Banca de QUALIFICAÇÃO: ANTONIO ANIZIO LEAL MACEDO NETO

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
DISCENTE : ANTONIO ANIZIO LEAL MACEDO NETO
DATA : 07/12/2018
HORA: 09:00
LOCAL: Auditório da Pós-Graduação (Setor de Solos)
TÍTULO:

VARIABILITY OF SOIL FERTILITY, NUTRITION AND PRODUCTIVITY IN TEAK CULTIVATION IN NORTHEAST OF PARÁ

PALAVRAS-CHAVES:

Tectona grandis L.f, Precision forestry, management zones.

PÁGINAS: 44
GRANDE ÁREA: Ciências Agrárias
ÁREA: Agronomia
SUBÁREA: Ciência do Solo
ESPECIALIDADE: Fertilidade do Solo e Adubação
RESUMO:

Tectona grandis L.f is the most important economic species of the genus Tectona. Due to the reduction of its availability of natural form, there was increase in its production of planted form. An alternative to homogeneous management is the varied management, which can be implemented through precision forestry. A widely used precision tool is geostatistics. As this tool generates many maps, one way to simplify them is multivariate statistics, through principal component analysis. With the main components, it is possible to determine the management zones of an area through the algorithm "Fuzzy c-means". The objective of this work is to generate areas for the management of a teak crop with the help of multivariate statistics and geostatistics, in order to contribute to the fertilization recommendation. The work was carried out at the São Luiz farm, located in the municipality of Captão Poço, which is approximately 75 km from the municipality's headquarters. The farm has grown teak (Tectona grandis) for 20 years. A systematic sampling grid of 155 geo-referenced points was used to collect soil samples and yield. The collected samples were placed in plastic bags, identified and sent to the laboratory, for determination in the soil of pH, Ca, Mg, K, SB, Al, H + Al, P melich, B, and for the calculation of the variables ; M.O, effective CTC, CTC pH 07, m%, v%, samples of DBH, Height and volume were also determined. It was possible to fit most of the variables of the study into one of the most used semivariogram models in precision forestry. After the semivariograms were adjusted, the kriging maps of the isolated variables were made. Through multivariate statistics, the first three main components were selected, representing 79.58% of the data variability. Geostatistical analyzes were also performed on the main components. The three selected components were used to determine the management zones, through the algorithm "Fuzzy c-means". Three were the number of zones suitable for work. After the determination of the zones it was possible to relate each of the 155 observations of the soil variables to one of the zones. Zone 1 corresponded to 9 of the samples, zone 2 to 43 and zone 3 to 103, then a variance analysis was performed to obtain the means of the soil variables in each of the zones and, afterwards, these averages were submitted tukey test so that it is possible to see if they are statistically different. The application of the cluster analysis showed different results from the homogeneous analysis. Zone 01 was the one that showed the best soil fertility results. Through cluster analysis it is possible to obtain results not shown by univariate statistics.

MEMBROS DA BANCA:
Externo ao Programa - 1492115 - ANTONIO VINICIUS CORREA BARBOSA
Presidente - 2874082 - GILSON SERGIO BASTOS DE MATOS
Externo à Instituição - JAVIER DIAS PITA - IFPA
Interno - 2121042 - SUZANA ROMEIRO ARAUJO