Searching for Caatinga and accurate maps of tropical dry forests: Comments on Zappi et al. (2025) and Hardy et al. (2025)/Martin et al. (2025) from an “ecosystems uncertain” perspective

Accurately mapping major terrestrial ecosystems is important for monitoring anthropogenic impacts, but there is widespread disparity among maps. A recent Opinion Article in Restoration Ecology highlighted an example where inaccurate mapping could lead to misguided restoration interventions if Caatinga was omitted from, but savanna included in, an earlier Review Article about non‐native invasive plants in tropical dry forests. The review authors' reply confirmed that Caatinga was actually included, but did not respond about including savannas. Neither original nor commenting authors explored reasons for inaccurate mapping. The aim of this Opinion Article is to address whether studies located in savannas and Caatinga (and Caatinga‐like ecosystems) should have been included in the review, and explore reasons for inaccurate delineation of dry tropical/subtropical ecosystems. Savannas are grassy ecosystems characterized by C 4 graminoids, variable woody cover and frequent fires. Caatinga‐like thicket ecosystems are non‐grassy, non‐pyric, mostly below 3 m, succulent‐rich and frequently spiny. Both have unlayered woody overstoreys and vigorous resprouting responses to fire or herbivory. Neither savanna nor Caatinga should have been included because both have characteristic sets of plant functional traits not characteristic of forests.