Development Of Pedotransfer Functions Research Articles

Scaling of Kostiakov–Lewis equation and estimation of scaling factors at field scale

ABSTRACT Seventy-two infiltration tests were conducted to model soil water infiltration variability using scaling in the first and third terraces of Yangling District, respectively. The relationship between the scaling factor of the Kostiakov–Lewis equation and soil properties for multiple scales was determined through two-dimensional continuous wavelet transform (2D-CWT) and path analyses, and the pedotransfer functions (PTFs) based on the support vector machine (SVM) and back propagation (BP) neural network were developed for estimating the scaling factor. The results indicated that when scaling factor was calculated using the least square method (FS ), the best match was achieved between the predicted infiltration and measured values. In the first and third terraces, the results of 2D-CWT and path analyses indicated that FS were positively correlated with sand content, silt content, and soil organic matter content. However, FS was negatively correlated with bulk density in the first terrace, and which was negatively correlated with clay content and initial water content in the third terrace. The accuracy of SVM-based PTF is higher than BP-based PTF, the mean absolute value of relative errors of SVM-based PTF were 11.3% and 8.16% for the first and third terrace, respectively. Therefore, the SVM is preferred for PTF development.

Developing pedotransfer functions using Sentinel-2 satellite spectral indices and Machine learning for estimating the surface soil moisture

To estimate the surface soil moisture (SM) using a combination of new spectral indices and methods of Random Forrest (RF) and Multiple Linear Regression (MLR), 11 pedotransfer functions (PTF1-11) were developed by combining basic soil properties (clay, silt/sand, and bulk density) and spectral indices of Sentinel-2 satellite. In this study, 124 surface soil samples were randomly taken from three regions including Telo in Tehran province, Ivaneki in Semnan province, and Borujerd in Lorestan province, Iran. The results showed that the accuracy of the RF method was considerably higher compared to the MLR method. The SM was better estimated using water spectral indices (such as Normalized Difference Water Index (NDWI) and Surface Water Capacity Index (SWCI)), along with the basic properties of soil as inputs of PTF7. In the training and testing steps, the Root Mean Square Error (RMSE) decreased from 0.041 and 0.05 (cm3 cm−3) in PTF1 to 0.028 and 0.039 in PTF7, respectively. The average values of RMSE, Akaike Information Criterion (AIC), coefficient of determination (R2) and Relative Improvement (RI) of the RF method were 0.028 (cm3 cm−3), −559, 0.79 and 0.001, and 0.038 (cm3 cm−3), −279, 0.73 and −0.006 for the training and testing steps of all PTFs, respectively. While, these values for the MLR method were 0.032 (cm3 cm−3), −542, 0.69 and 0.0003, and 0.043 (cm3 cm−3), −269, 0.63 and 0.002 for the training and testing steps, respectively. Due to the low values of MBE, it was possible to disregard the overestimation of the results. The evaluation of results for predictor importance indicated that among the basic properties, clay percent has a significant effect on estimation of SM. These results show that spectral indices alone are not suitable estimators for SM estimation. It suggests using basic soil properties and spectral indices to estimate the SM.

Development of pedotransfer functions for predicting soil bulk density: A case study in Indonesian small island

Development of pedotransfer functions for predicting soil bulk density: A case study in Indonesian small island

Estimating Atterberg limits of soils from reflectance spectroscopy and pedotransfer functions

