Tree-related Microhabitats as Indicators of Anthropogenic Impact: Comparing Conservation and Recreation Areas in Yedigöller National Park (Türkiye)

Schlüsselbegriffe: Waldbiodiversität, Habitatbaum, Nationalpark, Erholungsnutzungseinfluss, Baummikrohabitate

Download as PDF

See below the issue 3/2026 as E-Paper or have a look at our E-Paper archive dating back to 1955.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Abstract

Tree-related microhabitats (TreMs) are distinct structures on living or dead trees that provide critical habitats for numerous species and serve as valuable indicators of forest biodiversity. This study represents the first assessment of TreMs in a national park in Türkiye, Yedigöller National Park, with the aim of comparing areas with different levels of human impact to evaluate effects on TreMs. Fieldwork was carried out in 10 study sites (5 conservation, 5 recreation) with one 625 m² square plot established at each site, where tree species, diameter, health status, and TreMs types were recorded. In total, 272 trees and 2020 TreMs were documented. Although recreational areas hosted a higher overall number of TreMs, conservation areas exhibited greater average abundance and higher ecological value (EcoVal), reflecting the rarity and long formation time of many TreMs. Both living and standing dead wood (snags) were identified as essential for sustaining TreMs diversity, with dead wood harboring particularly rich and valuable TreMs assemblages. Habitat trees, regardless of being alive or dead, played a central role in supporting diverse and high-value microhabitats. These findings from Yedigöller National Park highlight the importance of TreMs for biodiversity conservation and underline the need to balance conservation priorities with recreational use in protected forests to ensure the long-term ecosystem integrity.

Zusammenfassung

Baummikrohabitate (TreMs) sind charakteristische Strukturen an lebenden oder toten Bäumen, die zahlreichen Arten essenzielle Lebensräume bieten und als wertvolle Indikatoren der Waldbiodiversität dienen. Die vorliegende Studie stellt die erste Untersuchung von TreMs in einem Nationalpark in der Türkei dar, Nationalpark Yedigöller, mit dem Ziel, Schutz- und Erholungsgebiete zu vergleichen, um die Auswirkungen menschlicher Aktivitäten auf TreMs zu bewerten. Die Feldarbeiten wurden in 10 Quadraten an 10 Untersuchungsflächen (5 Schutzgebiete, 5 Erholungsgebiete) mit jeweils einem 625 m² großen quadratischen Plot durchgeführt, wobei Baumarten, Durchmesser, Gesundheitszustand und TreMs-Typen erfasst wurden. Insgesamt wurden 272 Bäume und 2020 TreMs dokumentiert. Obwohl in den Erholungsgebieten eine höhere Gesamtzahl von TreMs festgestellt wurde, wiesen die Schutzgebiete eine größere durchschnittliche Häufigkeit und einen höheren ökologischen Wert (EcoVal) auf, was auf die Seltenheit und lange Entwicklungszeit vieler TreMs zurückzuführen ist. Es zeigte sich, dass sowohl lebendes als auch totes Holz für den Erhalt der TreMs-Diversität unerlässlich ist, wobei Totholz besonders artenreiche und wertvolle TreMs-Gemeinschaften beherbergt. Habitatbäume, unabhängig davon, ob lebend oder tot, spielten eine zentrale Rolle bei der Unterstützung vielfältiger und ökologisch hochwertiger Mikrohabitate. Diese Ergebnisse aus dem Yedigöller-Nationalpark unterstreichen die Bedeutung von TreMs für den Biodiversitätsschutz und heben die Notwendigkeit hervor, Naturschutzprioritäten mit Erholungsnutzung in Schutzwäldern in Einklang zu bringen, um die langfristige Integrität der Ökosysteme zu gewährleisten.

1 Introduction

Protected forests are essential for biodiversity conservation, yet many face mounting pressure from recreational activities and management interventions. National parks, which generally contain a considerable amount of forest cover, are keystones of the global protected area network (Arcese and Sinclair 1997; Nelson and Serafin 1997). While there is still a need for effective tools to understand and monitor their ecological performance (Gaston et al. 2008), conflicts about land-use within and around the national parks persist more commonly in developing countries (Hough 1988) like Türkiye.

Tree-related microhabitats (TreMs)—including cavities, injuries, bark pockets, and fungal bodies—are distinct structures on living or dead trees that provide essential habitat for diverse taxonomic groups from invertebrates to vertebrates, fungi, and plants, during their life cycles (Grossmann et al. 2018; Larrieu et al., 2018; Bütler et al. 2020; Asbeck et al. 2021). Their diversity and abundance have emerged as powerful indicators of forest biodiversity (Larrieu and Cabanettes 2012; Michel and Winter 2009). Research demonstrates that TreMs abundance correlates strongly with tree size and vitality status: habitat trees (typically large, old trees) and dead wood harbor significantly more microhabitats than smaller, living trees (Paillet et al. 2019; Asbeck et al. 2020). The ecological value (EcoVal) of TreMs, based on their rarity and development time, provides a standardized metric for assessing habitat quality and conservation effectiveness (Kraus et al. 2018).

TreM-based assessments are increasingly operationalized through standardized field methodologies, including the Marteloscope approach, which enables systematic tree-level recording of microhabitat types and ecological values across permanent forest plots (Bütler et al. 2013; Kraus & Krumm 2013; Kraus et al. 2018). While TreMs research has been extensively conducted in temperate and boreal forests in Central Europe and North America, as well as in Western and Central Mediterranean  forests (Dutta et al. 2025) -particularly Spain, France, and Italy (see Integrate+ Network; Kraus et al. 2022), studies in the Eastern Mediterranean, Anatolian, and Middle Eastern regions remain relatively scarce despite their ecological significance. Recent work in Iranian Oriental beech forests has highlighted the importance of tree species identity and DBH as key determinants of TreM occurrence (Khanalizadeh et al. 2020), while a broader investigation across Europe and Iran identified stand dynamics and tree aging as primary drivers of microhabitat formation rates, underscoring the need for predictive models to guide retention forestry decisions (Courbaud et al. 2022). Similarly, studies in Mediterranean oak and Italian mountain forests have confirmed that large trees and standing dead wood disproportionately support microhabitat diversity (Regnery et al. 2013; Marziliano et al. 2021). The Western Black Sea Region of Anatolia, part of the Euro-Siberian phytogeographical region and home to transitional forests where European temperate and Colchic elements converge, represents a particularly understudied system in this regard. These forests, dominated by tree taxa such as Fagus orientalis, Carpinus betulus, Quercus spp., Abies spp., and Pinus spp., face increasing recreational and management pressure that may affect TreM assemblages in ways not yet documented.

Yedigöller National Park in Türkiye provides an instructive example of protected areas with dual conservation-recreation functions while offering an ideal setting to investigate TreMs patterns in Black Sea region forests. The park‘s mixed deciduous-coniferous forests experience a stark gradient of human impact: recreational facilities are clustered around a few of the park’s seven avalanche-formed lakes, concentrating visitor pressure into relatively small zones with measurable ecological impacts (Kırca and Altınçekiç 2018). In these areas, forest understory seedlings often fail to establish due to intensive trampling, while soil compaction from concentrated use leads to widespread exposure of tree roots at the surface (Kırca and Altınçekiç 2018). According to park management, additional interventions, including pruning to minimize branch-fall risks to visitors, are regularly applied in picnic and camping areas. Furthermore, dead branches and wood are frequently used in illegally lit fires, resulting in the near absence of dead wood in recreational sites. The park‘s rugged topography generates pronounced spatial heterogeneity in accessibility over short distances, allowing high-impact recreational zones to exist in close proximity to areas with minimal disturbance on less accessible terrain. Thus, although Yedigöller is legally designated as unmanaged forest within a national park, it remains subject to management interventions such as pruning for visitor safety as well as direct anthropogenic pressures (trampling, soil compaction, vandalism) and indirect impacts (dead wood removal for fires) (Kırca 2015; pers. comm. with park management; pers. obs.).

