Predição de fósforo adsorvido utilizando a suscetibilidade magnética em solos sob cultivo de cana-de-açúcar
Carregando...
Data
2018-08-01
Autores
Batista, Katharine Viana
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
O solo submetido as operações agrícolas intensa e repetitivas RESUMO - O fósforo é considerado um dos nutrientes que limitam o desenvolvimento da cultura de cana-de-açúcar em solos brasileiros. Além disso, apresenta grande variabilidade espacial devido aos atributos que influenciam sua adsorção ao solo. O objetivo do trabalho foi mapear o fósforo adsorvido (Pads)com o auxílio da susceptibilidade magnética (SM) e definir a proporção amostral para a caracterização da dependência espacial do Pads. Foram coletadas 371 amostras na profundidade de 0-0,25 m com separação entre as amostras de 145 m a 174 m, abrangendo uma área total de aproximadamente 870 ha. Em cada ponto amostral determinou-se a suscetibilidade magnética (SM), o fósforo adsorvido (Pads). Os dados foram analisados a partir da estatística descritiva e da geoestatística. Na utilização da técnica da cokrigagem de Pads com o auxílio da SM sendo que uma a cada 2.6, assim foram estabelecidas as seguintes proporções amostrais (3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 ha) e foram obtidos os mapas de cokrigagem. A acurácia das estimativas foi realizada utilizando a validação externa e a Raiz Quadrada do Erro Médio (RMSE). Os valores de alcance para as variáveis estudadas variaram de 1861,78 a 2081,68m. A partir desse resultado conclui-se que o Pads apresenta dependência espacial com a SM, indicando seu uso no mapeamento e na identificação de áreas com diferentes potenciais de adsorção de fósforo, e interpolação com co-krigagem utilizando a SM funciona até a proporção amostral 6 pontos por hectare, ou seja, para cada ponto de Fósforo adsorvido pode-se utilizar até 5 pontos de suscetibilidade magnética.
Phosphorus is considered one of the nutrients that limit the development of sugarcane cultivation in Brazilian soils. In addition, it is an element that presents great spatial variability due to the attributes that influence it is adsorption to the soil. The objective of this study was to map the adsorbed phosphorus with the aid of magnetic susceptibility and to define the sample proportion for the characterization of the spatial dependence of the Pads. A total of 371 samples were collected at a depth of 0-0.25 m with a separation between samples of 145 m and 174 m, covering a total area of approximately 870 ha. At each sampling point the magnetic susceptibility (SM), the adsorbed phosphorus (Pads) was determined. Data were analyzed from descriptive statistics and geostatistics. In the use of the cokriging technique of Pads with the aid of SM, one of each being 2.6, the following sample proportions were established (3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 ha), and the cokriging maps were obtained. The accuracy of the estimates was performed using the external validation and the Square Root of the Mean Error (RMSE). The range values for the studied variables ranged from 1861.78 to 2081.68m. From this result, it is concluded that the Pads presents spatial dependence with SM, indicating their use in mapping and identifying areas with different phosphorus adsorption potentials, and interpolation with co-kriging using the SM works up to the sample proportion 6 points per hectare, that is, for each point of adsorbed phosphorus can be used up to 5 points of magnetic susceptibility.
Phosphorus is considered one of the nutrients that limit the development of sugarcane cultivation in Brazilian soils. In addition, it is an element that presents great spatial variability due to the attributes that influence it is adsorption to the soil. The objective of this study was to map the adsorbed phosphorus with the aid of magnetic susceptibility and to define the sample proportion for the characterization of the spatial dependence of the Pads. A total of 371 samples were collected at a depth of 0-0.25 m with a separation between samples of 145 m and 174 m, covering a total area of approximately 870 ha. At each sampling point the magnetic susceptibility (SM), the adsorbed phosphorus (Pads) was determined. Data were analyzed from descriptive statistics and geostatistics. In the use of the cokriging technique of Pads with the aid of SM, one of each being 2.6, the following sample proportions were established (3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 ha), and the cokriging maps were obtained. The accuracy of the estimates was performed using the external validation and the Square Root of the Mean Error (RMSE). The range values for the studied variables ranged from 1861.78 to 2081.68m. From this result, it is concluded that the Pads presents spatial dependence with SM, indicating their use in mapping and identifying areas with different phosphorus adsorption potentials, and interpolation with co-kriging using the SM works up to the sample proportion 6 points per hectare, that is, for each point of adsorbed phosphorus can be used up to 5 points of magnetic susceptibility.
Descrição
Palavras-chave
Geoestatística, Krigagem ordinária, proporção amostral