Is the estimation of soil organic carbon using the colour space model, based on visible spectroscopy range, a reliable approach? DOI
James Kobina Mensah Biney, Jakub Houška, Nasem Badreldin

et al.

Soil Use and Management, Journal Year: 2024, Volume and Issue: 40(4)

Published: Oct. 1, 2024

Abstract Traditionally, soil colour attributes have been determined using the Munsell Colour Chart (MCC). However, lack of standardization with this method has made it more difficult to assess properties, particularly organic carbon (SOC). In contrast, reflectance spectroscopy (RS) across visible (Vis, 400–800 nm), near‐infrared (NIR, 800–2500 nm) and Vis–NIR (350–2500 spectral regions recognized as a reliable approach for predicting SOC. As result, scientists increasingly adopted RS obtain parameters, addressing limitations MCC. because techniques analysis is typically limited VIS range, key information from NIR are often neglected or eliminated. This study examined effectiveness VIS‐based in estimating SOC compared VIS, ranges. Fifteen parameters were derived spectrum, 12 indices calculated these parameters. Three multivariate models such random forest (RF), Cubist support vector machine regression (SVMR) used prediction, along various preprocessing algorithms remove artefacts. The results indicated that, ( R 2 = .54) .45), pre‐processed data produced most accurate .72). suggests that range alone lacks adequate information, likely affecting accuracy dataset, solely region. Although introduction slightly improved .47), still less than those obtained both ranges even .54). findings highlight need caution when methods estimation, high levels not necessarily restricted

Language: Английский

A Novel Model for Soil Organic Matter and Total Nitrogen Detection Based on Visible/Shortwave Near-Infrared Spectroscopy DOI Creative Commons
Jiangtao Qi, Peng Cheng,

Junbo Zhou

et al.

Land, Journal Year: 2025, Volume and Issue: 14(2), P. 329 - 329

Published: Feb. 6, 2025

Soil organic matter (SOM) and total nitrogen (TN) are critical indicators for assessing soil fertility. Although laboratory chemical analysis methods can accurately measure their contents, these techniques time-consuming labor-intensive. Spectral technology, characterized by its high sensitivity convenience, has been increasingly integrated with machine learning algorithms nutrient monitoring. However, the process of spectral data remains complex requires further optimization simplicity efficiency to improve prediction accuracy. This study proposes a novel model enhance accuracy SOM TN predictions in northeast China’s black soil. Visible/Shortwave Near-Infrared Spectroscopy (Vis/SW-NIRS) within 350–1070 nm range were collected, preprocessed, dimensionality-reduced. The scores first nine principal components after partial least squares (PLS) dimensionality reduction selected as inputs, measured contents used outputs build back-propagation neural network (BPNN) model. results show that processed combination standard normal variate (SNV) multiple scattering correction (MSC) have best modeling performance. To stability this model, three named random search (RS), grid (GS), Bayesian (BO) introduced. demonstrate Vis/SW-NIRS provides reliable PLS-RS-BPNN achieving performance (R2 = 0.980 0.972, RMSE 1.004 0.006 TN, respectively). Compared traditional models such forests (RF), one-dimensional convolutional networks (1D-CNNs), extreme gradient boosting (XGBoost), proposed improves R2 0.164–0.344 predicting 0.257–0.314 respectively. These findings confirm potential technology effective tools prediction, offering valuable insights application sensing information.

Language: Английский

Citations

1
Research on Soil Erosion Based on Remote Sensing Technology: A Review DOI Creative Commons
Jiaqi Wang, Jiuchun Yang, Zhi Li

et al.