This study examines the TreMs distribution in Yedigöller National Park, comparing conservation and recreational areas, to assess anthropogenic impacts on biodiversity indicators. We test the hypothesis that recreational areas exhibit reduced TreMs abundance and ecological value compared to conservation areas, and investigate whether habitat trees and dead wood remain key determinants of microhabitat occurrence under varying human pressure. Additionally, we evaluate how specific anthropogenic activities affect different TreMs types, providing insights into the mechanisms of human impact on forest microhabitats. This study contributes to the limited knowledge of TreM patterns in Black Sea region forests while informing management strategies that balance biodiversity conservation with sustainable recreational use in protected forests where multiple functions coexist.

2 Materials and Methods

2.1 Study Area and Sample Sites

Our study was conducted in Yedigöller National Park, located in the Western Black Sea Region of Türkiye, within the province of Bolu (31.7333° E – 31.7833° E and 40.9167° N – 40.9667° N). Designated in 1965, the park spans 1,631 hectares of mixed deciduous and coniferous forests dominated by Fagus orientalis, Carpinus betulus, Pinus nigra subsp. pallasiana, Quercus petraea, and Abies nordmanniana subsp. bornmuelleriana (Kırca 2015, Tokcan 2015), creating diverse microhabitat conditions suitable for our study objectives. These intact ecosystems provide refuge for a variety of plant and animal species (Beşkardeş 2012), contributing to the park’s conservation value (Kırca 2015). Forest stands are predominantly uneven-aged (97%), with 74% exhibiting double or multiple canopy layers, creating structurally diverse conditions favorable for TreMs development. The park receives approximately 200,000 visitors annually between April and mid-November, with recreational activities concentrated around the seven avalanche-formed lakes (Figure 1).

To assess human impact on the national park‘s ecological integrity, we selected 10 study sites representing a gradient of anthropogenic influence previously characterized by Kırca (2015) and Kırca and Altınçekiç (2018). Sites were classified into two categories based on hemeroby levels from Kırca (2015): conservation areas (n=5; hemeroby levels 8-9, indicating minimal human modification) and recreation areas (n=5; hemeroby levels 4-6, indicating moderate modification from recreational activities). Site selection followed a purposive sampling approach based on three criteria: 

(1) clear differentiation in hemeroby levels reflecting distinct human use intensities, 

(2) representation of the park‘s dominant forest types and structural characteristics, and 

(3) accessibility for comprehensive field surveys. 

Within each site, a single 625 m² (25 × 25 m) quadrat was established in locations representing modal stand conditions. In conservation areas, quadrats were positioned ≥50 m from roads and recreational infrastructure to minimize edge effects; in recreation areas, quadrats were placed within 20 m of trails or facilities to capture maximum anthropogenic impact. All quadrats avoided immediate trail edges (>5 m) to ensure safety and minimize observer disturbance. Information on management practices was obtained through consultation with park management staff and direct field observations during site surveys (2023).

Welch‘s t-tests confirmed no significant differences between area types in key topographic and structural variables (Table 1): elevation (recreation: 815 ± 53 m; conservation: 885 ± 118 m; t=-1.26, p=0.273), tree density (recreation: 460.8 ± 86.5 trees/ha; conservation: 409.6 ± 151.2 trees/ha; t=0.64, p=0.554), mean DBH (recreation: 33.65 ± 11.46 cm; conservation: 41.09 ± 22.46 cm; t=-0.68, p=0.527), and basal area (recreation: 49.34 ± 23.19 m²/ha; conservation: 65.68 ± 45.95 m²/ha; t=-0.71, p=0.490).

 

2.2 Tree Microhabitats, Habitat Trees and Vandalism

2.2.1 Tree inventory and Tree-Related Microhabitats

Within each 25 × 25 m (625 m²) square plot, we recorded data for every tree with DBH ≥7.5 cm, following Schuck et al. (2015b) and the Marteloscope methodology standard (Larrieu et al. 2018; Kraus & Krumm 2013). Recorded parameters included species identification, DBH, and health status (classified as alive or standing dead: only snags were recorded among dead trees). Although shrub species (e.g., Rhododendron spp.) were present in the study area, their stem diameters remained below the 7.5 cm DBH threshold and were therefore excluded from the analysis.

TreMs were assessed following the standardized catalogue of Kraus et al. (2016), which classifies microhabitats from broad categories (Level 1) to specific types (Level 3), and remains actively applied in recent studies on forest management and conservation (Fröhlich et al. 2025; Przepióra et al. 2025; Visick et al. 2025). We applied the 2016 catalogue to maintain compatibility with the Integrate+ decision support system, which automates EcoVal scoring, and to ensure comparability with the broader European TreM literature; our fieldwork preceded publication of the revised catalogue (Bütler et al. 2024). All TreM types were recorded at Level 3 following best practices (Larrieu et al. 2018), but results are presented at Level 2 to facilitate interpretation and ensure comparability with forest inventories and conservation frameworks such as Natura 2000 (Supplementary Table 1). We also documented a site-specific anthropogenic impact termed vandalism, characterized by human-caused trunk injuries from scraping and nailing; these records are presented separately to illustrate human impact patterns in recreational areas, but were not included in statistical analyses.

2.2.2 Habitat Trees and Vandalism

Habitat trees were defined as individuals with DBH ≥ 40 cm, following the operational threshold used in European forest management and TreM assessments. This threshold is widely used as an operational indicator of ecological value in European forest management (Niedermann-Meier et al. 2010).

Tree vandalism was frequently observed in recreation areas and consisted mainly of carved inscriptions and trunk injuries caused by visitors. Following the TreM framework, vandalism traces were classified into two size classes analogous to bark loss/exposed sapwood categories (IN1): VA1 (25–600 cm²) and VA2 (>600 cm²). Trees exhibiting both types were classified as VA1+VA2 (Figure 2).

2.2.3 TreM Categories Used to Assess Forestry Implementations and Anthropogenic Pressure

TreM types associated with cavity-related, direct and indirect anthropogenic effects were evaluated by comparing their frequencies between area types. We compared cavity-related TreMs including woodpecker cavities (CV12: ø 5-6 cm) and branch holes (CV31: ø ≥5 cm; CV32: ø ≥10 cm), which may develop following either branch pruning or natural branch fall (Figure 3-A). Since these structures form over 10-30+ years through progressive decay, we cannot determine whether any individual cavity originated from historical pruning, natural branch fall, storm damage, or other causes.

To identify direct anthropogenic effects, we examined direct injury-type TreMs that are more likely to reflect recent (<10 years) anthropogenic impacts, including fresh trunk injuries with limited decay progression. These included trunk and mould cavities with ground contact (CV23, ø ≥ 10 cm; CV24, ø ≥ 30 cm) and without ground contact (CV25, ø ≥ 30 cm/semi-open), together with bark loss (IN11, 25–600 cm², decay stage <3; IN12, >600 cm², decay stage <3) (Figure 3-B).

For indirect anthropogenic effects, discussions with foresters indicated that visitor activities such as walking, picnicking, or camping can cause soil compaction, exposing tree roots and eventually leading to the formation of trunk cavities with ground contact. Accordingly, we examined trunk and mould cavities with ground contact (CV21, ø ≥ 10 cm; CV22, ø ≥ 30 cm) and root buttress cavities (GR1) (Figure 3-C).

2.3 Data Analysis

All statistical analyses were conducted in R (R Core Team, 2023). TreM data were analyzed at the individual tree level using non-aggregated observations to preserve variability within the dataset. Mean (± standard deviation, SD) values were calculated only for descriptive summaries and graphical visualization. Two response variables were considered: 

(i) the number of TreMs per tree and 

(ii) the ecological value of TreMs (EcoVal). 

EcoVal was calculated following Kraus et al. (2018) by weighting TreMs according to their rarity (R), development time (D), and structural size (s), and was log-transformed (log10) where necessary to improve normality and reduce heteroscedasticity.

