Skip to content

hrlai/novelforestSG

Repository files navigation

Data and Model from the Novel Forests of Singapore

License: CC BY 4.0 R-CMD-check

About

Welcome to the R package novelforestSG! It contains the raw forest community data used in Lai et al. (2021) and Neo et al. (2017).

In addition, it provides a download_model() function to download the brms model fitted in Lai et al. (2021). This model object also contains input data that include the environmental/landscape variables of the 97 forest plots analysed by Lai et al. (2021). However, this does not include three plots in Neo et al. (2017) that had no woody stems. Environmental/landscape variables for all 100 plots, and more variables such as canopy cover and litter depth are now available.

Installation

From CRAN:

install.packages("novelforestSG")

Or install the development version (especially if the devel version > CRAN version as stated above):

install.packages("remotes")  # prerequisite
remotes::install_github("hrlai/novelforestSG")

Using the data

To access the raw data:

library(novelforestSG)
lapply(novelforest_data, head)

For more information, see ?novelforest_data.

To access the summarised data and environmental variables used in the Lai et al. (2021) analysis, first download the model object. The model object is too large (16.5 MB) to come with the package, but the download_model function will download the model from our GitHub development website:

mod <- download_model()

Then, extract the input data from the model object:

in_dat <- mod$data

In the input/summarised data, you will find the environmental variables as certain columns. These plot-level measurements can be matched to the stem-level raw data via plot names. See ?download_model for more details. Note that the predictor variables were log-transformed and then scale to zero mean and unit SD prior to modelling.

Dependencies

Because we analysed the data using the brms v2.10.0 package in R, it is highly recommended that you install brms to squeeze the most out of the model output:

install.packages("brms")

You may also need RTools or Xcode, depending on your operating system; see the brms homepage. This will take a few minutes so have a cup of hot beverage handy.

Issues

Please feel free to report any issues to our GitHub Issue page.

Erratum

While releasing additional data between v1.2.1 to v2.0.0, we realised that some observations were removed by mistake in the original dataset used in Lai et al. (2021). These omitted data are now included in >v2.0.0. We repeated the analysis using the corrected data in v2.0.0 and obtained extremely similar findings as reported Lai et al. (2021), which is reassuring! Please feel free to contact us if you have any concern.

Citation

We believe that the sharing of datasets is important for advancing ecology. When you use the data or model output in your original research or meta-analysis, we appreciate if the following papers are cited.

If you use the trees data:

Lai, H. R., Tan, G. S. Y., Neo, L., Kee, C. Y., Yee, A. T. K., Tan, H. T. W., & Chong, K. Y. (2021). Decoupled responses of native and exotic tree diversities to distance from old-growth forest and soil phosphorus in novel secondary forests. Applied Vegetation Science, 24, e12548. doi: 10.1111/avsc.12548

If you use the presence–absence data of all vascular plants, or the environmental/landscape variables:

Neo, L., Yee, A. T. K., Chong, K. Y., Kee, C. Y., & Tan, H. T. W. (2017). Vascular plant species richness and composition in two types of post-cultivation tropical secondary forest. Applied Vegetation Science, 20(4), 692–701. doi: 10.1111/avsc.12322

See the LICENSE file for license rights.

Contacts

Hao Ran Lai [email protected]
Kwek Yan Chong [email protected]
Alex Thiam Koon Yee [email protected]