Estimation Of Soil Properties Research Articles

Gaussian process regression for three-dimensional soil mapping over multiple spatial supports

This study investigates the complexity of spatial soil modelling, particularly focusing on the challenge of variable vertical support in traditional soil data collection. Traditional soil sampling, described in terms of horizons, often fails to accurately pinpoint the specific depths for specific soil properties. This gap is significant, as depth-specific data is crucial for a thorough understanding of soil formation processes and for assessing potential environmental impacts. In digital soil mapping (DSM), the prevalent reliance on standardised depth intervals and mass-preserving spline functions for data resampling results in a modelling approach that tends to disregard depth-related details, thereby introducing potential uncertainties.To address these limitations, this work explores how effectively Gaussian process regression (GPR) can model soil in 3D. This technique comprises two key components: a mean function and a semivariogram-like kernel. Unlike conventional methods that make a single prediction, GPR provides detailed probability distributions. This capability allows for the quantification of prediction uncertainty at various points, offering insights for decision-making and risk assessment purposes. Moreover, GPR has the capability to make volume or block estimates and assess associated uncertainties. Enabling volume-based predictions enriches the range of strategies available for land management.In this research, we employ GPR as a novel 3D soil modelling technique and compare its performance with traditional spline-based methods. The comparison is conducted through a case study on a farm in northern New South Wales, Australia, focusing on the 3D mapping of soil pH and electrical conductivity (EC). Our results demonstrate the ability of GPR to estimate soil properties across various volumes, utilising data from multiple vertical supports, thereby offering a more versatile approach for soil modelling in diverse spatial contexts.

Predicting the shear modulus and damping ratio of soils using machine learning models

In geotechnical engineering, the accurate estimation of fundamental soil properties, such as the shear modulus ratio (G/G max) and damping ratio (D), is crucial to design and analyze various structures subjected to dynamic loads. This study presents a comprehensive investigation on harnessing the power of machine learning techniques to precisely predict G/G max and D of granular soils. Using an extensive dataset gathered from cyclic triaxial and resonant column tests on diverse mixtures of sand and gravel, combined with previous research findings, a series of advanced machine learning algorithms including shallow neural networks, support vector regression, gradient boosting regression, and deep feed forward neural network (DFFNN) were developed. The proposed models elucidate various influential parameters, including the grading characteristics, void ratio, confining pressure, consolidation stress ratio, and specimen preparation techniques. The superiority of the DFFNN model in terms of accuracy and predictive performance was demonstrated through rigorous evaluation and comparison. This study contributes to a better understanding of soil behavior under dynamic conditions. It also provides a robust framework to employ machine learning in predicting G/G max and D of granular soils, thereby enhancing the efficiency and reliability of geotechnical designs and construction practices.

Evaluating performance of miniaturized spectrometers in predicting soil properties using multivariate statistical analysis

Despite the increasing use of portable, low-cost spectrometers in estimating soil properties, there is lack of documentation regarding the factors contributing to the lower performance of these spectrometers when compared to conventional ones. This study investigates potential factors influencing performance of the Nanoquest, a low-cost spectrometer, in estimating soil organic carbon (SOC) and total nitrogen (TN). To conduct the study, five different models (cubist, partial least squares regression, support vector machines, random forest, and generalised boosted models) were tested for the estimation SOC and TN and a fivefold cross-validation analysis was conducted for model hyperparameter optimization. The Nanoquest achieved a Lin’s concordance correlation coefficient (CCC) value of 0.84 and an R2 value of 0.74 for SOC. For TN, CCC values of 0.86 and an R2 value of 0.78 were obtained. To understand the impact of the spectral range and spectral resolution on SOC and TN estimation, the ASD spectra were digitally resampled to match the Nanoquest spectral range and resolution. This resampling resulted in a slight decrease in model performance for the spectral range and a more pronounced decrease for the spectral resolution.

