Visible-Near-Infrared Spectroscopy Prediction of Soil Characteristics as Affected by Soil-Water Content

2018 ◽  
Vol 82 (6) ◽  
pp. 1333-1346 ◽  
Author(s):  
Lashya P. Marakkala Manage ◽  
Mogens Humlekrog Greve ◽  
Maria Knadel ◽  
Per Moldrup ◽  
Lis W. de Jonge ◽  
...  
Keyword(s):  
Infrared Spectroscopy ◽  
Water Content ◽  
Soil Water ◽  
Soil Water Content ◽  
Near Infrared Spectroscopy ◽  
Near Infrared ◽  
Soil Characteristics

Near-infrared spectroscopy for soil water determination in small soil volumes

2004 ◽  
Vol 84 (3) ◽  
pp. 333-338 ◽  
Author(s):  
P. R. Bullock ◽  
X. Li ◽  
L. Leonardi
Keyword(s):  
Infrared Spectroscopy ◽  
Water Content ◽  
Soil Water ◽  
Soil Water Content ◽  
Near Infrared Spectroscopy ◽  
Near Infrared ◽  
Water Levels ◽  
Least Squares Regression ◽  
Volumetric Soil Water Content ◽  
Volumetric Soil Water

Critical soil water levels for soil microscale processes are difficult to determine because of variability in large soil volumes and lack of techniques for logging soil water contents in small soil volumes. This study tested nearinfrared (NIR) spectroscopy for soil water content determination. Five soil horizons with a range in soil texture, soil organic carbon, carbonates, pH and horizon depth, were tested at air-dry, field capacity and 0.1 MPa tension water content. Volumetric soil water content, determined using the standard method of oven-drying and soil bulk density, was compared to NIR absorbance in various combinations and wavelengths. The NIR spectra obtained with the probe in direct contact with the soil gave better results than when the probe was separated from the soil with a glass slide. The most reliable validation results were obtained using a multivariate partial least squares regression of the full spectrum with an r2 of 0.95 and RMSE of prediction of 6.4%. Smoothing and derivatives of the spectra did not improve the validation results. The relationships for absorbance at single wavelength segments, ratios, differences and area under the curve around the 1940 nm peak were good (r2 values near 0.85 ) but poorer than the results using the full spectra. The high correlation coefficients obtained with the wide variety of soils utilized in this study suggest that NIR absorbance is a practical method for determining volumetric soil water content for small soil volumes. Key words: Near-infrared spectroscopy, soil water, Near-infrared absorbance

The use of visible and near-infrared spectroscopy for the analysis of soil water repellency

2014 ◽  
Vol 65 (3) ◽  
pp. 360-368 ◽  
Author(s):  
I. Kim ◽  
R. R. Pullanagari ◽  
M. Deurer ◽  
R. Singh ◽  
K. Y. Huh ◽  
...  
Keyword(s):  
Infrared Spectroscopy ◽  
Soil Water ◽  
Near Infrared Spectroscopy ◽  
Near Infrared ◽  
Water Repellency ◽  
Soil Water Repellency

scholarly journals Separation of soil and canopy reflectance signatures of Mid German agricultural soils

2011 ◽  
Vol 51 (No, 7) ◽  
pp. 296-303 ◽  
Author(s):  
T. Behrens ◽  
K. Gregor ◽  
W. Diepenbrock
Keyword(s):  
Remote Sensing ◽  
Soil Properties ◽  
Water Content ◽  
Soil Water ◽  
Soil Water Content ◽  
Vegetation Index ◽  
Near Infrared ◽  
Field Experiments ◽  
Agricultural Soils ◽  
Soil Reflectance

Remote sensing can provide visual indications of crop growth during production season. In past, spectral optical estimations were well performed in the ability to be correlated with crop and soil properties but were not consistent within the whole production season. To better quantify vegetation properties gathered via remote sensing, models of soil reflectance under changing moisture conditions are needed. Signatures of reflected radiation were acquired for several Mid German agricultural soils in laboratory and field experiments. Results were evaluated at near-infrared spectral region at the wavelength of 850 nm. The selected soils represented different soil colors and brightness values reflecting a broad range of soil properties. At the wavelength of 850 nm soil reflectance ranged between 10% (black peat) and 74% (white quartz sand). The reflectance of topsoils varied from 21% to 32%. An interrelation was found between soil brightness rating values and spectral optical reflectance values in form of a linear regression. Increases of soil water content from 0% to 25% decreased signatures of soil reflectance at 850 nm of two different soil types about 40%. The interrelation of soil reflectance and soil moisture revealed a non-linear exponential function. Using knowledge of the individual signature of soil reflectance as well as the soil water content at the measurement, soil reflectance could be predicted. As a result, a clear separation is established between soil reflectance and reflectance of the vegetation cover if the vegetation index is known.

scholarly journals Classification of Granite Soils and Prediction of Soil Water Content Using Hyperspectral Visible and Near-Infrared Imaging

