About  |  News & Events  |  Products  |  Jobs  
HOME SCIENCE EDUCATION COLLABORATION SEARCH CIRES

Center for the Study of Earth from Space Center for the Study of Earth from Space
     CSES Home  |   Research  |   Education  |   Faculty  |   Facilities  |   Software  |   Links  |   Contact CSES

NASA/JPL contract No. 960983: 1997-2000

Identification and Mapping of Expansive Clay Soils in the Western U.S.

Using Field Spectrometry and AVIRIS Data

Results

Preliminary results in 1998: study of natural trench samples, Denver metropolitan area

Near-infrared reflectance spectroscopy of swelling soil samples with field spectrometers (Fieldspec (tm), ASD Inc.) shows that it is possible to discriminate among pure smectite and mixed smectite/illite layers samples. Figure 3 shows reflectance spectra of trench samples covering the whole range of variability in terms of mineralogy and swelling potential. Indeed, the spectra shown are associated with samples of mineralogy from pure montmorillonite (#96027, 96015), to smectite/illite layers with variable content in illite (#95014, 96017, 96024), to kaolinite-rich smectite/illite layers (#95022, 95017). Those samples have very variable swelling potential (see labels fig. 3): I=very high; II=high; III=intermediate; IV=low; V=none. The absorption band at 2.35 um provides a measure of the illite content. The more smectite-rich is the sample is indicative of higher swelling potential. Spectral detection of high water content is also possible, and is in saturated soils indicative of higher swelling potential. Spectral detection of kaolinite is also possible above 10% in clay fraction, and a high content of kaolinite will be indicative of a poor swelling potential. Those straightforward spectral observations are well correlated with mineralogical x-ray diffraction analyses and geotechnical engineering tests (Chabrillat et al., 1997). We were not able on this set of samples (mainly calcium-rich montmorillonites) to determine any variation in the location of the 2.2 um smectite band to discriminate between Na- or Ca- montmorillonite. We also used linear unmixing tools and statistical multivariate analyses (Unscrambler (c), Camo Inc.) for quantitative spectral determination of mineralogical composition.

Trench Samples
Figure 3. Laboratory reflectance spectra from field samples of swelling soils.





Privacy
Statement
An internationally recognized leader in innovative environmental science and research,
CIRES is jointly sponsored by the University of Colorado at Boulder
and the Office of Oceanic and Atmospheric Research at NOAA.
About | Science | Education | Collaboration | Accomplishments | News/Events
Products | Jobs | Search  |  Contact Us             [ page last modified: 01/25/05 ]



NOAA
Disclaimer