Dataset DOI: 10.5061/dryad.5qfttdzn3

Description of the data and file structure

Wood density (or wood specific gravity) is presented for 3,496 plants representing 750 species from central Panama. The data are presented in four tab-delimited text files. The file “WD_segments.txt” holds values for each segment for wood cores, which were broken into 5-cm segments, and the single value available for plants sampled with the branch and disc methods. For wood cores, the within-individual values enable studies of radial variation in wood density from the pith to the bark (see Hietz et al. 2013, Osazuwa-Peters et al. 2014, 2017). The file “WD_individuals.txt” holds individual-level values for each plant, with fresh volumes and dry masses summed over samples within individuals. The file “WD_species.txt” holds species-level mean wood densities and their standard deviations and sample sizes, where sample size refers to numbers of conspecific individuals. The file “Key_sp6.txt” provides full nomenclature for the six-character mnemonic used to identify species in the three previous files. We updated nomenclature using the World Checklist of Vascular Plants (WCVP) version 13 on 11 February 2026.

The 750 species include 741 with a valid species name. Seven of the nine remaining species are identified to genus, differ from congeners associated with a valid name, but cannot yet be associated with a valid name. One species is identified to family, and one species is unidentified. These nine species are included with temporary values in variables that identify species. Botanists at the Smithsonian Tropical Research Institute use these temporary values to identify trees tagged in permanent plots until the flowers or barcodes required for definitive species-level identification are available. We retain the species Guarea grandifolia, even though it has been synonymized with G. guidonia, because there are clearly two species in central Panama.

The references and related works include 36 publications that have used these wood density data. Values from the early campaigns were placed in the TRY database at its inception (Kattge et al. 2011, 2020). Some values have been published in electronic appendices with the articles in the reference list. Many values are being made publicly available for the first time here. This Dryad repository brings together all wood density data collected by the authors in central Panama.

References

Hietz, P, R Valencia, SJ Wright. 2013. Strong radial variation in wood density follows a uniform pattern in two neotropical rain forests. Functional Ecology 27: 684–692.

Kattge, J, et al. 2011. TRY - a global database of plant traits. Global Change Biology 17:2905-2935.

Kattge, J, et al. 2020. TRY plant trait database – Enhanced coverage and open access. Global Change Biology 26: 119-188.

Osazuwa-Peters, OL, SJ Wright, AE Zanne. 2014. Radial variation in wood specific gravity of tropical tree species differing in growth-mortality strategies. American Journal of Botany 101: 803-811.

Osazuwa-Peters, OL, SJ Wright, AE Zanne. 2017. Linking wood traits to vital rates in tropical rainforest trees: insights from comparing sapling and adult wood. American Journal of Botany 104: 1464-1473.

Files and variables

File: Key_sp6.txt

Description: 

Variables

• sp6: six-letter mnemonic used to identify species by the ForestGEO program. Nine temporary values identify trees that were not fully identified. The nine temporary values are ANIB1, COCC2, MYRC6, MYRRK2, NECTS1, NEEA6, SAP4, SAPISP and STEM1. Botanists at the Smithsonian Tropical Research Institute use these temporary values to identify trees tagged in permanent plots until the flowers or barcodes required for definitive species-level identification are available. There are no missing values.
• sp4: four-character mnemonic used widely at BCI. Missing values equal “NA”.
• family: plant family. Updated using the World Checklist of Vascular Plants version 13 on 11 February 2026. Takes the value ‘Unidentified’ for one unidentified tree. There are no missing values.
• genus: plant genus. Updated using the World Checklist of Vascular Plants version 13 on 11 February 2026. Takes the value ‘Unidentified’ for one unidentified tree and for one tree identified to family. There are no missing values.
• species: plant species. Updated using the World Checklist of Vascular Plants version 13 on 11 February 2026. We retain the species Guarea grandifolia, even though it has been synonymized with G. guidonia, because there are clearly two species in central Panama. Nine temporary values identify trees that were not fully identified. Botanists at the Smithsonian Tropical Research Institute use these temporary values to identify trees tagged in permanent plots until the flowers or barcodes required for definitive species-level identification are available. There are no missing values.
• subspecies: plant variety or subspecies. Updated using the World Checklist of Vascular Plants version 13 on 11 February 2026. The five non-missing values begin with ‘var.’ or ‘subsp.’ to distinguish varieties and subspecies. Missing values equal ‘NA’.
• author: the authority that described the species. Updated using the World Checklist of Vascular Plants version 13 on 11 February 2026. Takes the value “Unidentified” for the nine partially identified species. There are no missing values.
• habit_wcvp: plant habit or life form provided by the World Checklist of Vascular Plants. For values and definitions, see https://powo.science.kew.org/about-wcvp#lifeforms. Missing values equal “NA”.
• habit_panama: plant habit or life form in central Panama. Values are ‘Climbing’, ‘Free-standing’ and ‘Hemiepiphyte’. There are no missing values.

