Prediction of Soil-Available Potassium Content with Visible Near-Infrared Ray Spectroscopy of Different Pretreatment Transformations by the Boosting Algorithms

Jin, Xiu; Li, Shaowen; Wu, Zhang; Zhu, Juanjuan; Sun, Jia

doi:10.3390/app10041520

Cited by 31 publications

(14 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This paper used determination coefficients (R 2 ), the root mean square error (RMSE), and the ratio of performance to deviation (RPD) to evaluate the accuracy of model inversion. These three statistical indicators have been widely used for regression models of the Vis-NIR spectrum [22,53,76,77]. Regarding the stability and fitness of the R 2 reaction model, the closer R 2 approached 1, the higher the model fit and the more stable it was.…”

Section: Accuracy Comparisonmentioning

confidence: 99%

“…The total number of soil samples was 105, which were split into a calibration set and a validation set by the Kennard-Stone (KS) method at a proportion of 7:3 [53]. In reality, the levels of the soil properties were affected not only by the natural environment but also by human behaviors (i.e., before sample collection, there were fertilization activities in the field).…”

Section: Descriptive Statistics Of the Soil Nutrientsmentioning

confidence: 99%

“…In terms of model analysis methods, there are many methods, such as multiple linear regression (MLR), least squares regression, and partial least squares regression (PLSR), which are easy to use and act as simple model structures. Among them, PLSR is a commonly used method in hyperspectral regression modeling [17,30,44,[51][52][53]. However, in reality, correlations between wavelength variables and soil nutrients are rarely linear.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Evaluating Calibration and Spectral Variable Selection Methods for Predicting Three Soil Nutrients Using Vis-NIR Spectroscopy

Guo

Gao

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

Soil nutrients, including soil available potassium (SAK), soil available phosphorous (SAP), and soil organic matter (SOM), play an important role in farmland soil productivity, food security, and agricultural management. Spectroscopic analysis has proven to be a rapid, nondestructive, and effective technique for predicting soil properties in general and potassium, phosphorous, and organic matter in particular. However, the successful estimation of soil nutrient content by visible and near-infrared (Vis-NIR) reflectance spectroscopy depends on proper calibration methods (including preprocessing transformation methods and multivariate methods for regression analysis) and the selection of appropriate variable selection techniques. In this study, raw spectrum and 13 preprocessing transformations combined with 2 variable selection methods (competitive adaptive reweighted sampling (CARS) and the successive projections algorithm (SPA)) and 2 regression algorithms (support vector machine (SVM) and partial least squares regression (PLSR)), for a total of 56 calibration methods, were investigated for modeling and predicting the above three soil nutrients using hyperspectral Vis-NIR data (400–2450 nm). The results show that first-order derivatives based on logarithmic and inverse transformations (FD-LGRs) can provide better predictions of soil available potassium and phosphorous, and the best form of soil organic matter transformation is SG+MSC. CARS was superior to the SPA in selecting effective variables, and the PLSR model outperformed the SVM models. The best estimation accuracies (R2, RMSE) for soil available potassium, phosphorous, and organic matter were 0.7532, 32.3090 mg/kg; 0.7440, 6.6910 mg/kg; and 0.9009, 3.2103 g/kg, respectively, and their corresponding calibration methods were (FD-LGR)/SPA/PLSR, (FD-LGR)/SPA/PLSR, and SG+MSC/CARS/SVM, respectively. Overall, for the prediction of the soil nutrient content, organic matter was superior to available phosphorous, followed by available potassium. It was concluded that the application of hyperspectral images (Vis-NIR data) was an efficient method for mapping and monitoring soil nutrients at the regional scale, thus contributing to the development of precision agriculture.

show abstract

Section: Accuracy Comparisonmentioning

confidence: 99%

Section: Descriptive Statistics Of the Soil Nutrientsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Evaluating Calibration and Spectral Variable Selection Methods for Predicting Three Soil Nutrients Using Vis-NIR Spectroscopy

Guo

Gao

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Over the years, calibration models have been improved, more soil characteristics have been calibrated and validated, and many of these have gradually been introduced into routine soil testing (Tables 2 and 4). Although there is some debate regarding the reliability of the use of RPD to evaluate NIRS [65,66], it has been used for several years by NIR scientists working on agricultural products and has been adopted by soil scientists [67]. We choose to give the RPD values in addition to R 2 , the root mean squared error (RMSE), and bias.…”

