Prediction of the concentration of antimony in agricultural soil using data fusion, terrain attributes combined with regression kriging.

Potentially toxic elements in agricultural soils are primarily derived from anthropogenic and geogenic sources. This study aims to predict and map antimony (Sb) concentration in soil using multiple regression kriging in two distinct modeling approaches, namely Sb prediction using data fusion coupled with regression kriging (scenario 1) and Sb prediction using data fusion, terrain attributes, and regression kriging (scenario 2). Cubist regression kriging (cubist_RK), conditional inference forest regression kriging (CIF_RK), extreme gradient boosting regression kriging (EGB_RK) and random forest regression kriging (RF_RK) were the modeling techniques used in the estimation of Sb concentration in agricultural soil. The validation results suggested that in scenario 1, EGB_RK was the optimal modeling approach for Sb prediction in agricultural soil with root mean square error (RMSE) = 1.31 and mean absolute error (MAE) = 0.61, bias = 0.37, and high coefficient of determination R2 = 0.81. Similarly, the EGB_RK was also the optimal modeling approach in scenario 2, with the highest R2 = 0.76, RMSE = 0.90, bias = 0.06, and MAE = 0.48 values than the other regression kriging modeling approaches. The cumulative assessment suggested that the EGB_RK in scenario 2 yielded optimal results compared to the respective modeling approach in scenario 1. The uncertainty propagated by the modeling approaches in both scenarios indicated that the degree of uncertainty during the modeling process was distributed across the study area from a low to a moderate uncertainty level. However, cubist_RK in scenario 2 exhibited some elevated spots of uncertainty levels. As a result, the combination of data fusion, terrain attributes, and regression kriging modeling approaches produces optimal results with a high R2 value, minimal errors as well as bias. Furthermore, combining terrain attributes with data fusion is promising for reducing model error, bias and yielding high-accuracy predictions.

Optimal zinc level and uncertainty quantification in agricultural soils via visible near-infrared reflectance and soil chemical properties.

Zinc (Zn) is a vital element required by all living creatures for optimal health and ecosystem functioning. Therefore, several researchers have modeled and mapped its occurrence and distribution in soils. Nonetheless, leveraging model predictive performances while coupling information derived from visible near-infrared (Vis-NIR) and soils (i.e. chemical properties) to estimate potential toxic elements (PTEs) like Zn in agricultural soils is largely untapped. This study applies two methods to rapidly monitor Zn concentration in agricultural soil. Firstly, employing Vis-NIR and machine learning algorithms (MLAs) (Context 1) and secondly, applying Vis-NIR, soil chemical properties (SCP), and MLAs (Context 2). For the Vis-NIR information, single and combined pretreatment methods were applied. The following MLAs were used: conditional inference forest (CIF), partial least squares regression (PLSR), M5 tree model (M5), extreme gradient boosting (EGB), and support vector machine regression (SVMR) respectively. For context 1, the results indicated that M5-MSC (M5 tree model-multiplicative scatter correction) with coefficient of determination (R2) = 0.72, root mean square error (RMSE) = 21.08 (mg/kg), median absolute error (MdAE) = 13.69 and ratio of performance to interquartile range (RPIQ) = 1.63 was promising. Regarding context 2, CIF with spectral pretreatment and soil properties [CIF-DWTLOGMSC + SCP (conditional inference forest-discrete wavelet transformation-logarithmic transformation-multiplicative scatter correction-soil chemical properties)] yielded the best performance of R2 = 0.86, RMSE = 14.52 (mg/kg), MdAE = 6.25 and RPIQ = 1.78. Altogether, for contexts 1 and 2, the CIF-DWTLOGMSC + SCP approach (context 2) was the best Zn model outcome for the agricultural soil. The uncertainty map revealed a low to high error distribution in context 1, and a low to moderate distribution in context 2 for all models except CIF, which had some patches with high uncertainty. We conclude that a multiple optimization approach for modeling Zn levels in agricultural soils is invaluable and may provide fast and reliable information needed for area-specific decision-making.

Combination of enrichment factor and positive matrix factorization in the estimation of potentially toxic element source distribution in agricultural soil.