Differences in TreM abundance and EcoVal among area types (conservation vs. recreation), tree species, vitality classes (alive vs. standing dead wood), and habitat tree categories (DBH ≥ 40 cm vs. DBH < 40 cm) were evaluated using linear models (LM) or generalized linear models (GLM), depending on the distribution of the response variable. For count data, Poisson models were initially fitted and negative binomial models were applied when overdispersion was detected. Pairwise comparisons were performed using estimated marginal means (emmeans package).

Because Abies nordmanniana subsp. bornmuelleriana occurred only in conservation areas, species-level comparisons were restricted to Fagus orientalis, Carpinus betulus, and Quercus petraea, which were present in both area types. To account for potential confounding effects of tree size, additional models included DBH as a continuous predictor together with habitat tree classification, area type, and tree species. Interactions between DBH and habitat tree category were included in models testing EcoVal responses. Relationships between DBH and TreM abundance were further visualized using simple linear regression.

Multivariate differences in TreM composition between conservation and recreation areas were assessed using Principal Component Analysis (PCA) based on centered and scaled variables. TreM types with zero variance were excluded prior to analysis, and interpretation focused on the first two principal components (PC1 and PC2).

Vandalism categories (VA1, VA2, and VA1+VA2) and selected TreM types associated with cavity-related processes, forestry interventions, and anthropogenic impacts were evaluated descriptively by comparing occurrence frequencies between area types. These analyses were considered exploratory because the origin and timing of many TreMs could not be attributed with certainty to specific historical events.

Graphical outputs combined raw observations with mean ± SD values to visualize both variability and central tendency. EcoVal was displayed on a logarithmic (log10) scale where appropriate. Statistical significance was assessed at α = 0.05. Analyses were conducted using the ggplot2 (Wickham, 2016), emmeans (Lenth, 2024), and vegan (Oksanen et al. 2019) packages.

3 Results

3.1 Effects of Management Context on Tree Microhabitat Abundance, Ecological Value and Composition

While the total number of TreMs (conservation area = 983; recreation area = 1,037) and their cumulative EcoVal scores (conservation area = 10,177; recreation area = 10,611) were higher in recreation areas (Supplementary Table 1), mean-based comparisons revealed a different pattern. The average number of TreMs per tree (7.7 ± 6.4 in conservation areas vs. 7.2 ± 4.7 in recreation areas) and mean EcoVal (79.5 ± 90.2 vs. 73.7 ± 85.8, respectively) were both higher in conservation areas (Figure 4; Supplementary Table 2). This indicates that, on a per-tree basis, trees in conservation areas support richer and more valuable microhabitat structures. 

No significant difference was detected in TreM abundance between area types (p = 0.478), whereas EcoVal values were significantly higher in conservation areas (p < 0.001) (Figure 4; Supplementary Figure 1). The three most frequently recorded TreM types were identical in both area types: GR1 (root buttress cavities), IN1 (bark loss/exposed sapwood), and EP3 (epiphytic crypto- and phanerogams) (Figure 5; Supplementary Table 1).

Principal Component Analysis (PCA) revealed significant differences in microhabitat composition between conservation and recreation areas for both TreM abundance and EcoVal. The first two principal components explained 56.3% of the variation in TreM abundance data and 53.9% of the variation in EcoVal data (Figure 6). PC1 accounted for the largest share of variation in both analyses (36.6% and 36.8%, respectively) and was primarily associated with CV4 (dendrotelms and water-filled holes) and GR2 (witches’ brooms), which showed the strongest loadings on this axis (Supplementary Table 3). Pairwise comparisons based on PCA scores confirmed significant differences between conservation and recreation areas for both TreM abundance (R² = 0.592, p = 0.021) and EcoVal (R² = 0.527, p = 0.031).

3.2 Species-Specific Patterns of Tree Microhabitats

A total of eight tree species were recorded across the study area, with marked differences in species composition between conservation and recreation areas (Table 2; Supplementary Table 4). In general, trees in conservation areas tended to have larger diameters and higher TreM abundance and EcoVal values on a per-tree basis, despite often occurring at lower densities.

To ensure comparability between area types, species-level analyses were restricted to Fagus orientalis, Carpinus betulus, and Quercus petraea, which occurred in both conservation and recreation areas (Figure 7). For F. orientalis and C. betulus, total TreM numbers were generally higher in recreation areas because of greater tree abundance. However, mean TreM abundance and EcoVal per tree were significantly higher in conservation areas (p < 0.01), indicating that individual trees supported richer and more valuable microhabitat assemblages. This pattern was particularly pronounced for C. betulus.

A similar trend was observed for Q. petraea, for which both TreM abundance and EcoVal were higher in conservation areas, although differences were not statistically significant (TreM abundance: p = 0.087; Figure 7; Supplementary Table 4). Given the low number of Q. petraea individuals in recreation areas (n=2), these results should be interpreted with caution.

The distribution of TreM types also varied among species and area types (Supplementary Table 5; Figure 8). Fagus orientalis and Carpinus betulus supported the highest diversity and abundance of microhabitats, whereas Quercus petraea exhibited lower values. Conservation areas were characterized by higher frequencies of epiphytic microhabitats (EP1–EP3) and cavity-related structures (CV2–CV5), while recreation areas showed a greater occurrence of injury-related microhabitats (IN1–IN3). These patterns suggest that both tree species and management context influence the abundance, ecological value, and composition of tree-related microhabitats.

To further assess whether these area-type differences could be explained by tree size alone, we conducted analyses including DBH as a continuous predictor alongside area type and habitat tree classification (see section 3.4).

 

3.3 Effects of Tree Vitality on Tree Microhabitat Abundance and Ecological Value

Out of the 272 sampled trees, 259 were alive and 13 were standing dead wood (snags), with dead wood being considerably less frequent in recreation areas (2 individuals) than in conservation areas (11 individuals), likely reflecting both safety-related snag removal by park management and illegal harvesting for campfires in recreational zones. Mean-based comparisons showed that living trees exhibited similar TreM abundance and EcoVal values in conservation and recreation areas, with no significant differences between area types (Figure 9; Supplementary Table 6).

In contrast, standing dead wood supported substantially higher TreM abundance and EcoVal on a per-tree basis, particularly in conservation areas. Despite representing only a small proportion of the sampled trees, snags exhibited significantly higher TreM abundance and ecological value than living trees. These results highlight the disproportionate ecological importance of standing dead wood for the development and maintenance of tree-related microhabitats.

3.4 Habitat Trees, Tree Diameter and Ecological Value

Of the 272 sampled trees, 79 were classified as habitat trees (DBH ≥ 40 cm), including 38 individuals in conservation areas and 41 in recreation areas (Figure 10; Supplementary Table 9).

TreM abundance did not differ significantly between habitat trees and trees with DBH < 40 cm. In both area types, smaller trees exhibited slightly higher mean TreM numbers, although these differences were not consistently significant (Supplementary Table 9).

In contrast, habitat trees consistently exhibited higher EcoVal values than smaller trees despite having comparable or lower TreM abundance. This difference was particularly pronounced and statistically significant in conservation areas (p < 0.001; Supplementary Table 11), indicating that habitat trees support microhabitats of greater ecological importance even when overall TreM abundance is similar (Figure 10).

Regression analyses further showed that DBH was not significantly related to TreM abundance (p = 0.103), and neither habitat tree classification, area type, nor species identity significantly affected TreM abundance (p > 0.29). Pairwise comparisons likewise revealed no significant differences in TreM abundance between habitat trees and smaller trees within either area type (p = 0.12) (Figure 11A).

In contrast, EcoVal increased significantly with tree diameter (p < 0.001), following a significant positive relationship (Figure 11B; see caption for model parameters) . Although habitat tree classification itself was not a significant predictor (p > 0.60), area type and species identity remained significant after accounting for DBH (p < 0.01). A significant interaction between DBH and habitat tree classification (p < 0.001) further indicated that the relationship between tree size and ecological value differed between tree categories (Figure 11B).

