Soil texture prediction using portable X-ray fluorescence spectrometry and visible near-infrared diffuse reflectance spectroscopy

Benedet, Lucas; Faria, Wilson Missina; Silva, Sérgio Henrique Godinho; Mancini, Marcelo; Demattê, José Alexandre Melo; Guilherme, Luiz Roberto Guimarães; Curi, Nilton

doi:10.1016/j.geoderma.2020.114553

Cited by 45 publications

(18 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on the results of this study, for the SOC, total K, total P, and available K, the models using the combined data from both sensors did not achieve obviously better results than those achieved from a single sensor. This conclusion was consistent with the conclusions of some previous studies (Benedet et al, 2020;Ng et al, 2019). Specifically, the SOC can be accurately estimated by combining the VisNIR and PXRF data using the model fusion (R 2 = .78, RPIQ = 2.06, and RPD = 2.12) and model averaging methods (R 2 = .78, RPIQ = 2.07, and RPD = 2.12), while it can also be estimated by the VisNIR data alone with almost equal accuracy (R 2 = .77, RPIQ = 2.05, and RPD = 2.10).…”

Section: Comparison Of the Separate And Combined Use Of The Pxrf And Visnir Datasupporting

confidence: 94%

“…In recent decades, proximal soil sensing techniques, especially visible and near‐infrared (VisNIR) spectroscopy and portable X‐ray fluorescence (PXRF) spectrometry, have become attractive alternatives for the estimation of soil properties due to their rapid data acquisition, high efficiency, satisfactory accuracy, and low cost (Andrade, Faria, et al., 2020; Benedet et al., 2020; Nawar et al., 2019; O'Rourke, Minasny, et al., 2016; Wang et al., 2015; Weindorf et al., 2014; Zhang & Hartemink, 2019). Visible and near‐infrared spectroscopy estimates soil properties mainly by directly establishing calibration models between soil reflectance and measured soil properties (Adeline et al., 2017; Conforti et al., 2018; Summers et al., 2011), whereas PXRF usually predicts soil properties from the correlation between the soil properties and element concentrations derived from PXRF spectrometry (Andrade, Silva, et al., 2020; Mancini et al., 2019; Nawar et al., 2019).…”

Section: Introductionmentioning

confidence: 99%

“…Satisfactory results have been reported for both groups of studies. The methods of combining the two sensors consist of three main types: (a) concatenation (Benedet et al., 2020; Wan et al., 2020; Wang et al., 2013, 2015; Weindorf et al., 2016; Xu et al., 2019; Zhang & Hartemink, 2019) (in certain models such as convolutional neural network [CNN], the two types of data are used as inputs with two channels; Ng et al., 2019), (b) model fusion (Chakraborty et al., 2015; Weindorf et al., 2016), and (c) model averaging (O'Rourke, Minasny, et al., 2016; O'Rourke, Stockmann, et al., 2016; Xu et al., 2019). All three combination methods have been reported to be effective with excellent prediction results.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Prediction of multiple soil fertility parameters using VisNIR spectroscopy and PXRF spectrometry

Liu

Wang

Xiao

et al. 2021

Soil Science Soc of Amer J

View full text Add to dashboard Cite

Visible and near-infrared (VisNIR) spectroscopy and portable X-ray fluorescence (PXRF) spectrometry are successful in estimating soil attributes. However, it is still necessary to investigate their performance in estimating soil fertility properties. In this study, for the soil fertility properties of soil organic C (SOC), available N, total P, available P, total K, and available K, the performance of using VisNIR and PXRF individually and combined, three data fusion methods (concatenation, model fusion and model averaging), and two modeling methods was evaluated. The results showed that combining VisNIR and PXRF did not substantially improve the prediction accuracy of SOC, total P, total K and available K. Using VisNIR alone could predict them well, achieving the best prediction in total K (R 2 = .90, ratio of performance to interquartile distance [RPIQ] = 5.08, and residual prediction deviation [RPD] = 3.13), followed by SOC (R 2 = .77, RPIQ = 2.05, and RPD = 2.10), total P (R 2 = .63, RPIQ = 2.35, and RPD = 1.64), and available K (R 2 = .53, RPIQ = 2.04, and RPD = 1.41). Both available N and available P were poorly estimated by all the methods proposed in this study. Although PXRF alone could predict total P, total K, and available K with acceptable accuracy, it did not perform well on SOC (R 2 < .32, RPIQ < 1.16, and RPD < 1.19). Hence, for the precision application of organic and NPK fertilizers, VisNIR is supposed to be the optimal way to rapidly obtain soil information.

show abstract

Section: Comparison Of the Separate And Combined Use Of The Pxrf And Visnir Datasupporting

confidence: 94%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Prediction of multiple soil fertility parameters using VisNIR spectroscopy and PXRF spectrometry

