Análise espacial da vegetação arbórea em parcela de estepe estacional savanicola, no Parque Estadual do Espinilho, RS

• Autor(a) principal: Silveira, Cássio Thomas da
• Data de Publicação: 2018
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Manancial - Repositório Digital da UFSM
• Texto Completo: http://repositorio.ufsm.br/handle/1/13021
• topic: Floresta de precisão; Geoestatística; Kernel; Processo pontual marcado; Precision forest; Geostatistics; Punctual process marked; CNPQ::CIENCIAS AGRARIAS::AGRONOMIA

In the Forest sector, the Precision Forest (PF) approach seeks market space to assist forest management by optimizing available resources and spatially locating silvicultural practices, management and harvesting, aiming at increasing production in the same area and reducing the costs of based on spatial analysis of production and the factors influencing it. These spatial analyzes can also be used to better understand other types of vegetation, such as Espinilho, using the PF approach for this study. This work aims to combine the geotechnologies, with the Geographic Information System (GIS) and spatial analysis, to the forest inventory to describe the spatial variability and associations of the arboreal component in the Parque Estadual do Espinilho (PESP), an approach not yet used for this type of this vegetation. The objective was to create a spatial database and analyze it with geostatistics, kernel density functions and punctual process marked to verify the spatial behavior of native tree species found in a sample unit at PESP, located in Barra do Quaraí, Rio Grande do Sul, Brazil. The inventory data were obtained from a postdoctoral work by Watzlawick et al. (2014), arranged in printed field fiches. The GstarCAD, ArcGis, GS +, and RStudio software were designed for the design of the unit and subunits, shapefiles generation and density mapping, semivariogram adjustment and interpolation by punctual ordinary kriging and punctual process marked with the K function of Ripley. There were 578 trees distributed in three species, Acacia caven (Molina) Molina, Prosopis affinis Spreng., Prosopis nigra (Griseb.) Hieron and dead individuals. Semivariograms were adjusted and interpolations were generated for the total number of trees, species and P. affinis counted in the subunits. For these same variables and for A. caven and dead trees, Kernel functions maps were generated from their own coordinates. The marked punctual process showed that A. caven, P. nigra and dead trees presented complete spatial independence without some kind of structuring. Already P. affinis, with a larger number of individuals (552 - 95.50%), presented a grouped behavior among their individuals. In the punctual process marked categorical with two marks, used to verify if the interaction between species, no type of spatial dependence was found. However, P. affinis tends to repel A. caven and associate with dead individuals.