Over the past few decades, wildfire activity has increased globally. In the Central Zagros Mountains of Iran, widespread oak decline has significantly altered fuel structures and raised ecological concerns. However, the specific contribution of long-term vegetation degradation to wildfire burn severity remains under-quantified. This study addresses this gap by investigating whether decadal declines in the Normalized Difference Vegetation Index (NDVI) are associated with higher burn severity, while assessing their relative importance against short-term pre-fire environmental conditions, fuel types, topography, and human access gradients. Focusing on a major wildfire in June 2024 in the Kashkan Watershed, we mapped burn severity using the differenced Normalized Burn Ratio (dNBR) derived from Sentinel-2 imagery. A comprehensive suite of predictors was evaluated, including the long-term NDVI trend (May- September 2014-2023), pre-fire Normalized Difference Moisture Index (NDMI), Land Surface Temperature (LST), fuel categories based on ESA WorldCover, slope, aspect, spatial texture contrast, and distance to roads and settlements. A stratified sampling design supported statistical inference using ordinary least squares regression and predictive modeling via Random Forest and XGBoost algorithms with five-fold cross-validation. XGBoost demonstrated the highest predictive performance (cross-validated R2 = 0.70; RMSE = 0.11), outperforming Random Forest (R2 = 0.64) and linear regression (R2 = 0.34). Fuel moisture (NDMI) and thermal stress (LST) emerged as the primary drivers of severity, followed by land cover, topography, and spatial texture. While the long-term NDVI decline showed a small but consistent positive association with dNBR at the landscape scale, the results highlight a distinct multi-scale mechanism. In this dynamic, pre-fire moisture and temperature establish the broad watershed-scale environmental susceptibility, whereas specific fuel types and their spatial configurations drive the local-level amplification or damping of burn severity. These findings provide a framework for proactive vulnerability mapping in Zagros oak woodlands, suggesting that integrating dynamic fire-weather data and field-based fuel measurements could further refine future predictability.
Accurate volume measurement of stacked timber logs is essential for the timber industry. Photo-optical Automated Volume Measuring Systems (AVMS) offer efficient alternatives to manual methods but require verification to ensure reliability. This paper presents a simple, reliable, and easily implementable verification method for AVMS, based on measuring the wood stack front surface area. The proposed method utilizes reference models, calibrated using traceable working standards, to simulate timber log stacks with defined contours. This involves comparing the AVM’s measured area to the reference model’s area, enabling the determination of area measurement errors. The study investigates the selection of reference instruments, proposes a comprehensive AVMS verification procedure, and evaluates the method’s metrological characteristics. Laboratory tests involving 110 AVMS units revealed a repeatability of 0.46% and a total uncertainty of approximately 1.1%. The method successfully identified shortcomings in commercially available AVMS, including non-compliance with manufacturer-declared errors and design imperfections related to camera limitations. The developed method can be used in industry, metrological supervision, and scientific research to ensure accurate and reliable measurements of wood stacks and the long-term reliability of AVMS. This study also highlights the need to further improve photo-optical measurement systems to increase their reliability and accuracy.