Sensors ◽  
2020 ◽  
Vol 20 (6) ◽  
pp. 1611 ◽  
Author(s):  
Hwan-Hui Lim ◽  
Enok Cheon ◽  
Deuk-Hwan Lee ◽  
Jun-Seo Jeon ◽  
Seung-Rae Lee
Keyword(s):  
Water Content ◽  
Soil Water ◽  
Soil Water Content ◽  
Near Infrared ◽  
Classification Model ◽  
Projection Algorithm ◽  
Support Vector ◽  
Ann Model ◽  
Landslide Hazards ◽  
Combining Data

Soil water content is one of the most important physical indicators of landslide hazards. Therefore, quickly and non-destructively classifying soils and determining or predicting water content are essential tasks for the detection of landslide hazards. We investigated hyperspectral information in the visible and near-infrared regions (400–1000 nm) of 162 granite soil samples collected from Seoul (Republic of Korea). First, effective wavelengths were extracted from pre-processed spectral data using the successive projection algorithm to develop a classification model. A gray-level co-occurrence matrix was employed to extract textural variables, and a support vector machine was used to establish calibration models and the prediction model. The results show that an optimal correct classification rate of 89.8% could be achieved by combining data sets of effective wavelengths and texture features for modeling. Using the developed classification model, an artificial neural network (ANN) model for the prediction of soil water content was constructed. The input parameter was composed of Munsell soil color, area of reflectance (near-infrared), and dry unit weight. The accuracy in water content prediction of the developed ANN model was verified by a coefficient of determination and mean absolute percentage error of 0.91 and 10.1%, respectively.

Towards a robust water content determination of freeze-dried samples by near-infrared spectroscopy

2010 ◽  
Vol 676 (1-2) ◽  
pp. 34-40 ◽  
Author(s):  
Holger Grohganz ◽  
Delphine Gildemyn ◽  
Erik Skibsted ◽  
James M. Flink ◽  
Jukka Rantanen
Keyword(s):  
Infrared Spectroscopy ◽  
Water Content ◽  
Near Infrared Spectroscopy ◽  
Near Infrared ◽  
Content Determination ◽  
Freeze Dried

scholarly journals Prediction of water content and soluble solids content of ‘manalagi’ apples using near infrared spectroscopy

2021 ◽  
Vol 922 (1) ◽  
pp. 012062
Author(s):  
K Kusumiyati ◽  
Y Hadiwijaya ◽  
D Suhandy ◽  
A A Munawar
Keyword(s):  
Infrared Spectroscopy ◽  
Water Content ◽  
Near Infrared Spectroscopy ◽  
Near Infrared ◽  
Quality Attributes ◽  
Soluble Solids ◽  
Coefficient Of Determination ◽  
Soluble Solids Content ◽  
Orthogonal Signal Correction ◽  
Solids Content

Abstract The purpose of the research was to predict quality attributes of ‘manalagi’ apples using near infrared spectroscopy (NIRS). The desired quality attributes were water content and soluble solids content. Spectra data collection was performed at wavelength of 702 to 1065 nm using a Nirvana AG410 spectrometer. The original spectra were enhanced using orthogonal signal correction (OSC). The regression approaches used in the study were partial least squares regression (PLSR) and principal component regression (PCR). The results showed that water content prediction acquired coefficient of determination in calibration set (R2cal) of 0.81, coefficient of determination in prediction set (R2pred) of 0.61, root mean squares error of calibration set (RMSEC) of 0.009, root mean squares of prediction set (RMSEP) of 0.020, and ratio performance to deviation (RPD) of 1.62, while soluble solids content prediction displayed R2cal, R2pred, RMSEC, RMSEP, and RPD of 0.79, 0.85, 0.474, 0.420, and 2.69, respectively. These findings indicated that near infrared spectroscopy could be used as an alternative technique to predict water content and soluble solids content of ‘manalagi’ apples.

scholarly journals Elaboration of near infrared spectroscopy model for determination of water content in freeze-dried immunological veterinary medicinal products

2016 ◽  
Author(s):  
Krzysztof Bulenger ◽  
Dorota Marta Krasucka ◽  
Bogumił Biernacki ◽  
Jakub Szumiło ◽  
Beata Cuvelier
Keyword(s):  
Infrared Spectroscopy ◽  
Water Content ◽  
Near Infrared Spectroscopy ◽  
Near Infrared ◽  
Residual Water ◽  
Medicinal Products ◽  
Karl Fischer Titration ◽  
Karl Fischer ◽  
Veterinary Medicinal Products

Residual water is a critical parameter in assessing the quality of immunological veterinary medicinal products (IVMPs). In majority of the laboratories the Karl Fischer titration (KFT) is used for the determination of water content in IVMPs. However, the transfer of IVMP into titration cell without affecting the baseline drift and repeatability seems to be the main problem when using this method. In turn, Near Infrared Spectroscopy (NIRS) allows measurement of closed vials, therefore eliminating the impact of atmospheric conditions on the sample. The aim of the study was to create a calibration model based on the reference method (Karl Fischer titration) and its optimization. Five different IVMPs designated for two animal species (dogs and rabbits) were used. The model was constructed on the basis of 49 samples tested, each in triplicate (n=147). The spectra were divided in two sets: calibration and validation. Proper selection of samples and their processing allowed to obtain a model of high quality (Q-value>0.6).

Sign in / Sign up

Export Citation Format

Share Document