ANÁLISE MULTIFRACTAL APLICADA A SISTEMAS AMBIENTAIS DO ESTADO DO MARANHÃO (BRASIL)

Detalhes bibliográficos
Ano de defesa: 2021
Autor(a) principal: SILVA, Suena dos Santos lattes
Orientador(a): SIQUEIRA, Glécio Machado lattes
Banca de defesa: SIQUEIRA, Glécio Machado lattes, RODRIGUEZ LEIVA, Jairo Osvaldo lattes, AQUINO JÚNIOR, José lattes
Tipo de documento: Dissertação
Tipo de acesso: Acesso aberto
Idioma: por
Instituição de defesa: Universidade Federal do Maranhão
Programa de Pós-Graduação: PROGRAMA DE PÓS-GRADUAÇÃO EM GEOGRAFIA (PPGGEO)
Departamento: DEPARTAMENTO DE GEOCIENCIAS/CCH
País: Brasil
Palavras-chave em Português:
Palavras-chave em Inglês:
Área do conhecimento CNPq:
Link de acesso: https://tedebc.ufma.br/jspui/handle/tede/4106
Resumo: The Cerrado Biome is a heterogeneous system, with different phytophysiognomies and in anthropogenic areas, different uses are verified in the landscape. Given the above, it is necessary to understand the complexity of the Cerrado through mathematical models that consider scale variability as multifractal analysis. In multifractal analysis, variability is evaluated using more than one exponent to describe the properties of scales, thus allowing to characterize the behavior or structures of objects, and phenomena in a temporal or spatial series. Thus, the aim of this work is to characterize the multifractality of different systems in the Cerrado Biome of Maranhão (Brazil). The specific objectives are: 1) To analyze the multifractality of soil water content and soybean yield in a no-tillage system (Mata Roma, Maranhão, Brazil); 2) Determine the reflectance multifractality (NDVI) of plots with different vegetation formations in the Cerrado at Parque Estadual do Mirador (Maranhão, Brazil). The studied systems were: soybean production area (Glycine max L.) in the municipality of Mata Roma (Maranhão, Brazil); and experimental plots with different Cerrado vegetation types in Parque Estadual do Mirador (PEM – Municipality of Mirador, Maranhão, Brazil). The experiment in soybean cultivation was carried out in a commercial production area of the Unha de Gato Farm with 193.6 ha and has been managed under no- tillage system with soybean cultivation since 2004. In a transect containing 128 points, with spacing of 10 m between points and total length of 1270 m. Altitude (m), soybean yield (kg ha-1), plant height (m) and water volumetric moisture (%) measured by TLO - Transmisión Line Oscillator were sampled. Subsequently, the data were evaluated using descriptive statistics and multifractal analysis. In the Mirador State Park, three plots were installed in an area with natural Cerrado vegetation, comprising the following vegetation formations: thin Cerrado (Plot A), transition between thin Cerrado and dense Cerrado (Plot B) and typical Cerrado (Plot C). The experimental plots have 26,214.4 m2 (5,120 m x 5,120 m), being determined in altitude (m) and the NDVI (Normalized Difference Vegetation Index), which were evaluated by descriptive statistics and multifractal analysis. The main conclusions of this work are: 1) The multifractal analysis showed that the height of the soybean plants along the transept is the attribute with the greatest heterogeneity among the evaluated variables, expressed by the difference between the generalized dimensions (D5) and (D-5 ). 2) The singularity spectra for the series of variables under study expressed greater branch amplitude and width for height of soybean plants, corroborating the multifractal properties of generalized dimension that also indicated the greatest multifractality for the height of soybean plants, when compared to altitude (m), θ0-0.2 (%), θ0.2-0.4 (%), θ0-0.4 (%) and productivity (kg ha-1). 3) The soil water content measured in the layers (θ0-0.2 and θ0.2- 0.4) and the average soil water content (θ0-0.4) showed reduced spectra, indicating that they correspond to a monofractal or quasimonofractal. 4) The data showed that the NDVI in different Cerrado gradients has multifractal behavior in the landscape. 5) The vegetation indices showed variability in the scales in the study period (2015-2019), indicating that terrain features with less slope result in indices with lower multifractality. 6) The plot with typical Cerrado showed the greatest heterogeneity and asymmetry, reflecting the environmental complexity of this vegetation physiognomy.