Adapting Prediction Models to Bare Soil Fractional Cover for Extending Topsoil Clay Content Mapping Based on AVIRIS-NG Hyperspectral Data

The deployment of remote sensing platforms has facilitated the mapping of soil properties to a great extent. However, the accuracy of these soil property estimates is compromised by the presence of non-soil cover, which introduces interference with the acquired reflectance spectra over pixels. Therefore, current soil property estimation by remote sensing is limited to bare soil pixels, which are identified based on spectral indices of vegetation. Our study proposes a composite mapping approach to extend the soil properties mapping beyond bare soil pixels, associated with an uncertainty map. The proposed approach first classified the pixels based on their bare soil fractional cover by spectral unmixing. Then, a specific regression model was built and applied to each bare soil fractional cover class to estimate clay content. Finally, the clay content maps created for each bare soil fractional cover class were mosaicked to create a composite map of clay content estimations. A bootstrap procedure was used to estimate the standard deviation of clay content predictions per bare soil fractional cover dataset, which represented the uncertainty of estimations. This study used a hyperspectral image acquired by the Airborne Visible/Infrared Imaging Spectrometer-Next Generation (AVIRIS-NG) sensor over cultivated fields in South India. The proposed approach provided modest performances in prediction (Rval2 ranging from 0.53 to 0.63) depending on the bare soil fractional cover class and showed a correct spatial pattern, regardless of the bare soil fraction classes. The model’s performance was observed to increase with the adoption of higher bare soil fractional cover thresholds. The mapped area ranged from 10.4% for pixels with bare soil fractional cover >0.7 to 52.7% for pixels with bare soil fractional cover >0.3. The approach thus extended the mapped surface by 42.4%, while maintaining acceptable prediction performances. Finally, the proposed approach could be adopted to extend the mapping capability of planned and current hyperspectral satellite missions.

Using the surface scattering mechanism from dual-pol SAR data to estimate topsoil particle-sizefractions

Data extracted from Synthetic Aperture Radar (SAR) have been widely employed to estimate soil properties. However, these studies are typically constrained to bare soil conditions, as soil information retrieval in vegetated areas remains challenging. Polarimetric decomposition has emerged as a potentially useful method to separate the scattering contributions of different targets (e.g. canopy/leaves and the underlying soil), which is of significance for areas that are near-permanently covered in low-lying vegetation (e.g. grass) like Ireland – the study area for this investigation. Here, we test the surface scattering mechanism, derived from H-alpha dual-pol decomposition, together with other covariates, to estimate percentages of sand, silt, and clay, over vegetated terrain, using Sentinel 1 data (dual-pol C-band SAR). The statistical modelling approaches evaluated – linear regression (LRM) and tree-based regression models (machine learning) – explicitly consider the compositional nature of soil texture. When compared to the models fitted without surface scattering data, results showed that the inclusion of the surface scattering data improved estimates of silt and clay, with the compositional linear regression model, and estimates of sand and silt fractions with different tree-based models. While not without limitations, our study demonstrated that the polarimetric decomposition method, which is typically used for classification and segmentation purposes, could also be used for soil property estimation, broadening the application of this technique in microwave remote sensing studies.

Evaluation of soil fertility using combination of Landsat 8 and Sentinel‑2 data in agricultural lands.

