Fractal dimension and anisotropy of soil CO2 emission in a mechanically harvested sugarcane production area

Soil CO2 emissions (FCO2) are spatially dependent, and their spatial structure varies in different directions along the soil surface (anisotropy). This anisotropy, which can result from several pedological factors that are directly related to soil carbon dynamics, is affected by soil management. In this study, the anisotropies of the spatial variability of soil CO2 emissions and of other soil properties were determined for a sugarcane production area under mechanical harvest, when crop residues are left on soil surface, located in the northeastern part of the state of São Paulo, Brazil. The anisotropic characterization of variables was performed by deriving the fractal dimension (DF) from experimental semivariograms calculated at angles of 0°, 45°, 90° and 135° from the between crop line direction (0°). The mean FCO2 was 2.19 μmol m−2 s−1, and values were significantly lower in the 0° direction. A principal component analysis was applied to study soil properties and the first principal component was mainly related to soil physical properties and FCO2. A multiple regression analysis indicated that air-filled pore space (AFPS) was the main factor affecting the spatial variability of FCO2 in all directions. The AFPS DF values were significantly lower in the direction in which sugarcane crops were planted, indicating anisotropy of this property and greater homogeneity in this direction. Even after rainfall, there was no change in the structure of spatial variability as expressed by the values of DF. The results indicate that in sugarcane areas, several factors inherent to soil forming processes and management practices during harvest and seeding were responsible for the observed anisotropy, which affected soil CO2 emissions.

► The anisotropy of the spatial variability of soil CO2 emission was determined in a sugarcane area.
► The anisotropy was performed by deriving the fractal dimension in different scales and directions.
► The soil CO2 emissions were significantly lower in the crop line direction.
► Air-filled pore space was the main factor affecting the spatial variability of soil CO2 emission.
► Emissions were spatially dependent at a scale of 0.5 m in all directions and 2 m in crop line direction.