Maria Gabriela Ferreira da Mata, Departamento de Solos, Universidade Federal Rural do Rio de Janeiro, Seropedica, Brazil, Marcos Bacis Ceddia Sr., Soil Department, Federal Rural University of Rio de Janeiro, Seropédica, Brazil, Jose Guilherme Marinho Guerra, Agricultura Orgânica, Embrapa Agrobiologia, Seropedica, Brazil, Erika Pinheiro, Instituto de Agronomia - Departamento de Solos, Universidade Federal RURAL do Rio de Janeiro, Seropédica - RJ, Brazil and Ole Wendroth, N-122M Ag Science N., University of Kentucky, Lexington, KY
Spatial variability of soil organic matter content and its temporal behavior are not well understood at the field scale. The purpose of this study was to provide insights how soil organic carbon content changes during several years, and whether its spatial distribution remains rather stable or variable. The experiment was carried out at Seropédica city, Rio de Janeiro State, Brazil in a 1 ha organic vegetable production area. This area was subdivided into two areas, one (area 1) with vegetables being grown and the other (area 2) being used for biomass production for organic material applications in the first area. A regular 10-m grid was laid out for monitoring soil organic carbon (SOC) resulting in 131 sampling locations. Soil samples were obtained from 0 – 0.20 meters depth from year 2010 until 2014. The highest values of SOC were found in 2010 that were statistically different from subsequent years. The SOC showed a trend to increase in area 1 in subsequent years, despite not achieving the initial levels. The field average SOC decreased in 2011 and remained stable at subsequent sampling campaigns. Besides their decreases and increases, observations of SOC were regionalized and exhibited spatial structure. Therefore, semivariograms were modeled for each of the sampling periods. Due to the addition of biomass from area 2 SOC was not related to the spatial distribution of texture in area 1. However in area 2 SOC and soil texture were spatially related as manifested in their crossvariograms. Their common covariance structure allowed the creation of maps using ordinary kriging and cokriging. Co-regionalization based on multivariate geostatistics was helpful to evaluate the field-scale SOC dynamics and their temporal behavior.