Variabilidade na amostragem de serrapilheira em floresta estacional decidual no município de Silveira Martins/RS

• Autor(a) principal: Pereira, Ane Caroline da Silva
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Manancial - Repositório Digital da UFSM
• dARK ID: ark:/26339/00130000028z8
• Texto Completo: http://repositorio.ufsm.br/handle/1/17295
• topic: Serapilheira; Floresta estacional decidual; Variância; Amostragem; Geoestatística; Litterfallfall; Seasonal decidual forest; Variance; Sampling; Geostatistics; CNPQ::CIENCIAS AGRARIAS::RECURSOS FLORESTAIS E ENGENHARIA FLORESTAL

The objective of this work was to study the origin of the variance in litterfall production and the main contributing factors. The study was carried out in a fragment of Seasonal Decidual Forest in the municipality of Silveira Martins/RS. Sampling was systematic with 117 litterfall collectors with dimensions 0.5 x 0.5 m and depth of 3 cm, spaced 7.5 x 7.5 m, distributed in a grid of 9 rows by 13 columns, totaling 5,400 m² of sample area. The experiment was installed in the year 2015, however, the litterfall collections for the present work were given during the year 2018. Information was collected on the forest species present in the area and the litterfall production in each collector, as well as its position of the X and Y coordinates in the sample grid. The data of litterfall production were obtained through 5 collections, carried out in: January, May, August, November and, December. Studies were carried out on vegetation composition, description of litterfall production, spatial dependence and investigations on the interaction between litterfall production with surface height and also with the species of higher frequency in the study area. The analysis of the forest composition was performed by calculating the relative frequency (FR%) of the species. The description of the litterfall production was obtained by means of the survey of the average, median, standard deviation, variance and coefficient of variation of each collector (ton / ha). In a complementary manner, analyzes of classical statistics were performed, involving normality tests, sample adequacy calculations, variance analysis, covariance analysis and mean tests. The spatial dependence of the litterfall from each collection was tested through geostatistical analyzes using semivariograms and kriging interpolation maps. The bivariate Ripley K Function was used to investigate the interaction of litterfall production with surface height and species with the highest frequency in the area. The results indicated Ocotea puberula, Cupania vernalis, Parapiptadenia rigida and Nectandra megapotamica as the most frequent species in the study area. Regarding spatial dependence, only collections 3 and 5 indicated a pattern of spatial dependence. Investigations on litterfall association factors indicated that areas of low surface height tend to produce the largest volumes of litterfall while areas of high surface height tend to accumulate the lowest volumes of litterfall. In relation to the association with the most frequent species, litterfall production did not present any association with them, which can be explained by this being a natural process that varies with several factors, so that the population as a whole forest is more important than the frequency of species, since each one presents specific characteristics when it comes to the production of the materials that compose the litterfall. It is recommended to sample at lower distances for studies of spatial dependence involving the average production of litterfall.