Predicting Cation Exchange Capacity Research Articles

Application of proximal sensing approach to predict cation exchange capacity of calcareous soils using linear and nonlinear data mining algorithms

Application of proximal sensing approach to predict cation exchange capacity of calcareous soils using linear and nonlinear data mining algorithms

Developing Vis–NIR libraries to predict cation exchange capacity (CEC) and pH in Australian sugarcane soil

In sugarcane growing areas of Queensland Australia, management of soil condition requires information about soil chemical properties (i.e., cation exchange capacity [CEC], and pH), because they are used to make fertiliser recommendations. While laboratory analysis is cost prohibitive the development of visible near–infrared (Vis–NIR) spectroscopy libraries might be useful. The aims of this study was to compare: i) linear (i.e., Partial least squares regression [PLSR]) and machine learning (i.e., Cubist, Random Forest [RF] and Support vector machine [SVM]) algorithms in terms of their calibration strength (i.e., coefficient of determination (R2); ii) depth-specific (i.e., topsoil [0 – 0.3 m], subsurface [0.3 – 0.6 m], subsoil [0.6 – 0.9 m] and deep subsoil [0.9 – 1.2 m]) and multi-depth libraries; iii) prediction R2, agreement (Lin’s concordance correlation coefficient [LCCC]) and accuracy (ratio of performance to interquartile [RPIQ]); and iv) minimum number of calibration sample locations (i.e., n = 400, 350, …, 50). For depth-specific calibration for CEC, strong calibration R2 was achieved for the subsurface (i.e., RF [0.95], Cubist [0.93], PLSR [0.88] and SVM [0.81]), and for pH, the largest R2 was for deep subsoil (i.e., RF [0.91], followed by PLSR [0.80], Cubist [0.79] and SVM [0.56]), with subsurface, and deep subsoil similar, with topsoil R2 smaller. In terms of model prediction of depth-specific, Cubist was superior to PLSR, RF and SVM for CEC, and PLSR was the best for pH. For example, agreement for topsoil CEC was substantial for Cubist (0.80) and PLSR (0.80), but moderate for RF (0.69) and poor for SVM (0.59). For subsurface (0.9), subsoil (0.91) and deep subsoil (0.92), the agreements were perfect for CEC using Cubist. In terms of multi-depth calibration for CEC it was strong for RF (R2 = 0.94), Cubist (0.92), PLSR (0.81) and SVM (0.72), but for prediction agreement, Cubist (LCCC = 0.92) was perfect, with PLSR (0.85), RF (0.84) and SVM (0.83) substantial. This was also the case for pH (0.84) substantial. However, there are efficiencies in developing a multi-depth calibration. Moreover, the minimum number of calibration sample sites required was 300 (i.e., 1,200 samples) (1.78 sites/ha). The results have implications for using a Vis–NIR library to replace traditional soil laboratory analysis and for fertiliser recommendations for sugarcane soil.

Comparison of Different Machine Learning Methods for Predicting Cation Exchange Capacity Using Environmental and Remote Sensing Data.

This study was conducted to examine the capability of topographic features and remote sensing data in combination with other auxiliary environmental variables (geology and geomorphology) to predict CEC by using different machine learning models ((random forest (RF), k-nearest neighbors (kNNs), Cubist model (Cu), and support vector machines (SVMs)) in the west of Iran. Accordingly, the collection of ninety-seven soil samples was performed from the surface layer (0–20 cm), and a number of soil properties and X-ray analyses, as well as CEC, were determined in the laboratory. The X-ray analysis showed that the clay types as the main dominant factor on CEC varied from illite to smectite. The results of modeling also displayed that in the training dataset based on 10-fold cross-validation, RF was identified as the best model for predicting CEC (R2 = 0.86; root mean square error: RMSE = 2.76; ratio of performance to deviation: RPD = 2.67), whereas the Cu model outperformed in the validation dataset (R2 = 0.49; RMSE = 4.51; RPD = 1.43)). RF, the best and most accurate model, was thus used to prepare the CEC map. The results confirm higher CEC in the early Quaternary deposits along with higher soil development and enrichment with smectite and vermiculite. On the other hand, lower CEC was observed in mountainous and coarse-textured soils (silt loam and sandy loam). The important variable analysis also showed that some topographic attributes (valley depth, elevation, slope, terrain ruggedness index—TRI) and remotely sensed data (ferric oxides, normalized difference moisture index—NDMI, and salinity index) could be considered as the most imperative variables explaining the variability of CEC by the best model in the study area.

