Near‐Infrared Reflectance Spectroscopy Prediction of Soil Properties: Effects of Sample Cups and Preparation

Nduwamungu, Cargèle; Ziadi, Noura; Tremblay, Gaëtan F.; Parent, Léon‐Étienne

doi:10.2136/sssaj2008.0213

Cited by 61 publications

(58 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This agrees with the findings of Uldelhoven et al (2003) and Nduwamungu et al (2009b) who reported that the prediction of soil properties without a theoretical basis for NIRS calibration (i.e., secondary properties; P, K, Cu, Zn, and Mn) was generally less reliable compared with the prediction of primary properties. The difference is probably related to the correlation of the reference methods to near-infrared reflectance spectra.…”

Section: Assessment Of the Performance Of Nirs Predictionssupporting

confidence: 92%

Opportunities for, and limitations of, near infrared reflectance spectroscopy applications in soil analysis: A review

Nduwamungu¹,

Ziadi²,

Parent³

et al. 2009

Can. J. Soil. Sci.

View full text Add to dashboard Cite

limitations of, near infrared reflectance spectroscopy applications in soil analysis: A review. Can. J. Soil Sci. 89: 531Á541. Near infrared reflectance spectroscopy (NIRS) is a cost-and time-effective and environmentally friendly technique that could be an alternative to conventional soil analysis methods. In this review, we focussed on factors that hamper the potential application of NIRS in soil analysis. The reported studies differed in many aspects, including sample preparation, reference methods, spectrum acquisition and pre-treatments, and regression methods. The most significant opportunities provided by NIRS in soil analysis include its potential use in situ, the determination of various biological, chemical, and physical properties using a single spectrum per sample, and an estimated reduction of analytical cost of at least 50%. Contradictory results among studies on NIRS utilisation in soil analysis are partly related to variations in sample preparation and reference methods. The following calibration statistics appear to be most appropriate for comparing NIRS performance across soil attributes: (i) coefficient of determination (r 2 ), (ii) ratio of performance deviation (RPD), (iii) coefficient of regression (b), and (iv) ratio of the standard error of prediction (SEP) to the standard error of the reference method (SER), i.e., the ratio of standard errors (RSE). Further investigations on issues such as (i) RSE guidelines, (ii) correlation between NIRS spectrophotometers, (iii) correlation of different reference methods for a given attribute to soil spectra, (iv) identification of key factors affecting the accuracy of NIRS predictions, and (v) efficient use of spectral libraries are required to enhance the acceptability of NIRS as a soil analysis technique and to make it more user-friendly. Standardized guidelines are proposed for the assessment of the accuracy of NIRS predictions of soil attributes.

show abstract

Section: Assessment Of the Performance Of Nirs Predictionssupporting

confidence: 92%

Opportunities for, and limitations of, near infrared reflectance spectroscopy applications in soil analysis: A review

Nduwamungu¹,

Ziadi²,

Parent³

et al. 2009

Can. J. Soil. Sci.

View full text Add to dashboard Cite

show abstract

“…Previous studies have suggested that models with R 2 > 0.8 provide acceptable or high accuracy levels for the prediction of soil properties whereas models with R 2 < 0.8 but > 0.6 provide only medium level predictions (e.g., Chang et al, 2001;Malley et al, 2004;Nduwamungu et al, 2009). In the present study, we adopted this classification into different R 2 classes and advise that the evaluation of the usefulness of a predictive model should be determined by the amount of uncertainty reduction it provides for a given purpose.…”

Section: Chemometric Analysesmentioning

confidence: 99%

Mid-Infrared and Total X-Ray Fluorescence Spectroscopy Complementarity for Assessment of Soil Properties

Towett

Shepherd

Sila

et al. 2015

Soil Science Society of America Journal

View full text Add to dashboard Cite

Diffuse reflectance Fourier transformed mid-infrared (DRIFT-MIR) spectroscopy can predict many soil properties but extractable nutrients are often predicted poorly. This paper tested the combined DRIFT-MIR and total X-ray fluorescence (TXRF) spectroscopy analysis for prediction of soil properties related to soil fertility. A total of 700 soil samples from 44 stratified randomly located 100-km 2 sentinel sites distributed across Sub-Saharan Africa (SSA) were analyzed for physicochemical composition using conventional reference methods, and compared with MIR and TXRF spectra using Random Forests (RF) regression algorithm and an internal out-of-bag (OOB) validation. Mid-infrared spectra resulted in good prediction models (R 2 > 0.80) for organic C and total N, Mehlich-3 Ca and Al, and pH. Moderately predicted (R 2 > 0.60) were extractable Mg, P sorption index, sand, silt, and clay. Models were less satisfactory (R 2 < 0.60) for Mehlich-3 extractable K, Mn, Fe, Cu, B, Zn, P, S, and Na, exchangeable acidity and electrical conductivity (Ecd). Inclusion of total element concentration data from TXRF analysis in the MIR RF models significantly reduced root mean square error of prediction by 70% for Ecd, 66% for Mehlich-3 Na, 61% for Mehlich-3 S, and 50% for Mehlich-3 B due to detection by TXRF of some saline soils that were not well predicted by MIR. Overall, both methods predicted soil properties that relate to nutrient buffering capacity, including some exchangeable bases, pH, P sorption capacity, clay content, and organic matter content, and fingerprint basic soil mineralogy. Thus, further research should test whether MIR and TXRF fingerprinting could better predict soil nutrient supply capacity, as determined by crop nutrient uptake potential, than conventional soil P, K and micronutrient tests.

