Patterns in macroecology are related to species occurrence across meaningful spatial and temporal scales. The dataset provided here reports species distribution data (presence-absence) for herbaceous plants across a number of European habitats (ecoregions). Species occurrence is accompanied by the corresponding plant's maximun stem height values. This dataset has been used to unveil patterns of herbaceous plant height clustering in mid-latitude European ecoregions.

Presence-absence data for herbaceous plants were drawn from Atlas Florae Europaeae (Jalas & Suominen, 1964-1999). Associated to each species, a dominant habitat (ecoregion) was assigned according to the WWF Biomes of the World classification. Each herbaceous species in an ecoregion was characterized by its maximum stem height. Mean height values were obtained for different sources. In order to correlate clustering patterns with productivity measures, actual evapotranspiration (AET) data is also provided. AET maps were obtained from data estimated through remote sensing (Mu et al., 2011), which are publicly available in the MODIS project website (http://www.ntsg.umt.edu/project/modis/mod17.php).

Plant distribution and trait data across Europe unveils a relation between plant height clustering and actual evapotranspiration. This clustering is most evident in mid-latitude ecoregions, where conditions for growth (reflected in actual evapotranspiration rates) are optimal. Away from this optimum, climate severity leads to non-significant height clustering in actual communities.

DATASETS:

ABUN.Herba.csv: Plant community data were drawn from Atlas Florae Europaeae (AFE, Jalas & Suominen, 1964-1999). The distribution of flora is geographically described using equally-sized grid cells based on the Universal Transverse Mercator projection and the Military Grid Reference System. Each cell was assigned to a dominant habitat type based on the WWF Biomes of the World classification (Olson et al., 2001), which defines different ecoregions. We consider each cell in an ecoregion to represent a species aggregation.

TRAIT.Herba.csv: Mean height values were obtained from the LEDA database (Kleyer et al., 2008) for as many species as there were available in the database. Most of the missing values were taken from (Ordonez et al., 2010), and some of them inferred using a MICE (Multivariate Imputation by Chained Equations) approach (Buuren & Groothuis-Oudshoorn, 2011). Based on plant growth forms, 2610 herbaceous species (aquatic, herbs, or graminoid) were considered in this dataset.

MODIS_AET_mean.bil: Mean (annual) actual evapotranspiration maps were obtained from data estimated through remote sensing (Mu et al., 2011), which are publicly available in the MODIS project website (http://www.ntsg.umt.edu/project/modis/mod17.php). This data was put into a .bil format using QGIS. A map for European ecoregions is included in the file ecoregions.bil.

CODE:

Python code for replicability of the results is provided.

randomize.py: performs randomization tests for each cell, by comparing with random samples taken from species in the corresponding ecoregion. It yields p-value distributions for each ecoregion, from which it is easy to compute height clustering indices. The variation of coexistence probabilities as function of competitive strengths can also be reproduced using the output file from this code.

evapo_data.py: calculates ecoregional mean and std for actual evapotranspiration from MODIS data. Similar code can be used to obtain gross primary productivity averages. This code can be used to correlate AET with clustering indices and latitude. Note that this code can be easily adapted to obtain AET by cell instead of by ecoregion, using latitude and longitude defining each AFE cell.