The study intended to assess the level of pollution of potential toxic elements (PTEs) at different soil depths and to evaluate the source contribution in agricultural soil. One hundred and two soil samples were collected for both topsoil (51), and the subsoil (51) and the content of PTEs (Cr, Cu, Cd, Mn, Ni, Pb, As and Zn) were determined using inductively coupled plasma-optical emission spectroscopy (ICP-OES). The concentrations of Zn and Cd in both soil horizons indicated that the current study levels were higher than the upper continental crust (UCC), world average value (WAV), and European average values (EAV). Nonetheless, the concentration values of PTEs such as Mn and Cu for EAV, As, Cu, Mn, and Pb for UCC, and Pb for WAV were lower than the average values of the corresponding PTEs in this study. The single pollution index, enrichment factor, and ecological risk revealed that the pollution level ranged from low to high. The pollution load index, Nemerow pollution index, and risk index all revealed that pollution levels ranged from low to high. The spatial distribution confirmed that pollution levels varied between the horizons; that is, the subsoil was considered slightly more enriched than the topsoil. Principal component analysis identified the PTE source as geogenic (i.e. for Mn, Cu, Ni, Cr) and anthropogenic (i.e. for Pb, Zn, Cd, and As). PTEs were attributed to various sources using enrichment factor-positive matrix factorization (EF-PMF) and positive matrix factorization (PMF), including geogenic (e.g. rock weathering), fertilizer application, steel industry, industrial sewage irrigation, agrochemicals, and metal works. Both receptor models allotted consistent sources for the PTEs. Multiple linear regression analysis was applied to the receptor models (EF-PMF and PMF), and their efficiency was tested and assessed using root-mean-square error (RMSE), mean absolute error (MAE), and R2 accuracy indicators. The validation and accuracy assessment of the receptor models revealed that the EF-PMF receptor model output significantly reduces errors compared with the parent model PMF. Based on the marginal error levels in RMSE and MAE, 7 of the 8 PTEs (As, Cd, Cr, Cu, Ni, Mn, Pb, and Zn) analysed performed better under the EF-PMF receptor model. The EF-PMF receptor model optimizes the efficiency level in source apportionment, reducing errors in determining the proportion contribution of PTEs in each factor. The purpose of building a model is to maximize efficiency while minimizing inaccuracy. The marginal error limitation encountered in the parent model PMF was circumvented by EF-PMF. As a result, EF-PMF is feasible and useful for apparently polluted environments, whether farmland, urban land, or peri-urban land.

Can in situ spectral measurements under disturbance-reduced environmental conditions help improve soil organic carbon estimation?

In situ visible and near-infrared (Vis-NIR) spectroscopy has proven to be a reliable tool for determining soil organic carbon (SOC) content with a small loss of precision as compared to laboratory measurements. The loss of precision is a result of disturbing external environmental factors that disrupt spectral measurements. For example, roughness, changes in weather conditions, humidity, temperature, human factors, spectral noise and especially soil water. It has been assumed that, in situ predictive capability could be improved if some of these factors are either minimized or eliminated during the in situ measurement. For this study, the prediction of SOC was carried out under two different in situ measurement conditions; less favourable environmental conditions (with disturbances) and more favourable site-specific conditions (disturbance-reduced conditions). The primary goal is to determine whether the estimate of SOC can be improved under more favourable site-specific conditions, as well as the impact of pre-treatment algorithms on both less and more favourable disturbed conditions. The study employed a large range of pretreatment algorithms and their combinations. Three separate multivariate models were used to predict SOC, namely Cubist, support vector machine regression (SVMR), and partial least squares regression (PLSR). The result clearly shows that reduced disturbing factors (i.e., drier and unploughed soil as well as noise reduction) result in an improvement of SOC prediction with in situ Vis-NIR spectroscopy. The best overall result was achieved with SVMR (R2CV = 0.72, RMSEPcv = 0.21, RPIQ = 2.34). Although the combination of pre-treatment algorithms resulted in an improvement, overall, these pre-treatment algorithms could not compensate for the factors affecting the measured spectra with disturbance. Though the obtained result is promising, further study is still needed to disentangle the impacts and interactions of various disturbing factors for different soil types.

