Near Infrared Spectra and Soft Independent Modelling of Class Analogy for Discrimination of Chernozems, Luvisols and Vertisols (original) (raw)
Related papers
Spectral analysis as an extra method to soil type discrimination
Agricultural Science and Technology, 2020
The purpose of the study was to test near infrared soil spectra as an extra method for three soil types (Fluvisols, Vertisols and Solonchaks) discrimination from different regions of South Bulgaria. The diffuse reflectance spectra of 177 soil samples (from the 0-20cm layers): 50 samples of Fluvisols soil type, 78 samples of Vertisols soil type and 48 samples of Solonchaks soil type were obtained using a Spectrum NIRQuest (OceanOptics, Inc.) working within the range from 900 to 1700 nm. Soft independent modelling of class analogy (SIMCA) was performed to classify samples according to their taxonomic classes. The results obtained showed that the soil samples are separated accurately according to their soil type based on their spectral information. All this could be used in the future studies related to the application of the NIRS method as a qualitative or quantitative method for soil analysis and also for the purposes of precision farming.
2020
This study aimed to predict soil properties using visible–near infrared (VIS-NIR) spectroscopy combined with partial least square regression (PLSR) modeling. Special emphasis was given to evaluating effect of pre-processing methods on prediction accuracy and important wavelengths. A total of 114 samples were collected and involved in chemical and spectral analyzes. PLSR model of each soil property was calibrated for all pre-processing methods using all samples, and leave-one-out cross-validation was used to make comparisons between them. Then, PLSR model of each best pre-processing method was calibrated using a 75% of all samples and correspondingly validated with the remaining a 25%. Model accuracy was evaluated based on coefficient of determination (R2), root mean-squared errors (RMSE), and residual prediction deviations (RPD). The high correlation coefficients were found between the tested soil properties and reflectance spectra. The pre-processing methods considerably improved pr...
Investigations into Soil Composition and Texture Using Infrared Spectroscopy (2–14 μ m)
Applied and Environmental Soil Science, 2012
The ability of thermal and shortwave infrared spectroscopy to characterise composition and texture was evaluated using both particle size separated soil samples and natural soils. Particle size analysis and separation into clay, silt, and sand-sized soil fractions was undertaken to examine possible relationships between quartz and clay mineral spectral signatures and soil texture. Spectral indices, based on thermal infrared specular and volume scattering features, were found to discriminate clay mineral-rich soil from mostly coarser quartz-rich sandy soil and to a lesser extent from the silty quartz-rich soil. Further investigations were undertaken using spectra and information on 51 USDA and other soils within the ASTER spectral library to test the application of shortwave, mid-and thermal infrared spectral indices for the derivation of clay mineral, quartz, and organic carbon content. A nonlinear correlation between quartz content and a TIR spectral index based on the 8.62 μm was observed. Preliminary efforts at deriving a spectral index for the soil organic carbon content, based on 3.4-3.5 μm fundamental H-C stretching vibration bands, were also undertaken with limited results.
Korean Journal of Soil Science and Fertilizer, 2014
This study investigates the prediction of soil chemical properties (organic matter (OM), pH, Ca, Mg, K, Na, total acidity, cation exchange capacity (CEC)) on 688 Korean soil samples using the visible-near infrared reflectance (VIS-NIR) spectroscopy. Reflectance from the visible to near-infrared spectrum (350 to 2500 nm) was acquired using the ASD Field Spec Pro. A total of 688 soil samples from 168 soil profiles were collected from 2009 to 2011. The spectra were resampled to 10 nm spacing and converted to the 1st derivative of absorbance (log (1/R)), which was used for predicting soil chemical properties. Principal components analysis (PCA), partial least squares regression (PLSR) and regression rules model (Cubist) were applied to predict soil chemical properties. The regression rules model (Cubist) showed the best results among these, with lower error on the calibration data. For quantitatively determining OM, total acidity, CEC, a VIS-NIR spectroscopy could be used as a routine method if the estimation quality is more improved.