Atterberg limits are broadly used for engineering and geology purposes as well as in agricultural and environmental applications. Laboratory methods used for their determination are, however, laborious, destructive and tool dependent. The aim of this study was to test the feasibility of using visible near-infrared spectroscopy (vis–NIRS) as a fast and accurate alternative to the conventional measurements of Atterberg limits (LL and PL) and the PI for 229 geographically diverse soil samples originating from 24 countries. Three types of calibration techniques including Partial Least Squares (PLS) regression, Artificial Neural Networks (ANN) and Support Vector Machines (SVM) were applied to the spectral data. The performance of the best vis–NIRS models was validated using 45 independent samples and compared with two existing and one newly developed pedotransfer functions (PTF). The application of SVM yielded marginally better predictive ability than PLS and ANN for all modelled properties. The SVM models estimated LL, PL, and PI with root mean squared error (RMSE) of 7%, 5% and 7%, respectively. The newly developed PTF gave slightly better estimations than the existing ones, with RMSE values of 8%, 6%, and 6%, respectively for LL, PL, and PI. Furthermore, in terms of the sample swelling class, the SVM model correctly classified 31 of the 45 samples, compared to 34 samples for the best PTF. The results indicate a great potential of vis–NIRS for reliable estimates of Atterberg limits for soil samples of large geographical and mineralogical diversity.

Evaluation and Development of Pedo-Transfer Functions for Predicting Soil Saturated Hydraulic Conductivity in the Alpine Frigid Hilly Region of Qinghai Province

In recent years, Pedo-Transfer Functions (PTFs) have become a commonly used tool to predict the hydraulic properties of soil. As an important index to evaluate the function of forest water conservation, the prediction of saturated hydraulic conductivity (KS) on the regional scale is of great significance to guide the vegetation construction of returning farmland to forest area. However, if the published PTFs are directly applied to areas where the soil conditions are different from those where the PTFs are established, their predictive performance will be greatly reduced. In this study, 10 basic soil properties were measured as input variables for PTFs to predict KS in the three watersheds of Taergou, Anmentan, and Yangjiazhai in the alpine frigid hilly region of Qinghai Province, China. The parameters of the eight published PTFs were modified by the least-squares method and new PTFs were also constructed, and their prediction performance was evaluated. The results showed that the KS of coniferous and broad-leaved mixed forests and broad-leaved pure forests in the study area were significantly higher than those of pure coniferous forests, and grassland and farmland were the lowest (p > 0.05). Soil Organic Matter plays an important role in predicting KS and should be used as an input variable when establishing PTFs. The Analysis-Back Propagation Artificial Neural Network (BP ANN) PTF that was established, with input variables that were, Si·SOM, BD·Si, ln2Cl, SOM2, and SOM·lnCl had a better predictive performance than published PTFs and MLR PTFs.

Land use effects on soil hydraulic properties and the contribution of soil organic carbon

Soil hydraulic properties (SHP) control water movement and storage and thus affect a wide range of biogeochemical processes and ecosystem services. The objectives of this study were to identify the interactions between long-term land use and soil type on SHP (e.g., available water content [AWC], hydraulic conductivity [K]), and the effects of soil organic carbon (SOC) content on these properties. Soil water release curves and (near) saturated hydraulic conductivity at three soil depths (0–7.5, 7.5–15, and 15–30 cm) were measured under three long-term (>20 years) land uses (irrigated pasture [IP], dryland pasture [DP], and irrigated cropping [IC]) in Canterbury, New Zealand. For each land use, three soil types with contrasting drainage characteristics were selected: well-drained Lismore (LIS) stony silt loam, moderately well-drained Templeton (TEM) silt loam, and poorly-drained Waterton/Temuka (WAT) clay loam. Compared to DP, the IP and IC soils had lower AWC and K. Significant interactions between land use and soil type were found for AWC but not for K. Bulk density, SOC, and clay content could explain half the variation in AWC but had limited influence on K. More variance in AWC (6–13%) and K (5–27%) could be explained by including categorical variables (i.e., land use, soil type, and soil depth) and their interactions with continuous variables, indicating the potential benefit of including these categorical variables when developing pedotransfer functions for AWC and K. This study suggests that many macropores (>30 μm) were poorly connected and contributed more to water storage than to conducting water. The commonly used soil matric potential of −10 kPa was suitable for defining field capacity in DP and IC soils but a lower matric potential (−33 kPa) was more appropriate for IP soil affected by livestock treading during grazing. Soil organic carbon mainly increased AWC by affecting water retention at soil matric potentials of between −40 and −1500 kPa (equivalent to pore size 0.2–7.5 μm). Therefore, soil carbon sequestration may be important to alleviating water stress in dry environments. Our study also suggests that the effects of land use and its interaction with soil type, soil matric potential at field capacity, and soil organic carbon on SHP need to be considered during soil parameterization of hydrological models.

