provides alternatives by combining classes. For example, using only raw empirical data is discouraged whenever there are less than ten observations per combination of parent states, which is a relaxed version of the untested 73 recommendation by Cain (2001) 56 . Concretely, when one parent has three states and the second has four, the empirical dataset should have at least 120 observations for the child node. When this is not reached but empirical data are available, users are redirected to options where literature results or expert opinion data are used as priors 72 before ‘belief updating’ is done with raw data as posteriors. The decision tree also implies a priority ranking in classes: whenever possible, own empirical data was prioritised over literature results, which was prioritised over expert opinion data. The common forest structure was exclusively quantified using empirical data. However, since all submodels relied on it and the canopy density node had four parents, additional data were retrieved. Its eight constituting nodes (plus ‘understory productivity’ and ‘soil fertility’ nodes) were reinforced with data from international forest networks that were designed to study forest biodiversity effects on ecosystem functioning, analogous to our network. These include 225 FORMICA 47 plots and the 41 TREEWEB 62 plots that were not yet included in our network, for a total of 430 plots. As a final submodel-specific procedure, a formal validation meeting was organised after submodels were operationalized in Netica. Results of sensitivity analyses and general submodel behaviour were scrutinised for inconsistencies by our expert partners 67 , allowing for final adaptations before merging them with other submodels. At last, all submodels were joined to the common structure and interactions were added. Some interactions were indicated as merely ‘implicit’ associations when two nodes were not directly comparable or already had the same parent node combinations. General model insights and shared forest nodes With 82 nodes (i.e. variables), 123 links and 3154 conditional probabilities, the final model is among the largest published BBNs in ecosystem service modelling 31 . Most nodes (n = 56; 68%) were exclusively quantified using our own empirical data. Ten were operationalized via logical equations and nine via quantitative literature findings. Expert opinion elicitation was used for eight nodes, but always as priors with posterior belief updating using empirical data when the minimal number of observations was lacking. The model building exercise yielded an extensive Supplementary Information document that is appended for maximal transparency and comprehensibility of the numerous choices underlying submodels. It contains information on general methodology, submodel-specific summaries and limitations, and node-specific details including variable definitions, state thresholds and key references. As only part of the full complexity of our final BBN can be exhibited here, this reference document should accompany the model’s usage and interpretation. Synthesis analyses Once the complete BBN was assembled, it was subjected to three main analyses. First, the explanatory power of the common forest characteristics was quantified using built-in sensitivity analyses, represented by the variance reduction of the expected real value of the child node due to a finding in the parent 69 . Second, we extracted the mutual information between health outcome nodes to identify trade-offs and synergies. This is quantified as the reduction in entropy in one node, measured in computational bits, due to a finding in the other node 69 . Because health outcome nodes are separated by long interaction chains that cause effect dilutions 56 , this yielded low values. Threshold values were established to distinguish robust associations from the bulk of negligible ones. If remaining associations were visible between nodes of different submodels in each stage until health outcomes, they were considered the strongest and most consistent associations. Those only emerging in function or service stages were considered secondary associations. Third and last, the magnitude of health outcomes was expressed relative to non-forest conditions and among forests. To illustrate potential inter-forest differences, four scenarios were considered based on two of the most influential and independent forest characteristics: monospecific or mixed stands with open (LAI 3) canopies. Per scenario, 10 000 cases were simulated in Netica to obtain means and uncertainty distributions. In about half of the forest-health pathways, health outcomes were explicitly expressed relative to control conditions, which could be urban environments (mental wellbeing – visual) or ambient, non-forest conditions (thermal comfort and air quality). In the other cases, suitable control conditions were generally less evident. For mental mental wellbeing via auditory stimuli, the baseline were measures taken prior to the intervention, which was already accounted for by looking at mental state changes (Δ). For medicinal plants and mushrooms, baselines can be assumed to represent no intake. For Lyme disease, baselines would be environments with risk absence such as highly urbanised locations. Declarations Acknowledgements This research was funded by the ERA-Net BiodivERsA project Dr.FOREST, with the national funders German Research Foundation (DFG − 428795724, Germany), French National Research Agency (ANR, France), Research Foundation – Flanders (FWO, Belgium), Austrian Science Fund (FWF, Austria) and National Science Center (NCN, Poland, project no. 2019/31/Z/NZ8/04032), as part of the 2018–2019 BiodivERsA call for research proposals. PDF received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (ERC Starting Grant FORMICA 757833). DL was supported by a postdoctoral fellowship of the FWO. TV was supported by the UGent GOA project (BOF20/GOA/009). QP would like to thank the Walloon forest service (SPW – DNF) for its support to the maintenance of the FORBIO-Gedinne experiment in the frame of the 5-yr research programme ‘Plan quinquennal de recherche et de vulgarisation forestières’. References Beute F et al (2023) How do different types and characteristics of green space impact mental health? A scoping review. People Nat 5:1839–1876 Hartig T, Mitchell R, de Vries S, Frumkin H (2014) Nature and Health. Annu Rev Public Health 35:207–228 Konijnendijk C, Devkota D, Mansourian S, Wildburger C (2023) Forests and Trees for Human Health: Pathways, Impacts, Challenges and Response Options. A Global Assessment Report . 232 WHO. Nature, Biodiversity and health: an overview of interconnections. 42 (2021) Reyes-Riveros R et al (2021) Linking public urban green spaces and human well-being: A systematic review. Urban Forestry Urban Green 61:127105 WHO (2022) Noncommunicable diseases progress monitor. World Health Organization Patel V et al (2018) The Lancet Commission on global mental health and sustainable development. Lancet 392:1553–1598 Chi D et al (2022) Residential Exposure to Urban Trees and Medication Sales for Mood Disorders and Cardiovascular Disease in Brussels, Belgium: An Ecological Study. Environ Health Perspect 130:057003 Iungman T et al (2023) Cooling cities through urban green infrastructure: a health impact assessment of European cities. Lancet 0 Barboza EP et al (2021) Green space and mortality in European cities: a health impact assessment study. Lancet Planet Health 5:e718–e730 van den Berg M et al (2015) Health benefits of green spaces in the living environment: A systematic review of epidemiological studies. Urban Forestry Urban Green 14:806–816 Nguyen P-Y, Astell-Burt T, Rahimi-Ardabili H, Feng X (2021) Green Space Quality and Health: A Systematic Review. Int J Environ Res Public Health 18:11028 De Frenne P et al (2019) Global buffering of temperatures under forest canopies. Nat Ecol Evol 3:744–749 Diener A, Mudu P (2021) How can vegetation protect us from air pollution? A critical review on green spaces’ mitigation abilities for air-borne particles from a public health perspective - with implications for urban planning. Sci Total Environ 796:148605 Ekkel ED, de Vries S (2017) Nearby green space and human health: Evaluating accessibility metrics. Landsc Urban Plann 157:214–220 Dallimer M et al (2012) Biodiversity and the Feel-Good Factor: Understanding Associations between Self-Reported Human Well-being and Species Richness. Bioscience 62:47–55 Aerts R, Honnay O, Van Nieuwenhuyse A (2018) Biodiversity and human health: mechanisms and evidence of the positive health effects of diversity in nature and green spaces. Br Med Bull 127:5–22 Brockerhoff EG et al (2017) Forest biodiversity, ecosystem functioning and the provision of ecosystem services. Biodivers Conserv 26:3005–3035 Haines-Young R, Potschin M (2010) The links between biodiversity, ecosystem services and human well-being. in Ecosystem Ecology (eds. Raffaelli, D. G. & Frid, C. L. J.) 110–139Cambridge University Press, 10.1017/CBO9780511750458.007 Tilman D, Isbell F, Cowles JM (2014) Biodiversity and Ecosystem Functioning. Annu Rev Ecol Evol Syst 45:471–493 Marselle MR et al (2021) Pathways linking biodiversity to human health: A conceptual framework. Environ Int 150:106420 Marselle MR (2020) Urban street tree biodiversity and antidepressant prescriptions. Sci Rep 11 Krayenhoff ES et al (2021) Cooling hot cities: a systematic and critical review of the numerical modelling literature. Environ Res Lett 16:053007 Han D, Shen H, Duan W, Chen L (2020) A review on particulate matter removal capacity by urban forests at different scales. Urban Forestry Urban Green 48:126565 Rozario K et al (2024) Perceived biodiversity: is what we measure also what we see and hear? 2024.04.03.587905 Preprint. at https://doi.org/10.1101/2024.04.03.587905 Gillerot L et al (2024) Forests are chill: The interplay between thermal comfort and mental wellbeing. Landsc Urban Plann 242:104933 Rey Gozalo G, Barrigón Morillas JM, Montes González D (2018) Atanasio Moraga, P. Relationships among satisfaction, noise perception, and use of urban green spaces. Sci Total Environ 624:438–450 Bourdin A et al (2023) Forests harbor more ticks than other habitats: A meta-analysis. For Ecol Manag 541:121081 Dudek T, Kasprzyk I, Dulska-Jeż A (2018) Forest as a place for recreation but also the source of allergenic plant pollen: to come or avoid? Eur J For Res 137:849–862 Park B-J et al (2011) Relationship between psychological responses and physical environments in forest settings. Landsc Urban Plann 102:24–32 Landuyt D et al (2013) A review of Bayesian belief networks in ecosystem service modelling. Environ Model Softw 46:1–11 Rozario K et al (2024) The more the merrier? Perceived forest biodiversity promotes short-term mental health and well-being—A multicentre study. People Nat 6:180–201 Gillerot L et al (2022) Forest structure and composition alleviate human thermal stress. Glob Change Biol 28:7340–7352 Steinparzer M, Haluza D, Godbold DL (2022) Integrating Tree Species Identity and Diversity in Particulate Matter Adsorption. Forests 13:481 Zellweger F et al (2020) Forest microclimate dynamics drive plant responses to warming. Science 368:772–775 Wong NH, Tan CL, Kolokotsa DD, Takebayashi H (2021) Greenery as a mitigation and adaptation strategy to urban heat. Nat Rev Earth Environ 2:166–181 Verheyen K et al (2024) Forest canopies as nature-based solutions to mitigate global change effects on people and nature. J Ecol Ampoorter E et al (2020) Tree diversity is key for promoting the diversity and abundance of forest-associated taxa in Europe. Oikos 129:133–146 Stojek K, Gillerot L, Jaroszewicz B (2022) Predictors of mushroom production in the European temperate mixed deciduous forest. For Ecol Manag 522:120451 Gori A, Nascimento LB, Ferrini F, Centritto M, Brunetti C (2020) Seasonal and Diurnal Variation in Leaf Phenolics of Three Medicinal Mediterranean Wild Species: What Is the Best Harvesting Moment to Obtain the Richest and the Most Antioxidant Extracts? Molecules 25:956 Grinfeder E et al (2022) Soundscape dynamics of a cold protected forest: dominance of aircraft noise. Landsc Ecol 37:567–582 Schulp CJE, Thuiller W, Verburg PH (2014) Wild food in Europe: A synthesis of knowledge and data of terrestrial wild food as an ecosystem service. Ecol Econ 105:292–305 Jansson M, Fors H, Lindgren T, Wiström B (2013) Perceived personal safety in relation to urban woodland vegetation – A review. Urban Forestry Urban Green 12:127–133 Lai D et al (2020) A comprehensive review of thermal comfort studies in urban open spaces. Sci Total Environ 742:140092 Blondeel H et al (2020) Light and warming drive forest understorey community development in different environments. Glob Change Biol 26:1681–1696 Gray JS (1998) Review The ecology of ticks transmitting Lyme borreliosis. Exp Appl Acarol 22:249–258 Vanneste T et al (2024) Trade-offs in biodiversity and ecosystem services between edges and interiors in European forests. Nat Ecol Evol 1–8. 10.1038/s41559-024-02335-6 Vanroy T et al (2024) The effect of forest structural complexity on tick-borne pathogens in questing ticks and small mammals. For Ecol Manag 562:121944 Bourdin A et al (2022) Forest Diversity Reduces the Prevalence of Pathogens Transmitted by the Tick Ixodes ricinus. Front Ecol Evol. 10.3389/fevo.2022.891908 Ostfeld RS, Keesing F (2012) Effects of host diversity on infectious disease. Annu Rev Ecol Evol Syst 43:157–182 Steinparzer M, Schaubmayr J, Godbold DL, Rewald B (2023) Particulate matter accumulation by tree foliage is driven by leaf habit types, urbanization- and pollution levels. Environ Pollut 335:122289 Cao Z, Zhou J, Li M, Huang J, Dou D (2023) Urbanites’ mental health undermined by air pollution. Nat Sustain 6:470–478 OECD (2022) Health at a Glance: Europe 2022: State of Health in the EU Cycle. Organisation for Economic Co-operation and Development Ballester J et al (2023) Heat-related mortality in Europe during the summer of 2022. Nat Med 29:1857–1866 Freer-Smith P et al (2019) Plantation forests in Europe: challenges and opportunities. 