3.5 Vandalism Patterns in Conservation and Recreation Areas

In the protected areas, we recorded VA1 type on 1 tree and VA2 type on 4 trees. The recreational areas showed significantly higher incidence: 52 trees with VA1 damage and 58 with VA2 damage. Additionally, 43 trees (1 in the protected areas and 42 in the recreational areas) exhibited both VA1 and VA2 types (Figure 12). The concentration of vandalism in recreation areas, particularly the high prevalence of VA2 damage, suggests that large-scale bark injuries may progressively develop into more complex cavity structures, potentially altering long-term TreMs composition in these zones.

3.6 Potential Effects of Forestry Implementations and Anthropogenic Pressure on Tree Microhabitats

Descriptive comparisons of selected TreM types associated with cavity formation, direct injury, and indirect anthropogenic effects revealed the following patterns. No significant differences were found in cavity-related TreMs between conservation and recreation areas (Figure 13-A; X-squared = 1.1965, df = 2, p-value = 0.5498; Supplementary Table 12). This result may reflect three possible explanations: 

(1) pruning does not substantially alter long-term cavity formation rates, 

(2) natural branch fall in conservation areas creates equivalent structures at similar frequencies, or 

(3) current cavity distributions reflect historical conditions rather than recent management practices. 

Given these causal ambiguities, we interpret this result cautiously as indicating no detectable difference in cavity-type TreM availability rather than as evidence regarding pruning impacts specifically.

Although not statistically significant (Figure 13-B; X-squared = 2.4747, df = 4, p-value = 0.6492), notable patterns emerged: CV23, IN11, and IN12 were approximately twice as common in recreation areas, while CV24 and CV25 occured at roughly four times the frequency (Supplementary Table 13). These patterns are suggestive of higher injury rates in recreation areas, though temporal ambiguity — even „fresh“ injuries may be 5-10 years old — limits causal interpretation. These structures could reflect visitor-caused damage, management interventions, or differential natural processes between area types.

Indirect anthropogenic effects showed the strongest and only statistically significant differences (Figure 13-C; X-squared = 26.893, df = 2, p-value = 1.446e-06). CV21 was significantly more prevalent in recreation areas, consistent with soil compaction from visitor pressure.  GR1 was more common in conservation areas (Supplementary Table 14), a pattern that may reflect earlier-stage root exposure that progressively develops into ground-contact cavities (CV21, CV22) under sustained anthropogenic pressure.

4 Discussion

4.1 Not All Microhabitats Are Equal: Higher Ecological Value in Conservation Areas

Our findings demonstrate that conservation and recreation areas in Yedigöller National Park differ not in the overall quantity of TreMs but in their ecological quality, with conservation areas supporting significantly higher EcoVal despite comparable or slightly lower total TreM counts. This distinction is ecologically meaningful: recreation areas accumulated higher frequencies of common, low-value microhabitat types such as bark loss and epiphytes, while conservation areas harbored structurally complex, slow-forming structures that contribute disproportionately to habitat value. This pattern is consistent with the broader literature showing that management disturbance reduces not only TreM abundance but preferentially suppresses rare, high-value microhabitat types that require long development times and structural complexity to form (Paillet et al. 2018; Asbeck et al. 2021). The mechanism likely involves multiple interacting pathways: safety-related pruning and tree removal reduce the structural substrate for complex cavity development, while soil compaction and trampling impair the decay processes that underlie many high-value TreM types. Conversely, the reduced disturbance in conservation areas allows trees to age, accumulate structural damage, and develop the morphological complexity that characterizes high-value microhabitat assemblages.

These findings are consistent with studies from Central European and Mediterranean forests demonstrating that reduced management intensity promotes the accumulation of structurally complex, high-value TreMs over time (Paillet et al. 2018; Regnery et al. 2013)). Importantly, our results extend this pattern to Black Sea region forests, where such comparisons have not previously been documented, suggesting that the relationship between disturbance regime and microhabitat quality may be broadly applicable across temperate forest types. The significant difference in EcoVal despite the absence of significant differences in raw TreM counts further underscores that abundance-based metrics alone are insufficient for assessing forest habitat quality—a finding with direct implications for biodiversity monitoring frameworks that currently rely on TreM counts rather than weighted ecological value scores.

An important methodological distinction emerges when comparing stand-level versus tree-level metrics. Although recreation areas contained marginally higher total TreM counts and slightly greater TreM density per hectare due to higher tree density, conservation areas exhibited higher  per tree TreM abundance and substantially higher EcoVal per tree. This divergence illustrates why raw stand-level TreM counts can be misleading as biodiversity indicators: a forest with higher total microhabitat numbers may nonetheless support lower habitat quality if those microhabitats are predominantly common, low-value types associated with recent disturbance. For conservation purposes, tree-level metrics weighted by ecological value are more informative because they capture habitat quality rather than quantity alone—a forest with fewer but structurally complex, high-value habitat trees may sustain more specialist species than one with many small-diameter trees bearing primarily recent wounds and common microhabitat types (Courbaud et al. 2022; Bianco et al. 2024). Our results suggest that areas with lower disturbance intensity are associated with the development of mature, ecologically valuable microhabitat assemblages (e.g., trunk cavities, dendrotelms, and epiphytic lichens and mosses), rather than simply higher microhabitat numbers—a pattern with direct implications for how TreM-based assessments are designed and interpreted in forest monitoring programmes.

4.2 Species Abundance Does Not Reflect Microhabitat Quality

Species identity influenced TreM abundance and diversity, with Fagus orientalis and Carpinus betulus supporting richer microhabitat assemblages than Quercus petraea across both area types. This is broadly consistent with species-specific patterns documented in Hyrcanian Oriental beech forests, where hornbeam showed higher microhabitat occurrence probabilities than beech (Khanalizadeh et al. 2020), underlining that broadleaved mixtures generally support higher microhabitat density than monospecific stands (Asbeck et al. 2021; Spînu et al. 2024).

Notably, despite being more abundant in recreation areas, both Fagus orientalis and Carpinus betulus exhibited significantly higher per-tree TreM abundance and EcoVal in conservation areas. This indicates that the same tree species can support markedly different microhabitat assemblages depending on the disturbance regime to which they are subjected, and that anthropogenic pressure—rather than differences in tree size or stand structural conditions—is the primary driver of microhabitat quality variation in Yedigöller. Species composition alone is therefore an insufficient proxy for microhabitat quality, and direct assessment of TreM ecological value is necessary to capture management-related differences in habitat provision (Courbaud et al. 2022). Broadleaved mixtures may provide the structural substrate for diverse microhabitat assemblages, but realizing that potential appears to require conditions of low disturbance intensity.

4.3 Dead Wood and Habitat Trees as Ecological Keystones Under Anthropogenic Pressure

Standing dead wood emerged as a disproportionately important microhabitat substrate despite its low overall occurrence in the study area. Snags supported substantially higher per-tree TreM abundance and EcoVal than living trees (Figure 9), consistent with the well-established role of dead wood as a driver of microhabitat diversity in temperate forests (Humphrey et al. 2005; Rondeux and Sanchez, 2010; Vuidot et al. 2011). This pattern reflects the progressive structural complexity that develops through decay: as wood deteriorates, it generates cavity types, fungal fruiting bodies, and saproxylic microhabitats that are largely absent from living trees (Siitonen 2012). The pronounced scarcity of snags in recreation areas (2 vs. 11 individuals in conservation areas) is likely attributable to two compounding anthropogenic factors: safety-related removal of potentially hazardous snags by park management near trails and facilities, and illegal harvesting of dead wood by visitors for campfires. The predominance of snags in conservation areas may partly reflect the presence of Abies nordmanniana subsp. bornmuelleriana, which occurred exclusively in those sites. Although aged living broadleaved trees generally host more TreMs in temperate forest ecosystems, dead conifer snags can contribute substantially to stand-level TreM richness through distinctive decay-related microhabitat types that differ in composition from those found on broadleaved species (Spînu 2023), potentially explaining the higher microhabitat values observed in conservation areas where Abies nordmanniana subsp. bornmuelleriana snags were present.