Growing concerns about climate change and biodiversity loss have increased the importance of research on vegetation ecology and ecosystem functioning for understanding forests and savannas. Functional diversity is a field of study that has expanded in recent years, aiming to explain ecosystem functions through organisms’ functional characteristics. The Cerrado, the second-largest biome in South America, is regarded as the world’s most biodiverse savanna and plays a crucial role in biomass accumulation. In this context, the present study aimed to investigate the role of functional traits in the accumulation of woody aboveground biomass (AGB) in the Cerrado by analyzing the relationship between Community Weighted Means (CWMs) of functional traits and functional diversity metrics across three distinct physiognomies (Cerrado Típico, Cerrado Denso, and Cerradão). We hypothesized that the relationships between biomass and functional diversity metrics would differ among physiognomies. To assess these relationships, we tested correlations between biomass and both CWMs and functional diversity metrics. We measured five traits for dominant species selected according to cumulative abundance criteria in each physiognomy: specific leaf area (SLA), wood density (WD), crown area (CA), maximum diameter (Dmax), and maximum height (Htmax). These traits were used to calculate the respective CWMs, along with five functional diversity metrics: Functional Richness (Fric), Functional Evenness (FEve), Functional Divergence (FDiv), Functional Dispersion (FDis), and Rao’s Quadratic Entropy (RaoQ). We used forest inventory data from the study areas to fit linear models for each physiognomy and mixed-effects models for the complete dataset, with physiognomic type as a random effect. The physiognomies differed in structural and taxonomic characteristics, displaying growth patterns proportional to vegetation stature. Functional differences were also observed: biomass in Cerradão was more strongly associated with functional traits, whereas biomass in Cerrado Denso and Cerrado Típico was more strongly associated with functional diversity metrics. The CWMs Dmax, WD, and CA, along with the diversity metrics Fric and FEve, were the predictors most frequently retained in the best-performing models. General models that included both CWMs and functional diversity metrics, or only CWM Dmax, outperformed models that included only Fric or excluded random effects.
Regeneration dynamics in forest ecosystems depend on seedlings’ ability to persist in shaded understories and, following canopy opening, to rapidly adjust physiologically to new, contrasting light and soil moisture conditions. In this study, we examined the real-time efficiency of the photosynthetic apparatus and the short- to intermediate-term acclimatory response of photosynthetic pigments on Nothofagus pumilio seedlings exposed to different light and soil moisture levels. We evaluated phenology, chlorophyll fluorescence and pigment content in leaves, on 2-3 years-old seedlings in a controlled greenhouse experiment with three levels of light intensity (low = 4%, medium = 26%, high = 64% of the natural incident irradiance) and two levels of soil moisture (optimal = 40-60%, excess = 80-100% of soil field capacity). We measured six samples per treatment (light intensity × soil moisture levels) monthly during one growing season (n = 216). Data were analyzed by multiple ANOVAs. Although maximum quantum yield remained stable (> 0.80) across treatments, seedlings exhibited differential eco-physiological responses primarily driven by light availability. Structural investment was synergistic and negatively affected by the light × soil moisture interaction, resulting in reduced leaf production, chlorophyll, and carotenoids during peak season under the high-light and excess-soil-moisture treatment. Seedlings under medium light showed optimal performance, with maximum pigment accumulation and high PSII efficiency without signs of chronic photoinhibition. N. pumilio seedlings decouple short-term functional photochemical efficiency from their structural investment strategy (pigment content), a mechanism that ensures survival under variable canopy cover but results in reduced growth under sub-optimal conditions.
Digitalization is increasingly recognized as a major driver of innovation in the forestry sector. However, the adoption of digital technologies remains uneven, particularly in structurally fragmented forestry systems such as those found in Southern Europe. This study aims to analyze and discuss stakeholders’ expectations, perceived barriers, and opportunities related to the adoption of digital technologies in Italian operational forestry, based on a participatory process that included a webinar, an online questionnaire, and a focus group, engaging a broad range of forestry stakeholders. Because the survey relied on a voluntary, non-probability sample, the findings should be interpreted as reflecting the perceptions and priorities of the surveyed stakeholders rather than as statistically representative estimates of the entire Italian forestry sector. The results reveal a growing interest in digital technologies across all operational domains, particularly in GIS-based decision support systems, remote sensing, product traceability, and precision harvesting. Nevertheless, stakeholders identified several persistent barriers, including high investment costs, limited interoperability among systems, insufficient digital skills, and weak connectivity in rural and mountainous areas. Training emerged as the most critical enabling factor, often considered even more important than direct investment in technologies. Overall, the evidence suggests that digitalization cannot advance without systematic investment in human capital, interoperability, and shared information infrastructures. Tools such as knowledge hubs, communities of practice, and targeted training programs may play a key role in bridging the gap between technological availability and operational implementation. Although rooted in the Italian context, the insights provided by this study offer relevant indications for digitalization policies and practices across Southern European countries and in forestry systems characterized by fragmented ownership structures and heterogeneous operational conditions.