Agriculture, Journal Year: 2024, Volume and Issue: 15(1), P. 18 - 18

Published: Dec. 25, 2024

Monitoring and assessing soil erosion is essential for reducing land degradation ensuring food security. It provides critical scientific insights developing effective policies implementing targeted preventive measures. The emergence of remote sensing technology has significantly bolstered research, empowering researchers to comprehensively accurately understand address erosion-related challenges. Consequently, become pivotal in research methodologies. In recent years, significant progress been made on erosion. This study aims encapsulate the current status advancements applications research. catalogs commonly used data sources introduces innovative methodologies detecting soil-erosion-related information utilizing technology. Furthermore, it delves into analysis acquisition methods factors influencing examines crucial role prevalent simulation prediction models. Additionally, this identifies existing challenges outlines prospects developmental directions emphasizing its potential contribute sustainable management practices environmental conservation efforts.

Language: Английский

Citations

4
Multi-spectral evaluation of total nitrogen, phosphorus and potassium content in soil using Vis-NIR spectroscopy based on a modified support vector machine with whale optimization algorithm DOI

Mochen Liu,

Yang Kuankuan,

Yinfa Yan

et al.

Soil and Tillage Research, Journal Year: 2025, Volume and Issue: 252, P. 106567 - 106567

Published: April 19, 2025

Language: Английский

Citations

0
Potential of EnMAP Hyperspectral Imagery for Regional-Scale Soil Organic Matter Mapping DOI Creative Commons
Yassine Bouslıhım, Abdelkrim Bouasria

Remote Sensing, Journal Year: 2025, Volume and Issue: 17(9), P. 1600 - 1600

Published: April 30, 2025

The emergence of new-generation hyperspectral satellites offers more potential for mapping soil properties. This study presents the first assessment EnMAP (Environmental Mapping and Analysis Program) imagery organic matter (SOM) prediction using actual spectral data from 282 samples. Different preprocessing techniques, including Savitzky–Golay (SG) smoothing, second derivative SG, Standard Normal Variate (SNV) transformation, were evaluated in combination with embedded feature selection to identify most relevant wavelengths SOM prediction. Partial Least Squares Regression (PLSR) models developed under different pre-treatment scenarios. best performance was obtained SNV top 30 bands (wavelengths) selected, giving R2 = 0.68, RMSE 0.34%, RPIQ 1.75. successfully identified significant prediction, particularly around 550 nm Vis–NIR region, 1570–1630 nm, 1600 2200 SWIR region. resulting predictions exhibited spatially consistent patterns that corresponded known soil–landscape relationships, highlighting properties despite its limited geographical availability. While these results are promising, this limitations ability PLSR extrapolate beyond sampled areas, suggesting need explore non-linear modeling approaches. Future research should focus on evaluating EnMAP’s advanced machine learning techniques comparing it other available products establish robust protocols satellite-based monitoring.

Language: Английский

Citations

0
Is the estimation of soil organic carbon using the colour space model, based on visible spectroscopy range, a reliable approach? DOI
James Kobina Mensah Biney, Jakub Houška, Nasem Badreldin

et al.

Soil Use and Management, Journal Year: 2024, Volume and Issue: 40(4)

Published: Oct. 1, 2024

Abstract Traditionally, soil colour attributes have been determined using the Munsell Colour Chart (MCC). However, lack of standardization with this method has made it more difficult to assess properties, particularly organic carbon (SOC). In contrast, reflectance spectroscopy (RS) across visible (Vis, 400–800 nm), near‐infrared (NIR, 800–2500 nm) and Vis–NIR (350–2500 spectral regions recognized as a reliable approach for predicting SOC. As result, scientists increasingly adopted RS obtain parameters, addressing limitations MCC. because techniques analysis is typically limited VIS range, key information from NIR are often neglected or eliminated. This study examined effectiveness VIS‐based in estimating SOC compared VIS, ranges. Fifteen parameters were derived spectrum, 12 indices calculated these parameters. Three multivariate models such random forest (RF), Cubist support vector machine regression (SVMR) used prediction, along various preprocessing algorithms remove artefacts. The results indicated that, ( R 2 = .54) .45), pre‐processed data produced most accurate .72). suggests that range alone lacks adequate information, likely affecting accuracy dataset, solely region. Although introduction slightly improved .47), still less than those obtained both ranges even .54). findings highlight need caution when methods estimation, high levels not necessarily restricted

Language: Английский

Citations

1