Development and testing of vehicle-mounted soil bulk density detection system. (April 2023)

• Record Type: Journal Article
• Authors: Meng, Chao; Yang, Wei; Wang, Dong; Hao, Ziyuan; Li, Minzan
• Is Part Of: Measurement. Volume 211(2023)
• Publication Date: 2023-04
• Subjects: Soil bulk density -- Electrical conductivity -- Soil mechanical resistance -- Vehicle-mounted detection system; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2023.112604
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: A vehicle mounted soil bulk density detection system was designed and developed. The entire process of measuring a single sampling point takes less than 10 s. Multivariate data fusion can accurately estimate soil bulk density. Soil mechanical resistance contributes the most to the estimation of soil bulk density. The system can also evaluate the degree of soil compactness. Abstract: Soil bulk density (BD) reflects the soil's ability to function for structural support, water and solute movement, and soil aeration. It is used to express soil physical and biological measurements on a volumetric basis for soil quality assessment and comparisons between management systems in precision agriculture. In order to solve the problem that it is difficult to directly obtain the BD with large sample size, this study introduces a self-developed vehicle-mounted BD detection system, which is a composite sensor. Through the collection and analysis of soil electrical conductivity (ECa), soil mechanical resistance (MR) and soil surface images (SIMG), BD is estimated. A systematic verification test was carried out in the experimental field in North China, the study depth was 0.2 m. The R 2 and RMSE between the estimated BD and actual BD measured by core sampling were 0.647 and 0.173 (using Multiple Linear Regression), 0.591 and 0.160 (using Ridge Regression), 0.719 and 0.090 (using Gradient Boosting Decision Tree). Soil moisture content (SMC) was included into the model of BD and R 2 of the model was increased from 0.719 to 0.775, RMSE was reduced from 0.090 to 0.073. The system can estimate the degree of soil compactness with R 2 = 0.708 and RMSE = 0.049. SMC can also be estimated from ECa and MR (R 2 = 0.481, RMSE = 1.658). These results show the potential of the vehicle-mounted BD detection system, which can provide the basis for fast, economical, and efficient soil physical property analysis.