Development of functions to predict soil hydraulic properties that account for solution sodicity and salinity

Saturated soil hydraulic conductivity (Ksat) and the soil water characteristic curve have been predicted from soil texture, bulk density, and other soil physical properties. These previous functions do not account for soil clay mineralogy or soil solution composition. The objectives of this study were to: (1) analyze the performance of previously developed pedotransfer functions (PTFs) on saline and sodic soils, (2) predict hydraulic properties of saline and sodic soils with a range of clay mineralogy, and (3) develop PTFs that account for clay mineralogy and solution composition. Most PTFs do not perform well on soils with smectitic clay mineralogy and usually over-estimate Ksat and under-estimate soil water characteristic on smectite clay dominant soils. Accounting for soil cation exchange capacity (CEC) as an indicator of soil clay types and solution composition improved the prediction of Ksat which had a geometric root mean squared error of 3.28 cm day−1 compared to 6.04 cm day−1 for the best performing PTF from the literature. A PTF we developed to predict soil water characteristic curve was able to predict soil water characteristic curve with a root mean squared error of 0.032 g g−1 compared to 0.082 g g−1 for the best performing previously published PTF. The PTFs developed in this study improved prediction of soil hydraulic properties on salt affected soils by using soil CEC to account for soil clay behavior and responsiveness to the soil solution composition. The PTFs reported here can easily be used to predict the Ksat and the soil water characteristic curve of soils in response to changes in irrigation water quality.

Investigating spatial relationships of soil friability and driving factors through co-regionalization with state-space analysis in a subtropical watershed

Soil friability (F) is one of the key indicators of soil structural conditions and can be used to assess the environmental impacts of different tillage and no-tillage practices on crop growth and yield. F can be affected by several driving factors such as soil (physical and chemical parameters) and topographic attributes and different land uses as well. Few studies have been carried out to identify those driving factors as well as to quantify their relationships with F at watershed scale. The objectives of this study were (i) to characterize the spatial variability of F and further soil and topographic attributes across a spatial transect of 11.2 km in length using classical statistics; (ii) to quantify the spatial relationships of F and driving factors using the state-space approach; and (iii) to compare the performance of state-space models in estimating F with those of corresponding linear and multiple regression models. The transect was established in the Micaela river watershed (MRW), Southern Brazil, with equidistantly distributed soil sampling points. Soil textural fractions (sand, clay and silt contents), organic matter content, bulk density, macroporosity, and F were determined in the 0−0.10 m layer at each sampling point. The digital elevation model provided elevation and soil slope as topographic attributes while a land use map was obtained from satellite images. All state-space models achieved better results in describing the spatial relationships of F and the further soil and topographic attributes than the corresponding multiple linear regression models. State-space analysis showed that soil slope could be used as a proxy to predict local variations of F in the MRW. The spatial covariance information should be included in the development of pedotransfer functions for estimating the spatial variation of F and the type of land use (as a soil structural indicator) should be investigated in future studies at watershed scale.