This study aimed to assess how the invasion of the shrub Sorbaria sorbifolia affects the soil seed bank in the urban forests of Ekaterinburg (Middle Urals, Russia). We compared species composition and abundance in the above-ground vegetation and seed banks across three forest vegetation types: those dominated by the invasive shrub S. sorbifolia, those dominated by the native shrub Rubus idaeus, and those without either species. In the presence of S. sorbifolia, the total cover of plants in the herb-dwarf shrub layer decreased by two orders of magnitude, the species richness decreased by 4-6 times, and the number of above-ground vegetation species decreased by 3-4 times. The species composition of the seed banks differed qualitatively from that of the above-ground vegetation and showed little relationship with the different vegetation types. While the species composition and richness of the seed bank remained unaltered by the S. sorbifolia invasion, its size was significantly decreased by 2.0-2.4 times. The ability to negatively affect the abundance of soil seed banks demonstrates that S. sorbifolia can be reasonably categorized as a transformer species. Therefore, this invasive plant requires special attention, as its further dispersal could have major consequences for native vegetation.
Native mycorrhizal fungal species can colonize the roots of tree seedlings and positively influence their growth and survival. This study evaluates the effects of three native ectomycorrhizal fungi on the growth and root colonization of black pine (Pinus nigra) seedlings from three distinct origins. For this purpose, the experiment included (i) the production of pure cultures of selected local mycorrhizal fungal species in the laboratory environment and (ii) the production of mycorrhizal black pine seedlings carried out by inoculating plants with these cultures. Pure cultures of three fungal species (Rhizopogon luteolus, Russula raoultii and Chroogomphus rutilus) were obtained from eastern Mediterranean black pine forests of Turkey and mixed into sterilized growth substrate (a mixture of humus, river sand and forest soil) for inoculation. Two other treatments, with no mycorrhizal inoculations, were also added to the experiments and considered as sterile and non-sterile control treatments. Twenty-two months after inoculation, seed origin significantly affected seedling growth parameters and survival rates. Among the tested provenances, Egirdir consistently exhibited the lowest growth performance. Mycorrhizal inoculation did not significantly influence shoot growth; however, it markedly enhanced survival rates, root length, and the extent of mycorrhizal root colonization. The mean mycorrhizal infection rate was 1.6% in the sterile control and 27% in the non-sterile control. In contrast, inoculation with pure cultures increased colonization to 61%, 64%, and 69% for R. raoultii, C. rutilus, and R. luteolus, respectively. Survival rates showed a significant interaction between origin and mycorrhizal inoculation. R. luteolus consistently resulted in the highest survival rates across all treatments, with the most pronounced effect observed in the Egirdir provenance, where survival increased from 63% to 81% following inoculation. The improved seedling survival of the Egirdir origin, which had the smallest seeds and weakest initial growth, suggests that the benefits of mycorrhizal symbiosis may be especially important for weaker seedlings and under unfavorable environmental conditions.
The restoration of a six-year-old, seasonally flooded riparian forest planted with woody species native to the Brazilian Cerrado biome within a buffer zone system was evaluated based on soil carbon and nitrogen cycling parameters and soil microbial communities. Soil organic carbon, humic and fulvic acids, phospholipid fatty acid profiles of total bacteria and fungi, and selected functional microbial groups, such as aerobic/anaerobic and denitrifying communities, were used as indicators of soil quality and fertility. Restoration was progressing towards the soil conditions of a preserved riparian forest (PS) compared to a disturbed site (DS). However, a higher soil ammonium concentration was observed at the experimental site (ES) and DS than at PS, which was attributed to impaired nitrification due to periodic flooding. Principal component analysis confirmed that soil organic matter, cation exchange capacity, NO3-, and the denitrifying, anaerobic/aerobic, actinomycetes, and Gram-negative groups are suitable indicators for riparian site restoration. In contrast, soil NH4+ accumulation emerged as the main environmental impact hindering riparian forest recovery.