Liu

Wang

Xiao

et al. 2021

Soil Science Soc of Amer J

View full text Add to dashboard Cite

show abstract

“…Recent studies have evaluated the combined use of different PSS techniques and data fusion approaches for soil characterization [ 39 , 40 , 41 , 42 ]. Some studies have demonstrated that merging datasets of both XRF and vis-NIR spectroscopies can improve the quality of predictive models for soil attributes, such as total carbon (TC) and total nitrogen (TN) [ 43 ], pH, CEC, and textural attributes [ 44 ], and extractable nutrients (ex-K and ex-Ca) [ 26 ]. In addition, a recent patent of a portable apparatus that allows for the characterization of soil attributes based on a combined use of XRF and vis-NIR sensors was published [ 45 ].…”

Section: Introductionmentioning

confidence: 99%

“…A third approach is to combine spectral data (denoted here as spectra fusion (SF)) in one matrix, which is subjected to linear or non-linear analysis. Furthermore, the majority of papers reporting the fusion of vis-NIR and XRF data focused on the prediction of one or limited number of soil attributes, e.g., soil textural attributes [ 44 ], TN and TC [ 43 ], textural attributes, pH [ 53 ], CEC [ 54 ], and chromium [ 48 ]. Although O’Rourke et al [ 26 ] combined the vis-NIR and XRF data for the analysis of a wide range of soil attributes, they have explored the averaging data fusion methods only.…”

Section: Introductionmentioning

confidence: 99%

Combined Use of Vis-NIR and XRF Sensors for Tropical Soil Fertility Analysis: Assessing Different Data Fusion Approaches

Tavares

Molin

Javadi

et al. 2020

Sensors

View full text Add to dashboard Cite

Visible and near infrared (vis-NIR) diffuse reflectance and X-ray fluorescence (XRF) sensors are promising proximal soil sensing (PSS) tools for predicting soil key fertility attributes. This work aimed at assessing the performance of the individual and combined use of vis-NIR and XRF sensors to predict clay, organic matter (OM), cation exchange capacity (CEC), pH, base saturation (V), and extractable (ex-) nutrients (ex-P, ex-K, ex-Ca, and ex-Mg) in Brazilian tropical soils. Individual models using the data of each sensor alone were calibrated using multiple linear regressions (MLR) for the XRF data, and partial least squares (PLS) regressions for the vis-NIR data. Six data fusion approaches were evaluated and compared against individual models using relative improvement (RI). The data fusion approaches included (i) two spectra fusion approaches, which simply combined the data of both sensors in a merged dataset, followed by support vector machine (SF-SVM) and PLS (SF-PLS) regression analysis; (ii) two model averaging approaches using the Granger and Ramanathan (GR) method; and (iii) two data fusion methods based on least squares (LS) modeling. For the GR and LS approaches, two different combinations of inputs were used for MLR. The GR2 and LS2 used the prediction of individual sensors, whereas the GR3 and LS3 used the individual sensors prediction plus the SF-PLS prediction. The individual vis-NIR models showed the best results for clay and OM prediction (RPD ≥ 2.61), while the individual XRF models exhibited the best predictive models for CEC, V, ex-K, ex-Ca, and ex-Mg (RPD ≥ 2.57). For eight out of nine soil attributes studied (clay, CEC, pH, V, ex-P, ex-K, ex-Ca, and ex-Mg), the combined use of vis-NIR and XRF sensors using at least one of the six data fusion approaches improved the accuracy of the predictions (with RI ranging from 1 to 21%). In general, the LS3 model averaging approach stood out as the data fusion method with the greatest number of attributes with positive RI (six attributes; namely, clay, CEC, pH, ex-P, ex-K, and ex-Mg). Meanwhile, no single approach was capable of exploiting the synergism between sensors for all attributes of interest, suggesting that the selection of the best data fusion approach should be attribute-specific. The results presented in this work evidenced the complementarity of XRF and vis-NIR sensors to predict fertility attributes in tropical soils, and encourage further research to find a generalized method of data fusion of both sensors data.

show abstract

Comparing Sand Content Measurements by the Pipette, Hydrometer, and Vis-NIR and MIR Spectroscopy

Stevenson,

Hartemink,

Zhang

2023

Progress in Soil Science

View full text Add to dashboard Cite

Soil texture prediction using portable X-ray fluorescence spectrometry and visible near-infrared diffuse reflectance spectroscopy

Cited by 45 publications

References 39 publications

Prediction of multiple soil fertility parameters using VisNIR spectroscopy and PXRF spectrometry

Prediction of multiple soil fertility parameters using VisNIR spectroscopy and PXRF spectrometry

Combined Use of Vis-NIR and XRF Sensors for Tropical Soil Fertility Analysis: Assessing Different Data Fusion Approaches

Comparing Sand Content Measurements by the Pipette, Hydrometer, and Vis-NIR and MIR Spectroscopy

Contact Info

Product

Resources

About