Prediction of nickel concentration in peri-urban and urban soils using hybridized empirical bayesian kriging and support vector machine regression.

Soil pollution is a big issue caused by anthropogenic activities. The spatial distribution of potentially toxic elements (PTEs) varies in most urban and peri-urban areas. As a result, spatially predicting the PTEs content in such soil is difficult. A total number of 115 samples were obtained from Frydek Mistek in the Czech Republic. Calcium (Ca), magnesium (Mg), potassium (K), and nickel (Ni) concentrations were determined using Inductively Coupled Plasma Optical Emission Spectroscopy. The response variable was Ni, while the predictors were Ca, Mg, and K. The correlation matrix between the response variable and the predictors revealed a satisfactory correlation between the elements. The prediction results indicated that support vector machine regression (SVMR) performed well, although its estimated root mean square error (RMSE) (235.974 mg/kg) and mean absolute error (MAE) (166.946 mg/kg) were higher when compared with the other methods applied. The hybridized model of empirical bayesian kriging-multiple linear regression (EBK-MLR) performed poorly, as evidenced by a coefficient of determination value of less than 0.1. The empirical bayesian kriging-support vector machine regression (EBK-SVMR) model was the optimal model, with low RMSE (95.479 mg/kg) and MAE (77.368 mg/kg) values and a high coefficient of determination (R2 = 0.637). EBK-SVMR modelling technique output was visualized using a self-organizing map. The clustered neurons of the hybridized model CakMg-EBK-SVMR component plane showed a diverse colour pattern predicting the concentration of Ni in the urban and peri-urban soil. The results proved that combining EBK and SVMR is an effective technique for predicting Ni concentrations in urban and peri-urban soil.

Multi-geochemical background comparison and the identification of the best normalizer for the estimation of PTE contamination in agricultural soil.

Identifying a suitable geochemical background level (GBL) and an appropriate normalizer is imperative for ensuring soil quality, health, and security. The objective of this study was to identify the appropriate normalizer and suitable GBL for determining PTE enrichment levels in agricultural soils and investigate if there are any statistical differences due to the GBL [World Average Value (WAV) European Average Value (EAV)] used. Forty-nine topsoil samples were obtained from seven agricultural communities in the Frdek-Mstek District (Czech Republic). Portable X-ray fluorescence was used to determine the total PTEs (Cr, Ni, Cu, Y, Ba, Th, As, Pb, and Zn) concentration levels in the soil. Correlation matrix analysis was used to determine the metallic relationship between the PTEs and the normalizers (Al, Fe, Ti, Zr, Sr and Rb). Pollution indices such as contamination factor (CF), geoaccumulation index (Igeo) and enrichment factor (EF) analysis were used to determine the most suitable GBL. Al, Fe, Sr, Ti and Rb strongly correlated with the CF, Igeo and EF, whereas WAV performed better than the other geochemical background (EAV). The results indicated that Rb was the suitable normalizer and WAV was the appropriate GBL for agricultural soil and provided a foundation for evaluating and surveilling soil quality and health in agricultural soil.

Using an ensemble model coupled with portable X-ray fluorescence and visible near-infrared spectroscopy to explore the viability of mapping and estimating arsenic in an agricultural soil.