10.36333/fs09 Planning improvements in natural resources management: Guidelines for using Bayesian networks to support the planning and management of development programmes in the water sector and beyond . (2001) Bratman GN et al (2019) Nature and mental health: An ecosystem service perspective. Sci Adv 5:eaax0903 Sohn JA, Saha S, Bauhus J (2016) Potential of forest thinning to mitigate drought stress: A meta-analysis. For Ecol Manag 380:261–273 Kilpatrick AM, Salkeld DJ, Titcomb G, Hahn MB (2017) Conservation of biodiversity as a strategy for improving human health and well-being. Phil Trans R Soc B 372:20160131 Jactel H, Moreira X, Castagneyrol B (2021) Tree Diversity and Forest Resistance to Insect Pests: Patterns, Mechanisms and Prospects. Annual Review of Entomology 66, null Baeten L et al (2013) A novel comparative research platform designed to determine the functional significance of tree species diversity in European forests. Perspect Plant Ecol Evol Syst 15:281–291 De Groote S et al (2017) Tree species identity outweighs the effects of tree species diversity and forest fragmentation on understorey diversity and composition. Plant Ecol Evol 150:229–239 Paquette A et al (2018) A million and more trees for science. Nat Ecol Evol 2:763–766 Nickmans H (2019) The nutrition of oak and beech trees along a tree diversity gradient Gillerot L, Landuyt D, De Frenne P, Muys B, Verheyen K (2024) Urban tree canopies drive human heat stress mitigation. Urban Forestry Urban Green 92:128192 Jensen FV, Nielsen TD (2007) Bayesian networks and decision graphs. Springer Aguilera PA, Fernández A, Fernández R, Rumí R, Salmerón A (2011) Bayesian networks in environmental modelling. Environ Model Softw 26:1376–1388 Marcot BG, Steventon JD, Sutherland GD, McCann RK (2006) Guidelines for developing and updating Bayesian belief networks applied to ecological modeling and conservation. Can J Res 36:3063–3074 Norsys Software Corporation. Netica Application User’s Guide (1998) Pascual U et al (2017) Valuing nature’s contributions to people: the IPBES approach. Curr Opin Environ Sustain 26–27:7–16 Uusitalo L (2007) Advantages and challenges of Bayesian networks in environmental modelling. Ecol Model 203:312–318 Kuhnert PM, Martin TG, Griffiths SP (2010) A guide to eliciting and using expert knowledge in Bayesian ecological models. Ecol Lett 13:900–914 Chen SH, Pollino CA (2012) Good practice in Bayesian network modelling. Environ Model Softw 37:134–145 Additional Declarations There is NO Competing Interest. Supplementary Files Fig.S1.pdf Fig.S2.pdf SupplementaryInformation.pdf Cite Share Download PDF Status: Published Journal Publication published 19 May, 2025 Read the published version in Nature Sustainability → 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-4669329","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":346991467,"identity":"1d80b725-128d-4d7c-94b0-3c8d34e8e5f8","order_by":0,"name":"Loic Gillerot","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8ElEQVRIiWNgGAWjYBACPgbGBiAlIcMH5lZARJnxaWGDauFhA/EOnIFoIKAFAiBaDrYRo4W9ufEDwy8LHjb23oePP847bNcv3X+AuaACjxaeg80SjH1Ah/EcNzY4uO1w8sw5hxmYZ5zBo0UisY2BsQeoRSKNTQKkxeBGMgMzbxsxWuSfAbXMOZxsD9byj4AWhh8gW9iAWhoO2xlIgLQ0EPBLYgPIL2nMBmeOpSdI3Eg2ODzjGG4t/OztDz98+FMnx89+jPFBRY21Pf+MxIePC2pwawGDRCTPJoKcdICABiD4g2DaE1Y9CkbBKBgFIw0AAOd/RtYXMQAqAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-0699-4478","institution":"Ghent University","correspondingAuthor":true,"prefix":"","firstName":"Loic","middleName":"","lastName":"Gillerot","suffix":""},{"id":346991468,"identity":"d8e384f9-59cd-4029-b2e5-d7cb64cc9fa2","order_by":1,"name":"Dries Landuyt","email":"","orcid":"","institution":"Ghent University","correspondingAuthor":false,"prefix":"","firstName":"Dries","middleName":"","lastName":"Landuyt","suffix":""},{"id":346991469,"identity":"1f74ecda-d6b4-4b32-8110-6d4808253e6e","order_by":2,"name":"Audrey Bourdin","email":"","orcid":"","institution":"University of Bordeaux","correspondingAuthor":false,"prefix":"","firstName":"Audrey","middleName":"","lastName":"Bourdin","suffix":""},{"id":346991470,"identity":"95f4d894-5d9c-4038-9543-3da84b3f6251","order_by":3,"name":"Kevin Rozario","email":"","orcid":"","institution":"German Centre for Integrative Biodiversity Research (iDiv)","correspondingAuthor":false,"prefix":"","firstName":"Kevin","middleName":"","lastName":"Rozario","suffix":""},{"id":346991471,"identity":"7d8ea394-72d3-4255-807d-036eda99b618","order_by":4,"name":"Taylor Shaw","email":"","orcid":"","institution":"University of Freiburg","correspondingAuthor":false,"prefix":"","firstName":"Taylor","middleName":"","lastName":"Shaw","suffix":""},{"id":346991472,"identity":"22ad6d07-e596-4243-b37a-fd5bb464e339","order_by":5,"name":"Matthias Steinparzer","email":"","orcid":"","institution":"BOKU University","correspondingAuthor":false,"prefix":"","firstName":"Matthias","middleName":"","lastName":"Steinparzer","suffix":""},{"id":346991473,"identity":"8f0e91b9-0ebc-42ea-8cbd-b29df4da1498","order_by":6,"name":"Katarzyna Stojek","email":"","orcid":"","institution":"University of Warsaw","correspondingAuthor":false,"prefix":"","firstName":"Katarzyna","middleName":"","lastName":"Stojek","suffix":""},{"id":346991474,"identity":"5ec07b05-6992-40e3-9e60-a2ced0ed3ef4","order_by":7,"name":"Tosca Vanroy","email":"","orcid":"","institution":"Ghent University","correspondingAuthor":false,"prefix":"","firstName":"Tosca","middleName":"","lastName":"Vanroy","suffix":""},{"id":346991475,"identity":"cd61a03c-f5b8-48ba-a17b-2c901de5cb46","order_by":8,"name":"Ana Gabriela Cuentas Romero","email":"","orcid":"https://orcid.org/0000-0001-7948-1115","institution":"University of Freiburg","correspondingAuthor":false,"prefix":"","firstName":"Ana","middleName":"Gabriela Cuentas","lastName":"Romero","suffix":""},{"id":346991476,"identity":"9c7fade8-c69a-425d-b585-a794a23d870b","order_by":9,"name":"Sandra Müller","email":"","orcid":"https://orcid.org/0000-0003-4289-755X","institution":"University of Freiburg","correspondingAuthor":false,"prefix":"","firstName":"Sandra","middleName":"","lastName":"Müller","suffix":""},{"id":346991477,"identity":"a6f73753-d6aa-46e5-bca0-a0e472b08a7e","order_by":10,"name":"Rachel Oh","email":"","orcid":"","institution":"German Centre for Integrative Biodiversity Research (iDiv)","correspondingAuthor":false,"prefix":"","firstName":"Rachel","middleName":"","lastName":"Oh","suffix":""},{"id":346991478,"identity":"4b615268-90db-4cf0-b603-d243ace2398a","order_by":11,"name":"Tobias Proß","email":"","orcid":"","institution":"Martin Luther University Halle-Wittenberg","correspondingAuthor":false,"prefix":"","firstName":"Tobias","middleName":"","lastName":"Proß","suffix":""},{"id":346991479,"identity":"4b7419b2-7e53-4159-9c51-ae02bcce38df","order_by":12,"name":"Damien Bonal","email":"","orcid":"","institution":"Université de Lorraine","correspondingAuthor":false,"prefix":"","firstName":"Damien","middleName":"","lastName":"Bonal","suffix":""},{"id":346991480,"identity":"4a515a7b-eca8-4088-a406-04d065502d04","order_by":13,"name":"Aletta Bonn","email":"","orcid":"https://orcid.org/0000-0002-8345-4600","institution":"Helmholtz-Centre for Environmental Research - UFZ, Friedrich Schiller University Jena \u0026 German Centre for Integrative Biodiversity Research (iDiv) Halle-Jena-Leipzig","correspondingAuthor":false,"prefix":"","firstName":"Aletta","middleName":"","lastName":"Bonn","suffix":""},{"id":346991481,"identity":"e10a78e9-a7f8-49bf-aef1-2de3e90c7da5","order_by":14,"name":"Helge Bruelheide","email":"","orcid":"https://orcid.org/0000-0003-3135-0356","institution":"Martin Luther University Halle-Wittenberg","correspondingAuthor":false,"prefix":"","firstName":"Helge","middleName":"","lastName":"Bruelheide","suffix":""},{"id":346991482,"identity":"59b7dc03-b5e8-4892-8d3e-b9fa76f66627","order_by":15,"name":"Douglas Godbold","email":"","orcid":"","institution":"BOKU university","correspondingAuthor":false,"prefix":"","firstName":"Douglas","middleName":"","lastName":"Godbold","suffix":""},{"id":346991483,"identity":"1fd5d5ab-e80b-452d-b5b7-df953cc0437a","order_by":16,"name":"Daniela Haluza","email":"","orcid":"https://orcid.org/0000-0001-5619-2863","institution":"Medical University of Vienna","correspondingAuthor":false,"prefix":"","firstName":"Daniela","middleName":"","lastName":"Haluza","suffix":""},{"id":346991484,"identity":"1fe72179-88a5-4d69-9442-9f800000732d","order_by":17,"name":"Herve Jactel","email":"","orcid":"","institution":"INRAE","correspondingAuthor":false,"prefix":"","firstName":"Herve","middleName":"","lastName":"Jactel","suffix":""},{"id":346991485,"identity":"520f6395-9455-4235-81d4-6ef3fe2370cd","order_by":18,"name":"Bogdan Jaroszewicz","email":"","orcid":"https://orcid.org/0000-0002-2042-8245","institution":"Faculty of Biology, University of Warsaw","correspondingAuthor":false,"prefix":"","firstName":"Bogdan","middleName":"","lastName":"Jaroszewicz","suffix":""},{"id":346991486,"identity":"32d21e97-66cd-4df4-a751-439cb5aa07db","order_by":19,"name":"Katriina Kilpi","email":"","orcid":"","institution":"Bos+","correspondingAuthor":false,"prefix":"","firstName":"Katriina","middleName":"","lastName":"Kilpi","suffix":""},{"id":346991487,"identity":"9d308e72-6ab5-41e6-ab9d-bca7ef8266cf","order_by":20,"name":"Melissa Marselle","email":"","orcid":"","institution":"University of Surrey","correspondingAuthor":false,"prefix":"","firstName":"Melissa","middleName":"","lastName":"Marselle","suffix":""},{"id":346991488,"identity":"1cb4c54d-10d5-4ce9-a918-ca925b4d147f","order_by":21,"name":"Quentin Ponette","email":"","orcid":"https://orcid.org/0000-0002-2726-7392","institution":"Université catholique de Louvain","correspondingAuthor":false,"prefix":"","firstName":"Quentin","middleName":"","lastName":"Ponette","suffix":""},{"id":346991489,"identity":"df4e3d26-5039-4b0b-b859-1c4f0585d831","order_by":22,"name":"Michael Scherer-Lorenzen","email":"","orcid":"https://orcid.org/0000-0001-9566-590X","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Michael","middleName":"","lastName":"Scherer-Lorenzen","suffix":""},{"id":346991490,"identity":"8825902a-0736-4cc9-a589-5570efc946ad","order_by":23,"name":"Pieter De Frenne","email":"","orcid":"https://orcid.org/0000-0002-8613-0943","institution":"Ghent University","correspondingAuthor":false,"prefix":"","firstName":"Pieter","middleName":"","lastName":"De Frenne","suffix":""},{"id":346991491,"identity":"dcbe209a-ec85-4aa7-a39c-1705292f0b3a","order_by":24,"name":"Bart MUYS","email":"","orcid":"","institution":"KU Leuven","correspondingAuthor":false,"prefix":"","firstName":"Bart","middleName":"","lastName":"MUYS","suffix":""},{"id":346991492,"identity":"ecf1eec1-3f33-40db-83df-a211fb45cc23","order_by":25,"name":"Kris Verheyen","email":"","orcid":"https://orcid.org/0000-0002-2067-9108","institution":"Ghent University","correspondingAuthor":false,"prefix":"","firstName":"Kris","middleName":"","lastName":"Verheyen","suffix":""}],"badges":[],"createdAt":"2024-07-01 16:10:27","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4669329/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4669329/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41893-025-01547-3","type":"published","date":"2025-05-19T04:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":63777176,"identity":"96f1c4f7-4840-428c-bcd1-f8e482f3429f","added_by":"auto","created_at":"2024-09-02 09:03:47","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":325129,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eStudy design\u003c/strong\u003e. \u003cstrong\u003ea\u003c/strong\u003e, Locations of the ten study sites, which are either young plantations (light-green tree icon) or mature natural forests (dark-green). \u003cstrong\u003eb\u003c/strong\u003e, Overview of forest characteristics that were sampled on all sites and shared among multiple pathways. A ‘stand’ refers to a contiguous group of trees that is relatively uniform in terms of age, structure, tree diversity and composition. \u003cstrong\u003ec\u003c/strong\u003e, Overview of model structure with causal relationships between forest characteristics, followed by forest-health pathways using the Ecosystem Services Cascade model framework. Outcomes are classified per health domain \u003cem\u003esensu \u003c/em\u003eMarselle et al.\u003csup\u003e21\u003c/sup\u003e. Complete pathways for each outcome are found in the Supplementary Information.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-4669329/v1/d5b9d6d276ce93333fdd1f8b.png"},{"id":63777174,"identity":"471d0d77-8156-4c07-8247-cfb8684f30cd","added_by":"auto","created_at":"2024-09-02 09:03:47","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":121966,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSensitivities of forest-health pathways to characteristics of tree diversity, structure and composition\u003c/strong\u003e. Based on sensitivity analyses, yielding percentages of variance reduced in the real value of the response due to a finding in predictors. The stand-type node is excluded here, as it is a nominal variable functioning as a grouping factor. Illustrated is the large dependency in explanatory power of selected forest variables on considered pathways. This is a selection of most relevant variables (n = 39 out of 82), effectively capturing most influential links except for two absent yet strong links: canopy density to PM\u003csub\u003e10\u003c/sub\u003e- (16.2% variance explained) and PM\u003csub\u003e100\u003c/sub\u003e filtration capacities (8.5%). The x-axis is square root-transformed. The term ‘affect’ refers to a display of emotions. Symbols represent forest-health pathways as shown in Fig. 1c. For variable definitions, see Supplementary Information.