Potential Application of Laser‐Induced Breakdown Spectroscopy (LIBS) Data for the Determination of Cation Exchange Capacity (CEC) of Agricultural Soils

AbstractThe potential for Laser Induced Breakdown Spectroscopy (LIBS) technique to predict cation exchange capacity (CEC) values in agricultural soils, based on a partial least square (PLS) regression model, was investigated by analysing soils (N= 225) using a portable LIBS system. The prediction from the validation set was 0.75 and the RMSE was 5.4 cmol+/kg. Analysis of the PLS loading weights showed that emission lines that correlate with the exchangeable cations (Ca, Mg, K and Na) and clay (Al and Si) are significant. Good quality predictions could be obtained from the model as described by the ratio of performance to deviation (RPD) values calculated for calibration (RPDc =1.99) and validation (RPDv=2.02). The ratio of performance to interquartile distance (RPIQ) for calibration (RPIQc=3.20) and validation (RPIQv=3.32) indicated excellent prediction capability. It has been demonstrated that LIBS spectra coupled with multivariate data analysis could be used to predict CEC of agricultural soils which is considered fast and a chemical‐free technique.

Modelling of soil cation exchange capacity for some soils of east gharaf lands from mid-Mesopotamian plain (Wasit province/Iraq)

This work was carried out in an area located at the mid-Mesopotamian plain, Wasit province/Iraq, and nearly 15 km. south of Kut city and aimed to develop a pedotransfer function to aid in predicting cation exchange capacity value using readily measured soil physicochemical properties. Soil data were subjected to descriptive statistics (minimum, maximum, standard deviation, the coefficient of variation, the coefficient of skewness and kurtosis). Three different combinations of soil physicochemical attributes bulk density (Bd), porosity (p), volumetric moisture content (Pv), texture class (T.C), total available water content (Taw), soil color (col.), clay (C), sand (S), silt (Si), cation exchange capacity (CEC), soil reaction (pH), electrical conductivity (Ec), sodium (Na+), potassium (K+), calcium (Ca+2), magnesium (Mg+2), organic matter (O.M), sodium adsorption ratio (SAR) and exchangeable sodium percentage (Esp) were used in developing models. Consequently, 25 mathematical models were generated. The generated model was cross-validated by means of coefficient of determination (R2), standard estimated error (SEE), root mean squared error (RMSE), mean absolute error (MAE), relative error (RE). In addition, the accuracy and precision of the constructed models were checked by assessing the strength of the relationship between observed and estimated values of cation exchange capacity. The results showed that the model physicochem.1.1 (Y = a * TC + b * Bd + c * T.a.w + d * col. + e * SAR + f * EC + g * Ca+2 + h) was best in predicting the value of soil cation exchange capacity with R2 = 0.98, SEE = 2.243, RMSE = 0.709, MAE = 0.510, RE = 0.016, r = 0.99, S = 0.745.

Short communication: Predicting cation exchange capacity from hygroscopic moisture in agricultural soils of Western Europe

Soil cation exchange capacity (CEC) depends on the extent and negative charge density of surfaces of soil mineral and organic components. Soil water sorption also depends on the extent of such surfaces, giving thus way to significant relationships between CEC and hygroscopic moisture (HM) in many soils. In this work, we explored whether CEC could be accurately predicted from HM in agricultural soils of Mediterranean and humid temperate areas in Western Europe. For this purpose, we examined 243 soils across a wide variation range of their intrinsic properties. Soil CEC was determined using 1 M ammonium acetate at pH 7 and HM at an equilibrium air relative humidity (RH) of 43% (HM43). Most of the variation of soil CEC was explained by HM43 through a linear function (CEC = 1.4 + 0.78HM43; R2 = 0.962; standard deviation = 2.30 cmolc/kg). Coefficients of the regression equation were similar for subgroups of soils differing in moisture regime, clay mineralogy, carbonate content and organic carbon content. Therefore, soil hygroscopic moisture measurements at a fixed RH level provided a simple, robust, inexpensive method for predicting soil CEC.

Using the nonparametric k-nearest neighbor approach for predicting cation exchange capacity

The objectives of this study were to apply a k-NN approach to predict CEC in Iranian soils and compare this approach with the popular artificial neural network model (ANN). In this study, a data set of 3420 soil samples from different parts of Iran was used. Two different sets of cheaper-to-measure soil attributes were selected as potential predictors. The first set consisted of clay, silt, sand and organic carbon (OC) contents. The second data set was constructed using OC and clay contents. Two ‘design-parameter’ parameters should be optimized before application of the k-NN approach. Results showed that the algorithm efficiency is not dependent on these parameters. A wide range of suboptimal values around the optimal values may cause a slight error in terms of estimation accuracy. However, the optimal settings of the design-parameters depend on the size of the development/reference data set. In both k-NN and ANN models, the higher number of input variables can relatively improve the estimation of CEC. But this improvement was not statistically significant at the 0.05 level. Furthermore, the results showed that increasing the size of the reference data set to a certain amount (N=1200) reduced the estimation error significantly in terms of root-mean-squared residuals (RMSE). However, no significant difference in the accuracy of k-NN and ANN methods was detected in the reference data set sizes for N>1200. Results showed no significant difference between this approach and ANN models, suggesting the competitive advantage of the k-NN technique over other techniques to develop pedotransfer functions (PTFs), for example, the redevelopment of PTFs is not necessarily required as new data become available.

