Representative elementary area for soil bulk density measurements of samples collected in volumetric rings by CT image analyses

• CT allowed a new insight about the representative elementary area (REA) evaluation.
• Representative area intervals (RAI) were defined by using CT and VR methods.
• 2D images permitted to evaluate detailed soil bulk density variations along depth.
• Three different schemes of REA and RAI evaluation were analyzed.

The objective of this study was to define soil bulk density (ρs) representative elementary areas (REA) of samples collected in volumetric rings. A new analysis procedure was proposed to define representative area intervals (RAI). A sandy clay loam (Geric Ferralsol – GF) and two clayey (Eutric Nitosol – EN and Rhodic Ferralsol – RF) soils were analyzed. Six samples of each soil were used to determine the ρs employing the volumetric ring method (VR) and computed tomography (CT). Tomography images were obtained with a first generation system equipped with a 241Am gamma-ray source. Consecutive areas from 1.2 to 678.8 mm2 were selected in the images, according to three different Schemes: concentric areas starting from the center (Scheme 1), the lower (Scheme 2) and the upper (Scheme 3) borders of the image. Two types of analyses were adopted to investigate the REA: considering CT (Analysis 1) and VR results (Analysis 2) as references. For the RAI, ρs results obtained via both methods were considered simultaneously (Analysis 3). The ρs REA from Analysis 1, according to Schemes 1, 2 and 3, were 432.0; 393.3 and 334.0 mm2 (GF), 473.1; 457.4 and 393.3 mm2 (RF) and 349.7; 457.4 and 334.0 mm2 (EN), respectively. It was not possible to define REA using Analysis 2, however, ideal areas established from the RAI coincided with the REA found for GF, according to the three Schemes, and for EN, according to Schemes 1 and 2. These results confirm the reliability of the new type of analysis proposed.