Increasing concentrations of potentially toxic elements (PTE) in agricultural soils remain a major source of public concern. Monitoring PTEs in an agricultural field with no history of contaminants necessitate adequate analysis utilizing a robust model to accurately uncover hidden PTEs. Detecting and mapping the distribution of soil properties using portable X-ray fluorescence (pXRF) and proximal sensing techniques is not only rapid, but also relatively inexpensive. In this study, an ensemble model, consisting of partial least square regression (PLSR), support vector machine (SVM), random forest (RF) and cubist, was used for the prediction and mapping of soil As content in an agricultural field with no history of pollution. The datasets were collected using pXRF and field spectroscopy techniques. The main goal was to compare the ensemble model to each of the calibration techniques in terms of prediction accuracy of As content in such a field. Other components [e.g., soil organic carbon (SOC), Mn, S, soil pH, Fe] that are known to influence As levels in the soil were also retrieved to assess their correlation with soil As. The models were evaluated using the root mean squared error (RMSECV), the coefficient of determination (R2CV) and the ratio of performance to interquartile range (RPIQ). In terms of prediction accuracy, the ensemble model outperformed each of the individual techniques (R2CV = 0.80/0.75) and obtained the least error margin (RMSECV = 1.91/2.16). Overall, all the predictive techniques were able to detect both low and high estimated values of soil As within the study field, but with the ensemble model resembling the measurements better. The ensemble model, a promising tool as demonstrated by the current study, is highly recommended to be included in future studies for more accurate estimation of As and other PTEs in other agricultural fields.

A geostatistical approach to estimating source apportionment in urban and peri-urban soils using the Czech Republic as an example.

Unhealthy soils in peri-urban and urban areas expose individuals to potentially toxic elements (PTEs), which have a significant influence on the health of children and adults. Hundred and fifteen (n = 115) soil samples were collected from the district of Frydek Mistek at a depth of 0-20 cm and measured for PTEs content using Inductively coupled plasma-optical emission spectroscopy. The Pearson correlation matrix of the eleven relevant cross-correlations suggested that the interaction between the metal(loids) ranged from moderate (0.541) correlation to high correlation (0.91). PTEs sources were calculated using parent receptor model positive matrix factorization (PMF) and hybridized geostatistical based receptor model such as ordinary kriging-positive matrix factorization (OK-PMF) and empirical Bayesian kriging-positive matrix factorization (EBK-PMF). Based on the source apportionment, geogenic, vehicular traffic, phosphate fertilizer, steel industry, atmospheric deposits, metal works, and waste disposal are the primary sources that contribute to soil pollution in peri-urban and urban areas. The receptor models employed in the study complemented each other. Comparatively, OK-PMF identified more PTEs in the factor loadings than EBK-PMF and PMF. The receptor models performance via support vector machine regression (SVMR) and multiple linear regression (MLR) using root mean square error (RMSE), R square (R2) and mean square error (MAE) suggested that EBK-PMF was optimal. The hybridized receptor model increased prediction efficiency and reduced error significantly. EBK-PMF is a robust receptor model that can assess environmental risks and controls to mitigate ecological performance.

Source apportionment, contamination levels, and spatial prediction of potentially toxic elements in selected soils of the Czech Republic.

The sustenance of humans and livestock depends on the protection of the soil. Consequently, the pollution of the soil with potentially toxic elements (PTEs) is of great concern to humanity. The objective of this study is to investigate the source apportionment, concentration levels and spatial distribution of PTEs in selected soils in Frýdek-Místek District of the Czech Republic. The total number of soil samples was 70 (topsoil 49 and 21 subsoils) and was analysed using a portable XRF machine. Contamination factor and the pollution index load were used for the assessment and interpreting the pollution and distribution of PTEs in the soils. The inverse distance weighting was used for the spatial evaluation of the PTEs. The results of the analysis showed that the area is composed of low-to-high pollution site. PTEs displayed spatial variation patterns. The average PTE concentration decreases in this Fe > Ti > Ba > Zr > Rb > Sr > Cr > Y>Cu > Ni > Th order for the topsoil and also decreases in this Fe > Ti > Zr > Ba > Rb > Sr > Cr > Y > Cu > Ni > and Th order for the subsoil. These PTEs Cr, Ni, Cu, Rb, Y, Zr, Ba, Th, and Fe were far above the baseline European average value and the World average value level, respectively. The source apportionment showed the dominance of Cr, Ni, Rb, Ti, Th, Zr, Cu, Fe in the topsoil, while the subsoil was dominated by all the PTEs (factor 1 to 6) except Ba. The study concludes that indiscriminate human activities have an enormous effect on soil pollution.

Absorption Features in Soil Spectra Assessment.

