Demographic and structural variability modulate growth dynamics in European beech primary forests
Data files
Apr 24, 2026 version files 70.72 MB
-
Begovic_et_al._GCB-5265646_dataset.zip
70.71 MB
-
README.md
7.56 KB
Abstract
Intensifying droughts and heatwaves are increasing hydraulic constraints on European beech (Fagus sylvatica L.) and driving recent tree vitality losses at its warmer and drier range margins. Network-wide growth models predict widespread beech growth declines under future warming, yet these models rarely account for intraspecific demographic variability, potentially underestimating the adaptive capacity of natural beech forests. We leveraged a network of 530 plots and ~ 11,000 trees from primary beech forests covering a broad environmental gradient in central and southeastern Europe to assess how demographic variability in growth dynamics modulates forest productivity. We explored demographic differences in temporal and spatial variation in radial growth patterns and growth sensitivity to climate warming. Previous summer maximum temperatures and climatic water balance were the dominant climatic constraints of radial growth in xeric regions, whereas growth in mesic regions was more closely associated with current spring and summer conditions. Large trees exhibited stronger sensitivity to prior-summer heat stress, while small trees had stronger associations with spring and summer moisture availability. Severe summer droughts caused pronounced multi-year growth legacies across the demographic strata only in xeric regions. Long-term growth trend analysis revealed substantial regional heterogeneity, with growth declines in mid-sized and large trees contrasted by gains in small and mid-sized young (mesic regions) and old trees (xeric regions). Recent growth declines were linked to rising summer temperatures and declining water balance across the forest strata. Overall, productivity losses outweighed gains across the study network under climate warming. Stand structural complexity and demographic variability in growth dynamics were key drivers of ecosystem productivity, but their positive effects diminished under severe climatic stress. Our findings highlight the critical role of demographic and structural heterogeneity in shaping beech growth dynamics under climate warming, and caution against inferring future ecosystem productivity trajectories without explicitly accounting for demographic variability.
GENERAL INFORMATION ABOUT DATASETS
1.Title of Dataset:
Data from: Begovic_et_al._GCB-5265646_dataset.zip
2. Author Information
Krešimir Begović1, Jakob Pavlin1, Thomas Langbehn1, Kristyna Svobodová Langbehn1,2, Jakub Kašpar3, Andrei Popa4,5, Thomas A. Nagel1,6, Jeňýk Hofmeister1, Pavel Janda1, Miloš Rydval1, Daniel Kozak1, Martin Mikoláš1, Stjepan Mikac7, Miroslav Svoboda1
- Faculty of Forestry and Wood Sciences, Czech University of Life Sciences Prague, Czechia
- Faculty of Environmental Sciences, Czech University of Life Sciences Prague, Czechia
- Department of Forest Ecology, Landscape Research Institute, Lidická 25/27, 602 00 Brno, Czech Republic
- National Institute for Research and Development in Forestry “Marin Drăcea”, Bucharest, Romania
- Faculty of Silviculture and Forest Engineering, Transilvania University of Brasov, Brasov, Romania
- Biotechnical faculty, University of Ljubljana
- Department of Forest Ecology and Silviculture, Faculty of Forestry, University of Zagreb, Croatia
Corresponding author: Ing. Krešimir Begović, Ph.D.,
Email: begovic@fld.czu.cz
Phone no.: (+420) 77-604-5944
Department of Forest Ecology, Faculty of Forestry and Wood Sciences,
Czech University of Life Sciences Prague
Kamýcka 129, 165 21 Prague, Czech Republic
3. Date of data collection
2014-2024
4. Geographic location of data
The research sites are part of an international network of permanent forest inventory plots established in the dominant mountain ranges of central and southeastern Europe (i.e., REsearch on MOuntain TEmperate network; see www.remoteforests.org), and include the Carpathian Mts. (Slovakia, Romania), Dinaric Mts. (Croatia, Bosnia, Albania), Balkan Mts. (Bulgaria) and the Jizera Mts. (Czechia).
5. Information about funding sources that supported the collection of the data
Funding for this research was provided by the Czech Science Foundation Grant (project no. GA15-14840S, 21-27454S, 24-12210K, and 25-18519S).
