Data from: Application of a trait‐based species screening framework for vegetation restoration in a tropical coral island of China

1. Selecting suitable species for vegetation restoration presents a
notable challenge for land managers and scientists. Recently developed
trait‐based approaches may be an effective means of overcoming this
challenge. However, we lack a trait‐based species screening model that can
be used to select potential species for restoration of degraded
ecosystems. 2. Here, we developed a species screening model based on
quantitative trait‐based theory and a maximum entropy algorithm. The
objective was to select more species that have comparable restoration
abilities to the target species that have high survival rates for
vegetation restoration based on species’ functional traits. Thus, species
diversity will be improved to facilitate restoration. We also developed a
software platform that can be used to implement the model. We then applied
our model and software platform to select species for restoration efforts
in a tropical coral island which is part of Hainan Island, China. 3. As a
prerequisite, we started with three target species which have high
potential for restoring the island. Likewise, 66 non‐native species were
selected as the potential species pool. For each species, we identified
and measured 28 traits that are strongly associated with harsh
environments. Harsh environments are those with drought stress, high
temperatures, intensive UV radiation, lack of real soil and nutrients, and
high salinity and alkalinity. Then, our software platform was used to run
the species selection model. Finally, 12 out of 66 species being
identified as suitable species for restoration. 4. We transplanted
seedlings of all 66 species to the island to monitor seedlings survival.
We found that the 12 species identified from our model had high survival
rates, which ranged from 86% to 91%. In contrast, the mean survival rate
for species not identified from our model was less than 40%. These results
suggest that our species screening procedure was appropriate for selecting
candidate species for use in vegetation restoration. 5. We show that by
using species natural history information, as well as functional trait
data, candidate species for restoration efforts can be successfully
identified in a timely manner. Importantly, our proposed method is faster
and less costly than more commonly used ‘trial-and-error’ method. The most
time-consuming aspect of our approach is the need to measure the
functional traits of target and potential species. Ultimately, we provide
a protocol for using functional traits to quickly select a large number of
suitable species for restoring degraded ecosystems. We expect that this
work will be important for future vegetation restoration efforts in
tropical islands, and perhaps other ecosystems as well. We also expect
that our model will help prevent the invasive species and promote specific
ecosystem functions.