Modelagem de distribuição das formações vegetais do Estado do Paraná: passado, presente e futuro

Abstract

Deforestation is currently the greatest threat to terrestrial ecosystems, but climate changes can have equal or even greater effect on species extinction. The mosaic vegetation that cover Paraná State (Southern Brazil) comprising forests, grasslands, and savannas was result of past climate fluctuations during the Quaternary. It makes Paraná an interesting field of study about the dynamic of vegetation distribution in response to climate, being able to provide valuable data about the response of terrestrial ecosystems to future climate changes. Species Distribution Modeling have been widely used to generate this kind of information. Here, we used Maxent to generate distribution models of the vegetation types of Paraná in five temporal scenarios: Last Glacial Maximum (LGM), Mid Holocene (MID), current time, and optimistic and pessimistic projections for 2070. We divided the results into three sections. In the first section of the results, we present tutorials designed to pre-process environmental variables in a fast way using open source tools, as R and QGIS. In the second section, we evaluated changes in vegetation between temporal scenarios. Models of past show a domain of subtropical grasslands during LGM, with forest expansion during the MID, which is consistent with paleopalinological data. Optimistic future projection shows a forest expansion, mainly over Steppe and Savanna. In the pessimistic future projection, almost the entire Paraná is covered with suitable areas for only a vegetation type: Semideciduous Seasonal Forest. In the third section, we evaluated the efficiency of protected areas in the protection of climatically stable areas, and we proposed priority areas considering climatically stable areas, protected areas and native vegetation remnants. CSAs loss can reach 25% in an optimistic climate changes scenario and thus would decrease by 5.2% the PA efficiency. In a pessimistic scenario, CSA loss can reach 99% and thus would decrease by 80% the PA efficiency. Some conservation priority areas identified here overlap with priority areas identified in other researches and with areas in need of restoration according to the new Brazil’s Forest Code. Planning for new PAs should consider a variety of biological criteria, including the identification of CSAs that could ensure the long-term functionality of these areas. Landscape modifications caused by climate changes can induce irreversible ecological and economic impacts both in the short-term, due to species loss, and in the long-term, due to loss of areas that have the potential to act as refuges in a climate change scenario, reinforcing the need for policies to reduce gases that contribute to global warming.