Article
Soil Science
Wartini Ng, Budiman Minasny, Edward Jones, Alex McBratney
Summary: This study evaluates different strategies to improve the model accuracy of a regional spectral library for soil organic carbon prediction and found that local and localized models perform better than other strategies when more than 20 local samples are used.
Article
Agricultural Engineering
M. I. S. Verissimo, C. Soares, C. F. Moreirinha, M. T. S. R. Gomes
Summary: In this study, a method based on NIR spectroscopy was used to predict the pour point and ethanol content of biodiesel mixtures. The results showed that this method has good prediction capability and is faster and more convenient compared to the traditional ASTM D97-08 procedure.
BIOMASS & BIOENERGY
(2023)
Article
Soil Science
Yongsheng Hong, Muhammad Abdul Munnaf, Angela Guerrero, Songchao Chen, Yaolin Liu, Zhou Shi, Abdul Mounem Mouazen
Summary: Spectral techniques, including the fusion of visible-to-near-infrared and mid-infrared absorbance, can improve the estimation of soil organic carbon. The use of continuous wavelet transform and optimal band combination strategies contribute to the accuracy of the prediction models. Among the investigated models, the combination of visible-to-near-infrared and mid-infrared using the optimal band combination fusion at a specific scale yielded the best prediction.
SOIL & TILLAGE RESEARCH
(2022)
Article
Forestry
Miao Long, Tianxiang Yue, Zhe Xu, Jiaxin Guo, Jie Luo, Xi Guo, Xiaomin Zhao
Summary: The rapid quantitative assessment of soil organic carbon (SOC) is crucial for understanding SOC dynamics and developing management strategies in forest ecosystems. Visible and near-infrared spectroscopy is an efficient and inexpensive technique widely used for predicting SOC content. By comparing different spiking strategies, a cost-effective and accurate method was developed for local-scale SOC assessment in target forest areas using a large soil spectral library.
Article
Environmental Sciences
Xianglin Zhang, Jie Xue, Yi Xiao, Zhou Shi, Songchao Chen
Summary: Soil visible and near-infrared (Vis-NIR, 350-2500 nm) spectroscopy has been proven as an alternative to conventional laboratory analysis. The study evaluated seven variable selection algorithms and three predictive algorithms in predicting soil properties using a regional soil Vis-NIR spectral library. The results showed that Cubist outperformed partial least squares regression (PLSR) and random forests (RF) in most soil properties when using the full spectra. The study provides valuable insights for predicting soil information using spectroscopic techniques and variable selection algorithms.
Article
Chemistry, Multidisciplinary
Xiaomi Wang, Can Yang, Mengjie Zhou
Summary: This study combines the PLS-MARS method and the MVARC-R-KS method to construct a nonlinear prediction model for soil organic matter, achieving local optimization considering spatial heterogeneity. Through a case study in Jianghan Plain, China, the proposed method shows superior accuracy and robustness compared to existing methods.
APPLIED SCIENCES-BASEL
(2021)
Article
Soil Science
Mervin St. Luce, Noura Ziadi, Raphael A. Viscarra Rossel
Summary: This article introduces a new approach, GLOBAL-LOCAL, which uses large soil spectral libraries to develop spectroscopic models that can fit locally. The method improves the accuracy of predicting soil organic carbon content and performs better at local scales. It also reduces analytical costs and improves predictions on local scales.
Article
Multidisciplinary Sciences
Divo Dharma Silalahi, Habshah Midi, Jayanthi Arasan, Mohd Shafie Mustafa, Jean-Pierre Caliman
Summary: The study introduced robust kernel partial least square regression methods to address potential issues caused by outliers and high leverage points in high-dimensional spectral datasets. Improvements on GM6 estimators were made to enhance the model's robustness and efficiency.
Article
Automation & Control Systems
J. Haritha, R. S. Valarmathi, M. Kalamani
Summary: In this study, the effect of different concentrations of urea on mid-infrared transmittance spectra in soil was analyzed using Partial Least Square Regression and Support Vector Machine algorithms. Results show that the Support Vector Machine model provides better prediction accuracy compared to the Partial Least Square Regression model.
INTELLIGENT AUTOMATION AND SOFT COMPUTING
(2022)
Article
Soil Science
Anatol Helfenstein, Philipp Baumann, Raphael Viscarra Rossel, Andreas Gubler, Stefan Oechslin, Johan Six
Summary: Efficient calibration sampling schemes and accurate modeling strategies were developed in this study to estimate soil carbon content in two drained peatland regions in Switzerland. By utilizing the spectral information in the SSL, a modeling approach combining RS-LOCAL and SSL was found to be effective in improving prediction accuracy, reducing field and laboratory work, and minimizing bias and uncertainty.