Predicting Cation Exchange Capacity by Artificial Neural Network and Multiple Linear Regression Using Terrain and Soil Characteristics

This research was an attempt to investigate the capability of using attributes derived from digital elevation model together with easily/readily obtainable soil properties for estimating Cation Exchange Capacity (CEC) of soils. The study area was located in hilly lands of Lordegan region, west of Iran. Soil samples were collected from 130 points with 0-30 cm depth. Modeling was performed applying the artificial neural networks and regression pedotransfer functions and three models with different inputs (soil properties, topographic properties and both of them) were used. The results showed the efficacy of models when using soil and topographic attributes alone, do not have much difference, while utilization of both soil and topographic characteristics was improved the precision of models in both methods which was more sensible for neural networks. Moreover, the results of quantifying variables importance employing connection weight method were implied that four parameters comprising topographic wetness index, slope, surface curvature and aspect had the highest contribution in CEC variation in the study area. Furthermore, the results of current research support this idea that soil attributes are highly controlled by topographic properties and affected by water movement, hydrological and erosional processes and also microclimate variation caused by topographic features. Thus, topographic data could be used with high confidence in order to predict soil characteristics such as CEC in Lordegan region.

Using support vector machines to predict cation exchange capacity of different soil horizons in Qingdao City, China

AbstractAgricultural, environmental and ecological modeling requires soil cation exchange capacity (CEC) that is difficult to measure. Pedotransfer functions (PTFs) are thus routinely applied to predict CEC from easily measured physicochemical properties (e.g., texture, soil organic matter, pH). This study developed the support vector machines (SVM)‐based PTFs to predict soil CEC based on 208 soil samples collected from A and B horizons in Qingdao City, Shandong Province, China. The database was randomly split into calibration and validation datasets in proportions of 3:1 using the bootstrap method. The optimal SVM parameters were searched by applying the genetic algorithm (GA). The performance of SVM models was compared to those of multiple stepwise regression (MSR) and artificial neural network (ANN) models. Results show that the accuracy of CEC predicted by SVM improves considerably over those predicted by MSR and ANN. The performance of SVM for B horizon (R2 = 0.85) is slightly better than that for A horizon (R2 = 0.81). The SVM is a powerful approach in the simulation of nonlinear relationship between CEC and physicochemical properties of widely distributed samples from different soil horizons. Sensitivity analysis was also conducted to explore the influence of each input parameter on the CEC predictions by SVM. The clay content is the most sensitive parameter, followed by soil organic matter and pH, while sand content has the weakest influence. This suggests that clay is the most important predictor for predicting CEC of both soil horizons.

Developing cation exchange capacity and soil index properties relationships using a neuro-fuzzy approach

Artificial intelligence methods are employed to predict cation exchange capacity (CEC) from five different soil index properties, namely specific surface area (SSA), liquid limit, plasticity index, activity (ACT), and clay fraction (CF). Artificial neural networks (ANNs) analyses were first employed to determine the most related index parameters with cation exchange capacity. For this purpose, 40 datasets were employed to train the network and 10 datasets were used to test it. The ANN analyses were conducted with 15 different input vector combinations using same datasets. As a result of this investigation, the ANN analyses revealed that SSA and ACT are the most effective parameters on the CEC. Next, based upon these most effective input parameters, the fuzzy logic (FL) model was developed for the CEC. In the developed FL model, triangular membership functions were employed for both the input (SSA and ACT) variables and the output variable (CEC). A total of nine Mamdani fuzzy rules were deduced from the datasets, used for the training of the ANN model. Minimization (min) inferencing, maximum (max) composition, and centroid defuzzification methods are employed for the constructed FL model. The developed FL model was then tested against the remaining datasets, which were also used for testing the ANN model. The prediction results are satisfactory with a determination coefficient, R 2 = 0.94 and mean absolute error, (MAE) = 7.1.

Using distance metrics to determine the appropriate domain of pedotransfer function predictions

The development of pedotransfer functions continues unabated, in part due to the inability of users to identify functions suitable to their soil of interest. In this paper, a protocol is proposed that allows users to determine the similarity between a functions calibration data and a subject of interest. The protocol uses standard distance metrics, Standardized Euclidean and Mahalanobis distances, to determine the distance from the mean of the calibration data and the subject. Distances exceeding a designated cutoff limit are deemed distinct from the calibration data and as such unsuitable for function application. An example is provided whereby a function to predict Cation Exchange Capacity is trained on a local dataset and applied to a second dataset with large pedodiversity. Those subjects suitable for functions application are identified via the proposed protocol. The proposed protocol is dependent on the provision of robust estimates of mean, standard deviation and covariance matrix of the calibration data.