Development of near infrared spectral library of Danish soils
This paper describes the methodology for the establishment of a Danish soil Near InfraRed Spectroscopy (NIRS) library. In order to make an efficient application of NIRS in the field for soil property mapping, it is necessary to establish a NIRS library for global calibration. Representative 3,534 samples from a 7 km grid sampling were chosen to cover variability in the geographical area of Denmark. Partial least square regression was used to build a regression model between SOC and the spectra. The data set was divided into three subsets: calibration, validation and prediction. Outliers were removed from spectral and reference data. The calibration result for non-organic soils was R 2 =0.81, RMSE=0.22, while validation was R 2 =0.81, RMSE=0.22 and RPD=2.3. This indicates good prediction abilities. The results were also tested on independent datasets, with the following results: R 2 =0.82, RMSEP=0.21 and RPD 2.4. Likewise, the calibration result for organic soils was R 2 =0.82, RMSE=...
Soil characterization by near-infrared spectroscopy and principal component analysis
REVISTA CIÊNCIA AGRONÔMICA, 2021
This research aimed to use principal component analysis (PCA) as an exploratory method for spectral data of soil absorbance from the Baturité Massif and Central Hinterland (Ceará State, Brazil) to verify the potential of the technique in soil characterization. We analyzed 46 soil samples from different areas (native and cultivated). Each sample was analyzed in two particle sizes: 2 and 0.2 mm. We obtained spectral data by near-infrared spectroscopy (NIR), selecting the 1,360-2,260 nm range (2,376 variables). We evaluated three data pretreatment methods: multiplicative scatter correction (MSC), first derivative, and second derivative of the Savitzky-Golay filter. The absorption bands observed were: 1,414 nm (C-H stretching and deformation combination), 1,450 nm (O-H associated with the carbon chain), 1,780 nm (second overtone of C-H), 1,928 nm (O-H associated with molecular water), and 2,208 nm (C-H stretch and C=O combination). The best pretreatment was verified using only the multiplicative scatter correction (MSC). Two principal components explained 98% of the data variability, being the first principal component (PC1) related to the characteristic band of moisture, with negative values in the 1,928 nm region, while the second principal component (PC2) was related to the total organic matter (OM) originating from the C-H, C=O, and N-H bonds, wavelength region 1,414 nm. The PCA allowed characterizing the samples in terms of moisture and OM contents, with emphasis on soils under irrigated agroforestry system with higher values of moisture and OM, while the soil in degradation process presented lower values for these attributes. The NIR spectroscopy, associated with data processing methods (PCA and MSC), allows identifying changes in soil attributes, such as moisture and OM.
Journal of Arid Environments, 2010
Reflectance spectroscopy can be used to nondestructively characterize materials for a wide range of applications. In this study, visible-near infrared reflectance spectroscopy (VNIR) was evaluated for prediction of diverse soil properties related to four different soil series of the Entisol soil group within a single field in northern Turkey. Soil samples were collected from 512 locations in a 25 Â 25 m sampling grid over a 32 ha (800 Â 400 m) area. Air-dried soil samples were scanned at 1 nm resolution from 350 to 2500 nm, and calibrations between soil physical and chemical properties and reflectance spectra were developed using cross-validation under partial least squares regression (PLSR) and multivariate adaptive regression splines (MARS). Raw reflectance and first derivative reflectance data were used separately and combined for all samples in the data set. Data were additionally divided into two random subsets of 70 and 30% of the full data, which were each used for calibration and validation. Overall, MARS provided better predictions when under cross-validation. However, PLSR and MARS results were comparable in terms of prediction accuracy when using separate data sets for calibration and validation. No improvement was obtained by combining first derivative and raw data. Strongest correlations were obtained with exchangeable Ca and Mg, cation exchange capacity, and organic matter, clay, sand, and CaCO 3 contents. When soil data were classified into groups, VNIR spectroscopy estimated class memberships well, especially for soil texture. In conclusion, VNIR spectroscopy was variably successful in estimating soil properties at the field scale, and showed potential for substituting laboratory analyses or providing inexpensive co-variable data.