Tree-level models are commonly used to estimate the mean and total forest aboveground biomass (AGB). However, tree biomass predictions can vary significantly between models due to limited understanding of how calibration datasets influence their performance. The aim was to recommend a sample size for a dataset to calibrate tree biomass models that estimate mean AGB per unit area with the best possible precision and accuracy. The methodology consisted of (i) simulating an Amazonian forest using the Monte Carlo Method (MCM) and m out of n Bootstrap, (ii) estimating the average forest biomass using models calibrated with datasets of varying sample sizes, (iii) analyzing the precision and accuracy of the estimate of the mean AGB in Mg ha-1 per unit area using the models calibrated with different sample sizes, and (iv) relating characteristics of the calibration datasets with the errors produced by the models. Our findings indicate that increasing sample size improves the precision of biomass estimates, with notable gains at 200 trees. The study highlights that larger samples better represent tree diversity and support reliable biomass modeling, which is essential for effective forest management. We conclude that a minimum calibration sample size of 200 trees is sufficient to optimize biomass predictions while balancing practical sampling constraints. These results can aid in developing forest management strategies and improving carbon credit calculations in tropical ecosystems.
Soil erosion is a critical factor contributing to land degradation, and its dynamics in urban environments is influenced by a combination of anthropogenic activities and climate change. This study evaluates the impact of climate change on soil erosion in Belgrade’s urban landscape using the Revised Universal Soil Loss Equation (RUSLE) model and climate projections from the EURO-CORDEX dataset. The analysis was conducted for the reference period (2001-2019) and projected for three future time frames (2016-2035, 2046-2065, 2081-2100) under two greenhouse gas emission scenarios (RCP4.5 and RCP8.5). The results indicate that the intensity of soil erosion will vary depending on the climate scenario. In the reference year 2019, the average annual soil loss was 4.15 t ha-1 y-1. In future periods, soil erosion is expected to increase under the RCP4.5 scenario, with values ranging from 4.2 to 4.25 t ha-1 y-1. In contrast, the RCP8.5 scenario shows more pronounced fluctuations, ranging from 3.71 to 4.39 t ha-1 y-1. The spatial analysis shows that the most vulnerable areas are those with steep slopes and degraded vegetation cover, while the expansion of impervious surfaces in urban areas further exacerbates the risks of surface runoff and erosion. These findings underscore the urgent need for tailored urban planning strategies and sustainable land management practices in urban areas. The preservation of vegetation cover, the implementation of soil protection measures, and the integration of climate projections into urban development planning can substantially mitigate the adverse effects of erosion. This study provides valuable insights that can support more effective decision-making in land conservation and urban planning in the context of climate change.
The intensification of wind disturbances increases the importance of maintaining tree mechanical stability in Northern European forests. Soil preparation during forest regeneration may influence long-term stability by modifying rooting conditions. We assessed the effects of site conditions, defined as combinations of soil type (e.g., freely draining mineral, waterlogged mineral, and drained peat soils) and site-specific soil preparation methods (disc trenching and mounding), on the mechanical stability of mid-aged Scots pine (Pinus sylvestris L.) stands in hemiboreal forests of northeastern Europe (Latvia). A static tree-pulling test was used to quantify resistance to primary and secondary failure, as well as the post-primary-failure stability margin. Trees growing on unprepared freely draining mineral soils exhibited the highest resistance to both primary and secondary failure, whereas trees on waterlogged mineral and peat soils showed reduced stability and a predominance of uprooting. Mounding on waterlogged sites did not consistently enhance long-term mechanical stability. Overall, soil physical properties and their interaction with tree size outweighed the long-term effects of soil preparation alone, highlighting the need for site-specific, climate-smart forest management strategies to enhance wind resistance in hemiboreal forests.