Article
Environmental Sciences
Everson Cezar, Tatiane Amancio Alberton, Evandro Freire Lemos, Karym Mayara de Oliveira, Liang Sun, Luis Guilherme Teixeira Crusiol, Marlon Rodrigues, Amanda Silveira Reis, Marcos Rafael Nanni
Summary: The quantification of soil organic matter (SOM) has been increasing in the Brazilian Cerrado region, where SOM content tends to be low. This study evaluated the performance of a local spectral model for SOM prediction using spectroradiometry. The results showed that recalibration of the local models improved the prediction accuracy, but further research is needed to improve the identification of SOM spatial variability.
Article
Chemistry, Applied
Wenfei Tian, Gengjun Chen, Guorong Zhang, Donghai Wang, Michael Tilley, Yonghui Li
Summary: Near-infrared spectroscopy can effectively predict the total phenolic content in whole wheat flour, avoiding the high cost and complexity of traditional methods, which is practical for wheat breeders.
Article
Agriculture, Multidisciplinary
Lixin Lin, Xixi Liu
Summary: The aim of this study was to investigate the potential of improving the accuracy of soil organic carbon estimation by combining the soil-moisture-index spectrum reconstruction method with partial least squares regression method.
PRECISION AGRICULTURE
(2022)
Article
Spectroscopy
Xiao-Wen Zhang, Zheng-Guang Chen, Feng Jiao
Summary: The dimensionality of near-infrared (NIR) spectral data is often large, and dimensionality reduction is crucial for increasing the model's performance. Laplacian Eigenmaps (LE) can preserve local neighborhood information but is disturbed by irrelevant information and multicollinearity. Random Frog (RF) algorithm can eliminate noise and collinearity. Hence, before using LE, RF is used to eliminate irrelevant information and reduce correlation, resulting in improved regression models' prediction accuracy and stability.
SPECTROCHIMICA ACTA PART A-MOLECULAR AND BIOMOLECULAR SPECTROSCOPY
(2023)
Article
Agriculture, Multidisciplinary
Fang Li, Li Xu, Tianyan You, Anxiang Lu
Summary: This study aims to improve the detection accuracy of key elements in the soil by combining X-ray fluorescence, near-infrared, and mid-infrared sensors. The XRF-MIR model using strategy (ii) showed better performance and more accurate predictions. Sensor fusion effectively enhances the accuracy of the spectrometer in detecting metal elements in the soil.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2021)
Article
Materials Science, Textiles
Lasse Christiansen, Yovko Ivanov Antonov, Rasmus Lund Jensen, Emmanuel Arthur, Lis Wollesen de Jonge, Per Moldrup, Hicham Johra, Peter Fojan
Summary: Fibre materials are commonly used for insulation in clothing and construction, and understanding the correlation between thermal conductivity and gas permeability can lead to improved energy efficiency and product quality. This study investigates this relationship for Rockwool, Kevlar, and polyester fibres, and finds that a transition zone exists where gas permeability and thermal conductivity change with compaction. This correlation can be used for rapid thermal conductivity assessment in various industries.
JOURNAL OF INDUSTRIAL TEXTILES
(2022)
Article
Soil Science
Yuting Fu, Lis W. de Jonge, Mogens H. Greve, Emmanuel Arthur, Per Moldrup, Trine Norgaard, Marcos Paradelo
Summary: The study revealed that grasslands exhibit the highest decomposition rate and stability among different land use types, with agricultural soils showing higher TBI values compared to seminatural soils. The inclusion of soil physicochemical properties in multiple linear regression analysis improved the prediction of stability in litter decomposition.
SOIL SCIENCE SOCIETY OF AMERICA JOURNAL
(2022)
Article
Soil Science
Peter Lystbaek Weber, Cecilie Hermansen, Trine Norgaard, Charles Pesch, Per Moldrup, Mogens H. Greve, Emmanuel Arthur, Lis Wollesen de Jonge
Summary: This study evaluated the particle density of soils from Southwest Greenland using a three-compartment model and compared it with pedotransfer functions and vis-NIR spectroscopic models. The results showed that Greenlandic soils have relatively high particle densities, and the accuracy of the three-compartment model was comparable to or better than traditional methods.
SOIL SCIENCE SOCIETY OF AMERICA JOURNAL
(2022)
Article
Soil Science
Mogens H. Greve, Kabindra Adhikari, Amelie Beucher, Goswin Heckrath, Bo V. Iversen, Maria Knadel, Mette B. Greve, Anders B. Moller, Yi Peng, Yannik E. Roell, Gasper L. Sechu
Article
Soil Science
Raphael A. Viscarra Rossel, Thorsten Behrens, Eyal Ben-Dor, Sabine Chabrillat, Jose Alexandre Melo Dematte, Yufeng Ge, Cecile Gomez, Cesar Guerrero, Yi Peng, Leonardo Ramirez-Lopez, Zhou Shi, Bo Stenberg, Richard Webster, Leigh Winowiecki, Zefang Shen
Summary: Spectroscopic measurements of soil samples are reliable and cost-effective for estimating soil properties. Machine learning is becoming more powerful in extracting information from spectra, and methods for interpreting the models exist.
