Is the estimation of soil organic carbon using the colour space model, based on visible spectroscopy range, a reliable approach?

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00027073%3A_____%2F24%3AN0000089" target="_blank" >RIV/00027073:_____/24:N0000089 - isvavai.cz</a>

Výsledek na webu

<a href="https://bsssjournals.onlinelibrary.wiley.com/doi/epdf/10.1111/sum.13147" target="_blank" >https://bsssjournals.onlinelibrary.wiley.com/doi/epdf/10.1111/sum.13147</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1111/sum.13147" target="_blank" >10.1111/sum.13147</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Is the estimation of soil organic carbon using the colour space model, based on visible spectroscopy range, a reliable approach?

Popis výsledku v původním jazyce

Traditionally, soil colour attributes have been determined using the Munsell Colour Chart (MCC). However, the lack of standardization with this method has made it more difficult to assess soil properties, particularly soil organic carbon (SOC). In contrast, reflectance spectroscopy (RS) across the visible (Vis, 400-800 nm), near-infrared (NIR, 800-2500 nm) and Vis-NIR (350-2500 nm) spectral regions has been recognized as a more reliable approach for predicting SOC. As a result, soil scientists have increasingly adopted RS to obtain soil colour parameters, addressing the limitations of the MCC. However, because RS techniques for soil colour analysis is typically limited to the VIS range, key information from the NIR and Vis-NIR regions are often neglected or eliminated. This study examined the effectiveness of the VIS-based colour approach in estimating SOC compared with spectroscopy in the VIS, NIR and Vis-NIR ranges. Fifteen soil colour parameters were derived from the VIS spectrum, and 12 colour indices were calculated from these parameters. Three multivariate models such as random forest (RF), Cubist and support vector machine regression (SVMR) were used for prediction, along with various preprocessing algorithms to remove artefacts. The results indicated that, compared with VIS spectroscopy (R2 = .54) and the VIS-based colour method (R2 = .45), the pre-processed Vis-NIR data produced the most accurate results (R2 = .72). This suggests that the VIS range alone lacks adequate information, likely affecting the accuracy of the VIS-based colour dataset, as it is derived solely from this region. Although the introduction of colour indices slightly improved the VIS-based colour approach (R2 = .47), the results were still less accurate than those obtained using both the Vis-NIR and NIR spectroscopy ranges or even the VIS range alone (R2 = .54). The findings of this study highlight the need for caution when using VIS-based colour methods for SOC estimation, as high SOC levels information is not necessarily restricted to the VIS region.

Název v anglickém jazyce

Is the estimation of soil organic carbon using the colour space model, based on visible spectroscopy range, a reliable approach?

Popis výsledku anglicky

Traditionally, soil colour attributes have been determined using the Munsell Colour Chart (MCC). However, the lack of standardization with this method has made it more difficult to assess soil properties, particularly soil organic carbon (SOC). In contrast, reflectance spectroscopy (RS) across the visible (Vis, 400-800 nm), near-infrared (NIR, 800-2500 nm) and Vis-NIR (350-2500 nm) spectral regions has been recognized as a more reliable approach for predicting SOC. As a result, soil scientists have increasingly adopted RS to obtain soil colour parameters, addressing the limitations of the MCC. However, because RS techniques for soil colour analysis is typically limited to the VIS range, key information from the NIR and Vis-NIR regions are often neglected or eliminated. This study examined the effectiveness of the VIS-based colour approach in estimating SOC compared with spectroscopy in the VIS, NIR and Vis-NIR ranges. Fifteen soil colour parameters were derived from the VIS spectrum, and 12 colour indices were calculated from these parameters. Three multivariate models such as random forest (RF), Cubist and support vector machine regression (SVMR) were used for prediction, along with various preprocessing algorithms to remove artefacts. The results indicated that, compared with VIS spectroscopy (R2 = .54) and the VIS-based colour method (R2 = .45), the pre-processed Vis-NIR data produced the most accurate results (R2 = .72). This suggests that the VIS range alone lacks adequate information, likely affecting the accuracy of the VIS-based colour dataset, as it is derived solely from this region. Although the introduction of colour indices slightly improved the VIS-based colour approach (R2 = .47), the results were still less accurate than those obtained using both the Vis-NIR and NIR spectroscopy ranges or even the VIS range alone (R2 = .54). The findings of this study highlight the need for caution when using VIS-based colour methods for SOC estimation, as high SOC levels information is not necessarily restricted to the VIS region.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

40104 - Soil science

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

SOIL USE AND MANAGEMENT

ISSN

0266-0032

e-ISSN

1475-2743

Svazek periodika

40

Číslo periodika v rámci svazku

4

Stát vydavatele periodika

GB - Spojené království Velké Británie a Severního Irska

Počet stran výsledku

16

Strana od-do

e13147

Kód UT WoS článku

001370298300001

EID výsledku v databázi Scopus

—