Remote Sensing (RS) and Geographic Information Systems (GIS) technologies are valuable tools for managing forest ecosystems today. This study employed a Binary Logistic Regression Model to analyze the environmental factors influencing the natural regeneration of Cedrus libani A. Rich. within its native habitat in the Jawbat Burghal forest of Syria. The effects of various environmental factors (spatial and forest structure, topography, and spectral indicators) were assessed both individually and collectively to understand their impact on the probability of C. libani seedling presence in Jawbat Burghal. The model’s accuracy significantly improved when considering the combined effects of these factors, reaching 94.3%, exceeding the accuracy of the other models evaluated (none of which surpassed 85%). Higher stone density, increased broadleaved tree cover, and favorable summer soil moisture were found to promote the natural regeneration of C. libani in Jawbat Burghal.
The carbon market for the forestry sector is internationally recognized as a policy tool to reduce greenhouse gas emissions, to support afforestation and to improve forest management activities, which would not be economically viable without the sale of carbon credits. The international market is regulated by standards based on Guidelines for national greenhouse gas inventories, and in many countries, domestic markets are emerging, managed by governments whose credits can be used to meet national climate targets under the UNFCCC. In Italy, the market is currently unregulated, despite the approval of a law to establish a national registry for agriculture- and forest-based carbon credits. An analysis of the voluntary carbon credit market for the Italian forestry sector from 2011 to 2024 revealed an active market, with credit transactions priced in line with other European domestic markets. Although volumes are lower, they have been growing in recent years and peaked in the period 2021-2022. Monitoring by the Carbon Monitoring Unit highlighted critical issues and verified the impact and future prospects of the market if it is regulated by the introduction of National Guidelines and the National Registry of agriculture and forestry carbon credits.
In the boreal forests of Central Siberia, a marked increase in tree mortality caused by the invasive four-eyed fir bark beetle (Polygraphus proximus Blandf.) has raised concerns about its impact on forest carbon dynamics. This study aims to quantify changes in coarse woody debris (CWD) stocks and their associated carbon fluxes under different levels of beetle infestation. Using systematic field surveys across infestation gradients, we measured CWD accumulation and decomposition by component (snags, logs, and stumps). Our results show that CWD stocks varied significantly, ranging from 68.1 to 179.9 Mg ha-1, with the highest amounts observed in the most heavily infested areas. In particular, the proportion of carbon stored in snags increased with infestation severity. Furthermore, our analysis of decomposition rates based on empirical data indicates accelerated carbon dioxide emissions from CWD in severely damaged stands. These findings underline the substantial impact of bark beetle outbreaks on the carbon storage capacity of fir forests and highlight the need for management strategies that account for pest effects on forest carbon budgets. This study contributes to a better understanding of pest-induced changes in forest ecosystems and supports the refinement of carbon dynamics models for boreal forests.
The western conifer seed bug (Leptoglossus occidentalis Heidemann, 1910) is an invasive alien species from North America that causes significant economic losses by damaging cones and reducing seed yield in pines. The pest was first detected in Turkey in 2009, coinciding with increasing seed losses in many conifers, particularly Stone pine, Pinus pinea L. This study aimed to assess the impact of L. occidentalis on seed loss in three-year-old P. pinea cones. Conducted in 2020-2021 in P. pinea stands of the Izmir/Bergama Kozak Plateau, the experiment involved 5103 seeds collected from 30 cones, both caged (excluded) and uncaged (exposed), on 15 trees over two years in a natural setting. The seeds were first categorized by damage level using X-ray imaging, then subjected to a flotation test to determine the floating and sinking ratios. Subsequently, the kernels were manually examined to classify damage, and their weights were measured. The results showed that seed loss in excluded cones was significantly lower than in exposed cones, highlighting the impact of a factor that cannot reach excluded cones, with L. occidentalis as the most likely cause. X-ray, seed flotation, and naked-eye methods yielded similar results in assessing overall seed loss. Seed damage rates in the exclusion treatments were 17.6%, 17.0%, and 17.0%, respectively, whereas in the exposition treatments they were 46.2%, 48.0%, and 53.0%, respectively. As of 2021, these levels of seed loss corresponded to an estimated economic loss of approximately & euro;55 per kg of seed. When damage levels were categorized, X-ray and naked-eye assessments produced inconsistent results. These findings suggest that X-ray, seed flotation, and naked-eye methods can be used interchangeably to determine overall seed damage. However, the naked-eye method appears to be more suitable for categorizing damage levels.
