Métodos de amostragem para quantificar indivíduos adultos do pequi Caryocar brasiliense Cambess
Loading...
Files
Date
Journal Title
Journal ISSN
Volume Title
Publisher
Ciência Rural
Abstract
O objetivo do trabalho foi avaliar o uso da Amostragem Adaptativa Cluster (AAC) na quantificação de indivíduos adultos de Caryocar brasiliense Camb. (Pequi), em comparação aos métodos de amostragem tradicionais. Foi feito um censo com mapeamento dos indivíduos adultos de pequi em uma área de cerrado de 36,5ha no Parque Estadual do Rio Preto/MG. O mapa gerado foi divido em unidades de 20×20m no qual foram testadas cinco alternativas de amostragem utilizando a Amostragem Casual Simples, Amostragem Sistemática e Amostragem Adaptativa Cluster. A comparação entre elas foi feita através do teste F de Graybill a 1% de significância, considerando os parâmetros precisão e exatidão. Todas as alternativas apresentaram boa exatidão, sendo a Amostragem Adaptativa Cluster, com condição de inclusão igual a 2, a mais precisa.
In this study adult individuals of Caryocar brasiliense (pequi) were quantifi ed by (AAC) and they were compared the traditional sampling methods. Census of adult individuals of pequi in a savannah area of 36.5ha in Black River State Park / MG was transformed in map. Thus map was divided into cells of 20x20m where fi ve alternative of sampling were tested using simple Random Sampling, Systematic Sampling and Adaptative Cluster Sampling. Comparision among them was done by F test Graybill 1% of signifi cance, considering precision and accuracy as parameters. All alternatives showed good accuracy, and the Adapatative Cluster Sampling with a condition of inclusion equal to 2 was most accurate.
In this study adult individuals of Caryocar brasiliense (pequi) were quantifi ed by (AAC) and they were compared the traditional sampling methods. Census of adult individuals of pequi in a savannah area of 36.5ha in Black River State Park / MG was transformed in map. Thus map was divided into cells of 20x20m where fi ve alternative of sampling were tested using simple Random Sampling, Systematic Sampling and Adaptative Cluster Sampling. Comparision among them was done by F test Graybill 1% of signifi cance, considering precision and accuracy as parameters. All alternatives showed good accuracy, and the Adapatative Cluster Sampling with a condition of inclusion equal to 2 was most accurate.
