The distribution of canopy heights in tropical rain forests directly affects carbon storage and the maintenance of biodiversity. We report here raw field data on annually-measured canopy height distributions over an old-growth tropical rain forest landscape in Costa Rica from 1999-2018. The data were taken at 231 points on a 5 x 5 m grid in 18 0.50 ha plots that were initially sited with a stratified random design across local gradients of soil nutrients and topography. In temporal duration, resampling interval and spatial extent the data are currently the most detailed available for any forested landscape. A digitized version of the data has been published (Clark, Clark and Kellner 2021) based on this database. The version published here is the definitive raw data, and in case of discrepancies between the digitized and raw version the raw data are the primary data source. The raw data also contain numerous field notes that clarify rare or odd field events that provide context for the digitized data.

This data set contains the raw data for a study of tropical rain forest canopy heights carried annually from 1999 to 2018.  The study was carried out at the La Selva Biological Station in Costa Rica (10^o 26’ N, 83 ^o 59’ W) in old-growth Tropical Wet Forest.  All research was carried in a network of 18 0.50 ha plots (50 x 100 m) that was sited using a strict stratified-random design based on local gradients of soil nutrients and topography (the CARBONO Project).  Plots were located in three contrasting edaphic conditions: flat areas on relatively more fertile old alluvial soils (A alluvion plots), flat ridgetops on less fertile residual soils (L loma plots), and steep residual soil slopes (P pendiente plots) (further explained in Clark et al. 2019).   Descriptions of forest structure and dynamics of these plots and the landscape, including analyses of the effects of the strong 1997-1998 El Niño event, have been reported in a series of publications (Clark et al. 2010, 2013, 2017, 2019). 

Beginning in 1999 canopy height measurements were taken annually in the CARBONO plots on a 5x5 m grid.  Nine plots were censused in 1999, after which all 18 plots were censused annually from 2000-2018.  The detailed protocol for canopy heights measurements in given in S1 Protocol.  At each of the 231 points per plot a 15-m extendable measuring pole was placed vertically at each point.  The pole was raised to the highest vegetation vertically above the point and vegetation height was measured to the nearest cm.  Verticality was measured with a Suunto clinometer.  If the highest vegetation was >15 m tall, the point was classified as “high canopy”.  If there was no vegetation above the point in an area estimated to be at least 2 x 2 m canopy height was classified as <2m. 

Every year we assessed the precision of the canopy height estimates by remeasuring an average of 123 points after a one to several day interval.  Precision remeasurements were done in plot A1 in 1999, thereafter in plot L4.  Of the points that were classified as >15 m height on the first census (N=1870), 99.4% were classified identically on the second assessment.  For the 571 points that measured <15 m at both measurements the median absolute difference in height was 0.15 m.  

On 19 May 2018 a powerful storm hit the field site (Rader et al. 2020), causing extensive tree mortality and associated changes in canopy structure in several of the CARBONO plots.  All the data for 2018 were taken after than storm, beginning on 21 May 2018.

Literature cited here:

Clark, D.B., D.A. Clark and S.F. Oberbauer.  2010.  Annual wood production in a tropical rain forest in NE Costa Rica linked to climatic variation but not to increasing CO₂.  Global Change Biology 16:747-759.  doi: 10.1111/j.1365-2486.2009.02004.x.

Clark, D.A., D.B. Clark and S.F. Oberbauer.  2013.  Field-quantified responses of tropical rainforest aboveground productivity to increasing CO2 and climatic stress, 1997-2009.  Journal of Geophysical Research-Biogeosciences 118:1–12, doi:10.1002/jgrg.20067

Clark, D.B., D. A. Clark, S.F. Oberbauer and James R. Kellner.  2017.  Multidecadal stability in tropical rain forest structure and dynamics across an old-growth landscape.  PLOS One https://doi.org/10.1371/journal.pone.0183819.

Clark, D.B., A. Ferraz, D. A. Clark, J. R. Kellner, S. G. Letcher and S. Saatchi.  2019.  ­Diversity, distribution and dynamics of large trees across an old-growth lowland tropical rain forest landscape.  PLOS One https://doi.org/10.1371/journal.pone.0224896

Rader AM, Cotrell A, Kudla A, Lum T, Henderson D, Krandikar H et al.  Tree functional traits as predictors of microburst-associated treefalls in tropical wet forests. Biotropica. 2020;00:1–5. https: //doi.org/10.1111/btp.12761

This data set contains the raw data for the canopy heights studies documented in the Related Works section. All digitized data on canopy heights were input from the field sheets archived here. In case of any discrepancy between the field sheets and the digitized data the field sheets are the original data source. In addition to being the original data, they also have a number of field notes that may clarify questions about the digitized data. To our knowledge there are no missing data, and all information necessary to interpret and use these data are given in the ReadMe.txt file. Each file also contains at the end copies of the protocols used by the field technicians each year (these vary very little from year to year). The original field data for the precision remeasurements (see Methods) are also included (plot A1 in 1999, plots L4 thereafter).

The data sheets all have the form shown below.  PLOT is the individual forest inventory plot (A1-A6, L1-L6, P1-P6 see Clark et al. 2017), Y is the Y coordinate in the grid (0-50), X is the X coordinate in the grid (0 - 100), ALTURA CM is the canopy height at that point in cm (to 1500 cm, values exactly 1500 are coded as 1499, 1500 code = >1500 cm), FECHA is the date, and COMENTARIOS are any additional field notes at that point.