Developing Pseudo Continuous Pedotransfer Functions for International Soils Measured with the Evaporation Method and the HYPROP System: II. The Soil Hydraulic Conductivity Curve

Direct measurement of unsaturated hydraulic parameters is costly and time-consuming. Pedotransfer functions (PTFs) are typically developed to estimate soil hydraulic properties from readily available soil attributes. For the first time, in this study, we developed PTFs to estimate the soil hydraulic conductivity (log(K)) directly from measured data. We adopted the pseudo continuous neural network PTF (PCNN-PTF) approach and assessed its accuracy and reliability using two independent data sets with hydraulic conductivity measured via the evaporation method. The primary data set contained 150 international soils (6963 measured data pairs), and the second dataset consisted of 79 repacked Turkish soil samples (1340 measured data pairs). Four models with different combinations of the input attributes, including soil texture (sand, silt, clay), bulk density (BD), and organic matter content (SOM), were developed. The best performing international (root mean square error, RMSE = 0.520) and local (RMSE = 0.317) PTFs only had soil texture information as inputs when developed and tested using the same data set to estimate log(K). However, adding BD and SOM as input parameters increased the reliability of the international PCNN-PTFs when the Turkish data set was used as the test data set. We observed an overall improvement in the performance of PTFs with the increasing number of data points per soil textural class. The PCNN-PTFs consistently performed high across tension ranges when developed and tested using the international data set. Incorporating the Turkish data set into PTF development substantially improved the accuracy of the PTFs (on average close to 60% reduction in RMSE). Consequently, we recommend integrating local HYPROPTM (Hydraulic Property Analyzer, Meter Group Inc., USA) data sets into the international data set used in this study and retraining the PCNN-PTFs to enhance their performance for that specific region.

Assessment of Various Pedotransfer Functions for the Prediction of the Dry Bulk Density of Cultivated Soils in a Semiarid Environment

ABSTRACT Bulk density (BD) is a key soil indicator for the assessment of the sustainability of production from agricultural land. However, it is difficult to obtain data on BD from field studies because sampling is an extremely labor-intensive, time-consuming and expensive task at the large scale needed to generate meaningful amounts of data. Therefore, the purpose of this study was to evaluate different pedotransfer functions (PTFs) for their usefulness in the estimation of the BD values of an intensely cultivated area through the use of different geostatistical methods. PTFs are estimation functions of certain soil properties derived from raw data for other soil properties. Different approaches have been used for the development of PTFs, including regression methods and artificial neural networks. In this study, the multivariate analysis methods, general and stepwise regression, were used for BD estimation. In addition, different learning algorithms were evaluated with artificial neural networks (ANN) for their efficacy in estimating the BD. Seven basic soil properties, namely, sand, silt, clay, organic matter, pH, EC and CaCO3, were used in the development of models. The estimation power of the general regression model (normalized root mean square error (NRMSE): 7.10%) was higher than that of stepwise regression (NRMSE: 19.93%). Additionally, the lowest NRMSE (6.74%) and the highest R2 for BD estimation determined with different learning algorithms through artificial neural networks (ANN) were obtained with the Levenberg-Marguardt algorithm. Moreover, for BD estimation, ANN performed better than the multivariate regression equations. Spatial distribution maps of soil BD were generated with the commonly used ordinary kriging method by utilizing the real BD values in combination with BD values obtained from estimation models. Maps of BD values produced by stepwise regression estimation deviated significantly from maps generated with real values whereas ANN-II (Bayesian regularization algorithm) values were closest to the real values and that was reflected in the increased accuracy of mapping.

Evaluation and Development of Pedotransfer Functions for Predicting Saturated Hydraulic Conductivity for Mexican Soils

In the present work, we evaluate the prediction capability of six pedotransfer functions (PTFs), reported in the literature, for the saturated hydraulic conductivity estimations (KS). We used a database with 900 measured samples obtained from the Irrigation District 023, in San Juan del Rio, Queretaro, Mexico. Additionally, six new PTFs were constructed for KS from clay percentage, bulk density, and saturation water content data. The results show, for the evaluated models, that one model presents an overestimation for KS > 0.5 cm h−1 values, three models have an underestimation for KS > 1.0 cm h−1, and two models have a good correlation (R2 > 0.98) but more than three parameters are necessary. Nevertheless, the last two models require 3–4 parameters in order to obtain optimization. On the other hand, the models proposed in this work have a similar correlation with fewer parameters. The fit is seen to be much better than using the existing ones, achieving a correlation of R2 = 0.9822 with only one variable and R2 = 0.9901 when we use two.

