Brazil nut journey under future climate change in Amazon

those that the species already occurred, causing the species to disperse to these neighboring pixels. We used four environmental change events, (one for each period up to the end of the cen- tury) with two (2) dispersal events, equivalent to one event every 10 years. The scattering func- tion for colonization of neighboring cells follows the following probabilistic values from the contiguous pixels: 1.0, 0.4, 0.16, 0.06, 0.03. This means that as the geographic departure of a particular pixel occurs, the likelihood of scattering events from that pixel decreases. The main result of this spatial population dynamics simulation is a map where it is possible to map all the events described above (persistence, extinction, and colonization) and in which period such an event occurred. In addition, a special class of pixels is created where it can occur, until the end of the evaluation period, the resumption to a previous level of suitability. This class can be interpreted as those pixels that initially lost suitability for the critical level of extinction, but that eventually became favorable by the end of the simulation. Influence of altitude on the future spatial dynamics of B. excelsa With the results of the spatialized simulations with MigClim for the different scenarios (extinc- tion, colonization, and persistence), we tested the hypothesis that the colonization events for new areas could be correlated with altitude, given a visual pattern detected in the analyzes. In this sense, we extracted the altitude values for each cell of the map of the study area and tested

the response of the simulations from a Pearson correlation analysis. Such analyzes were per- formed in R [33]. Results The maps in Fig 1 show the trend of loss/gain of suitability from the present until the year 2100. The positive values (blue tones) show gain in suitability and the negative ones (yellow and red tones) show loss. We observed a loss of suitability for Bertholletia excelsa in most of the Amazon basin. This pattern is persistent among the different future scenarios evaluated. Some areas would be affected by a greater loss of climate suitability than others. For example, in the Southwest Amazon moist forests, in the region of the northern trough of the Amazon River—Uatuma-Trombetas moist forests—and a portion of the southeast—Tapajo´s-Xingu moist forests (areas in red). On the other hand, it was possible to observe that to the west and north of the Amazon basin, around the region of Japura´-Solimoes-Negro moist forests, there would be a gain in suitability, as well as those areas with higher altitude (Peruvian Yungas, Gui- anan moist forests, Xingu-Tocantins-Araguaia moist forests). There are differences between future climatic scenarios that need to be highlighted. Although there is a tendency to reduce the generalized suitability for the Amazon basin, losing suitability would be more severe in the most pessimistic climate scenarios. This result is clear when the percentage lost via extinction in the different scenarios over time is evaluated (Table 1). All scenarios point to losses of more than 50% of the species’ distribution area due to

climate change by the end of the century. In the SSP3-7.0 scenario, the reduction of the avail- able area of the species could reach 94% of what is now the current distribution of Bertholletia excelsa. SSP2-4.5 would be the least severe among the future scenarios evaluated, with a loss of the original distribution area reaching 53.53% by the end of the century. Oppositely, the mod- els predict that the gain in suitability tends to be more pronounced also toward the most pessi- mistic scenarios. The graph in Fig 2 shows the relationship between the temporal trend of suitability and alti- tude. The results show that the colonization of the species to new areas may be related to a search for environments of higher elevation. This is evidenced by the positive values of the PLOS ONE Effects of climate change on Brazil nut PLOS ONE | https://doi.org/10.1371/journal.pone.0312308 November 13, 2024 5 / 13