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-4669329/v1/e6fb106c8a95928e75c81c3b.png"},{"id":63777175,"identity":"b9203e8f-4239-40f3-a747-e389ff87dfe9","added_by":"auto","created_at":"2024-09-02 09:03:47","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":70981,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSynergies and trade-offs between forest-health pathways\u003c/strong\u003e. Based on mutual information levels between nodes of different pathways. Thick full lines indicate that a minimum threshold level was reached at the health outcome stage, indicating the strongest and most consistent associations. Dotted lines indicate that the threshold was not reached in the health outcome stage, but for at least one pair of variables in the function or service stages. Selection of relevant forest characteristics is based on Fig. 2. See Supplementary Fig. S2 for a matrix of all mutual information values.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-4669329/v1/c41d28f277d4ce52a0f3c09f.png"},{"id":63777177,"identity":"4e12997b-5fca-4ba9-9e35-3854e276c82e","added_by":"auto","created_at":"2024-09-02 09:03:47","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":138524,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eHealth effect magnitudes relative to various baseline conditions and under contrasting tree diversity and canopy density\u003c/strong\u003e. Illustrates how changes in tree diversity and canopy density, e.g. through forest management, could lead to substantial health effect optimisation and how this compares to the absence of forests.\u0026nbsp; \u003cstrong\u003ea\u003c/strong\u003e, Health outcome relative to explicit non-forest controls when they were explicitly monitored for comparison. In the case of mental wellbeing via visual stimuli, controls were urban environments\u003csup\u003e32\u003c/sup\u003e. For the heat\u003csup\u003e33\u003c/sup\u003e and particulate matter\u003csup\u003e34\u003c/sup\u003e outcomes, controls were ambient, rural conditions. \u003cstrong\u003eb\u003c/strong\u003e, Outcomes without explicit control conditions. Red dots are means per forest type while grey horizontal bars are means per diversity level specifically to facilitate comparison. There is no expression of uncertainty for thermal comfort because it is estimated from an ordinal variable.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-4669329/v1/3417147fe3a0937c19df0c6c.png"},{"id":83110363,"identity":"44b64748-4b5a-47cb-a316-454badaa0a09","added_by":"auto","created_at":"2025-05-20 07:09:04","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1794500,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4669329/v1/b0b536de-4df0-467f-8cee-ce8e3386932a.pdf"},{"id":63777178,"identity":"ead7c76e-d710-4280-a22c-2e4af4b947f3","added_by":"auto","created_at":"2024-09-02 09:03:47","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":79192,"visible":true,"origin":"","legend":"","description":"","filename":"Fig.S1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4669329/v1/cb11d8e25596978ce5ca5659.pdf"},{"id":63777180,"identity":"de9454bc-389b-4888-8805-3edcd8f33ee3","added_by":"auto","created_at":"2024-09-02 09:03:47","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":431500,"visible":true,"origin":"","legend":"","description":"","filename":"Fig.S2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4669329/v1/5338433e821c30aafadbafdd.pdf"},{"id":63777181,"identity":"180225d9-ba04-4dad-abd8-29c6810fb6f7","added_by":"auto","created_at":"2024-09-02 09:03:48","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":4742317,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformation.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4669329/v1/1fe24401a7721c3ed8964f4f.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Forest biodiversity and structure modulate human health benefits and risks","fulltext":[{"header":"Introduction","content":"\u003cp\u003eA growing body of scientific literature hails contact with nature for its physical, mental and social wellbeing benefits\u003csup\u003e\u003cspan additionalcitationids=\"CR2 CR3 CR4\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. With 74% of deaths worldwide caused by non-communicable diseases (e.g., cardiovascular and respiratory conditions)\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e and an increasing global burden of mental health disorders\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e, preventive nature-based solutions can counterbalance multiple public health issues. Increasing the surface area of publicly accessible greenspaces has wide-ranging benefits observable at the scale of cities including, for example, reduced sales of mood disorder and cardiovascular medication\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e, heat mortality mitigation\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e and, more generally, annual preventable death reductions by 0.2\u0026ndash;5.5%\u003csup\u003e10\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eNot all greenspaces are equal, and identifying greenspace characteristics that optimise health outcomes can help offsetting the limitation posed by space restrictions\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. Firstly, the vegetation type will modulate health outcomes\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Reviews comparing greenspaces such as forests, grasslands and others suggest that forests are more strongly linked with a range of physical and mental benefits\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. With their high level of biomass and tall canopies, trees and forests may create immersive environments that are more strongly decoupled from ambient conditions than low-stature vegetation. Forests thereby foster cooler\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e and subjectively quieter\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e environments with \u0026ndash; in most circumstances \u0026ndash; better air quality\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e that benefit physical and mental health. A global evidence assessment affirms the supportive role of forests in the physical, mental, social and spiritual health of humans throughout all life stages\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eCompared to greenspace quantity, the quality of a given type of greenspace also affects human health but is much less studied\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Quantity is commonly assessed using well-defined proxies such as surface area of vegetated land-use classes or the Normalised-Difference Vegetation Index (NDVI)\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e, representing greenspaces as relatively homogenous entities\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. In contrast, greenspace quality is often inconsistently quantified and can encompass aspects of accessibility, safety, amenities and biodiversity\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. In health sciences, the latter is often proxied by indices that are merely indirectly related to biodiversity including, for instance, perceived bird, plant and butterfly species richness\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e or \u0026lsquo;naturalness\u0026rsquo;\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. These are poorly aligned with standards for rigorous ecological inventories established in the natural sciences\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. Yet, thorough quantification of biodiversity holds great promise. Five decades of biodiversity-ecosystem functioning (BEF) research have demonstrated that ecologically-relevant characteristics such as species diversity, functional traits and structural complexity greatly influence the functions and services that ecosystems provide\u003csup\u003e\u003cspan additionalcitationids=\"CR19\" citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e \u0026ndash; ultimately shaping health outcomes\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. Until now, the integration of BEF into biodiversity-health research has been limited, accentuating the need for interdisciplinary approaches.\u003c/p\u003e \u003cp\u003eThe biodiversity-health relationship is increasingly recognised, including by the Convention on Biological Diversity (CBD) and the World Health Organisation (WHO)\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. Still, the majority of nature-health studies to date overlook biodiversity effects and fail to unravel underlying mechanisms\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. Reported greening effects on reductions in heat mortality\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e, antidepressant prescriptions\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e and overall mortality\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e are compelling and invaluable to illustrate the magnitude of nature\u0026rsquo;s disease-prevention potential at large scales. However, the many factors co-varying with biodiversity in broad-scale longitudinal datasets limit attribution of causal effects\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e. These scientific efforts should be complemented with experimental studies with proper control conditions to understand the full complexity of nature-health relationships and their inherent context-dependency\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e. Identifying mechanisms and context-dependencies may inform \u003cem\u003ehow\u003c/em\u003e and \u003cem\u003ewhen\u003c/em\u003e management interventions can optimise health outcomes. For example, increasing tree canopy cover may reduce heat and air pollution levels in rural settings, but may aggravate these environmental stressors in cities depending on specific building and street design by locally trapping heat and particulate matter\u003csup\u003e\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. Interindividual differences can also alter health outcomes, with perceived rather than actual biodiversity generally better predicting mental wellbeing outcomes\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. Subjectivity may also affect physical wellbeing, with forest environments reducing perceived heat\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e and perceived noise\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e more strongly than expected from objective measurements. Nature and forests may also cause harm by enhancing exposure to disease-bearing vectors\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e and potentially allergenic pollen\u003csup\u003e\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. Such ecosystem disservices are rarely considered in some disciplines\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e, though physical risks related to wildlife exposure are thoroughly covered in the field of tropical medicine\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. At last, a handful of studies insinuate that nature-health outcomes may positively or negatively interact\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. How health benefits can be optimised and risks minimised, leveraging synergies and avoiding trade-offs between pathways, is largely unknown.\u003c/p\u003e \u003cp\u003eHere, we use a unique continental-scale interdisciplinary database gathered \u003cem\u003ein situ\u003c/em\u003e in 164 European forest stands (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea) by a team composed of ecologists, environmental psychologists, neuroscientists and public health experts to examine seven distinct forest-health pathways: i) mental wellbeing via visual stimuli; ii) mental wellbeing via auditory stimuli; iii) thermal comfort; iv) polyphenol content of medicinal plants; v) mushrooms and nutrition, vi) air quality, and vii) ticks and Lyme disease. The latter represents an important ecosystem disservice in Europe. Forest ecological characteristics were assessed using well-established forest inventory methods, enabling the attribution of health outcomes to recognized ecological variables (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb). The diversity in assembled datasets was synthesised using a Bayesian Belief Network (BBN), combined with an adapted version of the Ecosystem Services Cascade Model\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e providing the mechanistic framework. This flexible statistical approach allowed for the modelling of complex interaction chains, the detection of synergies and trade-offs and the inclusion of crucial but unsampled variables.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eEffects of forest characteristics on functions, services and health outcomes\u003c/h2\u003e \u003cp\u003eThe effect magnitude and direction of seven focal forest characteristics depended strongly on specific forest-health pathways and their considered stage along the Cascade Model (i.e. function, service or health outcome; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Canopy density was the most influential forest property overall, followed by stand density and tree diversity. Canopy height, proportion of deciduous trees and leaf size had relatively weak effects, and stem density was least influential. However, effect magnitude was highly dependent on the forest-health pathway. Increased tree diversity was a dominant driver for mental wellbeing via visual stimuli and mushroom pathways. Canopy density had a positive effect on the thermal comfort pathway yet a negative effect on medicinal plants. Increased canopy density led to increased risk of Lyme disease, mainly by strongly increasing relative humidity thereby supporting high tick population densities. Mental wellbeing via visual stimuli was almost not influenced by \u003cem\u003eactual\u003c/em\u003e forest characteristics (i.e., in opposition to what is subjectively perceived).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eChanges in canopy density, represented by Leaf Area Index (LAI), were explained by mean leaf size (6.2% of variance explained), followed by stand density (3.1%), tree diversity (1.3%) and canopy height (0.7%). We highlight these results for canopy density as a central driver, but statistics on associations between other forest characteristics can be found in the Supplementary Information (p. 11).