Today, remote sensing is widely used to estimate soil properties. Because it is an easy and accessible way to estimate soil properties that are difficult to estimate in the field. Based on this, to evaluate the soil fertility (SF), soil sampling was performed irregularly from the surface depth of 0-30cm in 216 points, 11 soil properties were measured, and the soil fertility index (SFI) was calculated by soil properties. Simultaneously, we combined satellite images of Landsat 8 and Sentinel-2 using the Gram-Schmidt algorithm. Finally, multiple linear regression SFI was calculated using satellite data, as well as the spatial distribution of SFI was obtained in very low, low, moderate, high, and very high classes. Our findings showed that the combination of Landsat 8 and Sentinel-2 data using the Gram-Schmidt algorithm has a higher correlation with SFI than when these data are individually. Therefore, combined Landsat 8 and Sentinel 2 data were used for SFI modeling. Using model selection procedure indices (including Cp, AIC, and ρc criteria), the visible range bands, notably blue (r = 0.65), green (r = 0.63), and red (r = 0.61), provide the best model for estimating SFI (R2 = 0.43, Cp = 3.34, AIC = -277.4, and ρc = 0.44). Therefore, these bands were used to estimate the SFI index. Also, the spatial distribution of the SIF index showed that the most significant area was related to the low class, and the lowest area belonged to the high and very high fertility classes. According to these results, it can be concluded that using the combination of Landsat 8 and Sentinel 2 bands to estimate soil fertility index in agricultural lands can increase the accuracy of soil fertility estimation.

Evaluating mobile applications for estimating soil properties: Quality of current apps, limitations and future directions

The estimation of soil properties is an important factor in agriculture. With the advancement of smartphones and artificial intelligence, estimating soil properties such as the Munsell soil color, pH, soil organic matter and moisture is tangible, alleviating the need for laboratories, which are time-consuming and costly. While there are several smartphone applications available in different app stores that claim to provide this functionality, an evaluation of the limitations and quality is required for these apps. The objective of this study is to identify existing soil property estimation apps, devise a soil apps rating scale to evaluate the functionality along with the software quality characteristics, and provide design guidelines for improving apps of this type. A thorough search was conducted in two prominent app stores: the Google Play and Apple App store. The selected apps were then rated by four individual raters. A total of 337 apps were found in the search and from this list, 32 selected apps were reviewed using the devised app rating scale. The rating scale evaluates the apps against ten sub-scales: app metadata, aesthetics, general app features, performance and efficiency, usability, functionality, subjective quality, transparency, user comments, and the perceived impact of the app on users. The inter-rater and intra-rater reliability are calculated along with the internal consistency of the rating scale to ensure the validity of the results. The evaluation results that returned the highest-ranking mean sub-scales were performance, usability, and aesthetics. Adversely, the lowest-ranging sub-scales were functionality and general features. There was also an evident absence of the functionality to calculate any type of soil properties in all but four of the apps reviewed. Several key design considerations were identified, such as the need for automated estimation of soil properties, an improvement of general features, and the inclusion of a shared soil database.

Estimating soil organic carbon content at variable moisture contents using a low-cost spectrometer

Research-grade spectrometers such as ASD are widely used in the lab to estimate soil properties, but they are bulky, heavy, and not easily deployable to measure field soils. The newer FT-NIR spectrometers are compact, lightweight, and robust, suitable for developing portable sensors for emerging applications such as field-based soil carbon stock assessment. In this study, we investigated the usefulness of an FT-NIR spectrometer (NanoQuest) for estimating SOC content while correcting for the effect of soil moisture using External Parameter Orthogonalization (EPO), and its performance was compared to that of ASD. To develop EPO transformation, five levels of soil moisture were used at 0, 0.07, 0.13, 0.18, 0.24, and 0.30 g g−1. We tested two modeling approaches: Partial Least Squares Regression (PLSR) and Support Vector Regression (SVR). The results showed that EPO was more effective in correcting for the moisture effect as samples became drier. ASD gave a better performance in estimating SOC with SVR (R2: 0.17 to 0.84, RMSE: 6.1 to 3.9 g C kg−1, bias: −0.3 to 0.1 g C kg−1) after EPO transformation. NanoQuest gave slightly lower, but still satisfactory performance in SOC estimation (R2: 0.17 to 0.70, RMSE: 9.2 to 5 g C kg−1, bias: −0.3 to 0.1 g C kg−1). EPO substantially reduced the bias of the SOC models for both ASD and NanoQuest. This study demonstrates the usefulness of low-cost FT-NIR spectrometers for SOC measurement at varying moisture contents and their great potential for field-deployable soil sensor development.

