Study location: Ilha do Cardoso, São Paulo, Brasil, (25° 03' S, 47° 53' W)
If re-using this data, please cite the version which is archived on Zenodo
The data is in the data directory, stored as rectangular CSV files. There are four files with data, and a fifth containing the metadata:
abundance.csv: Abundances of species in bromeliads. Data is in a "long" format, with one row for every bromeliad x species combination.environmental_variables.csv: Environmental variables measured on each bromeliad. One row per bromeliad species.macroinvertebrate_names_mass.csv: Traits and notes at the level of each morphospecies. One row for every distinct morphospecies found.predator_names.csv: A list of all taxa known to be predators. This also maps morphospecies to higher taxonomic levels, where that is necessary.metadata_dictionary.csv: describes the variables in each dataset in detail
In 2008, each bromeliad was dissected and washed to remove invertebrates. We passed this water through two sieves (150 and 850 µm), which removed particulate organic matter without losing any invertebrates. All invertebrates were counted and identified to the lowest taxonomic level possible. The body length of all individuals was measured when possible for small and medium-sized taxa (< 1cm final instar) and always for large-bodied taxa (> 1 cm final instar).
Acknowledgements: GQR is grateful to the Brazilian Council for Research, Development and Innovation (CNPq) and São Paulo Research Foundation (FAPESP) for research fund and productivity grant. DSS was supported by a NSERC Discovery grants and an E.W.R. Steacie Memorial fellowship.