Developing pedotransfer functions to harmonize extractable soil phosphorus content measured with different methods: A case study across the mainland of France

Phosphorus (P) is a nutrient essential to living organisms and ecosystems. Accurate information regarding extractable soil P is necessary for agricultural management and environmental quality. Direct measurements of extractable soil P at large scales are usually impeded by considerable time, labour, and economic resources required for implementation. To meet agronomic and environmental monitoring needs, multiple extraction methods have been developed worldwide to estimate the different components of soil P. In France, three extraction methods are used, namely the Dyer method for acidic soils, Joret-Hébert for calcareous soils, and Olsen for all soils. Therefore, it is difficult to compare data obtained nationwide for monitoring purposes. Consequently, it is of significant importance to develop pedotransfer functions (PTFs) to harmonise extractable soil P data obtained from different extraction methods with the assistance of other easily available predictors from soil information systems. In this study, we used an extensive dataset from the French soil-monitoring programme for the calibration and evaluation of PTFs. We implemented the partial least squares regression to relate extractable P measured by the Dyer or Joret-Hébert method to extractable P determined by the Olsen method considering 14 soil properties (total P2O5, pH, cation exchange capacity (CEC), CaCO3, soil texture (clay, silt and sand contents), total organic carbon, and exchangeable Fe, Al, CaO, Mn, MgO, and K2O). We constructed patrimonial models by selecting the most important predictors. According to the results of 10 iterations cross-validation, the average R2, root mean-square error (RMSE), and mean error (ME) of the PTF of calcareous soils were 0.66, 25.81, and −0.11 mg kg−1, whereas those of acidic soils were 0.70, 24.02, and −0.87 mg kg1, respectively. The Joret-Hébert P2O5, silt, pH, total P2O5, CEC, and K were the most important predictors for estimating Olsen P2O5 in calcareous soils, whereas Dyer P2O5, exchangeable Al, K, and pH were the most important predictors for estimating Olsen P2O5 in acidic soils. We observed that the explanatory power of the soil properties was more important in calcareous than in acidic soils. As expected, the proxies of Olsen P2O5, namely, Dyer P2O5 and Joret-Hébert P2O5, were the most important variables in modelling Olsen P2O5 variations. In addition, the relationship between Olsen P2O5 and Dyer P2O5 was much stronger than that between Olsen P2O5 and Joret-Hébert P2O5. The results confirmed the feasibility of estimating extractable P in soil by PTFs that were constructed using statistical methods, such as partial least squares regression. The addition of more predictors that are related to agricultural practices and topography attributes may improve the prediction accuracy.

Development of Pedo-Transfer Functions for the Saturated Hydraulic Conductivity of Forest Soil in South Korea Considering Forest Stand and Site Characteristics

The saturated hydraulic conductivity (Ks) is one of the most important soil properties for many hydrological simulation models. Especially in South Korea, analyzing the Ks of the forest soil is essential for understanding the water cycle throughout the country, because forests cover almost two-thirds of the whole country. However, few studies have focused on the forest soil in the temperate climate zone on a nationwide scale. In this study, 1456 forest soil samples were collected throughout South Korea and pedo-transfer functions employed to predict the Ks were developed. The non-linearities of the soil and topographic features were considered with the pretreatment of variables, and the variance inflation factor was used for treating the multicollinearity problem. The forest stand and site characteristics were also categorized by an ANOVA and post hoc test due to their diversity. As a result, the Ks values were different for various forest stands and site characteristics, which was statistically significant. Additionally, the model performance was higher when both soil properties and topographic features were considered. The sensitivity analysis showed that the Ks was highly affected by the bulk density, sand fraction, slope, and upper catchment area. Therefore, the topographic features were as important in predicting the Ks as the soil properties of the forest soil.