Artificial Neural Network Approach for Predicting Cation Exchange Capacity in Soil Based on Physico-Chemical Properties

Cation exchange capacity (CEC) of soil is an important characteristic that determines the soil buffering capacity to maintain cationic nutrients and pollutants against leaching to the subsurface layers. Radial basis function neural networks (RBFN) were applied into the estimation of cation exchange capacity (CEC) from soil physicochemical properties. A total of 457 soil samples collected from surface to deep underground horizons under various climatic and geographic conditions were successfully compiled in the comprehensive modelling research. Soil horizon; pH; organic carbon content; and clay, silt, and sand contents were taken as the input variables. Four RBFN models applied in CEC estimates for A, B, C horizons as well as the aggregate soils from all horizons were developed, with high correlation coefficients of 0.9397, 0.8368, 0.8014, and 0.8335 between the predicted and measured values of the external test samples, respectively, after optimization of the network architectures. Furthermore, the influence of each input variable on the soil CEC was particularly evaluated and sequenced by sensitivity analysis, confirming that clay content, organic carbon content, and soil horizons are the most important determinants of CEC estimate. Due to the existing nonlinear and uncertain characteristics of natural soil properties demonstrated in this paper, the optimized RBFN models showed high superiority in large-scale data simulation of complicated soil samples compared with multiple linear regressions (MLR). All the results confirm that an artificial neural network is a precise, sensitive, and robust method in the quantitative study of CEC in widely distributed soil samples in the complex ecosystem.

Predicting Cation Exchange Capacity for Soil Survey Using Linear Models

Measuring the cation exchange capacity (CEC) for all horizons of every map unit component in a survey area is very time consuming and costly. The objective of this study was to develop CEC (pH 7 NH4OAc) prediction models that encompass most soils of the United States. The National Soil Survey Characterization database was used to develop the predictive models using general linear models. Data were stratified into more homogeneous groups based on the organic C content, soil pH, taxonomic family mineralogy class and CEC‐activity class, and taxonomic order. Models were developed for each strata or data group. Organic matter and noncarbonate clay contents were the main predictor variables used. Water at −1500 kPa was used in lieu of clay content on four groups. Results indicate that between 43 and 78% of the variation in CEC could be explained for the high organic C data groups; between 53 and 84% could be explained for the mineralogy groups; between 86 and 95% could be explained for the CEC‐activity class groups; and between 53 and 86% could be explained for the taxonomic orders. The same predictive model was applicable for Gelisols and Histosols. Inceptisols and Alfisols (>0.3% organic C) also shared the same model. In general, the mineralogy/CEC‐activity class equations had lower RMSEs than the taxonomic order equations. A decision tree, based on how the data was stratified, guides the selection of which model to use for a soil layer. Validation results indicated that the models, in aggregate, provide a reasonable estimate of CEC for most soils of the United States.

Neural network models to predict cation exchange capacity in arid regions of Iran

SummaryDesign and analysis of land‐use management scenarios requires detailed soil data. When such data are needed on a large scale, pedotransfer functions (PTFs) could be used to estimate different soil properties. Because existing regression‐based PTFs for estimating cation exchange capacity (CEC) do not, in general, apply well to arid areas, this study was conducted (i) to evaluate the existing models and (ii) to develop neural network‐based PTFs for predicting CEC in Aridisols of Isfahan in central Iran. As most researches have found a significant correlation between CEC and soil organic matter content (OM) and clay content, we also used these two variables for modelling of CEC. We tested several published PTFs and developed two neural network algorithms using multilayer perceptron and general regression neural networks based on a set of 170 soil samples. The data set was divided into two subsets for calibration and testing of the models. In general, the neural network‐based models provided more reliable predictions than the regression‐based PTFs.

Cation exchange capacity of two‐component container media predicted from laboratory analysis of components

Abstract A mathematical equation predicting cation exchange capacity (CEC) of pine bark‐sand container media from CEC of the individual components was formulated. The equation is the weighted sum of milliequivalents contributed by each component and is corrected for shrinkage due to mixing of components. Both measured and predicted CEC increased linearly with increasing percent volumetric bark in pine bark‐sand media. Regression equations describing measured and predicted CEC were not statistically different. The predictive equation was also tested on 6 non‐synthesized 2‐component media prepared from peat moss, perlite, pine bark, vermiculite, and sand. No statistical differences between measured and predicted CEC were obtained.