\u003c/p\u003e \u003cp\u003eThe more intermediate steps between two variables in the model, the less sensitive they were to each other. This dilution effect starts with the forest characteristics, proceeding via function to service and then to health outcome variables. However, this top-down approach largely conceals the predictive power of intermediate variables. This can be demonstrated with the mechanisms underlying mental wellbeing via visual stimuli. While tree diversity only explained 1.2% of the variance in perceived biodiversity, the latter was strongly explained by more intermediate variables: perceived density (6.7%) and perceived naturalness (5.4%). Furthermore, perceived biodiversity itself effectively determined restorativeness (13.9%), attention restoration (4.4%) and stress reduction (4.3%). For a complete understanding, sensitivities should thus be scrutinised case-by-case using the full model (Supplementary Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eTrade-offs and synergies in health outcomes\u003c/h2\u003e \u003cp\u003eBased on mutual information held among variables in function, service and health outcome stages, several synergies and trade-offs emerged. A quartet of pathways were consistently associated in function, service and health outcome stages: thermal comfort, medicinal plants, air quality and ticks and Lyme disease (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). This means that forest characteristics that effectively reduced heat stress also improved air quality through reducing air particulate matter. Similarly, medicinal plant production and Lyme risk reduction also synergized. However, these two pairs of pathways were antagonistic to each other. Forests that provide a high level of medicinal plant production and regulation of disease (reduced risk in Lyme) had a reduced capacity for microclimate and air quality regulation. Some weaker synergies and trade-offs were found with respect to the three remaining pathways, where especially both mental wellbeing pathways were relatively disconnected from other health outcomes.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eComparing overall forest effects to inter-forest variability\u003c/h2\u003e \u003cp\u003eWhen health outcomes were defined relative to non-forest conditions, forest presence generated substantial \u003cem\u003eabsolute\u003c/em\u003e health benefits \u0026ndash; regardless of forest characteristics (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea). Interpretation was more challenging when explicit controls were absent (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb), requiring a case-by-case consideration. For example, mental wellbeing outcomes were compared before and after being exposed to visual or acoustic stimuli for the mental wellbeing pathways.\u003c/p\u003e \u003cp\u003eOverall forest effects were compared to \u003cem\u003erelative\u003c/em\u003e differences caused by changing canopy density and tree diversity. Some health outcomes appeared to be nearly exclusively determined by forests in general, with very small changes generated by their ecological characteristics. This is most notably the case for anxiety reduction (visual), negative affect reduction (auditory), positive affect increase (visual and auditory) and PM\u003csub\u003e10\u003c/sub\u003e risk mitigation. For cases where forest characteristics did matter, tree diversity was most influential for mushroom health benefits. Yet, canopy density was again the strongest driver of variation: most evidently in microclimate (thermal comfort) and disease (Lyme) regulation, followed by medicinal plants- benefits and PM\u003csub\u003e2.5\u003c/sub\u003e risk mitigation.\u003c/p\u003e \u003cp\u003eAnalogous analyses of contrasted stand types (young plantations \u003cem\u003evs\u003c/em\u003e. mature natural forests) were yielding even smaller differences than shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. While changes in stand type affected canopy height very strongly, the latter had generally little influence on other variables (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eDrivers of forest functioning, services and health outcomes\u003c/h2\u003e \u003cp\u003eWe here report on the first trans-continental study linking forest biodiversity, functioning and health outcomes, capitalising on interdisciplinary approaches. For example, recent heat mitigation research reported which forest characteristics determine surface or air temperature and relative humidity of forest microclimates\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e, but few have connected this to indices relevant to human perception and health outcomes\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. In parallel, heat mitigation by \u0026lsquo;homogenous\u0026rsquo; greenspaces is already thoroughly reported\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e, but ignores ecological intricacies. By using a BBN to merge multiple data sources, this gap was effectively bridged for multiple forest-health pathways.\u003c/p\u003e \u003cp\u003eThe two most influential and independent forest characteristics were forest canopy density and tree diversity. Canopies are key drivers of ecosystem functioning\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e, creating a physical, chemical and biological filter to ambient conditions\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. Tree diversity is also a recognised determinant of ecosystem function, ecosystem service provision and non-tree biodiversity\u003csup\u003e\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e,\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. Beyond confirming these known consequences, we here show that they also shape human health impacts. Even though effect sizes are relatively small, it may make a substantial difference at the scale of populations at which greenspaces can prevent well over hundreds of deaths\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe process of disentangling mechanistic pathways via the Ecosystem Service Cascade Model also revealed dependencies at spatiotemporal, societal and individual levels. At the spatiotemporal level, the importance of forests to human health was defined by the surrounding environmental conditions, where locations with high particulate matter pollution and periods with high heat loads benefited proportionally more from forest tempering effects\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. Likewise, seasonality determines production of mushrooms\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e and medicinal plants\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e, and the amount of songbird-generated biophony with potential mental benefits\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. Societal context also plays a crucial role: local legislation and overall foraging culture are determinants for potential health benefits from non-timber forest products\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e. For example, while mushroom and berry collection is common in e.g. Poland, Germany and France\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e, it is legally not allowed in Flanders, Belgium. Mental wellbeing is also determined by societal realities: forests and dense vegetation can provoke fear of assault by criminals or wild animals\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. This implies that our findings are best applicable to the world\u0026rsquo;s forests that are perceived as \u0026lsquo;safe\u0026rsquo;. At last, individual experiences may further shape health outcomes. Fear of crime varies with individual factors such as age, gender and culture\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. Mental wellbeing was relatively insensitive to actual changes in forest characteristics in our analysis. This is partly explained by the pivotal role of perceived (visual and auditory) biodiversity, which can be based on cues like species richness, greenness, acoustic complexity or functional traits like colourfulness\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e \u0026ndash; thereby recurrently deviating from actual biodiversity\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e. Half of variation in thermal comfort is determined by factors other than the physical microclimate, including physiological acclimatisation, biological sex, thermal preference and expectations\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eTrade-offs bar all-round optimisation\u003c/h2\u003e \u003cp\u003eHealth outcome interactions were most clear for heat mitigation, medicinal plants, air quality and Lyme disease risk, with canopy density as the common denominator. The latter enhances mitigation of heat stress\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e and particulate matter\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e, but reduces understory light availability\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e which impedes production of medicinal plants. Closed canopies also create ideal conditions for tick activity and survival by maintaining relative humidity levels that rarely fall below 80%\u003csup\u003e46\u003c/sup\u003e. Whereas canopy density is the most influential forest characteristic, corroborating a recent study on ecosystem services\u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e, it can inadvertently lead to unwanted side-effects. This does not apply to tree diversity: though less influential, it always returned positive effects \u0026ndash; except for a small increase in Lyme disease risk. Note that the effect of increasing tree diversity and forest complexity on Lyme risk in Europe is contested, with reports of both negative\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e and positive\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e outcomes.\u003c/p\u003e \u003cp\u003eSome other, weaker, interactions were observed or theoretically expected. A separate analysis found enhanced levels of thermal and mental wellbeing to mutually reinforce each other, where forest environments incite this synergy\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. This non-linear, subtle interaction did not clearly emerge here (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Other indirectly explored interactions include the increase in questing tick nymphs with higher levels of medicinal plants, because ticks use understory vegetation to reach vertebrate hosts\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e,\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e. At the same time, more lush understory vegetation affected perceived density, which was a driver for perceived biodiversity and thus potentially determines mental wellbeing. Greater bird diversity increases biophony, but also influences tick nymphal infection prevalence as birds are suitable hosts for multiple \u003cem\u003eBorrelia\u003c/em\u003e strains\u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e. Some interesting interactions could not be explored. High rainfall frequency and abundance may increase mushroom biomass\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e, but also improve the air filtering capacity of the forest by cleaning the leaves and rendering them more sticky\u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e. Given that air pollution may worsen mental health\u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e, filtering of particulate matter by forests may have additional mental health effects.\u003c/p\u003e \u003cp\u003eAgain, this emphasises the need to consider results in the light of local needs. In a theoretical region with low-to-absent Lyme prevalence and foraging regulations restricting medicinal plant harvesting, silvicultural interventions may aim for closed canopies to foster the synergy in heat and particulate matter improvements. Prioritisation of target health outcomes should follow the health impact, the number of people affected and forest effect magnitude given the local context. For instance, in 2022, air pollution and heat caused an estimated 300 000\u003csup\u003e53\u003c/sup\u003e and 60 000\u003csup\u003e54\u003c/sup\u003e deaths in Europe, respectively, which could be effectively mitigated by forests\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. In contrast, despite widespread malnutrition in Europe\u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e, harvesting mushroom and medicinal plants may only have marginal health impacts compared to industrial sources of food and medicine production\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e. Still, these may have great importance locally and have recreational co-benefits not addressed here\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eSuboptimal forests are better than no forests\u003c/h3\u003e\n\u003cp\u003eDespite the large variability, our findings indicate that forests generally have net benefits to health, regardless of their biodiversity and structure (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Mental wellbeing effects were especially little influenced by \u003cem\u003eactual\u003c/em\u003e measures because they essentially revolve around \u003cem\u003eperceived\u003c/em\u003e forest characteristics, whereas predicted heat mortality strongly varied between forest types. Yet, even the least optimal forest stands led to strong heat reductions. Indeed, even young plantations have been found to reduce perceived temperatures by 10\u0026deg;C under hot conditions\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. In fact, the whole variety of forest conditions was generally unequivocally favourable to human health, except for ticks which are especially abundant in forests compared to any other habitat\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAltering forest characteristics is thus less consequential than the mere presence of forests, which has two positive implications. First, studies considering forests as homogenous entities certainly ignore plenty of variation, yet their findings may be roughly applicable to a wide range of forest conditions. Second, increasingly prevalent reforestation efforts, despite being mostly constituted of young plantation forests\u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e, may already yield substantial health benefits to people.