Quantification of macropore flow in Danish soils using near-saturated hydraulic properties

Larger pores allowing macropore flow can rapidly transport contaminants as well as nutrients from the soil surface to the subsoil and groundwater. Modeling of the involved macropore flow is complex due to large number of required parameters. One way to deal with this difficulty is the precise evaluation of the soil hydraulic conductivity function using robust pedotransfer functions for the near-saturated part. Aiming to develop pedotransfer functions to estimate saturated and unsaturated hydraulic conductivities [Ks and K(h)] as well as water contents at specific pressure heads, we used Gaussian Process Regression (GPR), a non-parametric machine learning model, to obtain van Genuchten (1980) retention and conductivity parameters for Danish soils based on soil texture and organic matter. We defined K10 as K(h) at − 10 cm pressure head. The difference between the logarithm of Ks and K10 [log(Ks) − log(K10)] was denominated log(Kjump) and was assumed to convey the potential degree of macropore flow considering only soil hydraulic properties. Macropore flow in Denmark was evaluated through the developed PTFs for log(Ks) and log(K10) with an average RMSE of 0.635 and 0.594 (for K in cm d−1) under a confidence interval obtained by bootstrapping. Dynamics of pressure head in an initially saturated soil column with a zero-flux top boundary condition and a 2 m deep groundwater level were simulated using the Hydrus-1D model. The result of this modeling showed that soils in the eastern part of Denmark had a higher likelihood to experience macropore flow. Finally, using the spatial distribution of meteorological data, a macropore flow map of Denmark was produced. Precipitation was a dominating factor on macropore flow in sandy clay loam and sandy loam soils when compared to sandy and loamy sand soils. Finally, maps of relative risk classes of macropore flow qualitatively differentiated land areas in Denmark for vulnerability of nutrient loss and groundwater contamination.

Development of pedotransfer functions for water retention in tropical mountain soil landscapes: spotlight on parameter tuning in machine learning

Abstract. Machine-learning algorithms are good at computing non-linear problems and fitting complex composite functions, which makes them an adequate tool for addressing multiple environmental research questions. One important application is the development of pedotransfer functions (PTFs). This study aims to develop water retention PTFs for two remote tropical mountain regions with rather different soil landscapes: (1) those dominated by peat soils and soils under volcanic influence with high organic matter contents and (2) those dominated by tropical mineral soils. Two tuning procedures were compared to fit boosted regression tree models: (1) tuning with grid search, which is the standard approach in pedometrics; and (2) tuning with differential evolution optimization. A nested cross-validation approach was applied to generate robust models. The area-specific PTFs developed outperform other more general PTFs. Furthermore, the first PTF for typical soils of Páramo landscapes (Ecuador), i.e., organic soils under volcanic influence, is presented. Overall, the results confirmed the differential evolution algorithm's high potential for tuning machine-learning models. While models based on tuning with grid search roughly predicted the response variables' mean for both areas, models applying the differential evolution algorithm for parameter tuning explained up to 25 times more of the response variables' variance.

Pedotransfer functions: the role of soil chemical properties units conversion for soil classification

Chemical soil analysis data can be expressed by weight (i.e., gravimetric basis) or volume (i.e., volumetric basis) of the fine earth (sieved ≥2 mm), resulting in different units, cmolc kg-1 and cmolc dm-3, respectively. The research problem is that the difference between methods to express the same soil properties hinders the comparison of results and database or dataset standardization. This paper aims to develop pedotransfer functions (PTF) to obtain the density of fine earth, which will then be used [...]