Beyond their ecological significance, dead wood structures at different decomposition stages may serve as tangible educational features that enhance visitor appreciation of natural forest processes (Rathmann et al. 2020; Sacher et al. 2022). Evidence suggests that when visitors are effectively informed about the ecological functions of dead wood, acceptance of its retention increases, potentially reducing illegal harvesting and supporting broader conservation goals (Gundersen and Frivold 2011). Integrating interpretive signage explaining the role of snags and decaying wood into visitor management strategies could therefore serve the dual purpose of habitat protection and environmental education—particularly relevant in Yedigöller where illegal firewood collection represents a direct and documented threat to this critical microhabitat substrate.

Habitat trees similarly supported microhabitats of substantially higher ecological value than smaller trees, with differences particularly pronounced in conservation areas (Figure 10). This finding is consistent with the well-documented positive relationship between tree diameter and microhabitat ecological value across European temperate forests, where large, old trees function as ecological keystones by accumulating rare, slow-forming structures over decades (Großmann and Pyttel 2019; Kozák et al. 2023; Spînu et al. 2024). The strong positive DBH-EcoVal relationship observed in our regression analysis (Figure 11B) confirms this pattern in Black Sea region forests, extending it to a system where such relationships had not previously been quantified. Critically, area type and species identity remained significant predictors of EcoVal even after accounting for DBH, indicating that the conservation-recreation contrast reflects genuine differences in microhabitat quality independent of tree size structure. This suggests that management regime shapes microhabitat quality through mechanisms beyond simply determining which tree sizes are present, likely including the preservation of structural complexity, decay processes, and long formation-time microhabitat types that are disrupted by safety interventions and visitor pressure (Martynova et al. 2021). The retention of large trees and promotion of natural aging are therefore essential management priorities, particularly given that safety-related interventions in recreation areas most commonly target the largest, oldest individuals—precisely those with the highest microhabitat value (Ministerium für Landwirtschaft, Umweltschutz und Raumordnung 2004; Mölder et al. 2020; Müller et al. 2020).

These findings collectively underscore that dead wood retention and habitat tree safeguarding are ecologically essential priorities in protected forests where anthropogenic pressure operates through both direct removal and indirect disturbance pathways. The management implications of these findings are developed further in the recommendations section below.

4.4 Management Interventions, Anthropogenic Pressure, and Microhabitat Development

Many TreM types develop over decades through complex processes, making it impossible to retrospectively attribute current microhabitat presence to specific historical management actions versus natural processes; the patterns reported here should therefore be interpreted as reflecting cumulative long-term effects of divergent management regimes rather than direct causal links.

Our exploratory analyses comparing specific TreM types between area types revealed complex patterns that require careful interpretation. While pruning is regularly conducted in recreation areas for visitor safety (park management, pers. comm.), no significant difference in cavity-type microhabitats was detected between area types. This reflects a fundamental methodological challenge in retrospective TreM studies: cavities develop slowly over 10-30+ years through progressive decay following branch loss, making it impossible to determine whether a current cavity originated from historical pruning, natural branch fall, storm damage, or other causes (Remm and Lõhmus, 2011). Our results therefore neither confirm nor deny pruning impacts on cavity formation; rather, they indicate that any such effects are not detectable above background variation from natural processes. This could mean: 

(1) pruning and natural branch fall create functionally equivalent structures at similar rates, 

(2) pruning intensity is insufficient to create detectable differences, or 

(3) current cavity distributions reflect historical conditions predating recent management regimes.

More direct evidence of anthropogenic impacts comes from fresh injury patterns and soil compaction effects. Recreation areas showed higher frequencies of recent trunk injuries (IN11, IN12)—though not statistically significant—and significantly higher frequencies of ground-contact cavities potentially linked to soil compaction (CV21). Combined with the substantially higher vandalism rates documented in section 3.5, these patterns suggest ongoing anthropogenic alteration of microhabitat composition, even when long-term cavity development shows no clear signal. Our study can document current TreM composition differences but cannot definitively attribute specific slow-forming TreMs to particular historical causes, as a limitation inherent to cross-sectional assessments of microhabitat systems with multi-decadal formation timescales.

Importantly, these shifts in microhabitat frequency are not necessarily associated with higher ecological value. Frequently recorded types in recreation areas, such as IN1 and GR1, are characterized by relatively low EcoVal scores, whereas less frequent but structurally complex microhabitats such as CV2 contribute disproportionately higher ecological value. This indicates that anthropogenic pressures may increase the occurrence of certain microhabitats while simultaneously reducing overall habitat quality.

Taken together, these results demonstrate that forestry practices and anthropogenic pressures do not simply reduce microhabitat availability but instead modify the processes and pathways through which TreMs are formed. Differences between conservation and recreation areas should therefore be interpreted as shifts in microhabitat composition and development dynamics rather than irreversible degradation—a distinction with important implications for restoration potential.  In Turkish national parks, where recreation is generally concentrated in a limited number of highly frequented sites—as in Yedigöller—adaptive management measures targeting the identified drivers of microhabitat quality loss may be particularly effective in helping recreation areas approximate the ecological functions observed in conservation zones. 

4.5 Vandalism: A Visitor-Induced Disturbance 

Our study documented vandalism as a novel anthropogenic TreM type concentrated almost exclusively in recreation areas, with 52 trees showing VA1 damage, 58 showing VA2 damage, and 42 exhibiting both types in recreation areas, compared to just 6 affected trees in conservation areas. The scale of this disturbance is ecologically significant: large-scale bark injuries of this kind may compromise the long-term microhabitat potential of affected trees by accelerating uncontrolled decay, facilitating pathogen entry, and interfering with the natural development of other TreM types. Previous research has highlighted how mechanical injuries caused by humans can disrupt tree health and reduce their habitat potential, and degrade the ecosystem services they provide (Cole and Marion 1988; Sánchez et al. 2021). Beyond its direct ecological effects, the prevalence of vandalism reflects insufficient visitor engagement with forest biodiversity values—a challenge that is not unique to Yedigöller but is particularly acute in national parks where recreational concentration creates sustained pressure on ecologically sensitive zones. Addressing vandalism therefore requires both regulatory enforcement and proactive visitor education strategies, which we discuss further in the management recommendations below.

4.6 Conservation Implications and Management Recommendations

Our findings have direct implications for conservation planning in Yedigöller National Park and comparable protected areas where recreation and biodiversity conservation functions coexist. While recreational activities need not be incompatible with conservation goals, our results demonstrate that their concentration in ecologically sensitive zones (e.g., areas around lakes) can substantially reduce microhabitat quality through multiple interacting pathways. Effective management therefore requires addressing both the structural drivers of microhabitat loss (dead wood removal, habitat tree loss, soil compaction) and the behavioral drivers (vandalism, illegal firewood collection) identified in this study.

Dead wood retention and habitat tree safeguarding should be treated as non-negotiable priorities in recreation area management. Where natural dead wood is scarce, supplementary measures such as selective girdling or on-site retention of material from necessary safety operations can help restore substrate availability (Çolak et al. 2011). Safety concerns regarding large or dead trees should be addressed through individual risk assessment rather than precautionary removal, as blanket removal disproportionately eliminates the highest-value microhabitat structures. Pruning practices should similarly be modified to retain cut material on site as habitat substrate where visitor safety permits.

Visitor education and engagement represent equally critical components of an effective management strategy. Evidence suggests that informed visitors show greater acceptance of dead wood retention and natural forest processes, reducing pressure on these critical substrates (Gundersen and Frivold 2011; Rathmann et al. 2020). More broadly, fostering meaningful engagement with forest biodiversity (e.g., interpretive infrastructure, guided experiences, and citizen science opportunities) has been shown to increase conservation awareness and participation among park visitors (Cosquer et al. 2012; Loyau and Schmeller 2017; McKinley et al. 2017). In Yedigöller, where vandalism and illegal firewood collection represent documented and ongoing threats, such engagement strategies could complement regulatory enforcement in addressing visitor-induced disturbance.

