Design, construction and evaluation of a miniature soil bin plus predicting the measured parameters during primary tests using ANFIS

A clear understanding of soil-machine interaction is utilised in many areas, such as rational design and performance optimisation of soil-engaging tools/implements. This research developed a new soil bin to investigate the interaction between soil-narrow tines and soil failure. The new soil bin consisted of a chassis, a bin, a variable speed carriage and a bolt and nut type power transmission system between the motor and carriage. The design criteria of the new soil bin were based on the drive system's immediate acceleration of the carriage unit. A curved chisel tine was tested to evaluate the system's capabilities. Three parameters, including lift height, failure side area and forward failure distance, were investigated at two forward speeds of 0.037 and 0.05 m/s and three rake angles of 5°, 10° and 20°. An analysis of variance (ANOVA) test revealed the effect of rake angle on the lift height, failure side area, and forward failure distance was significant (P < 0.01). However, the forward speed did not have any significant influence on parameters. Also, the lift height and failure side area increased significantly (P < 0.01) by increasing the rake angle, while the forward failure distance decreased. Regarding soil failure, the results were in harmony with Godwin and Spoor's model. The regression and ANFIS models were developed to predict the output parameters by considering input parameters. The R2 values and ANFIS models were 0.4895, 0.7264, 0.9856, and 0.9999, 1, 1, respectively, for lift height, side area, and forward distance. Therefore, ANFIS approach was more accurate for predicting soil failure parameters.

Soil parameters affecting the levels of potentially harmful metals in Thessaly area, Greece: a robust quadratic regression approach of soil pollution prediction

The behavior and possible contamination risk due to the presence of potentially harmful metals (PHM) were studied based on 2250 soil samples that were collected in a 5-year period (2013-2017) from the plain of Thessaly (prefectures of Karditsa, Trikala, and Larissa). The vertical distribution of metals was also investigated from sample profiles at three depths 0-30, 30-60, and 60-90cm. The soils of the sampling belong to four taxonomy soil orders that are dominant in the studied area (Alfisols, Inceptisols, Endisols, and Vertisols). In a novel approach, robust quadratic regression analysis on multiple variables was used to define prediction models of the concentrations of two metals: Fe which is an essential metal and the toxic Cd. Linear and quadratic regression formulae were estimated based on the iteratively reweighted least squares robust regression approach in an effort to eliminate the impact of the outliers. These formulae define how several soil properties affect the distribution of the considered metals in each soil order. The evaluation of the estimated regression equations based on the R² metric indicates that they constitute a useful, reliable, and valuable tool for managing, describing, and predicting the pollution in the studied area.

Quantitative estimation of soil properties using hybrid features and RNN variants

Estimating soil properties is important for maximizing the production of crops in sustainable agriculture. The hyperspectral data next input depends upon the previous one, and the current techniques do not take advantage of this sequential nature of hyperspectral signatures. The variants of RNN can learn the short-term and long-term dependencies between data. This paper proposes a deep learning hybrid framework for quantifying the soil minerals like Clay, CEC, pH of H₂O, Nitrogen, Organic Carbon, Sand of European Union from the LUCAS library. The hyperspectral signatures contain the data in the range of 400-2500 nm captured from the FOSS spectroscope in the laboratory. As hyperspectral data is high dimensional, Principal Component Analysis and Locality Preserving Projections are utilized to form the hybrid features, which have low dimensions containing the local and global information of the original dataset. These hybrid features are passed on to Long Short Term Memory Networks, a deep learning framework for building an effective prediction model. The effectiveness of the prepared models is demonstrated by comparing it to existing state-of-the-art techniques.

Pollution assessment of potentially toxic elements in soils of different taxonomy orders in central Greece

Four hundred fifty soil samples of the orders of Alfisols, Inceptisols, Endisols, and Vertisols from Karditsa, Trikala, and Larissa (Central Greece) were collected over a three-year period. In these samples we analyzed potentially toxic elements (PTEs) and soil properties known to affect their mobility. High regression coefficients were observed between soil pH and PTE concentrations in Alfisols, reflecting that soil pH is the dominant characteristic influencing PTEs. In Inceptisols, there was a significant interaction among the studied PTEs, probably due to PTEs having the same origin. The Endisol samples had high sand content and electrical conductivity values, resulting in high availability of all studied PTEs. In Vertisols, clay content proved to be the most important parameter influencing PTE levels. Factor analysis was also used in order to clarify the possible sources of metals in the studied areas.

Spatial and vertical distribution of mercury in upland forest soils across the northeastern United States

Assessing current Hg pools in forest soils of the northeastern U.S. is important for monitoring changes in Hg cycling. The forest floor, upper and lower mineral horizons were sampled at 17 long-term upland forest sites across the northeastern U.S. in 2011. Forest floor Hg concentration was similar across the study region (274 ± 13 μg kg(-1)) while Hg amount at northern sites (39 ± 6 g ha(-1)) was significantly greater than at western sites (11 ± 4 g ha(-1)). Forest floor Hg was correlated with soil organic matter, soil pH, latitude and mean annual precipitation and these variables explained approximately 70% of the variability when multiple regressed. Mercury concentration and amount in the lower mineral soil was correlated with Fe, soil organic matter and latitude, corresponding with Bs horizons of Spodosols (Podzols). Our analysis shows the importance of regional and soil properties on Hg accumulation in forest soils.

Sorption kinetics of diuron on volcanic ash derived soils

Diuron sorption kinetic was studied in Andisols, Inceptisol and Ultisols soils in view of their distinctive physical and chemical properties: acidic pH and variable surface charge. Two types of kinetic models were used to fit the experimental dates: those that allow to establish principal kinetic parameters and modeling of sorption process (pseudo-first-order, pseudo-second-order), and some ones frequently used to describe solute transport mechanisms of organic compounds on different sorbents intended for remediation purposes (Elovich equation, intraparticle diffusion, Boyd, and two-site nonequilibrium models). The best fit was obtained with the pseudo-second-order model. The rate constant and the initial rate constant values obtained through this model demonstrated the behavior of Diuron in each soil, in Andisols were observed the highest values for both parameters. The application of the models to describe solute transport mechanisms allowed establishing that in all soils the mass transfer controls the sorption kinetic across the boundary layer and intraparticle diffusion into macropores and micropores. The slowest sorption rate was observed on Ultisols, behavior which must be taken into account when the leaching potential of Diuron is considered.