EUROPEAN JOURNAL OF SOIL SCIENCE
(2022)
Review
Instruments & Instrumentation
Maria Knadel, F. Castaldi, R. Barbetti, E. Ben-Dor, A. Gholizadeh, R. Lorenzetti
Summary: Visible-near-infrared-shortwave-infrared (VNIR-SWIR) spectroscopy is a promising sensing technique for soil information. This paper reviews mathematical techniques for reducing or removing the effects of soil moisture content from spectra and provides examples of their effectiveness. The advantages and disadvantages of different techniques are summarized, and further research is recommended.
APPLIED SPECTROSCOPY REVIEWS
(2023)
Article
Soil Science
Anders Bjon Moller, Goswin Heckrath, Cecilie Hermansen, Trine N. orgaard, Lis Wollesen de Jonge, Mogens Humlekrog Greve
Summary: Managing soil phosphorus is crucial for agriculture and the environment. This study mapped the phosphorus sorption capacity (PSC) of Danish soils using a pedotransfer function based on Al-o and Fe-o. The main factors for predicting Al-o were parent material, topography, and precipitation, while soil texture, organic matter, and wetland areas were the main factors for Fe-o. The accuracy of the predictions was moderate, but the uncertainties were largest in wetland areas.
Article
Soil Science
Emmanuel Arthur, Markus Tuller, Trine Norgaard, Per Moldrup, Chong Chen, Hafeez Ur Rehman, Peter Lystbaek Weber, Maria Knadel, Lis Wollesen de Jonge
Summary: Soil specific surface area (SA) is influenced by clay content, organic carbon (OC) content, and clay mineralogy. The contribution of OC to SA varies depending on the clay type and the measurement method used. In this study, the contribution of OC to SA was quantified for different soil types using EGME and water adsorption techniques. The results showed that OC had a positive contribution to SA, except in montmorillonite-rich soils. The contribution of OC to SAH2O was higher than to SAE, and was influenced by the clay mineralogy.
Article
Environmental Sciences
Sonia Akter, Lis Wollesen de Jonge, Per Moldrup, Mogens Humlekrog Greve, Trine Norgaard, Peter Lystbaek Weber, Cecilie Hermansen, Abdul Mounem Mouazen, Maria Knadel
Summary: This study aimed to evaluate the feasibility of using visible near-infrared spectroscopy (vis-NIRS) for estimating the soil sorption coefficient (K-d) of glyphosate. Results showed that vis-NIRS had higher predictive ability for K-d compared to pedotransfer functions (PTFs) on the combined dataset, but PTFs provided slightly better estimations on the samples from Denmark and Greenland. However, the differences in prediction accuracy between vis-NIRS and PTF were statistically insignificant. Considering the multiple advantages of vis-NIRS, it can serve as a faster and easier alternative to PTFs for estimating glyphosate K-d.
Article
Spectroscopy
Alex Wangeci, Daniel Aden, Mogens H. Greve, Maria Knadel
Summary: Laser-induced breakdown spectroscopy (LIBS) is a promising technique for rapid and cost-effective determination of soil properties. This study investigated the effect of different sample pretreatments on pelletization of soil samples and prediction accuracy of texture and soil organic carbon (SOC) content. The results showed that sample pretreatment, such as sieving or milling, influenced the quality of soil pellets and the accuracy of texture and SOC prediction.
SPECTROCHIMICA ACTA PART B-ATOMIC SPECTROSCOPY
(2023)
Article
Environmental Sciences
Irene Navarro, Adrian de la Torre, Paloma Sanz, Isabelle Baldi, Paula Harkes, Esperanza Huerta-Lwanga, Trine Norgaard, Matjaz Glavan, Igor Paskovic, Marija Polic Paskovic, Nelson Abrantes, Isabel Campos, Francisco Alcon, Josefina Contreras, Abdallah Alaoui, Jakub Hofman, Anne Vested, Mathilde Bureau, Virginia Aparicio, Daniele Mandrioli, Daria Sgargi, Hans Mol, Violette Geissen, Vera Silva, Maria Angeles Martinez
Summary: Pesticide residues are commonly found in indoor dust, with lower levels in dust from organic farms. Some of the pesticides found are no longer approved for use and have toxic effects on human health and the environment.
SCIENCE OF THE TOTAL ENVIRONMENT
(2023)
Article
Agriculture, Multidisciplinary
Franck Albinet, Yi Peng, Tetsuya Eguchi, Erik Smolders, Gerd Dercon
Summary: This study demonstrates the use of a Convolutional Neural Network (CNN) model trained on a large soil spectral library to accurately determine and predict exchangeable potassium (Kex) content in soil. The CNN model outperforms the baseline model and provides important spectral features for predicting Kex.
ARTIFICIAL INTELLIGENCE IN AGRICULTURE
(2022)