\u003c/p\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eStrengths and weaknesses define avenues for future research\u003c/h2\u003e \u003cp\u003eWhile our modelling effort brings about insights on largely uncharted scientific terrain, limitations require consideration before extrapolating our conclusions. From the technical side, BBNs were elemental to linking forest characteristics with their health effects owing to the capacity to model complex causal pathways with heterogeneous data sources. Yet, they have inherent weaknesses. First and foremost, the mandatory discretisation of continuous variables is disadvantageous for environmental modelling, where such variables are omnipresent. This leads to significant information loss that can partly be compensated for by increasing the number of states \u0026ndash; though this ramps up the number of required observations to populate Conditional Probability Tables\u003csup\u003e\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e. We compensated for this by using expert opinion as priors, which improves information content for state combinations that are not covered by empirical data\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. Despite these measures, a clear effect size dilution was visible especially for long causal chains. While partly due to discretisation, this incidentally also reflects real-life processes: as the myriad of unmeasured confounders and aforementioned dependencies modulate nature-health relationships, small final effects are expected. Forest-health benefits and their modulation by forest characteristics should therefore not be exaggerated.\u003c/p\u003e \u003cp\u003eFrom the conceptual side, a major strength and singularity is the direct integration of multiple forest-health pathways. Yet, our selection of forest-health effects is not comprehensive. Others include, for example, reduction of noise stress, modulation of pollen allergy prevalence and fostering of spiritual wellbeing\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. These may create additional synergies and trade-offs that we have not covered. Next, our analysis focuses on local scales (plot- or stand level) but some pathways require consideration of landscape-scale dynamics. Most notably, highly mobile vertebrates such as ungulates and birds are likely influenced by characteristics of whole forest complexes. At last, outcomes should be cautiously interpreted in light of assumptions listed in the Supplementary Information. In sum, pathways assume that users interact intimately with the forest environment: our \u0026lsquo;virtual guinea pig\u0026rsquo; is consistently exposed to forest microclimate, atmosphere, soundscape and visual aspects and regularly consumes non-timber forest products. These assumptions may be naive in many cases, especially for urbanites, and resonates with the challenge of quantifying nature exposure, accounting for the frequency, duration and type of nature contact\u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn short, future research may turn towards following avenues: i) directly comparing multiple forest-health pathways and their interactions, ii) considering mechanistic processes via frameworks like the Cascade model and coupling statistical approaches like Structural Equation Models or BBNs, iii) incorporating relevant undesirable health effects, and iv) investigating how biodiversity-health relationships apply to (semi-)urban environments.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eManagement recommendations and conclusions\u003c/h2\u003e \u003cp\u003eDespite the case-specificity and complex pathway interactions, general patterns stand out. Firstly, any European forest is likely to have beneficial effects regardless of how it is managed, certainly compared to urban environments\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. Any type of forest is also likely to be more beneficial than low-stature vegetation for heat and particulate matter reduction, mushroom production and mental wellbeing, though forests host more ticks\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eYet, forest managers and policy makers have several degrees of freedom to enhance health effects and mitigate risks, with potentially considerable impact at the public health level. Our findings suggest that targeting canopy density (leaf area index) and stand density (basal area) are the most effective tools to modulate physical health outcomes in particular. Management strategies aiming at densification such as continuous cover forestry may enhance heat buffering and air filtering capacities, but may lead to an increased risk of Lyme and reduced medicinal plant yield. This clear trade-off requires careful consideration of local priorities regarding these four health outcomes. Canopy cover changes will also affect forest functioning itself, with potential cascading effects. For example, thinning may lead to increased water availability for certain tree species under drought\u003csup\u003e\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e, but rapid canopy opening may also perturb understory biodiversity\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. In contrast, increasing tree diversity represents a relatively safe, no-regret intervention with net advantages\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e and important co-benefits for biodiversity conservation\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e. In that sense, this may make it a compelling all-round management strategy in many situations with a win-win for biodiversity and human health, especially if local public health priorities are difficult to identify. Diverse forests are especially appealing given their higher resilience to multiple global change drivers and insect pests\u003csup\u003e\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e, which may secure provision of health benefits in the longer term\u003csup\u003e\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e. In sum, despite rather low direct health impacts, increasing forest tree diversity emerges as a robust strategy for promoting human health whereas altering forest density can be used to maximise specific benefits according to local contexts.\u003c/p\u003e \u003c/div\u003e"},{"header":"Materials and methods","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\n \u003ch2\u003eStudy sites and design\u003c/h2\u003e\n \u003cp\u003eWe deployed an international network of ten forest sites along a West-East gradient in Central Europe (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003ea), covering oceanic to sub-continental climates. Of these ten sites, four comprised mature forests, while six comprised young plantations within tree diversity experiments. The mature forest stands reflected natural conditions that varied in tree diversity levels. These included the Białowieża (PL), Hainich (DE) and MIXLOR (FR) sites\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e, and stands of the TREEWEB network\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e. All were designed to obtain an even distribution of species per species richness level, and to minimise variation in environmental factors such as soil properties and topography that could covary with species composition. The young plantations\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e63\u003c/span\u003e\u003c/sup\u003e were aged nine to 17 years at the time of sampling. These were all located within a small perimeter using a uniform planting design that standardised the density and spatial arrangement of different species, where only tree diversity varied with negligible covariation of other environmental factors. These included B-TREE (AT), BIOTREE (DE), three FORBIO sites (BE) and ORPHEE (FR). Both site types were specifically designed to study BEF relationships, enabling the quantification of causal effects of forest ecological characteristics on human health.\u003c/p\u003e\n \u003cp\u003eSelection of forest stands revolved around tree species diversity as one of the major determinants supporting overall forest biodiversity and functioning\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e18\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. At each site, an equal number of monospecific and mixed forest plots was selected. Mixed plots were always composed of a randomised selection of species present as monospecific stands. All mixed plots contained three target tree species, which was the diversity level that all sites had in common. Non-target species were often present in mature natural stands but were never dominant, having established after plot demarcation\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e. Our selection covers a total of 164 permanent forest plots and 18 target tree species. Participating researchers were encouraged to cover as many of these plots as possible to generate directly comparable forest-health data for identical plots (\u0026ldquo;all measurements on all plots\u0026rdquo; strategy). However, due to COVID-related fieldwork impediments, most researchers failed to cover the full selection (see Supplementary Information for number of plots per pathway).\u003c/p\u003e\n \u003cp\u003eThe same set of forest ecological characteristics was sampled over all plots (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eb). Central herein was tree diversity, expressed as a Shannon index. Unlike species richness, Shannon Diversity accounts for species\u0026rsquo; relative abundances, here based on species-specific basal area shares\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e64\u003c/span\u003e\u003c/sup\u003e. A second characteristic receiving special attention in our analyses is canopy density, proxied by the Leaf Area Index via indirect sampling (using hemispherical photographs). Details on remaining characteristics are found in the Supplementary Information.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\n \u003ch2\u003eForest-health pathways\u003c/h2\u003e\n \u003cp\u003eThis meta-analysis synthesises the published and unpublished works of six separate doctoral studies that gathered empirical data and insights for seven forest-health pathways, covering both mental and physical outcomes that can be considered services or disservices. Pathways are listed and shortly described in Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e and described in more detail in the Supplementary Information.\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eTable 1 | Overview of submodels and their constituting studies\u003c/strong\u003e. Submodels represent the seven forest-health pathways which have all been covered by multiple studies using interdisciplinary approaches. We briefly discuss the setup of studies of which data populated the final model, and some key findings. Included references refer to readily published studies done within the overarching project. The number of sites refers to the network shown in Fig. 1a and the symbols refer to forest-health pathways as shown in Fig. 1c. \u0026nbsp;\u003c/p\u003e\u0026nbsp;\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"654\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.091743119266056%\" valign=\"top\"\u003e\n \u003cp\u003eOutcome\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.45565749235474%\" valign=\"top\"\u003e\n \u003cp\u003eResearchers\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.345565749235472%\" valign=\"top\"\u003e\n \u003cp\u003eSetup\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"45.10703363914373%\" valign=\"top\"\u003e\n \u003cp\u003eKey findings\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.091743119266056%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003cimg src=\"data:image/png;base64,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\" alt=\"image\"\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eA1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.45565749235474%\" valign=\"top\"\u003e\n \u003cp\u003eKR, RRYO, MRM \u0026amp; ABON\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.345565749235472%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cem\u003eIn situ\u003c/em\u003e experiment involving 223 participants subjected to non-forest baseline conditions and different peri-urban forests with contrasting levels of tree species richness\u003csup\u003e32\u003c/sup\u003e. Complemented with a lab-based EEG experiment.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"45.10703363914373%\" valign=\"top\"\u003e\n \u003cp\u003eBiophysical structure has weak effects, with only tree diversity and understory productivity having limited influence on short-term mental health and wellbeing. Mainly subjectively perceived forest characteristics, and especially perceived biodiversity, fostered positive \u0026nbsp;mental wellbeing effects\u003csup\u003e25\u003c/sup\u003e. We found little congruence to actual biodiversity.\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.091743119266056%\"\u003e\n \u003cp\u003e\u003cimg src=\"data:image/png;base64,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\" alt=\"image\"\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eA2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.45565749235474%\" valign=\"top\"\u003e\n \u003cp\u003eTS, SM, KR, AGCR, RRYO, ABON \u0026amp; MSL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.345565749235472%\" valign=\"top\"\u003e\n \u003cp\u003eMonitoring of bird song diversity and abundance using audiologgers at eight sites. Complemented with a lab-based experiment using similar psychological questionnaires to submodel A1.