File: WD_individuals.txt

Description: 

Variables

• sp6: six-letter mnemonic used to identify species by the ForestGEO program. The file “Key_sp6.txt” links sp6 to full nomenclature. There are no missing values.
• site: locally recognized names of sites. Missing values equal “NA”.
• plot: names of permanent ForestGEO plots and “NPK” for an experimental plot in the Barro Colorado Nature Monument. Plants from the “NPK” plot were from unmanipulated, control areas. Missing values equal “NA” for plants outside permanent plots.
• tag: integer tag numbers attached to trees in ForestGEO and NPK plots. Missing values equal “NA” for untagged plants.
• latitude: decimal degrees north latitude. Missing values equal “NA”.
• longitude: decimal degrees west longitude. Missing values equal “NA”.
• altitude: altitude in meters. Most values are missing and equal “NA”.
• datecoll: date sample collected in YYYYMMDD format. Missing values equal “NA”.
• dbh: diameter at breast height of 1.3 m. Units are millimeters. Missing values equal “NA”
• method: Values are “core”, “disc” and “branch”. For “core”, we collected samples with a 5-mm diameter increment borer to the center (when possible) of trees with dbh > 100 mm (several smaller trees were cored). For “disc”, we used saws to collect wedge-shaped samples with their tip at the center of the trunk from palm trees recently knocked down by another falling tree. For “branch”, we collected approximately 10-cm long and 1-cm diameter sections of tree branches or liana stems. There are no missing values.
• ind: A character variable that identifies individual plants within for each campaign. Values of ind are repeated across campaigns. Therefore, the variables campaign and ind must be combined to identify unique individuals. There are no missing values.
• campaign: identifies field campaigns. Values are “brassfield”, “corteva”, “gaps”, “hcml2001”, “hcml2004”, “levinson”, and “palms”. There are no missing values.
• wd60: calculated as the sum of mass60c divided by the sum of freshvolume from the file “WD_segments.txt” for each plant. Units are g cm^-3. Missing values equal “NA”.
• wd100: calculated as the sum of mass100c divided by the sum of freshvolume from the file “WD_segments.txt” for each plant. Units are g cm^-3. Missing values equal “NA”.

File: WD_species.txt

Description: 

Variables

• sp6: six-letter mnemonic used to identify species by the ForestGEO program. The file “Key_sp6.txt” links sp6 to full nomenclature. There are no missing values.
• wd60.mean: mean wood density for samples dried at 60 °C. Calculated as the average over all individuals in the file “WD_individuals.txt” for each species. Missing values equal “NA”.  
• wd60.sd: standard deviation of wood density for samples dried at 60 °C. Calculated as the standard deviation over all individuals in the file “WD_individuals.txt” for each species. Missing values equal “NA”.  
• wd60.n: the number of individuals used to calculate the values in the variables wd60.mean and wd60.sd for each species. Missing values equal “NA”.  
• wd100.mean: mean wood density for samples dried at 100 °C. Calculated as the average over all individuals in the file “WD_individuals.txt” for each species. Missing values equal “NA”.
• wd100.sd: standard deviation of wood density for samples dried at 100 °C. Calculated as the standard deviation over all individuals in the file “WD_individuals.txt” for each species. Missing values equal “NA”.
• wd100.n: the number of individuals used to calculate the values in the variables wd100.mean and wd100.sd for each species. Missing values equal “NA”.  

