Carpathian tree-ring network for European beech and Norway spruce
Data files
Apr 02, 2024 version files 7.23 MB
-
README.md
-
Svoboda_Carpathian_plot_chronologies.csv
-
Svoboda_Carpathian_plots_metadata.csv
Abstract
Basic ecological theory suggests that a tradeoff between competitiveness and stress tolerance dictates species range limits at regional extents. However, empirical support for this key theory remains deficient because the necessary spatial and temporal coverage and scalability of field observations have rarely been achieved. We harnessed an extensive dendroecological network (>22,000 tree-ring samples from 816 forest inventory plots) to disentangle competition-limited from climate-limited growth in both overstory and understory trees. Growth synchrony among trees thereby served as an integral metric of climate sensitivity, an approach that we justify in supplementary analyses of growth responses to temperature, precipitation, and the standardized precipitation-evapotranspiration index. Sampling plots were arranged along elevational climate and vegetation gradients throughout the Carpathian Mountains, ranging from mixed-species lowland forests to coniferous forests at high elevations. With mixed-effect modelling, we also identified non-climatic factors (stand characteristics, species diversity, and disturbance history) that modulate spatial patterns in the growth rate and synchrony of European beech (Fagus sylvatica L.) and Norway spruce (Picea abies (L.) Karst.). Beech exhibited reduced growth and increased climate sensitivity towards higher elevations but performed better when species diversity was higher. The growth of spruce increased towards its lower range boundary, but understory cohorts grew poorly under interspecific competition. Overall, climate sensitivity was lower in more productive stands with benign climatic conditions and in recently disturbed sites with reduced stand density. These contrasting performances at mid-elevations where the two species overlap (900 – 1300 m a.s.l.) reflect their evolutionary history, which enables them to be competitive (beech) or cold-stress tolerant (spruce). This history will affect interactions between the two species under climate warming and shape macroecological patterns in the Carpathian ecoregion and likely other parts of Europe. Our findings point to a growing advantage of competitively stronger species in montane and subalpine vegetation zones.
README: Carpathian tree-ring network for European beech and Norway spruce
https://doi.org/10.5061/dryad.q2bvq83ss
This dataset contains tree-ring width records and associated site metadata from a network of sampling sites and plots that are distributed throughout the Carpathian Mountains. These sites are part of the REMOTE forest network and encompass primary forest locations in Slovakia, Ukraine, and Romania. Sites were identified in collaboration with park managers, land owners, and other local experts. Further details on site selection and sampling protocols can be found in Schurman et al. 2019, Global Change Biology, and Schurman et al. 2024, Ecography.
Description of the data and file structure
The data are organized in two primary files that contain the plot-level metadata ("Svoboda_Carpathian_plots_metadata.xlsx") and tree-ring chronologies ("Svoboda_Carpathian_plots_chronologies.xlsx"). Both files are in a standard Microsoft Excel format to facilitate broad use and conversion into other formats (e.g., ASCII text).
Metadata:
Each row in the file corresponds to one sampling plot (830 total). The columns contain the following information:
- Country where the sites and plots are located.
- Stand (= name of the site)
- Forest type (= either spruce or beech dominated)
- Plot ID (format: country_site_plot, e.g. ROM_FA0_001)
- Latitude (in latlon format, WGS84 geodetic reference system and datum)
- Elevation (in meters above sea level)
- Basal area of the plot (in square meters per hectare)
- Stem density of the plot (in stems per hectare)
Tree-ring data:
The data represent plot-level tree-ring chronologies for the dominant tree species (either European beech or Norway spruce). The data are arranged in a long format and sorted according to the Plot ID that corresponds to the respective column in the metadata. Each row represents an individual year. Two different chronologies are presented, along with the sample depth:
- std (= standard chronology, dimensionless index with a mean of 1)
- res (= residual chronology, dimensionless index with a mean of 1)
- samp_depth (= number of measurement series that were averaged into the site-level chronology)
A "chronology" represents the mean plot-level interannual growth variability from all sampled trees. To build a chronology, measurement series from each tree is first "detrended" to remove unwanted low-frequency trends that are influenced by stem geometry and age. In this dataset, this detrending was done with Friedman's super smoother (Friedman 1984, Stanford University California's Lab for Computational Statistics). Next, all detrended series are averaged into a plot-level chronology (std) using Tukey's bi-weight robust mean. This chronology often contains significant temporal autocorrelation that researchers may want to remove for specific analyses (e.g., for climate-growth associations). This is done through autoregressive (AR) modelling and the resulting residual chronology (res).
Sharing/Access information
Data sharing and use are not restricted, but researchers are encouraged to contact the data PI for more information regarding plot locations and details (Miroslav Svoboda, svobodam@fld.czu.cz).