Periodic TreM monitoring should be integrated into park management planning to track microhabitat recovery in response to management interventions and to provide an evidence base for adaptive management decisions. Given that TreMs develop over decadal timescales, establishing baseline assessments and regular monitoring cycles is essential for detecting management-related change and informing long-term conservation strategies (Winter and Möller 2008). Our study provides such a baseline for Yedigöller, and the methodological framework that combines TreM abundance with EcoVal assessment offers a transferable approach for protected area monitoring across the broader Black Sea region and Eastern Mediterranean.

5 Conclusions

This study provides the first assessment of TreMs in a Turkish national park, demonstrating that conservation and recreation areas differ not in the overall quantity of microhabitats but in their ecological quality—a distinction that abundance-based metrics alone would fail to detect. Our findings confirm that anthropogenic pressure, operating through both direct disturbance and indirect management interventions, shapes microhabitat quality independently of tree species composition and size structure. Standing dead wood and habitat trees emerged as critical ecological keystones whose disproportionate contribution to microhabitat value is severely undermined by recreational pressure and safety-related management in Yedigöller. The novel documentation of vandalism as an anthropogenic TreM type further indicates the extent of visitor-induced impacts on forest microhabitat structure in heavily used protected areas.

Beyond their local significance, these findings contribute to the limited knowledge of TreM patterns in Black Sea region forests and demonstrate the applicability of EcoVal-weighted assessment as a sensitive tool for detecting management-related differences in habitat quality that raw TreM counts obscure. Future research should extend TreM assessments to species-level responses to examine whether the microhabitat quality differences documented here translate into detectable differences in the abundance and diversity of TreM-dependent species groups, including cavity-nesting birds, saproxylic invertebrates, and epiphytic cryptogams (Remacha et al. 2016; Asbeck et al. 2021). Long-term monitoring of TreM dynamics in both area types would further enable assessment of recovery trajectories under improved management and provide the temporal perspective necessary for understanding microhabitat formation and loss in these transitional forest ecosystems.

Acknowledgements

We are deeply grateful to many individuals who voluntarily joined our fieldwork, including Ali Görkem Ballı, Ayşenaz Çolak, Fatma Nur Erdem, İlayda Kökçü, Mehmet Ali Atay, Mehmet Ali Derse, Mehtap Fırtına, Merve Nur Turan, Serpil Ürkmez, Süreyya Yağmur Toparlak, Ufuk Recep Kılıç, and Ümmü Akçay. We would also like to acknowledge the General Directorate of Nature Conservation and National Parks (Türkiye) and 9th Regional Directorate for providing research permissions. Special thanks are due to Fatih Kaya, Chief of Abant Lake National Park, and İnci Hurmadaloğlu, Chief of Yedigöller National Park, for their invaluable technical assistance. We also thank Nur Bahar Usta for her support with statistical analyses. Lastly, we have been using ChatGPT for assistance with English editing. This study was financially supported by the Rufford Foundation (grant no. 39922-2) and carried out in collaboration with the Eastern Mediterranean Research Society. 

Data availability statement

The data underlying this article will be shared on reasonable request to the corresponding author.

References

Arcese, P., and A. R. E. Sinclair. 1997. „The role of protected areas as ecological baselines.“ Journal of Wildlife Management 61: 587-602. https://doi.org/10.2307/3802167

Asbeck, T., J. Großmann, Y. Paillet, N. Winiger, and J. Bauhus. 2021. „The use of tree-related microhabitats as forest biodiversity indicators and to guide integrated forest management.“ Current Forestry Reports 7 (1): 59-68. https://doi.org/10.1007/s40725-020-00132-5

Asbeck, T., C. Messier, and J. Bauhus. 2020. „Retention of tree-related microhabitats is more dependent on selection of habitat trees than their spatial distribution.“ European Journal of Forest Research 139 (6): 1015-1028. https://doi.org/10.1007/s10342-020-01303-6

Beşkardeş, V. 2012. „Investigations on avifauna of Yedigöller and Yeşilöz Wildlife Reserves.“ Turkish Journal of Forestry 13: 28-34. https://doi.org/10.18182/tjf.84109

Bianco, G., Hemp, A., Schleuning, M., 2024. “Variation in the diversity and composition of tree-related microhabitats across climate and human impact gradients on a tropical mountain.” Ecological Indicators 169, 112927. https://doi.org/10.1016/j.ecolind.2024.112927

Bütler, R., T. Lachat, L. Larrieu, and Y. Paillet. 2013. „Habitat trees: Key elements for forest biodiversity.“ Joensuu: European Forest Institute.

Bütler, R., T. Lachat, F. Krumm, D. Kraus, and L. Larrieu. 2020. Field guide to tree-related microhabitats: Descriptions and size limits for their inventory. Birmensdorf: Swiss Federal Institute for Forest, Snow and Landscape Research WSL.

Bütler, R., Lachat, T., Krumm, F., Kraus, D., Larrieu, L., 2024. Field guide to tree-related microhabitats. Descriptions and size limits for their inventory in temperate and Mediterranean forests. (2nd ed.). Birmensdorf: Swiss Federal Institute for Forest, Snow and Landscape Research WSL.

Cole, D.N., Marion, J.L., 1988. “Recreation impacts in some riparian forests of the Eastern United States.” Environmental Management 12, 99–107. https://doi.org/10.1007/BF01867381

Cosquer, A., R. Raymond, and A.-C. Prevot-Julliard. 2012. “Observations of Everyday Biodiversity: A New Perspective for Conservation?” Ecology and Society 17 (4). https://www.jstor.org/stable/26269197.

Courbaud, B,, Larrieu, L., Kozak, D. et al. 2022. “Factors Influencing the Rate of Formation of Tree-Related Microhabitats and Implications for Biodiversity Conservation and Forest Management.” Journal of Applied Ecology 59 (2): 492–503. https://doi.org/10.1111/1365-2664.14068.

Çolak, A. H., M. Tokcan, and S. Kırca. 2011. Ölü Ağaç – Yaşayan Ölüler. Batı Karadeniz Ormancılık Araştırma Müdürlüğü-Bolu, Çevre ve Orman Bakanlığı.

Dutta, T., Larrieu, L., and Schuck, A. 2025. “Who Is Using Tree-Related Microhabitats (TreMs)?” Biological Conservation 307 (July): 111180. https://doi.org/10.1016/j.biocon.2025.111180.

Fröhlich, A., Bidziński, K., Jankowska-Jarek, M., Swearer, S., Ciach, M., 2025. Noise, light pollution, and human presence predict forest habitat degradation: A key agent in biodiversity decline. Ecological Applications 35, e70131. https://doi.org/10.1002/eap.70131

Gaston, K. J., S. F. Jackson, L. Cantú-Salazar, and G. Cruz-Piñón. 2008. „The ecological performance of protected areas.“ Annual Review of Ecology, Evolution, and Systematics 39: 93-113. https://doi.org/10.1146/annurev.ecolsys.39.110707.173529

Grossmann, J., J. Schultze, J. Bauhus, and P. Pyttel. 2018. „Predictors of microhabitat frequency and diversity in mixed mountain forests in South-Western Germany.“ Forests 9 (3): 104. https://doi.org/10.3390/f9030104

Großmann, J., and P. Pyttel. 2019. “Mikrohabitate Und Baumdimension Als Grundlage Der Habitatbaum-Auswahl Im Bergmischwald.” Natur Und Landschaft 12 (December). https://doi.org/10.17433/12.2019.50153759.531-541.

Gundersen, V. and L. H. Frivold. 2011. “Naturally Dead and Downed Wood in Norwegian Boreal Forests: Public Preferences and the Effect of Information.” Scandinavian Journal of Forest Research 26 (2): 110–19. https://doi.org/10.1080/02827581.2010.536567.

Hebert, J., P. Lejeune, J. F. van Belle, and Y. Schenkel. 1996. „Indice du martelage sur l‘exploitation mecanisee des premieres eclaircies resineuses.“ Silva Belgica 103: 9-14.

