Tree dataset
Tree abundance data was obtained
from the Forest Inventory and Analysis (FIA) program. The FIA program
uses a systematic grid of permanent ground plots (approximately 1 plot
per 24.28 km2) surveyed periodically across the entire
United States (Burril et al. , 2018). We used the R package rFIA
(Stanke et al. , 2020) to query the FIA dataset for tree species
in all plots located in the continental United States, including Alaska.
We obtained estimates of tree per
acre (TPA) as a measure of each species’ relative abundance sampled in
the most recent inventory until 2016, the last year for which
environmental data is available (see below). Finally, we dropped species
with less than 150 observations to minimize type I error in QR (Cadeet al. , 2005). Our final database comprised 401,033 observations
belonging to 186 species (identified to the species level) and
distributed in 89,369 ground plots. The database showed a range of 160
records (Quercus incana ) to 19,763 records (Acer rubrum )
per species.