File: WD_segments.txt

Description: 

Variables

• sp6: six-letter mnemonic used to identify species by the ForestGEO program. The file “Key_sp6.txt” links sp6 to full nomenclature. There are no missing values.
• site: locally recognized names of sites. Missing values equal “NA”.
• plot: names of permanent ForestGEO plots and “NPK” for an experimental plot in the Barro Colorado Nature Monument. Plants from the “NPK” plot were from unmanipulated, control areas. Missing values equal “NA” for plants outside permanent plots.
• tag: integer tag numbers attached to trees in ForestGEO and NPK plots. Missing values equal “NA” for untagged plants.
• latitude: decimal degrees north latitude. Missing values equal “NA”.
• longitude: decimal degrees west longitude. Missing values equal “NA”.
• altitude: altitude in meters. Most values are missing and equal “NA”.
• datecoll: date sample collected in YYYYMMDD format. Missing values equal “NA”.
• dbh: diameter at breast height of 1.3 m. Units are millimeters. Missing values equal “NA”
• method: Values are “core”, “disc” and “branch”. For “core”, we collected samples with a 5-mm diameter increment borer to the center (when possible) of trees with dbh > 100 mm (several smaller trees were cored). For “disc”, we used saws to collect wedge-shaped samples with their tip at the center of the trunk from palm trees recently knocked down by another falling tree. For “branch”, we collected approximately 10-cm long and 1-cm diameter sections of tree branches or liana stems. There are no missing values.
• diam1.branch: Diameter of one end of a branch after removing bark. Units are mm. Takes non-missing values when the variable method equals “branch”. Missing values equal “NA”.
• diam2.branch: Diameter of the other end of the branch. Units are mm. Takes non-missing values when the variable method equals “branch”. Missing values equal “NA”.
• length.branch: The length of a branch or stem segment. Units are cm. Takes non-missing values when the variable method equals “branch”. Missing values equal “NA”.
• letter.core: Values are lower case letters a through n. Cores were broken into pieces approximately 5 cm long. Values preserve the position of each piece from just beneath the bark (letter.core = “a”) toward the center of the tree. Together, the variables letter.core and length.core can be used to quantify the position of each piece of each core along a radius from just beneath the bark toward the center of the tree. Takes non-missing values when the variable method equals “core”. Missing values equal “NA”.
• length.core: The length of a piece from a core. Units are mm. Together, the variables letter.core and length.core can be used to quantify the position of each piece of each core along a radius from just beneath the bark toward the center of the tree. Takes non-missing values when the variable method equals “core”. Missing values equal “NA”.
• freshvolume: The volume of the sample determined by water displacement. Units are cm³. There are no missing values.
• mass60c: The mass of the sample after drying to constant mass at 60 °C. Units are grams. Missing values equal “NA”.
• mass100c: The mass of the sample after drying to constant mass at 100 °C. Units are grams. Missing values equal “NA”.
• wd60: calculated as the variable mass60c divided by the variable freshvolume. Units are g cm^-3. Missing values equal “NA”.
• wd100: calculated as the variable mass100c divided by the variable freshvolume. Units are g cm^-3. Missing values equal “NA”.
• ind: A character variable that identifies individual plants within for each campaign. Values of ind are repeated across campaigns. Therefore, the variables campaign and ind must be combined to identify unique individuals. There are no missing values.
• campaign: identifies field campaigns. Values are “brassfield”, “corteva”, “gaps”, “hcml2001”, “hcml2004”, “levinson”, and “palms”. There are no missing values.

Code/software

The four tab-delimited text files can be opened with most software.

Access information

Other publicly accessible locations of the data:

• None for the full data set.

Data was derived from the following sources:

• Not relevant.