Métodos de processos pontuais para análise espacial de espécies arbóreas nativas de um fragmento florestal da região de Lavras-MG

Abstract

Many random phenomena can be expressed by occurrences identified by coordinates located in space. These phenomena are called point processes. When some attribute (mark) is associated with the coordinate, a marked point process is determined. In forest management of native species, it is extremely important to characterize not only the spatial configuration of each tree species, but also the interaction between different species. This characterization can help the researcher by providing information that includes competition, distribution, growth, mortality and coexistence of species in that space. Thus, the objective of this work was to use point process methods for spatial analysis of native tree species from a fragment of Montana semideciduous forest located in the region of Lavras - MG. For this, some methods of non-homogeneous marked and unmarked point processes will be used to analyze the first and second order effects, allowing an analysis of distribution, interaction and spatial dependence between points (trees) and marks (DAP and species). First-order effects (or global effects) characterize the expected number of occurrences per unit area (trend), while second-order effects (or small-scale local effects) characterize the spatial dependence structure of the process. Based on the results, it is possible to observe the potential of methods for analyzing punctual spatial processes to answer questions that help in the management, control and development of native tree species.