Estimating the spatial distribution of soil volumetric water content in an agricultural field employing remote sensing and other auxiliary data under different tillage management practices

AbstractKnowledge of soil volumetric water content (VWC) on agricultural soils as influenced by different soil management practices is important, but the measurement outputs of frequently used traditional sampling techniques are restricted to point‐based measurements with limited spatial coverage. Remote sensing (RS) techniques are therefore being explored because of their greater spatial and temporal availability as well as their ability to cover large‐scale areas. But a general limitation for RS data is the presence of vegetation cover, cloud cover and the effect of topography. To minimize the effects of these factors during field sampling, the use of spectral indices and terrain attributes have proven successful in the estimation of several soil properties; however, the impact of these approaches for the estimation and mapping of soil VWC, an important factor in crop growth and development, especially where different soil management practices are employed, remains limited. To contribute to the knowledge base of RS under varying soil systems, this study explores the possibility of combining in‐situ measurements with remotely sensed (explanatory variables) data obtained under four different tillage practices to produce an estimated soil VWC that represents the entire study field. The tillage practices used include reduced till (RT), strategic till (ST), no‐till (NT) and conventional till (CT). From these tillage plots, three explanatory datasets, namely Sentinel‐2 (S2), spectral indices (SI) and terrain attributes (TA), were collected as predictors. In addition, each of the explanatory variables was structured into four groups, representing each of the four tillage methods. The eXtreme Gradient Boosting (XGBoost) model was used, and the best results were selected based on the root mean squared error (RMSE), the coefficient of determination (R2) and the mean absolute error (MAE). Prior to soil VWC prediction, the Pearson correlation matrix was used to determine the linear relationship between each of the selected explanatory variables (S2 bands, SI and TA) and soil VWC. Furthermore, spatial distribution maps of soil VWC were constructed using the inverse distance weighting (IDW) interpolation technique. For soil VWC estimation, the TA outperformed the SI and S2 datasets (R2 = 0.84). Similarly, the spatial distribution maps obtained from the TA data show a study area with a large concentration of VWC compared with the other datasets. The study also identified CT as the tillage approach that most impacts soil VWC estimation because all predicted results were poor without the addition of data from the CT plot. According to our findings, using TA data collected from various tillage management systems to estimate soil VWC is very promising because they can be used as predictors to improve soil VWC estimation and mapping. This study demonstrates conclusively that remote sensing data collected from different tillage management systems can be used as predictors to enhance the estimation and mapping of soil VWC, complementing the basis for the development of reliable and consistent precision irrigation management systems.

Impact of Spectral Resolution and Signal-to-Noise Ratio in Vis–NIR Spectrometry on Soil Organic Matter Estimation

Recently, considerable efforts have been devoted to the estimation of soil properties using optical payloads mounted on drones or satellites. Nevertheless, many studies focus on diverse pretreatments and modeling techniques, while there continues to be a conspicuous absence of research examining the impact of parameters related to optical remote sensing payloads on predictive performance. The main aim of this study is to evaluate how the spectral resolution and signal-to-noise ratio (SNR) of spectrometers affect the precision of predictions for soil organic matter (SOM) content. For this purpose, the initial soil spectral library was partitioned into to two simulated soil spectral libraries, each of which were individually adjusted with respect to the spectral resolutions and SNR levels. To verify the consistency and generality of our results, we employed four multiple regression models to develop multivariate calibration models. Subsequently, in order to determine the minimum spectral resolution and SNR level without significantly affecting the prediction accuracy, we conducted ANOVA tests on the RMSE and R2 obtained from the independent validation dataset. Our results revealed that (i) the factors significantly affecting SOM prediction performance, in descending order of magnitude, were the SNR levels > spectral resolutions > estimation models, (ii) no substantial difference existed in predictive performance when the spectral resolution fell within 100 nm, and (iii) when the SNR levels exceeded 15%, altering them did not notably affect the SOM predictive performance. This study is expected to provide valuable insights for the design of future optical remote sensing payloads aimed at monitoring large-scale SOM dynamics.