The relationship between wildfires and protected areas remains controversial in Mediterranean regions, where several studies have reported a disproportionate incidence of fires within protected sites. Using data from the European Forest Fire Information System, 8.241 wildfires recorded in Spain between 2008 and 2025 were analyzed to assess whether wildfire incidence in Natura 2000 sites is primarily associated with protection status or with land-use structure, particularly forest land cover. Indicators of burnt area for total land and forest land at the national level and within the Natura 2000 network were calculated by aggregating wildfire perimeters and their distribution across forest land and Natura 2000 areas, and expressed as proportional indicators to enable direct comparison between spatial units. Additionally, correlation analyses between burnt area variables and Mann-Kendall tests to assess temporal trends in burnt area were applied. Natura 2000 sites accounted for 38.6% of the burnt area in Spain and 40.6% of the burnt forest land. Although the proportion of burnt area relative to total territory was higher within Natura 2000 than nationally (4.9% vs. 3.5%), this difference was largely explained by the higher share of forest land in protected sites (80.9% vs. 54.4%). When considering forest land only, the proportion of burnt area in Natura 2000 was slightly lower than the national average (5.4% vs. 5.9%). Significantly increasing trends in burnt area were detected in both protected and non-protected areas using the Mann-Kendall test. These results suggest that apparent differences in wildfire incidence between protected and non-protected areas are largely attributable to land-use composition rather than to protection status itself, highlighting the importance of explicitly accounting for landscape structure when interpreting wildfire statistics.
Accurate estimation of the spatial properties of forest soil is essential for sustainable land management. This research aimed to map key soil properties in temperate forests using hybrid machine learning models. The integration of predictive and geostatistical techniques has gained prominence in soil science, with hybrid approaches enhancing prediction accuracy. Precise soil maps serve as a foundational resource for effective soil management, motivating the development of more accurate and cost-efficient mapping methods. We used hybrid machine-learning techniques that incorporated Euclidean distance (Dis), remote sensing (RS) data, and digital elevation models (DEM) to improve spatial predictions. We hypothesized that integrating Euclidean distance data into predictive models would boost the accuracy of soil property maps. Model performance was evaluated using root-mean-square error (RMSE), coefficient of determination (R2), mean absolute error (MAE), and concordance correlation coefficient (CCC). The hybrid RF+GA+Dis model offered the highest accuracy for predicting soil organic carbon (RMSE = 1.810, R2 = 0.763, MAE = 1.883, CCC = 0.803), with similar improvements observed for soil nitrogen, organic carbon stock, bulk density, calcium carbonate, and soil texture. Hybrid models consistently demonstrated a superior performance over individual machine learning methods such as Random Forest (RF) and Genetic Algorithm (GA). Incorporating ancillary data, especially from DEM and RS, substantially ameliorated prediction accuracy for soil physicochemical properties. Among the tested models, those using a broader range of covariates demonstrated a superior performance, with RF+GA+Dis outperforming RF+Dis. These findings confirm that integrating machine learning with comprehensive spatial data is a cost-effective and reliable approach for generating high-resolution soil maps, facilitating precision land management and informed decision-making. This highlights the value of hybrid modeling and diverse covariates in advancing soil property prediction.