Development of pedotransfer functions by machine learning for prediction of soil electrical conductivity and organic carbon content

The pedotransfer function is a mathematical model used to convert direct soil measurements into known and unknown soil properties. It provides information for modelling and simulation in soil research, hydrology, environmental science and climate change impacts, including investigating the carbon cycle and the exchange of carbon between soils and the atmosphere to support carbon farming. In particular, the pedotransfer function can provide input parameters for landscape design, soil quality assessment and economic optimisation. The objective of the study was to investigate the feasibility of using a generalised pedotransfer function derived with a machine learning method to predict soil electrical conductivity (EC) and soil organic carbon content (OC) for different regional locations in the state of Victoria, Australia. This strategy supports a unified approach to the interpolation and population of a single regional soils database, in contrast to a range of pedotransfer functions derived from local databases with measurement sets that may have limited transferability. The pedotransfer function generation was based on a machine learning algorithm incorporating the Generalized Linear Mixed Model with interactions and nested terms, with Residual Maximum Likelihood estimation, and a predictor-frequency ranking system with step-wise reduction of predictors to evaluate the predictive errors in reduced models. The source of the data was the Victorian Soil Information System (VSIS), which is a database administered for soil information and mapping purposes. The database contains soil measurements and information from locations across Victoria and is a repository of historical data, including monitoring studies. In total, data from 93 projects were available for inputs to modelling and analysis, with 5158 samples used to derive predictors for EC and 1954 samples used to derive predictors for OC. Over 500 models were tested by systematically reducing the number of predictors from the full model. Five-fold cross-validation was used for estimation of model mean-squared prediction error (MSPE) and mean-absolute percentage error (MAPE). The results were statistically significant with only a gradual reduction in error for the top-ranked 50 models. The prediction errors (MSPE and MAPE) of the top ranked model for EC are 0.686 and 0.635, and 0.413 and 0.474 for OC respectively. The four most frequently occurring predictors both for EC and OC prediction across the full set of models were found to be soil depth, pH, particle size distribution and geomorphological mapping unit. The possible advantages and disadvantages of this approach were discussed with respect to other machine learning approaches.

Temporal variations of the hydraulic conductivity characteristic under conventional and conservation tillage

Conservation agriculture promotes as one of its three key principles the reduction of tillage intensity to minimize soil physical disturbances. The choice of tillage system alters hydraulic conductivity at (Ks) and near saturation. Drier regions of the hydraulic conductivity characteristic (HCC) may be affected as well. Temporal variations of the HCC along with water retention characteristics of tilled soils have come into the focus of researchers in recent years. It was shown that it may be preferable to account for such variations in soil water modelling.Here, the effects of a conventional tillage (CT), reduced tillage (RT) and no tillage (NT) system on the HCC were investigated on an experimental field with a Haplic Luvisol in Eastern Germany throughout part of a winter wheat growing season. This included changes following agricultural management such as stubble breaking and seedbed preparation. Hood infiltrometer measurements were conducted in the field and transient evaporation experiments in the laboratory were performed on undisturbed soil cores to describe the HCC over a wide range of pressure heads (h) from saturation to −1000 cm (pF 3).On tilled plots, Ks and hydraulic conductivity at h = −2 cm (K−2 cm) were more variable with time than observed spatial variability. Overall, Ks was significantly (p < 0.05) higher under RT compared to NT while for CT they ranged in between. Correlation and multiple linear regression indicated a distinctly different soil structure between tilled and untilled treatments. While bulk density and macro-and mesoporosity could explain some variability in Ks and K−2 cm on CT and RT, it was meaningless for the untilled soil. A denser soil matrix on NT with few conducting macro-and biopores was attributed to those findings. Under tillage, the loosened soil matrix and decaying organic matter mixed into the topsoil likely govern water transport at and near saturation. Over the entire HCC, variability decreased with drier conditions on NT while this was less pronounced on tilled soil, especially RT, indicating that tillage may affect not only soil macroaggregates but also the soil matrix. In this specific case, variability of the HCC with time was shown to be large for tilled plots, and modeling of soil water may benefit from an explicit consideration of these changes. Contrasts in soil structure under tilled and untilled soil should also be reflected in the future development of pedotransfer functions for the prediction of (near-) saturated K. Different soil physical property predictors may be needed for a valid estimation.