Estimation of Soil Characteristics from Multispectral Sentinel-3 Imagery and DEM Derivatives Using Machine Learning.

In this paper, different machine learning methodologies have been evaluated for the estimation of the multiple soil characteristics of a continental-wide area corresponding to the European region, using multispectral Sentinel-3 satellite imagery and digital elevation model (DEM) derivatives. The results confirm the importance of multispectral imagery in the estimation of soil properties and specifically show that the use of DEM derivatives improves the quality of the estimates, in terms of R2, by about 19% on average. In particular, the estimation of soil texture increases by about 43%, and that of cation exchange capacity (CEC) by about 65%. The importance of each input source (multispectral and DEM) in predicting the soil properties using machine learning has been traced back. It has been found that, overall, the use of multispectral features is more important than the use of DEM derivatives with a ration, on average, of 60% versus 40%.

Combining ground penetrating radar methodologies enables large‐scale mapping of soil horizon thickness and bulk density in boreal forests

AbstractForest soil properties must be observed with the appropriate resolution by depth and landscape area to understand biogeomorphological controls on soil carbon (C). These observations, particularly in boreal forests, have been limited because of the poor resolution and unavailability of physical soil sampling results, especially for soil bulk density measurements. Ground penetrating radar (GPR) has been demonstrated to non‐destructively and continuously estimate forest soil properties required in Cstock estimates, such as soil horizon thickness and soil bulk density, across small spatial scales and shallow depths. Yet, successful small‐scale forest GPR approaches represent a potential opportunity to obtain soil property estimates at relevant resolution and depth across forest landscapes, enabling improvement to much needed soil mapping and stock estimates. This review discusses the existing soil property studies that utilize ground penetrating radar (GPR) and explores how the adaptation of GPR methodology can contribute to investigating soils in forest landscapes. We have identified common GPR surveying practices, data processing steps and interpretation methods employed in multiple studies. These approaches have proven effective in obtaining higher‐resolution estimates of important soil properties, such as bulk density and horizon thickness, within small‐scale forest plots. By applying relevant findings in this review to our own boreal forest investigation across an 80 m hillslope transect, we provide recommendations on how to tailor GPR methodology for landscape‐scale estimates of soil horizon thickness and bulk density to examine forest soil property distribution. These findings should enable the future collection of soil datasets informing the distribution of soil C stocks and their relationship to landscape features, and thus their controls and responses to climate and environmental change.

Simultaneous estimation of multiple soil properties under moist conditions using fractional-order derivative of vis-NIR spectra and deep learning

The application of visible and near-infrared (vis-NIR) spectroscopy for predicting soil properties presents a cost-effective and time-efficient approach for evaluating various soil properties. However, the accuracy of these predictions can be adversely affected by soil moisture. Consequently, previous research focused on removing moisture effects from the spectra, resulting in the loss of valuable information on soil moisture. In this study, we propose a novel approach to directly predict both soil moisture and a range of soil properties using vis-NIR spectra. To achieve this, we employed the fractional-order derivative (FOD) technique in combination with a one-dimensional convolutional neural network (1D-CNN) to mitigate the influence of soil moisture and enhance the prediction accuracy of soil properties. We calibrated 1D-CNN, partial least squares regression (PLSR) and support vector machine (SVM) models using FOD spectra with various orders (ranging from 0 to 2 with 0.1 intervals). The results demonstrated that the FOD spectra consistently outperformed the original and integer-order differential spectra. Moreover, the 1D-CNN models exhibited superior performance compared to PLSR and SVM models in predicting soil moisture content (SMC), soil organic matter (SOM), sand, silt, clay, and iron contents. However, it is challenging to predict all six soil properties using a single 1D-CNN model with spectra of a single FOD order. Our findings identified the optimal FOD order for different soil properties. Specifically, a FOD order of 0.5 yielded the best results for predicting SMC and SOM, while an order of 0.7 was optimal for iron oxides, clay, silt, and sand contents. Consequently, we propose the combined use of FOD and 1D-CNN to estimate multiple soil properties simultaneously.

