|
Extensive_description_of_the_dataset
|
We assembled data on freestanding woody plant species from five tropical forest dynamics plots of 16–50 ha from the ForestGeo network (Anderson‐Teixeira et al. 2015, Davies et al. 2021). The five sites are Barro Colorado Island (BCI; Panama), Fushan (Taiwan), Luquillo (Puerto Rico), Pasoh (Malaysia), and Yasuní (Ecuador). In each forest, individuals < 1 cm dbh (hereafter seedlings) were mapped and measured in subplots of the ForestGeo plots, and all individuals > 1 cm dbh (hereafter trees) were mapped and measured following Condit (1998). All five ForestGeo plots have been re-censused approximately every five years (BCI: 1982-2015, Fushan: 2004-2013, Luquillo: 1992-2016, Pasoh: 1987-2010, and Yasuní: 1996-2008).
Data on growth and survival of trees ≥ 1 cm dbh:
In every census, each individual with at least one living stem ≥ 1 cm dbh was identified to species or morphospecies, determined as ‘alive’ and the stem with the largest dbh was measured with an accuracy of 1 mm following the protocol presented in Condit (1998). For each individual tree where the main stem survived a census interval, we calculated growth as the absolute annual increment in dbh (Δdbh)/t with t = time in years between two consecutive censuses.
To account for the dependence of growth and survival on tree size and light availability, we assigned each tree to the overstory or the understory layer, based on the perfect plasticity approach as described in Kambach et al. (2022). The crown area of each individual tree was approximated with the allometric equation crown area (m²) = a * dbh (mm)^b with allometric coefficients a and b being either calculated from the Tallo global tree allometry and crown architecture database (for species with > 10 observations, Jucker et al. 2022) or assigned from site-specific allometries (as listed in Kambach et al. 2024). Each ForestGeo plot was divided into subplots of 31.5 x 31.5 m (following Bohlman and Pacala 2012), in which we ranked all individuals from the largest to the smallest diameter. We successively assigned the largest individuals to the overstory canopy layer, until their cumulative crown area exceeded twice the subplot area. The remaining smaller individuals were then assigned to the understory canopy layer. Palms (Arecaceae), hemi-epiphytic, and unidentified species were retained for assignment to the overstory or understory layer but were omitted from all other analyses.
Data on growth and survival of seedlings < 50 cm height:
At all five ForestGeo plots, growth, survival and recruitment rates of seedling individuals < 1 cm dbh were measured according to ForestGeo protocols (described in Davies et al. 2021). We calculated growth and survival in two seedling classes: individuals < 20 cm height and individuals of 20–50 cm height. Seedling datasets are described in detail in Kambach et al. (2024). For each individual seedling that survived a census interval, we calculated growth as the absolute annual increment in height (Δheight)/t with t = time in years between consecutive censuses.
Data on seed production:
Seed fall was recorded in seed traps according to ForestGeo protocols that are described in Davies et al. 2021. Seed trap datasets are described in detail in Kambach et al. (2024).
Data on species traits:
Species-level trait data were assembled from primary and published sources as follows that are described in Kambach et al. (2024).Maximum dbh was calculated as the mean dbh of the six largest trees of each species at the respective ForestGeo plot.
Calculation of species-specific demographic rates:
For each of the five forests, we estimated species-specific mean annual growth and survival rates in four size classes (seedlings < 20 cm height, seedlings 20–50 cm height, trees ≥ 1 cm dbh in the understory and in the overstory canopy) as well as species-specific mean seed-to-seedling transition rates. We only calculated growth and survival rates for species and size classes when there were > 10 observations.
We transformed the observed annual growth rates to an approximate normal distribution using the Modulus transformation with λ = 0.4 (Eqn. 1) as described in John and Draper (1980) and Condit et al. (2017).
To calculate species-specific growth rates within each size class, we randomly selected (without replacement) up to 200 growth observations across all census intervals per species and size class. We used hierarchical Bayesian models with flat priors to calculate the mean transformed growth rate of species j in size class k according to the likelihood described in Kambach et al. (2024).
