Skip to main content

Census data from 65 tree plots in Panama, 1994-2015

Cite this dataset

Condit, Richard; Pérez, Rolando; Aguilar, Salomón; Lao, Suzanne (2022). Census data from 65 tree plots in Panama, 1994-2015 [Dataset]. Dryad.


These are data from 65 tree plots in Panama established over 1994-2014; 43 of the plots have been recensused, while 22 plots have just a single census. Details of census methods are described in Condit (1998) and Condit et al. (2013). The 65 plots here are mostly 1 ha in area, though several are 0.32 ha, one is 4 ha, and one is 6 ha. Those two larger censuses are the Sherman and Cocoli plots described in Condit et al. (2004). A companion data archive includes all data from the Barro Colorado 50-ha plot (Condit et al. 2019).

The PIs would like to be informed of papers resulting from the marena plot data. Depending on our level of interest and how much a paper depends on the plots, co-authorship might be requested.


  • Condit, R., Engelbrecht, B. M. J., Pino, D., Pérez, R., and Turner, B. L. 2013. Species distributions in response to individual soil nutrients and seasonal drought across a community of tropical trees. Proceedings of the National Academy of Sciences 110:5064–5068. pdf
  • Condit, R., 1998. Tropical Forest Census Plots: Methods and Results from Barro Colorado Island, Panama and a Comparison with Other Plots. Springer-Verlag, Berlin. pdf
  • Condit R., Perez, R., Aguilar, S., Lao, S., Foster, R., Hubbell, S.P. 2019. Complete data from the Barro Colorado 50-ha plot: 423617 trees, 35 years, 2019 version.
      • Condit, R., Aguilar, S., Hernandez, A., Pérez, R., Lao, S., Angehr, G., Hubbell, S., and Foster, R. 2004. Tropical forest dynamics across a rainfall gradient and the impact of an El Niño dry season. Journal of Tropical Ecology 20:51–72. pdf


There are 42 tables in R format, with 21 tables in the tree format and 21 more in the stem format. The tree tables are divided into six sets:

  1. The table marena1cns.tree1 has a record for every one of the 65 plots in one census (usually the most recent, but not always).
  2. Tables marena2cns.tree1 and marena2cns.tree2 have a record for 43 plots having at least two censuses, with tree1 having the earlier census, tree2 a later census. The two tables have exactly the same set of trees, including those not yet present in the first census and dead trees in the second, and the rows of the two tables have exactly matching order. The two marena2cns.tree tables are thus effectively a single giant table, with columns for every bothcensuses. They are divided into separate tables for easier access.
  3. Tables marena3cns.tree1, tree2, tree3 have records from the 11 plots having at least three censuses. As for marena2cns, every tree from all censuses is included, with rows in matching order.
  4. Tables marena4cns.tree1, tree2, tree3, tree4 have records from the six plots having at least four censuses.
  5. Tables marena5cns.tree1, tree2, tree3, tree4, tree5 have records from the one plot having at least five censuses.
  6. Tables marena6cns.tree1, tree2, tree3, tree4, tree5, tree6 have records from the single plot having six censuses.

Note that the six sets of tables have many of the same records. So marena1cns has every tree, including all trees appearing in marena2cns etc. Likewise, the tables marena2cns include every tree appearing in marena3cnsetc. Most users will want just one of the five sets, depending on the specific question to be addressed. If only a single census is needed, then marena1cns should be used and there is no need to consult marena2cns and above. If exactly two censuses are needed, then marena2cns should be use and none of the others.

The 21 stem tables form a parallel set to the tree tables, divided into the same six groups, but having one row for every stem ever censused (instead of one per tree). The Barro Colorado data are organized in the same way (Condit et al. 2019).

In the tree tables, if stemID matches for a given tree in two censuses, then it is certain that the same stem was measured in both censuses. If the stemID differs between censuses, it often means that the first stem broke and a new stem was measured. However, in trees with 2 or more stems, it was not always possible to identify stems with certainty across censuses prior to 2010, because stem tags were not used. This means that there are cases where the stemID changes even though the stem(s) really did persist. The cleanest way to assess growth is to use only those cases where the stemID matches between censuses. It is also necessary to check the HOM (height-of-measurement) in case it changes between censuses. In the stem tables, records on matching rows are guaranteed to be the same stem, but the HOM might change.

The 21 tree tables are zipped into the single file, and the 21 stem tables into

All species names, with additional taxonomic details, can be found in bci.spptable, archived with the BCI data at Condit et al. (2019). Though named for the BCI plot, it covers all species in the entire set of plots.

Columns in R Analytical Tables: Tree and Stem

  • treeID: The unique tree identifier in the database. Guarantees a tree match.
  • stemID: The unique stem identifier in the database. Guarantees a stem match.
  • tag: Tag number on the tree (occasionally negative where a tag was duplicated by mistake).
  • StemTag: Tag number on the individual stem, if present.
  • sp: The species mnemonic. See the R Species Table for full Latin names.
  • quadrat: Quadrat designation, as a 2-digit row number then 2-digit column number on a 20x20 m grid.
  • gx: The x coordinate within the plot, meters from the west border of the plot, always in [0,1000).
  • gy: The y coordinate, meters from the south border, always in [0,500).
  • dbh: Diameter (mm) of one stem on the tree, the stem whose stemID is given.
  • hom: The height-of-measure, meters above the ground, where the dbh was measured.
  • ExactDate: The date on which a tree was measured.
  • date: Integer date for easy calculation of time interval between censuses (the number of days since 1 Jan 1960).
  • codes: The codes describing the measurement as recorded in the field. See Condit (1998) for a description of field codes. For analyses, status codes should be used, not field codes.
  • status: A status code for the tree or stem. See the section below, Status Codes for Analyses.
  • DFstatus: Alternate stem status, redundant relative to status.
  • nostems: The number of living stems on the date of measurement.
  • agb: Above-ground-biomass of all stems on the tree, in Mg (= metric tons), or for the individual stem. Note that agb=0 for dead trees.
  • ba: Basal area of all stems on the tree, in square meters. Note that ba=0 for dead trees.
  • plot: Name of the plot.
  • census: Census number, 1-7.
  • plotID: A numeric plot identifier, not necessary for most users.