Hough, J. L. 1988. „Obstacles to effective management of conflicts between national parks and surrounding human communities in developing countries.“ Environmental Conservation 15: 129-136. https://doi.org/10.1017/S0376892900028939

Humphrey, J. W. 2005. „Benefits to biodiversity from developing old-growth conditions in British upland spruce plantations: A review and recommendations.“ Forestry 78: 33-53. https://doi.org/10.1093/forestry/cpi004

Humphrey, J., A.-L. Sippola, G. Lempériere, B. Dodelin, K.N.A. Alexander, and J. Butler. 2005. “Deadwood as an Indicator of Biodiversity in European Forests: From Theory to Operational Guidance.” In Monitoring and Indicators of Forest Biodiversityin Europe – From Ideas to Operationality, edited by M. Marchetti, 193–206. Deadwood as an Indicator of Biodiversity in European Forests:From Theory to Operational Guidance 51. Finland: European Forest Institute. https://jukuri.luke.fi/handle/10024/512187.

Khanalizadeh, A., Eshaghi Rad, J., Zahedi Amiri, G., Zare, H. Werner Rammer, and Manfred J. Lexer. 2020. “Assessing Selected Microhabitat Types on Living Trees in Oriental Beech (Fagus Orientalis L.) Dominated Forests in Iran.” Annals of Forest Science 77 (3): 1–13. https://doi.org/10.1007/s13595-020-00996-4

Kırca, S. 2015. „Determination of Nature Conservation and Restoration Priority Areas in Landscape Planning: Example of Yedigöller National Park (Bolu).“ İstanbul Üniversitesi / Fen Bilimleri Enstitüsü / Peyzaj Mimarlığı Ana Bilim Dalı, Istanbul.

Kırca, S., and H. Altınçekiç. 2018. „Use of Ecological Value Analysis for Prioritizing Areas for Nature Conservation and Restoration.“ Journal of the Faculty of Forestry Istanbul University 68: 22–35.

Kozák, D., M. Svitok, V. Zemlerová, M. Mikoláš, T. Lachat, L. Larrieu, Y. Paillet, et al. 2023. “Importance of Conserving Large and Old Trees to Continuity of Tree-Related Microhabitats.” Conservation Biology 37 (3): e14066. https://doi.org/10.1111/cobi.14066.

Kraus, D., R. Bütler, F. Krumm, T. Lachat, L. Larrieu, U. Mergner, Y. Paillet, T. Rydkvist, A. Schuck, and S. Winter. 2016. Catalogue of Tree Microhabitats – Reference Field List. Integrate+ Technical Paper. 16p.

Kraus, D., and F. Krumm. 2013. Integrative Approaches as an Opportunity for the Conservation of Forest Biodiversity. European Forest Institute. http://www.integrateplus.org.

Kraus, D., A. Schuck, F. Krumm, R. Bütler Sauvain, H. Cosyns, B. Courbaud, and G. Wilhelm. 2018. Seeing Is Building Better Understanding - The Integrate+ Marteloscopes. Integrate+ Technical Paper, European Forest Institute. https://informar.eu/marteloscope-sites.

Kraus, D., Schuck, A., Bebi, P., Blaschke, M., Bütler, R., Flade, M., Heintz, W., Krumm, F., Lachat, T., Larrieu, L., Lehnerova, L., Levin, M., Mergner, U., Pach, M., Paillet, Y., Pyttel, P., Rydkvist, T., Santopuoli, G., Sever, K., Sturm, K., Vandekerkhove, K., Winter, S., Witz, M., Winnock, M., Marcandella, M., Roth, I., Jakob, A., Dickele, R., Hofmann, G., Ruis-Eckhardt, D., Boschen, S., Schölmerich, U., Leder, B., Guericke, M., Merkel, H., Löffler, D., Schusser, S., Runkel, M., Steinmetz, A., Marx, K.H., Mongelluzzi, A., Wilshusen, F., Jean-Jacques Boutteaux, Loic Duchamp, Dericbourg, N., Rouyer, E., Csikos, V., Sweeny, K., Steichen, D., Leytem, M., Konczal, S., Stereńczak, K., Kazimirovic, M., Šebeň, V., Zapater, T.B., Van Der Maaten-Theunissen, M., Pommer, A., Nord-Larsen, T., Fuhr, M., Luc-Olivier Delebeque, Lidón Martínez Navarro, Lasala, D., S.M. Waez-Mousavi, Kiomars Sefidi, Abellanas, B., Heintz, W., 2022. Spatially explicit database of tree related microhabitats (TreMs). https://doi.org/10.15468/OCOF3V

Larrieu, L., A. Cabanettes, A. Brin, et al. 2014. „Tree Microhabitats at the Stand Scale in Montane Beech–Fir Forests: Practical Information for Taxa Conservation in Forestry.“ European Journal of Forest Research 133: 355–367. https://doi.org/10.1007/s10342-013-0767-1.

Larrieu, L., and A. Cabanettes. 2012. „Species, Live Status, and Diameter Are Important Tree Features for Diversity and Abundance of Tree Microhabitats in Subnatural Montane Beech–Fir Forests.“ Canadian Journal of Forest Research 42: 1433–1445. https://doi.org/10.1139/x2012-077.

Larrieu, L., Y. Paillet, S. Winter, R. Butler, D. Kraus, F. Krumm, T. Lachat, A. K. Michel, B. Regnery, and K. Vandekerkhove. 2018. „Tree Related Microhabitats in Temperate and Mediterranean European Forests: A Hierarchical Typology for Inventory Standardization.“ Ecological Indicators 84: 194–207. https://doi.org/10.1016/j.ecolind.2017.08.051.

Leclerc, D., and M. Millot. 1998. „Techniques pour une Gestion Multifonctionnelle Durable des Forêts de Montagne: Cas des Alpes en France et en Italie.“ Revue Forestière Française 50: 67–81. https://doi.org/10.4267/2042/5583

Lenth, R. 2024. emmeans: Estimated Marginal Means, aka Least-Squares Means. R package version 1.10.5, https://github.com/rvlenth/emmeans

Loyau, A., and D. S. Schmeller. 2017. “Positive Sentiment and Knowledge Increase Tolerance towards Conservation Actions.” Biodiversity and Conservation 26 (2): 461–78. https://doi.org/10.1007/s10531-016-1253-0.

Martynova, M., R. Sultanova, D. Khanov, E. Talipov, and R. Sazgutdinova. 2021. “Forest Management Based on the Principles of Multifunctional Forest Use.” Journal of Sustainable Forestry 40 (1): 32–46. https://doi.org/10.1080/10549811.2020.1734025.

Marziliano, P. A., Antonucci, S., Tognetti, R. et al. 2021. “Factors Affecting the Quantity and Type of Tree-Related Microhabitats in Mediterranean Mountain Forests of High Nature Value.” Forest Ecology. iForest - Biogeosciences and Forestry 14 (3): 250. world. https://doi.org/10.3832/ifor3568-014.

McKinley, D. C., A. J. Miller-Rushing, H. L. Ballard, R. Bonney, H. Brown, S. C. Cook-Patton, D. M. Evans, et al. 2017. “Citizen Science Can Improve Conservation Science, Natural Resource Management, and Environmental Protection.” Biological Conservation, The role of citizen science in biological conservation, 208 (April): 15–28. https://doi.org/10.1016/j.biocon.2016.05.015.

McMullin, R. T., and Y. F. Wiersma. 2019. “Out with OLD Growth, in with Ecological continNEWity: New Perspectives on Forest Conservation.” Frontiers in Ecology and the Environment 17 (3): 176–81. https://doi.org/10.1002/fee.2016.

Michel, A. K., and S. Winter. 2009. „Tree Microhabitat Structures as Indicators of Biodiversity in Douglas-Fir Forests of Different Stand Ages and Management Histories in the Pacific Northwest, U.S.A.“ Forest Ecology and Management 257: 1453–1464. https://doi.org/10.1016/j.foreco.2008.11.027.

Ministerium für Landwirtschaft, Umweltschutz und Raumordnung. 2004. Waldbau-Richtlinie „Grüner Ordner“ der Landesforstverwaltung Brandenburg. Ministerium für Landwirtschaft, Umweltschutz und Raumordnung (MLUR) des Landes Brandenburg, p. 143.

