Data and code for article: Tree mortality increases with temperature across elevation gradients of the America’s tropical mountain forests - Dec 5, 2025 08:32

This is the code to replicate the analyses and results reported in the article entitled: "Tree mortality increases with temperature across elevation gradients of the America’s tropical mountain forests". The code uses as input data plot-level data, corresponding to forest census and trait data from 180 permanent plots encompassing 420 census intervals (1991–2024) across multiple elevational gradients (440–3,511 m a.s.l.) in seven countries. Input data description can be found at data repository: https://figshare.com/s/9b8b6ba0bea333c8a6fd?file=59397341 The code "TMF_mortality_full_code_JO_DB.R", is used for the following analyses 1) Section 1: PCA of four plant traits to generate PCA1 and PCA2 axes, to complete the analysis dataset 2) Section 2: Test of model consistency using DAGs. We also use Dagity functions to select appropriate covariates for statistical tests. 3) Section 3: Provides 3 examples of multilevel Bayesian models, one example with only 1 exposure, one example with an exposure and covariates, and a null model without exposures. The code "TMF_mortality_paper_figures_JO.R" generates all figures (from Bayesian models), used for the interpretation of results. Figures are generated using the input of all models assessed in our study. The model results are already stored in the file "MODEL_OUTPUTS", within the data folder. It is not necessary to run the codes in sequence; each code can be run separately with its own data input.