Literature search
A systematic literature search was conducted following the PRISMA guidelines (Liberati et al., 2009; Moher et al., 2009) and completed in March 2023. Three searches were carried out using ‘Web of Science’, with the search terms: i) fire* AND “species richness” AND plant*; ii) burn* AND “species richness” AND plant*; and iii) fire*AND “species richness” AND tree*. Our objective was to retain papers that were empirical field-based studies, conducted in the tropics or sub-tropics, i.e., 30o north to 30o south (Corlett, 2013), and that provided complete species lists for control (unburnt or sites sampled before a fire) and treatment sites (those with fires). We only selected studies with equal sampling effort in control and treatment sites, as such biases in study design can impact conclusions regarding fire impacts on biodiversity (Kelly et al., 2017).
Data collection process took place in five stages (Table 1). After collecting the papers from the initial search, titles were scanned to identify papers that could be used to understand the impacts of fire on plant diversity in tropical and sub-tropical locations. Duplicate papers were removed, and abstracts were then scanned. Papers were only accepted if the study met our criteria of being an empirical field-based study located in the tropics or sub-tropics. We then read each paper in full and removed those for which sampling effort was uneven across control (unburnt) and treatment (burnt) sites or did not provide a complete species list for each type of site. A list of retained papers is given at Appendix 1 .
Table 1: The selection stages, procedure, and total no. of papers obtained in the literature search.