Understanding the field‐scale variability of CO2 fluxes in space and time under different land‐use conditions is important for designing efficient sampling schemes for monitoring C loss from soil. The objectives of this study were to characterize the variability structure of CO2 fluxes and its change throughout a year, evaluate the temporal stability of spatial CO2 flux patterns, and quantify relationships with biotic and abiotic factors in crop and grass systems. In an 80‐ by 60‐m field divided into grass and crop systems, CO2 fluxes were measured with a photo‐acoustic analyzer at 60 locations 22 times during a year. Soil respiration was high when it was warm and wet and low under cool and wet conditions. Measurements of CO2 flux were spatially and temporally dependent for a longer distance and time in the grass system than in the crop system. Spatial patterns of soil respiration were temporally stable, and the larger the average spatial CO2 flux, the more obvious was the temporal stability. The variation of soil respiration was more pronounced in time than in space during the study period. At each location, soil temperature was the major factor controlling the temporal variability of CO2 flux; however, soil temperature did not explain the spatial CO2 flux pattern. Land use impacted the spatial and temporal variability dynamics of CO2 flux. These dynamics should be taken into account for experimental design, the detection of spatial and temporal associations with other variables, and C loss prediction.

Core Ideas Soil gas transport variables were greater in pasture than crop systems. Gas transport variables exhibited high spatial variation with higher CVs in the crop. Spatial variability of Ds/D0 was structured at all selected matric potentials. Ds/D0 revealed stronger spatial dependence in the crop and longer correlation in pasture. Spatial correlation length of Ds/D0 in the crop was controlled by soil water status. Soil gas diffusivity and air‐filled porosity are important soil variables indicating soil structure and aeration. Land use influences these soil gas transport variables, but their spatial patterns have not been sufficiently investigated. This study evaluated how land use affected soil gas diffusivity, air‐filled porosity, and pore continuity and quantified the spatial patterns of these soil gas transport variables in crop and pasture systems. Soil gas diffusivity was measured with a gas diffusion chamber as a function of air‐filled porosity in 60 soil cores taken from the two land‐use systems. Soil gas diffusivity, air‐filled porosity, and pore continuity were larger in the pasture than the crop system, indicating better soil aeration and a better developed soil structure, as observed under pasture relative to the crop system. Soil gas transport variables exhibited high spatial variability in both land‐use systems, with smaller CVs in the pasture than the crop system. Large spatial variability reflects the complexity of pore system organization at the sample scale and the possibility that the representative elementary volume was not captured at the sampling scale. Geostatistical analysis showed unique structured variability of these soil gas transport variables at the field scale, with longer correlation lengths in the pasture system and stronger spatial dependence in the crop system for soil gas diffusivity. The spatial patterns of soil gas transport variables and the influence of land use should be considered for experimental design and predictions of soil gas diffusivity.

Core Ideas Spatial variability of in situ Ks, K−1, K−5, and K−10 was decomposed into different scales using NA‐MEMD. Scale‐dependent relationships were observed for each K with six soil properties. Incorporating ANN for small‐scale variability of each K improves the estimation quality. Soil hydraulic conductivity near saturation (Kns) is affected by various soil properties operating at different spatial scales. Using noise‐assisted multivariate empirical mode decomposition (NA‐MEMD), our objective was to inspect the scale‐dependent interactions between Kns and various soil properties and to estimate Kns based on such relationships. In a rectangular field evenly across cropland and grassland, a total of 44 sampling points separated by 5 m were selected and measured for Kns at soil water pressure heads of −1, −5 and −10 cm. At each point, the saturated conductivity Ks was estimated using Gardner's exponential function, and six soil structural and textural properties were investigated. Decomposed into four intrinsic mode functions (IMFs) and a residue by NA‐MEMD, each K was found to significantly correlate with all six properties at one spatial scale at least. The variations in K were primarily regulated by soil structure, especially at the relatively small scales. Multiple linear regression (MLR) failed to regress either IMF1 or IMF2 of each K from the soil properties of the equivalent scales and only accounted for 13.7 to 43.6% of the total variance in calibration for the remaining half of the IMF1s and IMF2s. An artificial neural network was then adopted to estimate IMF1 and IMF2, and the corresponding results were added to the MLR estimates at other scales for which each K was estimated at the measurement scale. This prediction greatly outperformed the MLR modeling before NA‐MEMD and, on average, accounted for additional 74.4 and 73.4% of the total variance in calibration and validation, respectively. These findings suggest nonlinear correlations between K and the soil properties investigated at the small scales and hold important implications for future estimations of Kns and Ks as well as other hydraulic properties.