Simulação da condutividade hidráulica dos solos da sub-bacia do Rio Pardo – SP por meio de geoestatística multivariada
Carregando...
Data
2016-05-13
Autores
Nossack, Fábio Ávila [UNESP]
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
Os solos possuem alta variabilidade espacial devido aos diversos processos físicos, químicos e biológicos que atuam simultaneamente com diferentes intensidades na sua formação. O manejo uniforme dos solos não leva em conta tal variabilidade espacial, não sendo, portanto, a estratégia de manejo mais adequada. O uso do geoprocessamento, como ferramenta para o estudo da heterogeneidade dos solos, proporciona uma análise comportamental e uma visão espacial da realidade. Os fenômenos naturais são mais facilmente entendidos quando é possível a visualização da sua distribuição espacial e/ou temporal. Algumas das ferramentas que podem ser aplicadas para entender essas estruturas são os modelos geoestatísticos, em especial os modelos de simulação. Pela tentativa de reproduzir os mecanismos da realidade, esses modelos podem mostrar as características do solo em diferentes cenários ambientais. Pelos resultados das análises dos parâmetros físico hídricos da Sub-bacia da Cabeceira do Rio Pardo – SP, pode-se concluir: Os parâmetros que apresentaram maior variabilidade, correlação entre si e dependência espacial foram: teor de argila, umidade na capacidade de campo, umidade no ponto de murcha permanente e macroporosidade, que influenciam na condutividade hidráulica dos solos. A simulação sequencial direta de condutividade apresentou alto erro médio, o que indica o enviesamento dos dados e que o maior valor pode estar relacionado à pequena densidade amostral da área. A macroporosidade apresentou o melhor resultado e foi a covariável utilizada na 1 determinação das zonas de manejo. Foi verificada a viabilidade de utilização de covariáveis na determinação da condutividade hidráulica dos solos, o que acarreta em menor custo de amostragem geral para grandes áreas; constatou-se que 22,40 % da área são inaptas para a agricultura irrigada, 61,15% das áreas da sub-bacia são aptas e apresentaram classe média de permeabilidade e 16,45% das áreas também aptas apresentaram classe baixa. Foi comprovada a hipótese de que a simulação e co- simulação sequencial direta em ambiente multivariado descreveu o comportamento da permeabilidade dos solos na sub-bacia.
The soils have highly spatial variability due to various physical, chemical and biological processes that act simultaneously in their formation and with different intensities. Consequently, the uniform land management does not take into account such spatial variation, not being therefore more effective management strategy. The use of the GIS, as a tool for studying the dynamics of the soil, provides a behavioral analysis of spatial view and a soil reality. Natural phenomena are more easily understood when viewing the spatial distribution and/or time is possible. Some of the tools that can be applied to understand this dynamic are the geostatistical models, especially the simulation models. By trying to reproduce the mechanisms of fact, these models can show the soil behavior under different environmental scenarios. The results of the analysis of water physical parameters of the Sub-basin Headboard of Pardo River- SP, it can be concluded: The parameters that have greater variability, correlated and spatial dependence were clay content, humidity in the field capacity, moisture in point permanent wilting and macroporosity, that influence the hydraulic conductivity of the soil; direct sequential simulation of conductivity showed high mean error, indicating the bias of the data and that the greatest value may be related to the small sample density of the area; the macroporosity have had the best results anda was the co-factor chosen in the determination of management zones; it was verified the viability of the use of covariates in determining the hydraulic conductivity of the soil, resulting in lower overall cost for large sample areas; it was found that 22.4% of the area are unfit for irrigated agriculture, 61.15% of the areas of the sub-basin are suitable and showed middle permeability class and 16.45% of suitable areas also showed low class; the hypothesis that the simulation and co-simulation sequential directly, multivariable environment described the soil permeability behavior in sub-basin has been proven.
The soils have highly spatial variability due to various physical, chemical and biological processes that act simultaneously in their formation and with different intensities. Consequently, the uniform land management does not take into account such spatial variation, not being therefore more effective management strategy. The use of the GIS, as a tool for studying the dynamics of the soil, provides a behavioral analysis of spatial view and a soil reality. Natural phenomena are more easily understood when viewing the spatial distribution and/or time is possible. Some of the tools that can be applied to understand this dynamic are the geostatistical models, especially the simulation models. By trying to reproduce the mechanisms of fact, these models can show the soil behavior under different environmental scenarios. The results of the analysis of water physical parameters of the Sub-basin Headboard of Pardo River- SP, it can be concluded: The parameters that have greater variability, correlated and spatial dependence were clay content, humidity in the field capacity, moisture in point permanent wilting and macroporosity, that influence the hydraulic conductivity of the soil; direct sequential simulation of conductivity showed high mean error, indicating the bias of the data and that the greatest value may be related to the small sample density of the area; the macroporosity have had the best results anda was the co-factor chosen in the determination of management zones; it was verified the viability of the use of covariates in determining the hydraulic conductivity of the soil, resulting in lower overall cost for large sample areas; it was found that 22.4% of the area are unfit for irrigated agriculture, 61.15% of the areas of the sub-basin are suitable and showed middle permeability class and 16.45% of suitable areas also showed low class; the hypothesis that the simulation and co-simulation sequential directly, multivariable environment described the soil permeability behavior in sub-basin has been proven.
Descrição
Palavras-chave
Análise multivariada, Análise espacial, Simulação sequencial direta, Condutividade hidráulica, Multivariate analysis, Spatial analysis, Direct sequential simulation, Hydraulic conductivity