In the Mediterranean basin, extreme drought events are expected to become more frequent due to climate change. Consequently, selecting and registering forest reproductive material (FRM) adapted to such conditions is crucial to support effective forest restoration. This study aims to identify whether certain seed sources of Quercus exhibit traits potentially advantageous under ongoing climate change. An initial screening of seedlings from five ecologically diverse species from Puglia (southern Italy) was performed to evaluate early-stage responses to drought conditions simulating extreme Mediterranean summer events. Under semi-controlled conditions in a nursery, we evaluated the effect of water removal for 21 days on the growth, physiological, and metabolomic performance of FRM of two seed sources per five Mediterranean Quercus species. Morphological, physiological (SPAD, chlorophyll fluorescence), and metabolic traits (pigments and malondialdehyde) were measured throughout the experimental period and at its end. All Quercus species and their seed sources survived the water deficit and exhibited positive morphological and physiological responses. The effects of seed source were more pronounced in the "Cerris" section than in the evergreen "Ilex" section, making the former species good candidates for afforestation or restoration in Mediterranean areas and valuable material for further research. The high performance of Q. cerris, a species generally considered less drought-tolerant, highlights the Puglia region as a promising source of southern Mediterranean provenances with enhanced resistance to arid conditions.
Decomposition is essential for nutrient cycling, soil fertility, and productivity in tropical forest ecosystems. However, there is limited information on these aspects for some tropical forest tree species. The present study compared the decomposition rates of leaf litter of three tree species (Baphia nitida, Dacryodes klaineana, and Gluema ivorensis) and the influence of litter quality across three habitat types in the Ankasa Conservation Area, Ghana. Using the litterbag technique, a total of 432 bags were prepared and placed on the forest floor across the three habitats for 360 days. Litter samples were removed monthly to determine the mass remaining. At the beginning of the study, the litter chemistry of each species was determined using standard spectrophotometric methods. Differences in litter constituents and decomposition rates in the three species and habitats were evaluated using analysis of variance. Pearson product-moment correlation was used to explore the relationships between litter mass loss and the initial litter quality. The results showed variations (p<0.05) in litter quality among the three studied species, with nitrogen (4.95%) and lignin (1.63%) contents being highest in B. nitida, whereas phosphorous (0.97%), hemicellulose (31.85%) and C:N ratio (15.98) were more abundant in G. ivorensis. D. klaineana recorded the highest proportion of cellulose (65.3) and potassium (2.53). Decomposition rates averaged between 1.99 g year(-1) (in D. klaineana) to 2.35 g year(-1) (in B. nitida), although no statistical differences (F-[2]=0.252, p=0.0778) were found among the species. Habitat type and duration significantly influenced the decomposition rate (p > 0.05).
Interpopulation variation was investigated using seed samples originating from twenty-six European beech (Fagus sylvatica L.) populations across the Balkan Peninsula, a part of the species' distribution range characterized by high ecological heterogeneity in key climatic factors, such as temperature (5.8-10.6 degrees C), precipitation (648-1632 mm), and elevation (185-1410 m a.s.l.). The statistical significance of intrapopulation differences was confirmed by analysis of variance (ANOVA) for all seed traits analyzed: seed weight (g), length (mm), width (mm), thickness (mm), eccentricity and flatness indices, and germination capacity (%). Multivariate principal component analysis (PCA) was applied to examine seed traits in relation to environmental variables of the maternal site, such as mean temperature and precipitation in September and October (the seed maturation period), revealing distinct patterns of relationships among the variables studied. Seed traits were significantly positively correlated with mean temperatures of the maternal site in September and October, indicating that temperature during the seed-filling period affects seed mass. Germination capacity was associated with precipitation during the same period, though the correlation coefficient was not statistically significant; a shorter vector length in the PC biplot suggests a weaker contribution to population separation. Elevation of the site of origin showed a significant negative correlation with temperature, precipitation, and seed traits. Agglomerative hierarchical clustering analysis identified three distinct population clusters. Higher temperature and precipitation values did not necessarily result in higher seed trait values or higher germination percentages. The population with the highest seed mass exhibited the lowest germination capacity (32%) during seed maturation under the lowest precipitation. Conversely, the population characterized by the lowest seed mass showed a higher germination rate of 68% in environments with high precipitation. These results provide valuable insights into the reproductive ecology of European beech, suggesting that other factors beyond those analyzed here may have a more substantial influence on seed germination. The variation in seed traits across habitats that are either drier and hotter or colder and wetter, along the elevation gradient of the studied populations, paves the way for future research and breeding efforts to enhance the species' survival and reproductive success amid anticipated climate change scenarios.