\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"45.10703363914373%\" valign=\"top\"\u003e\n \u003cp\u003eForest characteristics, especially stand density, effectively explained variation in bird song diversity. Yet, further effects were strongly tempered since mental wellbeing is essentially driven by subjectively perceived acoustic diversity, analogous to submodel A1.\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.091743119266056%\"\u003e\n \u003cp\u003e\u003cimg src=\"data:image/png;base64,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\" alt=\"image\"\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eB\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.45565749235474%\" valign=\"top\"\u003e\n \u003cp\u003eLG, PDF, DL, BM \u0026amp; KV\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.345565749235472%\" valign=\"top\"\u003e\n \u003cp\u003eA thermal comfort index (Physiologically Equivalent Temperature) was measured for 14 months in eight rural sites\u003csup\u003e33\u003c/sup\u003e and in an urban setting\u003csup\u003e65\u003c/sup\u003e. How this was subjectively perceived by people was tested in a complementary experiment\u003csup\u003e26\u003c/sup\u003e.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"45.10703363914373%\" valign=\"top\"\u003e\n \u003cp\u003eCooling effects were very strongly explained by forest characteristics, with especially variables related to forest density and height being impactful, followed by multiple composition-related variables. Cooling capacity increased with increasing ambient heat stress. This was effectively experienced as such by participants. Mental wellbeing (A1) was found to positively interact with thermal comfort in forests\u003csup\u003e26\u003c/sup\u003e.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.091743119266056%\"\u003e\n \u003cp\u003e\u003cimg src=\"data:image/png;base64,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\" alt=\"image\"\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eC1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.45565749235474%\" valign=\"top\"\u003e\n \u003cp\u003eKS \u0026amp; BJ\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.345565749235472%\" valign=\"top\"\u003e\n \u003cp\u003eInventory of medicinal plants occurring over eight sites. The biologically active substances, polyphenols, were measured in a selection of species, and extrapolated at forest plot scales.\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"45.10703363914373%\" valign=\"top\"\u003e\n \u003cp\u003ePolyphenol content is mainly determined by medicinal plant productivity rather than plant diversity, and canopy density crucially determined productivity by altering the light regime. Soil fertility had little effect.\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.091743119266056%\"\u003e\n \u003cp\u003e\u003cimg src=\"data:image/png;base64,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\" alt=\"image\"\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eC2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.45565749235474%\" valign=\"top\"\u003e\n \u003cp\u003eKS \u0026amp; BJ\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.345565749235472%\" valign=\"top\"\u003e\n \u003cp\u003eContinuous and exhaustive sampling of mushrooms at 20 Białowieża plots over two years\u003csup\u003e39\u003c/sup\u003e. This was complemented by a single sampling campaign at five other sites.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"45.10703363914373%\" valign=\"top\"\u003e\n \u003cp\u003eBecause many macroscopic mushrooms are mycorrhizal and thus associate to specific tree species, tree diversity was a more influential variable than canopy cover. Health benefits are mostly driven by mushroom productivity and only little by mushroom diversity.\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.091743119266056%\"\u003e\n \u003cp\u003e\u003cimg src=\"data:image/png;base64,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\" alt=\"image\"\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.45565749235474%\" valign=\"top\"\u003e\n \u003cp\u003eMS, DH, BM \u0026amp; DG\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.345565749235472%\" valign=\"top\"\u003e\n \u003cp\u003eDetailed case-study at B-TREE with quantification of particulate matter deposition for different size classes (\u0026lt;100, \u0026lt;10, \u0026lt; 2.5\u0026micro;m)\u003csup\u003e34\u003c/sup\u003e. Meta-analysis on tree characteristics driving deposition rates\u003csup\u003e51\u003c/sup\u003e. Complemented with direct LAI measures at eight sites and PM monitoring at four of them.\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"45.10703363914373%\" valign=\"top\"\u003e\n \u003cp\u003eVariation in PM pollution mitigation by forests was mainly influenced by the canopy closure and its roughness, whereas mitigation becomes stronger as ambient pollution levels increase \u0026ndash; analogous to submodel B. \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"10.091743119266056%\"\u003e\n \u003cp\u003e\u003cimg src=\"data:image/png;base64,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\" alt=\"image\"\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"13.45565749235474%\" valign=\"top\"\u003e\n \u003cp\u003eABOU, TV \u0026amp; HJ\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"31.345565749235472%\" valign=\"top\"\u003e\n \u003cp\u003eDetailed case-study at ORPHEE where ticks were sampled and screened for various \u003cem\u003eBorrelia\u003c/em\u003e strains\u003csup\u003e49\u003c/sup\u003e. A second sampling campaign covered several other sites. A meta-analysis\u003csup\u003e28\u003c/sup\u003e on forest effects on ticks further defined submodel conceptualisation. Complemented with data from the Forester\u003csup\u003e48\u003c/sup\u003e project, to quantify links between forest characteristics, mouse- and deer abundance and tick-related variables (DON, NIP and DIN).\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"45.10703363914373%\" valign=\"top\"\u003e\n \u003cp\u003eForest characteristics were hypothesised risks posed by ticks and Lyme via three pathways: microclimate humidity (mediated by canopy density), understory vegetation and the vertebrate host community. Humidity levels and deer abundance strongly predicted tick densities (DON), whereas mouse abundance and understory productivity had only modest effects.\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003ch2\u003e\u0026nbsp;\u003c/h2\u003e\n \u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\n \u003ch2\u003eThe choice for Bayesian Belief Networks\u003c/h2\u003e\n \u003cp\u003eA BBN consists of a Directed Acyclic Graph, composed of a set of nodes (i.e. variables) connected via causal links. If node B is the causal agent of node A, node A is the child node and node B is a parent node. Each node is partitioned into a limited set of states that the node can reach. The probability of a node being in a particular state is based on so-called Conditional Probability Tables (CPT) which store probability distributions for all state combinations of parent nodes. BBNs do not generate p-values. Instead, the certainty of a given state (outcome) is related to the narrowness of the probability distribution. See Jensen \u0026amp; Nielsen (2007)\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e for a standard technical reference or Cain (2001)\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e for an applied introduction.\u003c/p\u003e\n \u003cp\u003eWe chose BBN as an analytical method as they provide a very flexible approach to integrate data from heterogeneous sources including expert knowledge, quantitative literature findings and empirical research\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. This was essential given that empirical data from its constituting studies \u0026ndash; despite interdisciplinary methodologies \u0026ndash; sometimes failed to cover the whole mechanistic pathway from forest until health outcome. Other advantages include the capacity to handle small datasets and missing data, the highly visual and interactive model building process, the fast processing times, and the very intuitive way of dealing with uncertainty\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e31\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e67\u003c/span\u003e\u003c/sup\u003e. BBNs also have disadvantages: feedback loops and cyclic processes cannot be handled, complex models with long interaction chains lead to \u0026lsquo;dilution\u0026rsquo; of effect sizes, random effects to account for e.g. spatial autocorrelation cannot be directly implemented, and generating child node CPTs becomes exponentially more cumbersome with each additional parent node\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e31\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e67\u003c/span\u003e\u003c/sup\u003e. Another limitation is the absent direct implementation of hierarchical data structures (e.g. random factors). Most importantly, continuous nodes need to be discretised into a limited number of states, leading to information loss\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e67\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\n \u003ch2\u003eModel building steps\u003c/h2\u003e\n \u003cp\u003eBBNs were built and adapted as individual research projects advanced. Model building was partitioned in three phases centred around three annual meetings (2021\u0026ndash;2023) during which protocols were presented and discussed prior to application. Protocols were adapted from existing guidelines\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e56\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e. Phases overlapped in time due to continuously updating insights as fieldwork and data analysis progressed in parallel (Supplementary Information p.5). BBNs were developed in Netica (v7.11)\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e69\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\n \u003ch2\u003ePhase I\u003c/h2\u003e\n \u003cp\u003eThe main goal of phase I was to create a Directed Acyclic Graph for each pathway/submodel. The framework we imposed was the Ecosystem Services Cascade Model by Haines-Young \u0026amp; Potschin\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e, previously adapted for BBNs\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e and here applied to health outcomes. This forced partners to disentangle their forest-health pathway into a mechanistic conceptual model with four stages. The first stage groups nodes defining the biophysical structure of the ecosystem (e.g. tree diversity, canopy height). This defines ecosystem functioning (e.g. heat buffering, particulate matter retention), which influences (dis-)service provisioning (e.g. medicinal plant productivity, attention restoration), which will ultimately influence the variables defining the health outcome (e.g. Lyme disease risk, positive affect). Partners avoided feedback loops and more than three parent nodes for the same child node whenever possible\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e. They prioritized nodes being empirically sampled and respected relationship causality: \u0026ldquo;\u003cem\u003eDraw arrows from node A to node B if changes in A will lead to changes in B, but not the opposite\u003c/em\u003e\u0026rdquo;. We acknowledge the more recent Nature\u0026rsquo;s Contributions to People framework\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e70\u003c/span\u003e\u003c/sup\u003e, but use ecosystem services to comply with the Cascade model.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\n \u003ch2\u003ePhase II\u003c/h2\u003e\n \u003cp\u003ePhase II aimed at identifying interactions between submodels and to define node\u0026rsquo;s state thresholds. Forest characteristics that were shared in multiple submodels were jointly defined and constitute the overarching model structure shown in Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003ec. This represents current knowledge in forest ecological interactions. Next, we explored interactions between submodels to increase model cohesion and the potential to unveil trade-offs and synergies. At last, nodes were partitioning according to specific state thresholds. The following guidelines were provided for continuous variables: i) define the node\u0026rsquo;s unit and potential range, ii) define the minimum number of possible states that still makes logical sense (maximum five)\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e, iii) define threshold values based on published guidelines (e.g., WHO air quality guidelines), recognized classifications (e.g., heat stress categories) or logical reasoning (e.g., increasing \u003cem\u003evs\u003c/em\u003e. decreasing anxiety)\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e71\u003c/span\u003e\u003c/sup\u003e, and iv) detailed reasoning, define state names whenever possible, and list key references. When obvious threshold values were lacking, data percentiles were used. When a lot of empirical data was available and a child node had few parents, the state number limitation could be relaxed because the minimum data amount is alleviated.