Prediction of Soil Hydraulic Conductivity at Saturation using Air Permeability at Any Individual Soil Water Content

Several studies aimed at linking hydraulic conductivity at saturation (Kw,s) to air permeability (Ka(θw)) of soil at given water content (θw) since it can be measured more rapidly and nondestructively than Kw,s especially regarding some new in situ technologies for Ka(θw) measurement. Following this, the current research aimed to develop a semi-theoretical relation between Kw,s and Ka(θw) using measured data in 27 soil samples. The Ka(θw) was measured at 12 different θw contents between 1.5 to 1,500 kPa suctions. Applying these measured data, we proposed a semi-theoretical function to predict Kw,s using Ka(θw) as input variable. The results showed that the proposed function was able to predict Kw,s using Ka(θw) at any individual θw content with really high accuracy consisting of R2 = 0.986 and evaluating error (ER) of the 2% between measured and predicted Kw,s. However, the outcomes revealed that Ka(θw) measurement at lower θw contents resulted in greater accuracy for proposed model. The pertinent section of the article applied multivariate linear regression (MLR) to develop pedo-transfer functions (PTFs) to estimate the parameters of the proposed model. The results revealed that the developed PTFs had relatively greater accuracy and reliability showing average determination coefficient (R2) of 0.807 and 0.729 for training and test datasets, respectively. However, more detailed investigation with wide range of soil parameters are needed for more general PTFs development.

Machine Learning Approaches to Develop Pedotransfer Functions for Tropical Sri Lankan Soils

Poor data availability on soil hydraulic properties in tropical regions hampers many studies, including crop and environmental modeling. The high cost and effort of measurement and the increasing demand for such data have driven researchers to search for alternative approaches. Pedotransfer functions (PTFs) are predictive functions used to estimate soil properties by easily measurable soil parameters. PTFs are popular in temperate regions, but few attempts have been made to develop PTFs in tropical regions. Regression approaches are widely used to develop PTFs worldwide, and recently a few attempts were made using machine learning methods. PTFs for tropical Sri Lankan soils have already been developed using classical multiple linear regression approaches. However, no attempts were made to use machine learning approaches. This study aimed to determine the applicability of machine learning algorithms in developing PTFs for tropical Sri Lankan soils. We tested three machine learning algorithms (artificial neural networks (ANN), k-nearest neighbor (KNN), and random forest (RF)) with three different input combination (sand, silt, and clay (SSC) percentages; SSC and bulk density (BD); SSC, BD, and organic carbon (OC)) to estimate volumetric water content (VWC) at −10 kPa, −33 kPa (representing field capacity (FC); however, most studies in Sri Lanka use −33 kPa as the FC) and −1500 kPa (representing the permanent wilting point (PWP)) of Sri Lankan soils. This analysis used the open-source data mining software in the Waikato Environment for Knowledge Analysis. Using a wrapper approach and best-first search method, we selected the most appropriate inputs to develop PTFs using different machine learning algorithms and input levels. We developed PTFs to estimate FC and PWP and compared them with the previously reported PTFs for tropical Sri Lankan soils. We found that RF was the best algorithm to develop PTFs for tropical Sri Lankan soils. We tried to further the development of PTFs by adding volumetric water content at −10 kPa as an input variable because it is quite an easily measurable parameter compared to the other targeted VWCs. With the addition of VWC at −10 kPa, all machine learning algorithms boosted the performance. However, RF was the best. We studied the functionality of finetuned PTFs and found that they can estimate the available water content of Sri Lankan soils as well as measurements-based calculations. We identified RF as a robust alternative to linear regression methods in developing PTFs to estimate field capacity and the permanent wilting point of tropical Sri Lankan soils. With those findings, we recommended that PTFs be developed using the RF algorithm in the related software to make up for the data gaps present in tropical regions.