Section: Near Infrared Spectroscopymentioning

confidence: 99%

Empower Innovations in Routine Soil Testing

et al. 2022

View full text Add to dashboard Cite

Conventional soil tests are commonly used to assess single soil characteristics. Thus, many different tests are needed for a full soil fertility/soil quality assessment, which is laborious and expensive. New broad-spectrum soil tests offer the potential to assess many soil characteristics quickly, but often face challenges with calibration, validation, and acceptance in practice. Here, we describe the results of a 20 year research program aimed at overcoming the aforementioned challenges. A three-step approach was applied: (1) selecting and establishing two contrasting rapid broad-spectrum soil tests, (2) relating the results of these new tests to the results of conventional soil tests for a wide variety of soils, and (3) validating the results of the new soil tests through field trials and communicating the results. We selected Near Infrared Spectroscopy (NIRS) and multi-nutrient 0.01 M CaCl2 extraction (1:10 soil to solution ratio; w/v) as broad-spectrum techniques. NIRS was extensively calibrated and validated for the physical, chemical, and biological characteristics of soil. The CaCl2 extraction technique was extensively calibrated and validated for ‘plant available’ nutrients, often in combination with the results of NIRS. The results indicate that the accuracy of NIRS determinations is high for SOM, clay, SOC, ECEC, Ca-CEC, N-total, sand, and inorganic-C (R2 ≥ 0.95) and good for pH, Mg-CEC, and S-total (R2 ≥ 0.90). The combination of the CaCl2 extraction technique and NIRS gave results that related well (R2 > 0.80) to the results of conventional soil tests for P, K, Mg, Na, Mn, Cu, Co, and pH. In conclusion, the three-step approach has revolutionized soil testing in The Netherlands. These two broad-spectrum soil tests have improved soil testing; have contributed to increased insights into the physical, chemical, and biological characteristics of soil; and have thereby led to more sustainable soil management and cropping systems.

show abstract

“…A larger RPIQ value indicates improved model performance. A similar classification can be used for RPIQ and RPD (Jin et al, 2020).…”

Section: Criteria For Model Selectionmentioning

confidence: 99%

Chemical Characterization of Volcanic Soils Using Near Infrared Spectroscopy (Nirs)

Ortega¹,

Valenzuela²,

Lagos³

et al. 2021

Chil. j. agric. anim. sci.

View full text Add to dashboard Cite

La cuantificación de la concentración de nutrientes en los suelos es importante para la implementación de estrategias de fertilización adecuadas y mejorar la productividad agrícola. En las últimas décadas se han desarrollado nuevos métodos instrumentales, tales como el uso de la espectroscopía de infrarrojo cercano (NIRS). Aunque su potencial ha sido reconocido por los científicos del suelo durante algunas décadas, el uso de NIRS para el análisis de rutina del suelo, según nuestro conocimiento, está poco desarrollado en Chile. Los modelos de calibración se desarrollaron utilizando la regresión de mínimos cuadrados parciales (PLSR). El objetivo de este estudio fue evaluar el uso de NIRS para la determinación de la concentración de calcio, magnesio, fósforo, potasio, pH (agua), materia orgánica, sodio, aluminio intercambiable, amonio y nitrato en suelos volcánicos del sur de Chile. El coeficiente de determinación en la calibración obtenida fluctuó entre 0,79-0,95 y 0,79-0,89, en Andisoles y Ultisoles, respectivamente; los valores de la desviación de predicción residual (RPD) variaron entre 2,1-4,4 y 2,2-3,1, en Andisoles y Ultisoles, respectivamente; y no fue posible generar modelos de calibración robustos para Al Exch, Na y NH 4 + en Andisoles y Al Exch, Mg, NO 3 y Na en Ultisoles. Aunque algunos de los modelos desarrollados presentaron elevados R 2 , esta metodología requiere una validación adicional que incluya un mayor número de muestras, una distribución espacial más amplia que cubra una variedad de condiciones climáticas y agrícolas.

show abstract

Prediction of Soil-Available Potassium Content with Visible Near-Infrared Ray Spectroscopy of Different Pretreatment Transformations by the Boosting Algorithms

Cited by 31 publications

References 32 publications

Evaluating Calibration and Spectral Variable Selection Methods for Predicting Three Soil Nutrients Using Vis-NIR Spectroscopy

Evaluating Calibration and Spectral Variable Selection Methods for Predicting Three Soil Nutrients Using Vis-NIR Spectroscopy

Empower Innovations in Routine Soil Testing

Chemical Characterization of Volcanic Soils Using Near Infrared Spectroscopy (Nirs)

Contact Info

Product

Resources

About