Spectral prediction of soil salinity and alkalinity indicators using visible, near-, and mid-infrared spectroscopy

Drought and the impacts of climate change have led to an escalation in soil salinity and alkalinity across various regions worldwide, including Iran. The Chahardowli Plain in western Iran, in particular, has witnessed a significant intensification of this phenomenon over the past decade. Consequently, modeling of soil attributes that serve as indicators of soil salinity and alkalinity became a priority in this region. To date, only a limited number of studies have been conducted to assess indicators of salinity and alkalinity through spectrometry across diverse spectral ranges. The spectral ranges encompassing mid-infrared (mid-IR), visible, and near-infrared (vis-NIR) spectroscopy were employed to estimate soil properties including sodium adsorption ratio (SAR), exchangeable sodium ratio (ESR), exchangeable sodium percentage (ESP), pH, and electrical conductivity (EC). Five distinct models were employed: Partial Least Squares Regression (PLSR), bootstrapping aggregation PLSR (BgPLSR), Memory-Based Learning (MBL), Random Forest (RF), and Cubist. The calibration and assessment of model performance were carried out using several key metrics including Ratio of Performance to Deviation (RPD) and the coefficient of determination (R2). Analysis of the outcomes indicates that the accuracy and precision of the mid-IR spectra surpassed that of vis-NIR spectra, except for pH, which exhibited a superior RPD compared to other properties. Notably, in the prediction of pH utilizing vis-NIR reflectance spectra, the BgPLSR model exhibited the highest accuracy and precision, boasting an RPD value of 2.56. In the domain of EC prediction, the PLSR model yielded an RPD of 2.64. For SAR, the MBL model achieved an RPD of 2.70, while ESR prediction benefited from the MBL model with an impressive RPD of 4.36. Likewise, the MBL model demonstrated remarkable precision and accuracy in ESP prediction, garnering an RPD of 4.41. The MBL model's efficacy in forecasting with limited datasets was notably pronounced among the models considered. This study underscores the valuable role of spectral predictions in facilitating the work of soil surveyors in gauging salinity and alkalinity indicators. It is recommended that the integration of spectrometry-based salinity and alkalinity predictions be incorporated into forthcoming soil mapping endeavors within semi-arid and arid regions.

A deep scalable neural architecture for soil properties estimation from spectral information

In this paper we propose an adaptive deep neural architecture for the prediction of multiple soil characteristics from the analysis of hyperspectral signatures. The proposed method overcomes the limitations of previous methods in the state of art: (i) it allows to predict multiple soil variables at once; (ii) it is based on a flexible neural architecture capable of automatically adapting to the spectral library under analysis. As an additional feature, our method permits to backtrace the spectral bands that most contribute to the estimation of a given variable. The proposed architecture is experimented on LUCAS, a large laboratory dataset and on a dataset achieved by simulating PRISMA hyperspectral sensor. Results, compared with other state-of-the-art methods confirm the effectiveness of the proposed solution. In fact, the proposed method achieves an overall R2 of about 0.75. In detail, pHCaCl2 and CaCO3 are predicted with an R2 above 0.90, pHH2O, OC, N, clay, sand, CEC, silt are predicted with an R2 between 0.89 and 0.70, K and P with an R2 respectively of 0.54 and 0.37. The proposed method overcomes, in terms of R2, the state of the art of about 15% on the LUCAS dataset and of about 24% on the PRISMA simulated dataset.

Different approaches to estimating soil properties for digital soil map integrated with machine learning and remote sensing techniques in a sub-humid ecosystem.

