If you are using the IKONOS 4m multispectral bands, then you will have coverage from around 0.4 to 0.8 microns, covering the visible and near-infrared spectral range (VNIR). The spectral bandwidth for the IKONOS sensor is relatively coarse but using this multispectral data you could potentially detect geological features which have distinct features in the visible-near infrared range (VNIR).
In the VNIR mineral absorption features are dominated by electronic processes occurring within transition metals (i.e. Fe, Mn, Cu, Ni, Cr etc.) of which the most important is iron (Gupta, 2002). Iron-bearing minerals display absorption features at wavelengths less than ~0.55 microns associated with Fe-O charge transfer and between 0.85 and 0.92 microns associated with crystal field effects. The Fe-O charge transfer feature is responsible for a steep decline in reflectance towards the blue end of the spectrum, which is common to all iron-bearing minerals. This absorption feature is most pronounced in iron oxides and hydroxides that display characteristic yellow, orange, red and brown colouration (Drury, 2001).
Applying this information, you could for example carry out a (Band 1 / Band 2) ratio to identify the intensity of the absorption features associated with ferric iron content. As the IKONOS sensor is lacking bands in the short wave infrared (SWIR; 1 to 2.5 microns), there is limited discriminatory ability, mostly restricted to the identification of iron bearing minerals as described above. However, you could apply other techniques such as a Principal Component Analysis or Decorrelation Stretch and qualitatively investigate lithological differences; however, spectral identification may remain difficult.
References:
Gupta, R. (2002). Remote Sensing Geology, 2nd Edn, Springer.
Drury, S. (2001). Image Interpretation in Geology, Blackwell Science.
- Badges
- Science topic
- Similar topics
- Analytical Chemistry
- Column