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec19\" class=\"Section2\"\u003e\n \u003ch2\u003ePhase III\u003c/h2\u003e\n \u003cp\u003eThe last phase quantified and joined submodels. The four major classes that can be combined in BBNs are raw empirical data, quantitative literature results, logical equations or expert opinion\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e31\u003c/span\u003e,\u003cspan class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e. Each has advantages and disadvantages. \u003cem\u003eEmpirical raw data\u003c/em\u003e are very easy to incorporate, especially when available for multiple nodes simultaneously, but the underlying counting algorithm requires sufficient data points for robust probability estimation. \u003cem\u003eQuantitative literature findings\u003c/em\u003e can include formulae with regression coefficients, or estimates with error distributions. When available, priority was given to robust meta-analyses. They were most often used to quantify health outcomes, given that multiple of our studies only gathered empirical data until the service stage. Their advantage is the relatively easy implementation in Netica, which integrates numerous error distribution functions, but finding published results with the same predictor set becomes more challenging as the number of parent nodes increases. \u003cem\u003eLogical equations\u003c/em\u003e usually link parent nodes via basic mathematical relationships. For example, the density of infected ticks (DIN) equals the density of questing ticks (DON) times the nymphal infection prevalence (NIP). They do not cause an effect dilution and are easy to incorporate, with few disadvantages. At last, \u003cem\u003eexpert opinion data\u003c/em\u003e are used when quantitative data are unavailable or not robust enough, which also represents their main advantage. We used a direct expert elicitation style based on recommendations by Kuhnert and colleagues\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e72\u003c/span\u003e\u003c/sup\u003e, assuming that our expert teams were made sufficiently familiar with probability theory through annual meetings and individual discussions. Here, experts had to estimate probability distributions for each combination of parent node states, thereby directly generating CPTs. The disadvantage is that this becomes a very cognitively-demanding endeavour as the number of parent nodes increases, which exponentially augments possible state combinations.\u003c/p\u003e\n \u003cp\u003eTo support partners in identifying data classes, a custom-made decision tree was presented (Supplementary Information p. 8). The decision tree aims to circumvent aforementioned weaknesses or provides alternatives by combining classes. For example, using only raw empirical data is discouraged whenever there are less than ten observations per combination of parent states, which is a relaxed version of the untested\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e73\u003c/span\u003e\u003c/sup\u003e recommendation by Cain (2001)\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e. Concretely, when one parent has three states and the second has four, the empirical dataset should have at least 120 observations for the child node. When this is not reached but empirical data are available, users are redirected to options where literature results or expert opinion data are used as priors\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e72\u003c/span\u003e\u003c/sup\u003e before \u0026lsquo;belief updating\u0026rsquo; is done with raw data as posteriors. The decision tree also implies a priority ranking in classes: whenever possible, own empirical data was prioritised over literature results, which was prioritised over expert opinion data.\u003c/p\u003e\n \u003cp\u003eThe common forest structure was exclusively quantified using empirical data. However, since all submodels relied on it and the canopy density node had four parents, additional data were retrieved. Its eight constituting nodes (plus \u0026lsquo;understory productivity\u0026rsquo; and \u0026lsquo;soil fertility\u0026rsquo; nodes) were reinforced with data from international forest networks that were designed to study forest biodiversity effects on ecosystem functioning, analogous to our network. These include 225 FORMICA\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e plots and the 41 TREEWEB\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e plots that were not yet included in our network, for a total of 430 plots.\u003c/p\u003e\n \u003cp\u003eAs a final submodel-specific procedure, a formal validation meeting was organised after submodels were operationalized in Netica. Results of sensitivity analyses and general submodel behaviour were scrutinised for inconsistencies by our expert partners\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e67\u003c/span\u003e\u003c/sup\u003e, allowing for final adaptations before merging them with other submodels. At last, all submodels were joined to the common structure and interactions were added. Some interactions were indicated as merely \u0026lsquo;implicit\u0026rsquo; associations when two nodes were not directly comparable or already had the same parent node combinations.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec20\" class=\"Section2\"\u003e\n \u003ch2\u003eGeneral model insights and shared forest nodes\u003c/h2\u003e\n \u003cp\u003eWith 82 nodes (i.e. variables), 123 links and 3154 conditional probabilities, the final model is among the largest published BBNs in ecosystem service modelling\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. Most nodes (n\u0026thinsp;=\u0026thinsp;56; 68%) were exclusively quantified using our own empirical data. Ten were operationalized via logical equations and nine via quantitative literature findings. Expert opinion elicitation was used for eight nodes, but always as priors with posterior belief updating using empirical data when the minimal number of observations was lacking.\u003c/p\u003e\n \u003cp\u003eThe model building exercise yielded an extensive Supplementary Information document that is appended for maximal transparency and comprehensibility of the numerous choices underlying submodels. It contains information on general methodology, submodel-specific summaries and limitations, and node-specific details including variable definitions, state thresholds and key references. As only part of the full complexity of our final BBN can be exhibited here, this reference document should accompany the model\u0026rsquo;s usage and interpretation.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec21\" class=\"Section2\"\u003e\n \u003ch2\u003eSynthesis analyses\u003c/h2\u003e\n \u003cp\u003eOnce the complete BBN was assembled, it was subjected to three main analyses. First, the explanatory power of the common forest characteristics was quantified using built-in sensitivity analyses, represented by the variance reduction of the expected real value of the child node due to a finding in the parent\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e69\u003c/span\u003e\u003c/sup\u003e. Second, we extracted the mutual information between health outcome nodes to identify trade-offs and synergies. This is quantified as the reduction in entropy in one node, measured in computational bits, due to a finding in the other node\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e69\u003c/span\u003e\u003c/sup\u003e. Because health outcome nodes are separated by long interaction chains that cause effect dilutions\u003csup\u003e\u003cspan class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e, this yielded low values. Threshold values were established to distinguish robust associations from the bulk of negligible ones. If remaining associations were visible between nodes of different submodels in each stage until health outcomes, they were considered the strongest and most consistent associations. Those only emerging in function or service stages were considered secondary associations. Third and last, the magnitude of health outcomes was expressed relative to non-forest conditions and among forests. To illustrate potential inter-forest differences, four scenarios were considered based on two of the most influential and independent forest characteristics: monospecific or mixed stands with open (LAI\u0026thinsp;\u0026lt;\u0026thinsp;1) or closed (LAI\u0026thinsp;\u0026gt;\u0026thinsp;3) canopies. Per scenario, 10 000 cases were simulated in Netica to obtain means and uncertainty distributions. In about half of the forest-health pathways, health outcomes were explicitly expressed relative to control conditions, which could be urban environments (mental wellbeing \u0026ndash; visual) or ambient, non-forest conditions (thermal comfort and air quality). In the other cases, suitable control conditions were generally less evident. For mental mental wellbeing via auditory stimuli, the baseline were measures taken prior to the intervention, which was already accounted for by looking at mental state changes (\u0026Delta;). For medicinal plants and mushrooms, baselines can be assumed to represent no intake. For Lyme disease, baselines would be environments with risk absence such as highly urbanised locations.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAcknowledgements\u003c/h2\u003e \u003cp\u003eThis research was funded by the ERA-Net BiodivERsA project Dr.FOREST, with the national funders German Research Foundation (DFG \u0026minus;\u0026thinsp;428795724, Germany), French National Research Agency (ANR, France), Research Foundation \u0026ndash; Flanders (FWO, Belgium), Austrian Science Fund (FWF, Austria) and National Science Center (NCN, Poland, project no. 2019/31/Z/NZ8/04032), as part of the 2018\u0026ndash;2019 BiodivERsA call for research proposals. PDF received funding from the European Research Council (ERC) under the European Union\u0026rsquo;s Horizon 2020 research and innovation programme (ERC Starting Grant FORMICA 757833). DL was supported by a postdoctoral fellowship of the FWO. TV was supported by the UGent GOA project (BOF20/GOA/009). QP would like to thank the Walloon forest service (SPW \u0026ndash; DNF) for its support to the maintenance of the FORBIO-Gedinne experiment in the frame of the 5-yr research programme \u0026lsquo;Plan quinquennal de recherche et de vulgarisation foresti\u0026egrave;res\u0026rsquo;.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eBeute F et al (2023) How do different types and characteristics of green space impact mental health? A scoping review. People Nat 5:1839\u0026ndash;1876\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHartig T, Mitchell R, de Vries S, Frumkin H (2014) Nature and Health. Annu Rev Public Health 35:207\u0026ndash;228\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKonijnendijk C, Devkota D, Mansourian S, Wildburger C (2023) \u003cem\u003eForests and Trees for Human Health: Pathways, Impacts, Challenges and Response Options. A Global Assessment Report\u003c/em\u003e. 232\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWHO. Nature, Biodiversity and health: an overview of interconnections. 42 (2021)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReyes-Riveros R et al (2021) Linking public urban green spaces and human well-being: A systematic review. Urban Forestry Urban Green 61:127105\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWHO (2022) Noncommunicable diseases progress monitor. World Health Organization\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePatel V et al (2018) The Lancet Commission on global mental health and sustainable development. Lancet 392:1553\u0026ndash;1598\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChi D et al (2022) Residential Exposure to Urban Trees and Medication Sales for Mood Disorders and Cardiovascular Disease in Brussels, Belgium: An Ecological Study. Environ Health Perspect 130:057003\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIungman T et al (2023) Cooling cities through urban green infrastructure: a health impact assessment of European cities. Lancet 0\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarboza EP et al (2021) Green space and mortality in European cities: a health impact assessment study. Lancet Planet Health 5:e718\u0026ndash;e730\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan den Berg M et al (2015) Health benefits of green spaces in the living environment: A systematic review of epidemiological studies. Urban Forestry Urban Green 14:806\u0026ndash;816\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNguyen P-Y, Astell-Burt T, Rahimi-Ardabili H, Feng X (2021) Green Space Quality and Health: A Systematic Review. Int J Environ Res Public Health 18:11028\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Frenne P et al (2019) Global buffering of temperatures under forest canopies. Nat Ecol Evol 3:744\u0026ndash;749\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDiener A, Mudu P (2021) How can vegetation protect us from air pollution? A critical review on green spaces\u0026rsquo; mitigation abilities for air-borne particles from a public health perspective - with implications for urban planning. Sci Total Environ 796:148605\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEkkel ED, de Vries S (2017) Nearby green space and human health: Evaluating accessibility metrics. Landsc Urban Plann 157:214\u0026ndash;220\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDallimer M et al (2012) Biodiversity and the Feel-Good Factor: Understanding Associations between Self-Reported Human Well-being and Species Richness. Bioscience 62:47\u0026ndash;55\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAerts R, Honnay O, Van Nieuwenhuyse A (2018) Biodiversity and human health: mechanisms and evidence of the positive health effects of diversity in nature and green spaces. Br Med Bull 127:5\u0026ndash;22\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrockerhoff EG et al (2017) Forest biodiversity, ecosystem functioning and the provision of ecosystem services. Biodivers Conserv 26:3005\u0026ndash;3035\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHaines-Young R, Potschin M (2010) The links between biodiversity, ecosystem services and human well-being. in \u003cem\u003eEcosystem Ecology\u003c/em\u003e (eds. Raffaelli, D. G. \u0026amp; Frid, C. L. J.) 110\u0026ndash;139Cambridge University Press, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1017/CBO9780511750458.007\u003c/span\u003e\u003cspan address=\"10.1017/CBO9780511750458.007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTilman D, Isbell F, Cowles JM (2014) Biodiversity and Ecosystem Functioning. Annu Rev Ecol Evol Syst 45:471\u0026ndash;493\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarselle MR et al (2021) Pathways linking biodiversity to human health: A conceptual framework. Environ Int 150:106420\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarselle MR (2020) Urban street tree biodiversity and antidepressant prescriptions. Sci Rep 11\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrayenhoff ES et al (2021) Cooling hot cities: a systematic and critical review of the numerical modelling literature. Environ Res Lett 16:053007\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHan D, Shen H, Duan W, Chen L (2020) A review on particulate matter removal capacity by urban forests at different scales. Urban Forestry Urban Green 48:126565\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRozario K et al (2024) Perceived biodiversity: is what we measure also what we see and hear? 2024.04.03.587905 Preprint. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003eat https://doi.org/10.1101/2024.04.03.587905\u003c/span\u003e\u003cspan address=\"at 10.1101/2024.04.03.587905\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGillerot L et al (2024) Forests are chill: The interplay between thermal comfort and mental wellbeing. Landsc Urban Plann 242:104933\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRey Gozalo G, Barrig\u0026oacute;n Morillas JM, Montes Gonz\u0026aacute;lez D (2018) Atanasio Moraga, P. Relationships among satisfaction, noise perception, and use of urban green spaces. Sci Total Environ 624:438\u0026ndash;450\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBourdin A et al (2023) Forests harbor more ticks than other habitats: A meta-analysis. For Ecol Manag 541:121081\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDudek T, Kasprzyk I, Dulska-Jeż A (2018) Forest as a place for recreation but also the source of allergenic plant pollen: to come or avoid? Eur J For Res 137:849\u0026ndash;862\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePark B-J et al (2011) Relationship between psychological responses and physical environments in forest settings. Landsc Urban Plann 102:24\u0026ndash;32\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLanduyt D et al (2013) A review of Bayesian belief networks in ecosystem service modelling. Environ Model Softw 46:1\u0026ndash;11\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRozario K et al (2024) The more the merrier? Perceived forest biodiversity promotes short-term mental health and well-being\u0026mdash;A multicentre study. People Nat 6:180\u0026ndash;201\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGillerot L et al (2022) Forest structure and composition alleviate human thermal stress. Glob Change Biol 28:7340\u0026ndash;7352\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSteinparzer M, Haluza D, Godbold DL (2022) Integrating Tree Species Identity and Diversity in Particulate Matter Adsorption. Forests 13:481\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZellweger F et al (2020) Forest microclimate dynamics drive plant responses to warming. Science 368:772\u0026ndash;775\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWong NH, Tan CL, Kolokotsa DD, Takebayashi H (2021) Greenery as a mitigation and adaptation strategy to urban heat. Nat Rev Earth Environ 2:166\u0026ndash;181\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVerheyen K et al (2024) Forest canopies as nature-based solutions to mitigate global change effects on people and nature. J Ecol\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAmpoorter E et al (2020) Tree diversity is key for promoting the diversity and abundance of forest-associated taxa in Europe. Oikos 129:133\u0026ndash;146\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStojek K, Gillerot L, Jaroszewicz B (2022) Predictors of mushroom production in the European temperate mixed deciduous forest. For Ecol Manag 522:120451\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGori A, Nascimento LB, Ferrini F, Centritto M, Brunetti C (2020) Seasonal and Diurnal Variation in Leaf Phenolics of Three Medicinal Mediterranean Wild Species: What Is the Best Harvesting Moment to Obtain the Richest and the Most Antioxidant Extracts? Molecules 25:956\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrinfeder E et al (2022) Soundscape dynamics of a cold protected forest: dominance of aircraft noise. Landsc Ecol 37:567\u0026ndash;582\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchulp CJE, Thuiller W, Verburg PH (2014) Wild food in Europe: A synthesis of knowledge and data of terrestrial wild food as an ecosystem service. Ecol Econ 105:292\u0026ndash;305\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJansson M, Fors H, Lindgren T, Wistr\u0026ouml;m B (2013) Perceived personal safety in relation to urban woodland vegetation \u0026ndash; A review. Urban Forestry Urban Green 12:127\u0026ndash;133\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLai D et al (2020) A comprehensive review of thermal comfort studies in urban open spaces. Sci Total Environ 742:140092\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBlondeel H et al (2020) Light and warming drive forest understorey community development in different environments. Glob Change Biol 26:1681\u0026ndash;1696\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGray JS (1998) Review The ecology of ticks transmitting Lyme borreliosis. Exp Appl Acarol 22:249\u0026ndash;258\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVanneste T et al (2024) Trade-offs in biodiversity and ecosystem services between edges and interiors in European forests. Nat Ecol Evol 1\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41559-024-02335-6\u003c/span\u003e\u003cspan address=\"10.1038/s41559-024-02335-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVanroy T et al (2024) The effect of forest structural complexity on tick-borne pathogens in questing ticks and small mammals. For Ecol Manag 562:121944\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBourdin A et al (2022) Forest Diversity Reduces the Prevalence of Pathogens Transmitted by the Tick Ixodes ricinus. Front Ecol Evol. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fevo.2022.891908\u003c/span\u003e\u003cspan address=\"10.3389/fevo.2022.891908\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOstfeld RS, Keesing F (2012) Effects of host diversity on infectious disease. Annu Rev Ecol Evol Syst 43:157\u0026ndash;182\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSteinparzer M, Schaubmayr J, Godbold DL, Rewald B (2023) Particulate matter accumulation by tree foliage is driven by leaf habit types, urbanization- and pollution levels. Environ Pollut 335:122289\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCao Z, Zhou J, Li M, Huang J, Dou D (2023) Urbanites\u0026rsquo; mental health undermined by air pollution. Nat Sustain 6:470\u0026ndash;478\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOECD (2022) Health at a Glance: Europe 2022: State of Health in the EU Cycle. Organisation for Economic Co-operation and Development\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBallester J et al (2023) Heat-related mortality in Europe during the summer of 2022. Nat Med 29:1857\u0026ndash;1866\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFreer-Smith P et al (2019) Plantation forests in Europe: challenges and opportunities. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.36333/fs09\u003c/span\u003e\u003cspan address=\"10.36333/fs09\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e\u003cem\u003ePlanning improvements in natural resources management: Guidelines for using Bayesian networks to support the planning and management of development programmes in the water sector and beyond\u003c/em\u003e. (2001)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBratman GN et al (2019) Nature and mental health: An ecosystem service perspective. Sci Adv 5:eaax0903\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSohn JA, Saha S, Bauhus J (2016) Potential of forest thinning to mitigate drought stress: A meta-analysis. For Ecol Manag 380:261\u0026ndash;273\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKilpatrick AM, Salkeld DJ, Titcomb G, Hahn MB (2017) Conservation of biodiversity as a strategy for improving human health and well-being. Phil Trans R Soc B 372:20160131\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJactel H, Moreira X, Castagneyrol B (2021) Tree Diversity and Forest Resistance to Insect Pests: Patterns, Mechanisms and Prospects. \u003cem\u003eAnnual Review of Entomology\u003c/em\u003e 66, null\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaeten L et al (2013) A novel comparative research platform designed to determine the functional significance of tree species diversity in European forests. Perspect Plant Ecol Evol Syst 15:281\u0026ndash;291\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Groote S et al (2017) Tree species identity outweighs the effects of tree species diversity and forest fragmentation on understorey diversity and composition. Plant Ecol Evol 150:229\u0026ndash;239\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePaquette A et al (2018) A million and more trees for science. Nat Ecol Evol 2:763\u0026ndash;766\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNickmans H (2019) The nutrition of oak and beech trees along a tree diversity gradient\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGillerot L, Landuyt D, De Frenne P, Muys B, Verheyen K (2024) Urban tree canopies drive human heat stress mitigation. Urban Forestry Urban Green 92:128192\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJensen FV, Nielsen TD (2007) Bayesian networks and decision graphs. Springer\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAguilera PA, Fern\u0026aacute;ndez A, Fern\u0026aacute;ndez R, Rum\u0026iacute; R, Salmer\u0026oacute;n A (2011) Bayesian networks in environmental modelling. Environ Model Softw 26:1376\u0026ndash;1388\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMarcot BG, Steventon JD, Sutherland GD, McCann RK (2006) Guidelines for developing and updating Bayesian belief networks applied to ecological modeling and conservation. Can J Res 36:3063\u0026ndash;3074\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNorsys Software Corporation. Netica Application User\u0026rsquo;s Guide (1998)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePascual U et al (2017) Valuing nature\u0026rsquo;s contributions to people: the IPBES approach. Curr Opin Environ Sustain 26\u0026ndash;27:7\u0026ndash;16\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUusitalo L (2007) Advantages and challenges of Bayesian networks in environmental modelling. Ecol Model 203:312\u0026ndash;318\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKuhnert PM, Martin TG, Griffiths SP (2010) A guide to eliciting and using expert knowledge in Bayesian ecological models. Ecol Lett 13:900\u0026ndash;914\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen SH, Pollino CA (2012) Good practice in Bayesian network modelling. Environ Model Softw 37:134\u0026ndash;145\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Dr.FOREST, ecosystem service, nature-based solution, environmental health, environmental psychology","lastPublishedDoi":"10.21203/rs.3.rs-4669329/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4669329/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Forest risks and benefits to human health are widely recognised. Yet, variation across forest types and their ecological characteristics driving health effects remain underexplored. Based on empirical data from an interdisciplinary European forest network, we developed a Bayesian Belief Network to quantify seven causal pathways relating different forest types to physical and mental health. Results show that forests always generate net health benefits regardless of their ecological characteristics. Forest canopy density and tree species diversity emerge as key drivers, but their effect size and directionality are strongly pathway-dependent. Changes in forest canopy density can generate trade-offs. For example, forests optimised for heat buffering and air pollution mitigation may compromise medicinal plant yield and enhance Lyme disease prevalence. Tree diversity effects were weaker but more consistently positive. Forest management should therefore account for such trade-offs to tailor forest biodiversity and functioning to local public health needs of priority.","manuscriptTitle":"Forest biodiversity and structure modulate human health benefits and risks","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-09-02 09:03:42","doi":"10.21203/rs.3.rs-4669329/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-sustainability","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"natsustain","sideBox":"Learn more about [Nature Sustainability](http://www.nature.com/natsustain/)","snPcode":"","submissionUrl":"","title":"Nature Sustainability","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Research","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"abbf35dd-8fa4-453e-a540-03c8cdeddfb0","owner":[],"postedDate":"September 2nd, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":36806582,"name":"Biological sciences/Psychology"},{"id":36806583,"name":"Earth and environmental sciences/Ecology/Biodiversity"},{"id":36806584,"name":"Earth and environmental sciences/Ecology/Ecological epidemiology"},{"id":36806585,"name":"Earth and environmental sciences/Ecology/Forest ecology"}],"tags":[],"updatedAt":"2025-05-20T07:08:57+00:00","versionOfRecord":{"articleIdentity":"rs-4669329","link":"https://doi.org/10.1038/s41893-025-01547-3","journal":{"identity":"nature-sustainability","isVorOnly":false,"title":"Nature Sustainability"},"publishedOn":"2025-05-19 04:00:00","publishedOnDateReadable":"May 19th, 2025"},"versionCreatedAt":"2024-09-02 09:03:42","video":"","vorDoi":"10.1038/s41893-025-01547-3","vorDoiUrl":"https://doi.org/10.1038/s41893-025-01547-3","workflowStages":[]},"version":"v1","identity":"rs-4669329","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4669329","identity":"rs-4669329","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}