JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 108, NO. C2, 3051, doi:10.1029/2001JC000887, 2003
[11] While the previous observations provide a general picture of sea ice structure, many aspects of the microstructure which could affect light scattering remain uncertain due to the limited spatial resolution in the earlier studies. The highest resolution study [Perovich and Gow, 1996], for example, could only detect inclusions larger than 0.05 mm. Most of the earlier data were also obtained from horizontal thin sections [e.g., Perovich and Gow, 1991, 1996; Eicken, 1993: Light, 1995], restricting the information to inclusion cross sections and masking the three-dimensional nature of the microstructure. In addition, there is little information about how inclusions interact as temperature changes. Of particular interest are the formation and growth of bubbles within brine inclusions, the merging and coalescing of brine inclusions at warmer temperatures, and the precipitation characteristics of salt crystals. Such information is needed in the formulation of a more accurate structural model for sea ice.
[12] To address these shortcomings, a program was designed to obtain very high resolution imagery of natural sea ice samples under a wide range of temperatures, the primary focus being on first-year ice. Because of the need for accurate temperature control, measurements were carried out in a laboratory cold room, rather than in situ. Ice cores used in this study were extracted from shorefast ice near Point Barrow, Alaska in May 1994. The ice was 1.65–1.75 m thick with near-surface temperatures of about -5.5°C, increasing linearly with depth. Measured ice salinity profiles were typically C-shaped with values ranging from 7–9‰ near the surface, 4–5‰ in the interior, and about 10‰ near the bottom. Calculated brine volumes ranged from 5–12%. Gas volumes were about 4% in the upper 10 cm of the ice, but less than 1% below this [see Perovich et al., 1998b]. Cores, 10 cm in diameter, were removed from the upper meter of the ice, shipped to Seattle in dry ice, and stored in a laboratory freezer at -20°C for three years. While it is not known how shipping core samples in dry ice and subsequent long-term storage affect the ice microstructure, we proceeded to analyze these samples for their microstructural and optical properties.
[13] Cores were cut into thin sections for laboratory examination. To avoid complications caused by surface-related processes (e.g., melting/refreezing, brine drainage, infusion of air), samples were taken from the lower portions of the cores, corresponding to a depth of 60–80 cm in the ice. These samples had relatively small gas volumes and brine inclusions which tended to be isolated, unlike ice near the growth interface or upper surface. Because the vertical dimension of brine inclusions is often much larger than their horizontal dimension, vertically oriented, 8 × 8 cm slabs were cut from the cores and microtomed to a thickness of approximately 2 mm. The sections were prepared at -15°C to minimize brine drainage. Thin sections were sealed between clear glass plates which allowed them to be mounted vertically for monitoring the thermal evolution of the ice. The sealed glass plates kept the structure intact, even at quite warm temperatures where the porosity was high and the samples extremely fragile. Samples were mounted on a x-y translation stage, allowing them to be moved as much as 3 cm laterally and 3 cm vertically, and then returned to a predetermined position. This made it simple to revisit and photograph specific scenes in the ice as conditions changed. Image sizes were calibrated using a grid with ruled lines separated by 50 microns.
|
|
[14] To record structural changes, we used a black and white CCD video camera (Panasonic model WV-BP500) with 682 by 492 pixel resolution and a high magnification zoom lens (Leica MonoZoom 7 Optical System). The high magnification lens (coupled with a 3X amplifier) gave us the ability to resolve inclusions as small as 0.01 mm, considerably smaller than in previous studies. A majority of our images were made with the zoom lens set to record approximately 0.003 mm per pixel, while some higher resolution images were made at 0.002 mm per pixel. Our objective, however, was to monitor features and processes that are important to understanding the optical properties of the ice, not necessarily to document the smallest features of the microstructure. Images obtained by the CCD camera were fed to a high resolution monitor visible through the cold room window. This monitor was used to select and focus images. Images were also fed to a MacIntosh Power PC, equipped with a video frame grabber (Scion VG-5) and image processing software (NIH Image 1.57). To facilitate alignment and reduce the effects of vibrations, components in the cold room were mounted on a 2-m long optical rail (see Figure 3).
[15] Information on crystallography is best obtained from images recorded in polarized light, whereas, inclusions of brine and gas are best viewed in transmitted or reflected natural light. Experiments with both types of illumination indicated that high contrast images with uniform lighting could be obtained with transmitted light from a diffuse source. To achieve this, an incandescent light was aimed at a diffusely reflecting Spectralon® panel mounted directly behind the sample. The source was baffled to prevent direct radiation from illuminating or heating the sample.
Citation: Effects of temperature on the microstructure of first-year Arctic sea ice, J. Geophys. Res., 108(C2), 3051, doi:10.1029/2001JC000887, 2003.