Quantifying Soil Chemical Properties Using Near Infrared Reflectance Spectroscopy
Journal of the Arkansas Academy of Science, 2007
Methodologies for determining soil chemical properties have evolved dramatically during the past century. Early geochemical analyses were conducted exclusively through the use of wet chemistry techniques that were relatively reliable but painstaking and subject to errors at various stages of analysis. Near infrared reflectance spectroscopy (NIRS) has emerged as a new approach for rapidly analyzing a variety ofmaterials including soils. In this study soil samples were taken from eight study areas across the Ozark Highlands of Arkansas, and NIRS calibration models were developed to determine the accuracy of using NIRS to analyze soils compared with standard soil chemical analysis protocols. Multivariate regression models were highly effective for analyzing several important elements. C and N models explained 92% and 88% of their variation, respectively, and Ca, Mg, P, and Mn models explained 72-88% ofthe variability in these elements. Models for C:N and pH explained 82% and 86% oftheir variability, respectively. Models for micronutrients Cu and Zn did not fit as well with 22% and 40% of their variability explained, respectively. Our findings suggest that additional NIRS calibration and modeling is promising for rapidly analyzing the chemical composition of soils, and it is desirable to develop model libraries that are calibrated for the soils of a given region.
Visible near-infrared reflectance spectroscopy as a predictive indicator of soil properties
2011
Reflectance spectroscopy can be used to nondestructively characterize materials for a wide range of applications. In this study, visible-near infrared reflectance spectroscopy (VNIR) was evaluated for prediction of diverse soil properties related to four different soil series of the Entisol soil group within a single field in northern Turkey. Soil samples were collected from 512 locations in a 25 Â 25 m sampling grid over a 32 ha (800 Â 400 m) area. Air-dried soil samples were scanned at 1 nm resolution from 350 to 2500 nm, and calibrations between soil physical and chemical properties and reflectance spectra were developed using cross-validation under partial least squares regression (PLSR) and multivariate adaptive regression splines (MARS). Raw reflectance and first derivative reflectance data were used separately and combined for all samples in the data set. Data were additionally divided into two random subsets of 70 and 30% of the full data, which were each used for calibration and validation. Overall, MARS provided better predictions when under cross-validation. However, PLSR and MARS results were comparable in terms of prediction accuracy when using separate data sets for calibration and validation. No improvement was obtained by combining first derivative and raw data. Strongest correlations were obtained with exchangeable Ca and Mg, cation exchange capacity, and organic matter, clay, sand, and CaCO 3 contents. When soil data were classified into groups, VNIR spectroscopy estimated class memberships well, especially for soil texture. In conclusion, VNIR spectroscopy was variably successful in estimating soil properties at the field scale, and showed potential for substituting laboratory analyses or providing inexpensive co-variable data.
Journal of Productivity and Development, 2018
Historically, our understanding of the soil and assessment of its quality and function has been gained through field survey and routine soil physicochemical laboratory analysis. Reflectance spectroscopy can be used to non-destructively characterize materials for a wide range of applications. Hyperspectral remote sensing data provide a rich source of information produced in the form of the spectrum which can be used to identify surface materials. In this study, Field Portable Hyperspectral Radiometer (FPHR) was evaluated in an attempt for prediction of diverse soil properties related to three different soil orders (Vertisols, Aridisols, and Entisols) across Lower Egypt. Eight pedons consisting of 34 samples were collected from different semi-arid areas. Soil horizonation and twelve soil attributes including clay, sand, silt, SOC, pH, EC, A.W, gypsum, CaCO 3 , Fe 2 O 3 , Al 2 O 3 , and SiO 2 were traditionally analyzed and then correlated with spectral reflectance of the spectrum range. Four bands (blue, green, red, and near-infrared) were calculated for prediction of these variables. The results showed that the variations in spectral reflectance for each horizon across the spectrum range (276-1093 nm) were matched well with those of morphologically described horizons in the field. Additionally, the correlation results of different soil variables were highly correlated with spectral reflectance at different band wavelengths. For example, clay content correlated negatively (r =-0.93) with reflectance at the green band while silt (r = 0.67 at the blue band) and sand (0.87 at the green band) correlated positively. Regression equations were fitted in graphs to attempt the quantification of the soil constituents from their reflectance values. The best predictive models were obtained for clay 0.79), CaCO 3 (R 2 = 0.79), gypsum (R 2 = 0.75), Fe 2 O 3 (R 2 = 0.71), sand (R 2 = 0.69), silt (R 2 = 0.54), and SOC (R 2 = 0.51) while the poor prediction was for EC and pH. The results concluded that the spectral reflectance of the spectrum had the potential to differentiate the soil horizonation and to predict the selected soil variable at different wavelength bands. Conclusively, FPHR was shown to be an effective tool for enhanced soil horizon differentiation and the acquisition of soil attributes information.