From a wide range of techniques appropriate to relate spectra measurements with soil properties, partial least squares (PLS) regression and support vector machines (SVM) are most commonly used. This is due to their predictive power and the availability of software tools. Both represent exclusively statistically based approaches and, as such, benefit from multiple responses of soil material in the spectrum. However, physical-based approaches that focus only on a single spectral feature, such as simple linear regression using selected continuum-removed spectra values as a predictor variable, often provide accurate estimates. Furthermore, if this approach extends to multiple cases by taking into account three basic absorption feature parameters (area, width, and depth) of all occurring features as predictors and subjecting them to best subset selection, one can achieve even higher prediction accuracy compared with PLS regression. Here, we attempt to further extend this approach by adding two additional absorption feature parameters (left and right side area), as they can be important diagnostic markers, too. As a result, we achieved higher prediction accuracy compared with PLS regression and SVM for exchangeable soil pH, slightly higher or comparable for dithionite-citrate and ammonium oxalate extractable Fe and Mn forms, but slightly worse for oxidizable carbon content. Therefore, we suggest incorporating the multiple linear regression approach based on absorption feature parameters into existing working practices.

Estimation of potentially toxic elements contamination in anthropogenic soils on a brown coal mining dumpsite by reflectance spectroscopy: a case study.

In order to monitor Potentially Toxic Elements (PTEs) in anthropogenic soils on brown coal mining dumpsites, a large number of samples and cumbersome, time-consuming laboratory measurements are required. Due to its rapidity, convenience and accuracy, reflectance spectroscopy within the Visible-Near Infrared (Vis-NIR) region has been used to predict soil constituents. This study evaluated the suitability of Vis-NIR (350-2500 nm) reflectance spectroscopy for predicting PTEs concentration, using samples collected on large brown coal mining dumpsites in the Czech Republic. Partial Least Square Regression (PLSR) and Support Vector Machine Regression (SVMR) with cross-validation were used to relate PTEs data to the reflectance spectral data by applying different preprocessing strategies. According to the criteria of minimal Root Mean Square Error of Prediction of Cross Validation (RMSEPcv) and maximal coefficient of determination (R2cv) and Residual Prediction Deviation (RPD), the SVMR models with the first derivative pretreatment provided the most accurate prediction for As (R2cv) = 0.89, RMSEPcv = 1.89, RPD = 2.63). Less accurate, but acceptable prediction for screening purposes for Cd and Cu (0.66 ˂ R2cv) ˂ 0.81, RMSEPcv = 0.0.8 and 4.08 respectively, 2.0 ˂ RPD ˂ 2.5) were obtained. The PLSR model for predicting Mn (R2cv) = 0.44, RMSEPcv = 116.43, RPD = 1.45) presented an inadequate model. Overall, SVMR models for the Vis-NIR spectra could be used indirectly for an accurate assessment of PTEs' concentrations.

Visible, near-infrared, and mid-infrared spectroscopy applications for soil assessment with emphasis on soil organic matter content and quality: state-of-the-art and key issues.

Visible near-infrared (Vis-NIR) reflection spectroscopy and mid-infrared (mid-IR) reflection spectroscopy are cost- and time-effective and environmentally friendly techniques that could be alternatives to conventional soil analysis methods. Successful determination of spectrally active soil components, including soil organic matter (SOM), depends on the selection of suitable pretreatment and multivariate calibration techniques. The objective of the present review is to critically examine the suitability of Vis-NIR (350-2500 nm) and mid-IR (4000-400 cm(-1)) spectroscopy as a tool for SOM quantity and quality determination. Particular attention is paid to different pretreatment and calibration procedures and methods, and their ability to predict SOM content from Vis-NIR and mid-IR data is discussed. We then review the most recent research using spectroscopy in different calibration scales (local, regional, or global). Finally, accuracy and robustness, as well as uncertainty in Vis-NIR and mid-IR spectroscopy, are considered. We conclude that spectroscopy, especially the mid-IR technique in association with Savitzky-Golay smoothing and derivatization and the least squares support vector machine (LS-SVM) algorithm, can be useful in determining SOM quantity and quality. Future research conducted for the standardization of protocols and soil conditions will allow more accurate and reliable results on a global and international scale.