Today, data mining has become a relevant topic in digital soil mapping. In this current study, prediction of some soil properties and their spatial distribution were examined by machine learning algorithms (Support Vector Machine, Artificial Neural Network) using reflectance values of Triplesat satellite image bands in Vezirköprü district of Samsun province. The band data obtained from different wavelengths revealed positive correlations between the electrical conductivity and calcium carbonate equivalent contents of the soils. The support vector machine algorithm was the most successful to estimate the textural fractions, organic matter, electrical conductivity, and calcium carbonate equivalent contents of the soils using the bands obtained from satellite images. The mean absolute error for estimating sand, silt, and clay contents by support vector machine was 4.05%, 3.05%, and 3.66%, respectively. Texture classes were determined with an accuracy of 82% with support vector machine and 60% with artificial neural network. In all estimations, the highest percentage of error was for calcium carbonate equivalent content with very low estimation reliability. The mean absolute percentage of error values for this property are 101.13% and 51.61% for artificial neural network and support vector machine, respectively. Also, in both algorithms, the most successfully estimated soil property was clay fraction of soils. It was also investigated the spatial distribution of actual and estimated values using various interpolation methods (Kriging, inverse distance weighting-, radial basis function). Considering the spatial distributions, it was determined that the most successful method was kriging for sand, silt, and clay contents and inverse distance weighting for electrical conductivity, calcium carbonate equivalent, and organic matter contents. According to our findings, it is concluded that successful estimations and spatial distributions can be made by the support vector machine algorithm using band data from different wavelengths.

Development of Hungarian spectral library: Prediction of soil properties and applications

Updating soil information systems (SIS) requires advanced technologies to support the time and cost-effective and environment-friendly soil data. The use of mid- infrared (MIR) Spectroscopy as alternative to wet chemistry has been tested. The MIR spectral library is a useful technique for predicting soil attributes with high accuracy, efficiency, and low cost. The Hungarian MIR spectral library contained data on 2200 soil samples from 10 counties representing the first Soil Information and Mentoring System (SIMS) survey. Archived soil samples were prepared and scanned based on Diffuse Reflectance Infrared spectroscopy (DRIFT) technique and spectra data were saved in the fourier transform infrared (FTIR) spectrometer OPUS software. Preprocessed filtering methods, outlier detection methods and calibration sample selection methods were applied for spectral library. MIR calibration models were built for soil attributes using Partial Least Square Regression (PLSR) method. Coefficient determination (R2), The Root Mean Squared Error (RMSE) and Ratio of Performance to Deviation (RPD) were used to assess the goodness of calibration and validation models. MIR spectral library had the ability to significantly estimate soil properties such as SOC, CaCO3, sand, clay and silt through various scale models (national, county and soil type). The findings showed that our spectral library soil estimations are precise enough to provide information on national, county and soil type levels enabling a wide range of soil applications that demand huge amounts of data such as soil survey, precision agriculture and digital soil mapping.

Spectral fusion modeling for soil organic carbon by a parallel input-convolutional neural network