Mölder, A., M. Schmidt, T. Plieninger, and P. Meyer. 2020. „Habitat-Tree Protection Concepts Over 200 Years.“ Conservation Biology. https://doi.org/10.1111/cobi.13511.

Müller, J., J. Bauhus, M. Dieter, H. Spellmann, B. Möhring, S. Wagner, H. Wolf, et al. 2020. “Wege zu einem effizienten Waldnaturschutz in Deutschland: Stellungnahme Januar 2020.” Berichte über Landwirtschaft – Zeitschrift für Agrarpolitik und Landwirtschaft, January. https://doi.org/10.12767/buel.vi228.284.

Nelson, J. G., and R. Serafin. 1997. „Keys to Life.“ In National Parks and Protected Areas, edited by J. G. Nelson and R. Serafin, 1–7. Berlin and Heidelberg: Springer. https://doi.org/10.1007/978-3-642-60907-7_1.

Ní Dhubháin, Á. 2010. An Evaluation of Continuous Cover Forestry in Ireland. COFORD, Department of Agriculture, Fisheries and Food, Dublin.

Niedermann-Meier, S., M. Mordini, R. Bütler, and P. Rotach. 2010. „Habitatbäume im Wirtschaftswald: Ökologisches Potenzial und finanzielle Folgen für den Betrieb.“ Schweizerische Zeitschrift für Forstwesen 161 (10): 391–400. https://doi.org/10.3188/szf.2010.0391.

Oksanen, J., F. G. Blanchet, R. Kindt, P. Legendre, P. R. Minchin, R. B. O‘Hara, G. L. Simpson, P. Solymos, M. H. H. Stevens, H. Wagner, et al. 2019. vegan: Community Ecology Package. R Package Version 2.5-6. https://CRAN.R-project.org/package=vegan.

Paillet, Y., Archaux, F., Puy, S. du, Bouget, C., Boulanger, V., Debaive, N., Gilg, O., Gosselin, F., Guilbert, E., 2018. “The indicator side of tree microhabitats: A multi-taxon approach based on bats, birds and saproxylic beetles.” Journal Applied Ecology 55, 2147–2159. https://doi.org/https://doi.org/10.1111/1365-2664.13181

Paillet, Y., N. Debaive, F. Archaux, E. Cateau, O. Gilg, and E. Guilbert. 2019. „Nothing Else Matters? Tree Diameter and Living Status Have More Effects than Biogeoclimatic Context on Microhabitat Number and Occurrence: An Analysis in French Forest Reserves.“ PLOS ONE 14 (5): e0216500. https://doi.org/10.1371/journal.pone.0216500.

Przepióra, F., Lewandowski, P., Ciach, M., 2025. Spatial distribution of tree-related microhabitats in a primeval mountain forest: From natural patterns to landscape planning and forest management recommendations. Science of the Total Environment 960, 178319. https://doi.org/10.1016/j.scitotenv.2024.178319

R Core Team 2020. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/

Rathmann, J., P. Sacher, N. Volkmann, and M. Mayer. 2020. “Using the Visitor-Employed Photography Method to Analyse Deadwood Perceptions of Forest Visitors: A Case Study from Bavarian Forest National Park, Germany.” European Journal of Forest Research 139 (3): 431–42. https://doi.org/10.1007/s10342-020-01260-0.

Regnery, B., Y. Paillet, D. Couvet, and C. Kerbiriou. 2013. „Which Factors Influence the Occurrence and Density of Tree Microhabitats in Mediterranean Oak Forests?“ Forest Ecology and Management 295: 118–25. https://doi.org/10.1016/j.foreco.2013.01.009

Remm, J., Lõhmus, A., 2011. Tree cavities in forests – The broad distribution pattern of a keystone structure for biodiversity. Forest Ecology and Management 262, 579–585. https://doi.org/10.1016/j.foreco.2011.04.028

Remacha, C., J. A. Delgado, M. Bulaic, and J. Pérez-Tris. 2016. “Human Disturbance during Early Life Impairs Nestling Growth in Birds Inhabiting a Nature Recreation Area.” PLOS ONE 11 (11): e0166748. https://doi.org/10.1371/journal.pone.0166748.

Rondeux, J., and C. Sanchez. 2010. „Review of Indicators and Field Methods for Monitoring Biodiversity within National Forest Inventories. Core Variable: Deadwood.“ Environmental Monitoring and Assessment 164: 617–30. https://doi.org/10.1007/s10661-009-0917-6.

Sacher, P., J. Meyerhoff, and M. Mayer. 2022. “Evidence of the Association between Deadwood and Forest Recreational Site Choices.” Forest Policy and Economics 135 (February): 102638. https://doi.org/10.1016/j.forpol.2021.102638.

Sánchez, J.J., Marcos-Martinez, R., Srivastava, L., Soonsawad, N., 2021. Valuing the impacts of forest disturbances on ecosystem services: An examination of recreation and climate regulation services in U.S. national forests. Trees For. People 5, 100123. https://doi.org/10.1016/j.tfp.2021.100123

Schuck, A., D. Kraus, F. Krumm, and H. Schmitt. 2015a. Integrate+ Marteloscopes Calibrating Silvicultural Decision Making. Integrate+ Technical Paper No1. 12p.

Schuck, A., F. Krumm, and D. Kraus. 2015b. Integrate+ Marteloscopes - Description of Parameters and Assessment Procedures. Integrate+ Technical Paper No. 18. 16 p.

Siitonen, J., 2012. “Microhabitats” In Biodiversity in Dead Wood, Ecology, Biodiversity and Conservation, edited by Jogeir N. Stokland, Juha Siitonen, and Bengt Gunnar Jonsson, 150–182. Cambridge, Cambridge University Press. https://doi.org/10.1017/CBO9781139025843.008

Spînu, A.P., 2023. “Quality and temporal deveopment of habitat trees and their microhabitats in temperate European forests”. University of Freiburg, Freiburg im Breisgau, Germany.

Spînu, A.P., Nicolaie, M.A., Asbeck, T., Kozak, D., Paillet, Y., Cateau, E., Mikoláš, M., Svoboda, M., Bauhus, J., 2024. “Temporal Development of Microhabitats on Living Habitat Trees in Temperate European Forests”. Ecosystems 27, 690–709. https://doi.org/10.1007/s10021-024-00915-y

Thormann, J.-J., K. Allenspach-Schliessbach, H. Bugmann, M. Frehner, P. Junod, C. Rosset, and K. Kühne. 2019. „Bedeutung von Marteloskopen für Praxis und Lehre in der Schweiz.“ Schweizerische Zeitschrift für Forstwesen 170: 60–68. https://doi.org/10.3188/szf.2019.0060.

Tokcan, M. 2015. Yedigöller Milli Parkı ile Sarımustan Bölgesi’nin Hemerobi Dereceleri ve Bitki Toplumlarının Belirlenmesi, Doktora Tezi, İ.Ü. Fen Bilimleri Enstitüsü, Orman Botaniği Anabilim Dalı, 343 s.

Visick, O.D., Adams, I., Marzano, F.S., Ratnieks, F.L.W., 2025. Effect of bark harvest intensity on the formation of cavities and other microhabitats in cork oaks (Quercus suber). European Journal of Forest Research 144, 467–479. https://doi.org/10.1007/s10342-025-01770-9

Vuidot, A., Y. Paillet, F. Archaux, and F. Gosselin. 2011. „Influence of Tree Characteristics and Forest Management on Tree Microhabitats.“ Biological Conservation 144 (1): 441–50. https://doi.org/10.1016/j.biocon.2010.09.030

Wickham H. 2016. ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York. ISBN 978-3-319-24277-4, https://ggplot2.tidyverse.org.

Winter, S., and G. C. Möller. 2008. „Microhabitats in Lowland Beech Forests as Monitoring Tools for Nature Conservation.“ Forest Ecology and Management 255: 1251–61. https://doi.org/10.1016/j.foreco.2007.10.029

Supplementary Material