The use of near infrared (NIR) spectroscopy for predicting levels of degradation in laboratory soil block tests was investigated. Calibrations were developed for mass loss, compression strength, and exposure period using data measured from the prior methods, and untreated and mathematically treated (multiplicative scatter correction and first and second derivative) NIR spectra from various spans of wavelengths by partial least squares regression. Strong correlations were obtained from each study conducted, while calibrations developed from NIR spectra from the cross-sectional face of southern yellow pine presented the strongest predictions. Of them, calibrations for mass loss resulted in the strongest predictions. Calibrations constructed from spectra obtained from the radial face of southern yellow pine also produced strong predictions, where the strongest model was for exposure period. While, calibrations developed for cottonwood presented the weakest statistics, the strongest calibration found was for exposure period.
| Identifer | oai:union.ndltd.org:MSSTATE/oai:scholarsjunction.msstate.edu:td-4718 |
| Date | 01 May 2010 |
| Creators | Green, Benny |
| Publisher | Scholars Junction |
| Source Sets | Mississippi State University |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Theses and Dissertations |
Page generated in 0.0022 seconds