DATA & FILE OVERVIEW
1. File List
- GCB_beech_tree_data.txt
- GCB_beech_plot_data.txt
- GCB_beech_climate_data.txt
- README - GCB-5265646 Demographic and structural variability modulate growth dynamics in European beech primary forests.docx
2. Relationship between files, if important
/
3. Additional related data collected that was not included in the current data package
/
4. Are there multiple versions of the dataset? yes/no
no
METHODOLOGICAL INFORMATION
1. Description of methods used for collection/generation of data and processing
Field data collection followed protocols established within the REMOTE network to ensure consistency and cross-study comparability. In short, sampling plots were established using a stratified random design to capture the spatial variability across European beech primary forests. A 10-ha grid was overlaid across each forest area and random points were generated within each 3.4-ha grid cell. In case of the unsuitability of a randomly generated point due to difficult terrain (e.g., rocky terrain, steep slopes, etc.), an alternative point was randomly generated within a grid cell. This design established a hierarchical sampling structure, with plots nested within stands and stands nested within regions, thereby maximizing the representativeness of ecological gradients across biogeographic regions. Pairs of circular sampling plots (1500 m2) were established 40 m in each direction from the randomly generated points along the slope contours. Within each sampling plot, topography (i.e., slope, aspect, and elevation) and biometric data (i.e., species, tree position within the plot, canopy, and living status) were recorded. In total, we used dendrometric and structural data from 530 sampling plots nested within 33 distinct forest stands (Table 1).
Increment cores were obtained from all trees with DBH ≥ 10 cm within the inner 200 m2 circular subplot at each sampling plot. In addition, trees with DBH ≥ 20 cm and approximately 25% of non-suppressed trees with DBH ≥ 10 cm were cored outside of the inner subplot. All cores were collected at 1 m stem height perpendicularly to the dominant terrain slope to avoid reaction wood (for a more detailed description of sampling protocols, see Janda et al., 2025; Marchand et al., 2023; Pavlin et al., 2024). Cores were processed following the established protocols by Stokes & Smiley (1996), which involved air-drying, mounting on wooden boards, and sanding with progressively finer-grit sanding paper until annual growth rings were clearly demarcated. Annual ring-width measurements (RW) were obtained using the LinTab stereomicroscopes with a sliding measuring stage and integrated TSAP-WinTM measuring software, and were visually cross-dated based on the extreme growth year approach (Yamaguchi, 1991). Dating quality was validated using the CDendro software (v. 9.6; Larsson, 2015). After excluding cores that could not be reliably cross dated, and those missing more than 30 mm or 20 years to the pith, we retained 10 860 cores for further analysis (Begović et al., 2026).
DATA-SPECIFIC INFORMATION (USAGE NOTES)
The dataset contains information on individual tree-ring chronologies (_tree_data), individual plots (_plot_data) and climate data (_climate_data).
In GCB_beech_tree_data.txt, each row shows annual information for each individual tree. Each tree is characterized by a unique “tree_id”, with its’ associated “plotid” as the identifier of the plot, “stand” as the identifier of the stand, and “x_m” and “y_m” as the identifiers of tree position from the respective plot center. “status” identifies whether the tree is alive or dead (recently dead), “layer” whether the tree is dominant (11), co-dominant (12) or suppressed (13), “Age” defines tree age at a specific “year”, “incr_mm” shows tree-ring width, “dbh_growth” is the annual cumulative value of each TRW, “BAI_mm” is the basal area increment, “BA_mm” identifies basal area (cumulatively), and “missing_years” and “missing_mm” identify number of tree rings and width of missing years to the pith, respectively. “NA” denote non-available data, specifically in the BAI_mm column, where the basal area increment could not be calculated as the difference between current and previous year basal area value (i.e., the 1st year of an individual tree’s basal area growth). “Age_class” and “size_class” denote to which demographic class each individual tree is associated to (used in growth trend analysis).
The second table provides information about each plot and stand (GCB_beech_plot_data). “slope” represents the slope [˚], “aspect” represents the aspect [˚], while “altitude_m” represents the elevation of the respective plot. “lat” and “lng” denote each plots’ latitude and longitude, respectively.
Finally, the third table contains climate information used throughout data analysis (GCB_beech_climate_data). Mean (Tmean), maximum (Tmax), and minimum (Tmin) temperature, total precipitation (Prec) and average climatic water balance (water_balance) are shown for each month, year and stand over the period 1940-2020.
The file "README - GCB-5265646 Demographic and structural variability modulate growth dynamics in European beech primary forests.docx" is a Microsoft Word copy of this README.