The LAS technique uses the FieldSpec Pro FR portable spectroradiometer (Hatchell, 1999) to measure, in the visible and near-infrared (350-2500 nm), the spectrum of light reflected from a rock surface. Light is absorbed by minerals at and near the surface of the rock or sediments as a result of both electronic and vibrational processes (Clark et al., 1990; Clark, 1995), which create absorption troughs in the reflecting spectrum that can be used for mineral identification. H2O, Mg-OH, Al-OH, Fe-OH, and CaCO3 absorption bands (identifiable in the near-infrared) as well as overall trends in the spectrum are particularly useful for mineral identification. If LAS is accompanied by a suite of local ground-truth measurements (e.g., X-ray diffraction [XRD]), depths of characteristic absorption features can be calibrated, resulting in the ability to determine semiquantitative mineral concentrations. This approach is likely to work for sediments with two to four spectrally significant minerals; it may not succeed in more complex mineralogies.
In order to gain the most accurate spectral data, samples should be thoroughly dried. If samples are wet, the spectral signature of pore water dominates the spectral features of mineralogical origin. Also, wet sediments are often darker than dry sediments, which affects total reflectance readings (Balsam et al.,1998).
Figure F1 illustrates two different instrumentation setups used to generate LAS measurements during ODP core samples. The light source for both methods is a high-intensity light probe fitted with a quartz halogen bulb and built-in direct current stabilizer circuitry. Before a reflectance measurement is taken, spectral response of any light source is normalized to a 100% reflectance level using a Spectralon white BaSO4 calibration plate.
The first instrument configuration (Fig. F1A) is used when the sample is large enough to cover the entire 1-in-diameter opening in the bottom of the high-intensity light probe. The fiber-optic cable, which conveys the reflected light to the spectrometer, is inserted within the light probe. The sample, preferably powdered, is spread out on a platform. To take the measurement, the light probe is placed directly onto the sample and the reflected light is sent via the fiber optic cable to the spectrometer. The spectral data is then saved on a small laptop computer. This method requires additional time to clean the surface of the light probe after each reflectance measurement.
The second method (Fig. F1B) is used when there is not enough powdered sample to cover the entire light probe opening or when the sample is solid but irregular in shape. In these situations, the fiber-optic detector cable is attached to a specialized holder and pointed at the sample, which again is placed on a small platform. It is important that the light probe (used only as a stable light source in this case) is placed at the same angle and distance in relation to the sample as the fiber-optic sensor. This will maximize the amount of reflectance recorded by the spectrometer. This method requires movement of only the sample and involves no contamination issues (the light probe never comes into direct contact with the sample), resulting in a faster measurement time.
We prefer to use the instrumentation setup shown in Figure F1A because it prevents light from escaping and results in a more accurate total reflectance value. Even though this method requires additional time to perform, it is important to obtain an accurate total reflectance value for use in our mineral concentration calculations. During Leg 199, great care was taken to ensure that the light probe was clean before measuring the next sample.