show abstract

“…Among the factors that could result in obtaining different results between studies the origin of the samples (i.e., different geographical and agro-ecological regions), preparation of the samples (i.e., particle size) and calibration procedures are mentioned (Nduwamungu et al, 2009a;2009b). Volkan-Bilgili et al (2010) reported that, the best prediction models were obtained for organic matter (R 2 = 0.80), total carbonates (R NCal, número de muestras de calibración; F= finos (<0.5 mm); G= gruesos (<2 mm); min, mínimo; max, máximo; DS= desviación estándar; CV= coeficiente de variación [(DS/media) x 100].…”

Section: Resultsmentioning

confidence: 99%

“…Entre los factores que pudieran resultar en la obtención de diferentes resultados entre los estudios se mencionan la procedencia de las muestras (i.e., diferentes regiones geográficas y agroecológicas), la preparación de las muestras (i.e., tamaño de partícula) y los procedimientos de calibración (Nduwamungu et al, 2009a;2009b). Volkan-Bilgili et al (2010) reportaron que los mejores modelos de predicción que obtuvieron fueron para materia orgánica (R (2006) presentan una excelente revisión de literatura la cual muestra el potencial del análisis cuantitativo por infrarrojo.…”

Section: Resultsunclassified

Predicción de propiedades del suelo de importancia agronómica por espectroscopia de reflectancia de infrarrojo cercano

Corral

Wong

Villalobos

et al. 2017

Remexca

View full text Add to dashboard Cite

ResumenEl creciente interés por obtener información sobre las propiedades del suelo de forma rápida, precisa y a bajo costo ha resultado en un aumento en el uso de la espectroscopia de infrarrojo cercano (NIRS, por sus siglas en inglés) para esta aplicación. El principal objetivo de este estudio fue evaluar la capacidad del NIRS para la predicción de 12 propiedades de suelo de importancia agronómica en el norte de México. Las muestras fueron recolectadas en parcelas agrícolas de dos regiones denominadas como sitio 1, Baja California (BC, n = 128) y sitio 2, Chihuahua (n = 143). Los espectros de infrarrojo fueron obtenidos a partir de dos tamaños de partículas de suelo: gruesos (< 2 mm) y finos (< 0.5 mm), creando dos librerías. Se obtuvo una predicción acertada (RPD >1.4) para pH, CE, CEC, K, contenido de arena, arcilla y limo en el gr upo de muestras de la validación cruzada para al menos un sitio. Los resultados fueron consistentes con los obtenidos en el grupo de validación independiente, excepto para pH y limo los cuales tuvieron un RPD < 1.4. Las propiedades sin capacidad de predicción por NIRS fueron N-NO 3 , N-NH 4 , P, carbonatos totales (CO 3 ) y materia orgánica. Una ventaja importante del NIRS para análisis de suelos es que a partir de un espectro pueden determinarse muchas propiedades. Una de las limitaciones de este estudio fue la cantidad de muestras empleadas para la validación independiente, la cual consistió de 10% del total de muestras. AbstractThe growing interest in information about soil traits quickly, accurately and cheaply has resulted in an increase in the use of near-infrared spectroscopy (NIRS) for this application. The main objective of this study was to evaluate the ability of NIRS to predict soil traits on 12 agronomically important in northern Mexico. Samples were collected from agricultural plots of two regions designated as site 1, Baja California (BC, n= 128) and site 2, Chihuahua (n= 143). Infrared spectra were obtained from two sizes of soil particles: thick (<2 mm) and thin (<0.5 mm), creating two libraries. An accurate prediction (RPD> 1.4) for pH, EC, CEC, K content of sand, clay and silt in the group of cross-validation samples for at least one site was obtained. The results were consistent with those obtained in the group of independent validation, except for pH and silt which had a RPD <1.4. The traits without predictability by NIRS were N-NO 3 , N-NH 4 , P, total carbonates (CO 3 ) and organic matter. An important advantage of NIRS for soil analysis is that from a spectrum can be determined many traits. One limitation of this study was the number of samples used for independent validation, which consisted of 10% of total samples.

show abstract

Near‐Infrared Reflectance Spectroscopy Prediction of Soil Properties: Effects of Sample Cups and Preparation

Cited by 61 publications

References 37 publications

Opportunities for, and limitations of, near infrared reflectance spectroscopy applications in soil analysis: A review

Opportunities for, and limitations of, near infrared reflectance spectroscopy applications in soil analysis: A review

Mid-Infrared and Total X-Ray Fluorescence Spectroscopy Complementarity for Assessment of Soil Properties

Predicción de propiedades del suelo de importancia agronómica por espectroscopia de reflectancia de infrarrojo cercano

Contact Info

Product

Resources

About