Visible-to-near-infrared (vis–NIR) and mid-infrared (MIR) spectroscopy have been widely utilized for the quantitative estimation of soil organic carbon (SOC). The fusion of vis–NIR and MIR data can be hypothesized to provide accurate and reliable prediction for SOC because spectral data within a specific range of each individual sensor may lack important absorptive features associated with SOC. In this study, six data fusion strategies, principally direct concatenation-partial least squares regression (DC-PLSR), outer product analysis-PLSR (OPA-PLSR), OPA-competitive adaptive reweighted sampling-PLSR (OPA-CARS-PLSR), sequentially orthogonalized-PLSR (SO-PLSR), DC-convolutional neural network (DC-CNN), and parallel input-CNN (PI-CNN), were compared for the spectral estimations of SOC. The data fusion and individual sensor models were developed using soil samples collected from Zhejiang Province, East China, and scanned under laboratory conditions with both vis–NIR and MIR spectrophotometers. The validation results of vis–NIR (validation coefficient of determination [R2] = 0.63–0.73) were generally better than those of MIR (validation R2 = 0.45–0.59). For data fusion, the best validation accuracy was achieved by the PI-CNN (validation R2 = 0.84), followed in descending order by DC-CNN (validation R2 = 0.78), SO-PLSR (validation R2 = 0.73), OPA-CARS-PLSR (validation R2 = 0.69), OPA-PLSR (validation R2 = 0.66), and DC-PLSR (validation R2 = 0.64). The better performance of PI-CNN over DC-CNN demonstrates the necessity of using different sizes of convolutional kernels before feeding into the fully connected layers in the CNN network for fusing vis–NIR and MIR spectral data. The deep-learning fusion method based on PI-CNN can be considered an efficient tool for integrating data from multiple sensors for estimating soil properties in the field of soil spectral modeling.

Evaluation of a miniaturized portable NIR spectrometer for the prediction of soil properties in Mediterranean central Chile

Near-infrared (NIR) spectrometry to estimate soil properties has been evolving toward more portable and accessible equipment in recent years for the prediction of indicators of soil fertility in agroecosystems. There were two objectives of this study. First, to calibrate and validate a miniaturized portable NIR spectrometer NeoSpectra-Module 2.5 with a limited spectral range (1350–2500 nm) for the prediction of total nitrogen (TN), soil organic carbon (SOC), soil pH, clay content and exchangeable cations (Ca2+, Mg2+, K+ and Na+) in dry conditions. And second, to evaluate the use of an external orthogonalization parameter (EPO) to reduce the effect of soil moisture in samples with field moisture conditions collected from Mediterranean central Chile (33° SL to 36° SL). Soil classified as Mollisols, Alfisols and Inceptisols was sampled (n = 170) from 0 to 30 cm depth and scanned with the NeoSpectra spectrometer. The spectrometer showed good performance (R2 > 0.5 and/or RPIQ>1.4) for predictions of SOC and exchangeable cations. When the spectra data without transformation was used (raw spectra) versus transformed spectra, better performance in the predictions was obtained for TN, SOC, pH, clay content and Ca2+. The cubist model (CM) showed better performance than partial least squares (PLSR) for most assessed soil properties. The EPO procedure allowed us to decrease the error due to soil moisture in the prediction of some soil properties. Although the use of a miniaturized portable NIR spectrometer showed promising results for prediction of soil properties, future testing should increase the number of soil samples, vary the spectral data processing, and/or test new miniaturized portable NIR spectrometers with a wider spectral range.

Prediction of some selected soil properties using the Hungarian Mid-infrared spectral library

Routine soil chemical and physical laboratory analysis provides a better understanding of the soil by evaluating its quality and functions. Demands for the development of national Mid-infrared (MIR) spectral libraries for predicting soil attributes with high accuracy have risen substantially in the recent past. Such MIR spectral library is usually regarded as a fast, cheap and non-destructive technique for estimating soil properties compared to laboratory soil analysis. The main objective of this research was to assess the performance of the Hungarian MIR spectral library in estimating four soil properties namely: Cation Exchange Capacity (CEC), Exchangeable Mg and Ca and pH water at different scenarios. Archived soil samples were scanned and spectra data were saved in the Fourier transform infrared spectrometer OPUS software. Preprocessed filtering, outlier detection and calibration sample selection methods were applied to the spectral library. MIR calibration models were built for soil attributes using partial least square regression method and the models were validated with sample predictions. R2, RMSE and RPD were used to assess the goodness of calibration and validation models. MIR spectral library had the ability to estimate soil properties such as CEC and exchangeable Ca and Mg through various scale models (national, county and soil type). The findings showed that the Hungarian MIR spectral library for estimation of soil properties has the ability to provide good information on national, county and soil type scales at different levels of accuracy.