To calculate species-specific survival rates within each size class, we randomly selected (without replacement) up to 1,000 survival observations across all census intervals per species and size class. We used hierarchical Bayesian models with flat priors to calculate the mean annual survival rate of species j in size class k according to the likelihhod described in Kambach et al. (2024).
To calculate comparable seed-to-seedling transition rates, we first calculated, for each forest and census interval, the annual number of newly emerging seedlings per ha (scaled from the summed area of seedling plots) and the annual number of seeds captured per ha (scaled from the summed area of seed traps). Seed-to-seedling transition rates were then calculated by dividing the number of newly emerged seedlings per ha and year by the number of seeds captured per ha and year, averaged across all census intervals. Mean seed-to-seedling transition rates > 1 were replaced with a value of one (as described in Kambach et al. 2024).
List of flies:
- readme.txt: metadata file with descriptions of variables in each file
- bci_seedling_rates.csv: Barro Colorado's species-level mean growth and survival rates of seedlings < 20 cm tall
- fushan_rates.csv: Fushan's species-level mean annual seed numbers, growth and survival rates of trees ≥ 1 cm diameter at breast height (dbh), and seedlings < 50 cm tall
- pasoh_rates.csv: Pasoh's species-level growth and survival rates of seedlings < 50 cm tall
- yasuni_growth_above_1cm_dbh.csv: Yasuní's growth observations per species and size class ≥ 1 cm dbh
- yasuni_survival_above_1cm_dbh.csv: Yasuní's survival observations per species and size class ≥ 1 cm dbh
The original data providers chose not to disclose the species’ full names.
All files use UTF-8 character encoding and represent missing values as "NA".
References:
Anderson‐Teixeira, K. J., S. J. Davies, A. C. Bennett, E. B. Gonzalez‐Akre, H. C. Muller‐Landau, S. J. Wright, K. A. Salim, et. al. 2015. CTFS‐ForestGEO: a worldwide network monitoring forests in an era of global change. Global Change Biology 21 (2): 528–549.
Bohlman, S., and S. Pacala. 2012. A forest structure model that determines crown layers and partitions growth and mortality rates for landscape‐scale applications of tropical forests. Journal of Ecology 100 (2): 508–518.
Condit, R., R. Pérez, S. Lao, S. Aguilar, and S. P. Hubbell. 2017. Demographic trends and climate over 35 years in the Barro Colorado 50 ha plot. Forest Ecosystems 4 (1): 17.
Condit, R. S. 1998. Tropical Forest Census Plots: Methods and Results from Barro Colorado Island, Panama, and a Comparison with Other Plots. Berlin, New York: Springer.
Davies, S. J., I. Abiem, K. A. Salim, S. Aguilar, D. Allen, A. Alonso, K. Anderson-Teixeira, et al. 2021. ForestGEO: Understanding forest diversity and dynamics through a global observatory network. Biological Conservation 253: 108907.
John, J. A., and N. R. Draper. 1980. An alternative family of transformations. Applied Statistics 29 (2): 190–197.
Jucker, T., F. J. Fischer, J. Chave, D. A. Coomes, J. Caspersen, A. Ali, G. J. Loubota Panzou et al. 2022. Tallo: A global tree allometry and crown architecture database. Global Change Biology 28:5254–5268.
Kambach, S., R. S. Condit, S. Aguilar, H. Bruelheide, S. Bunyavejchewin, C.‐H. Chang‐Yang, Y.‐Y. Chen, et al. 2022. Consistency of demographic trade‐offs across 13 (sub)tropical forests. Journal of Ecology 110 (7): 1485–1496.
Kambach, S., H. Bruelheide, L. S. Comita, R. Condit, S. J. Wright, S. Aguilar et al. 2024. Putting seedlings on the map: Trade-offs in demographic rates between ontogenetic size-classes in five tropical forests. (Ecology accepted). |
|