This comment addresses a statement made in “A review of air–ice chemical
and physical interactions (AICI): liquids, quasi-liquids, and solids in
snow” by Bartels-Rausch et al. (Atmos. Chem. Phys., 14, 1587–1633,
Recently, Bartels-Rausch et al. (2014) published a review of the recent literature on air–ice interactions. In it, they state that “using XMT is difficult when liquid is present, due to the small difference in absorption of liquid solutions and of solid ice. Hence it seems likely that the liquid features documented by Obbard et al. (2009) and Murshed et al. (2008) are to a certain degree sea salts that have precipitated at their imaging temperature of 263 K”. This is misleading.
First of all, XMT (X-ray computed microtomography) is not necessarily “difficult when liquid is present”. The difference in X-ray absorption between brine and solid ice is easily detectable with Oxford Instrument's SkyScan 1172 high-resolution desktop micro-computed tomography system. The first scientists collecting XMT images of sea ice (Golden et al., 2007) doped laboratory saltwater solutions with CsCl in order to produce ice with enough X-ray contrast for their instrument's 8 bit camera. With the SkyScan 1172, however, the three phases – ice, brine and air – can be easily distinguished due to their inherently different X-ray attenuation characteristics and the range of intensities (4096) captured by the instrument's 12 bit camera. This is explained in our paper (Obbard et al., 2009) and is illustrated with a reconstructed grey-scale image of a horizontal slice of a sample of Amundsen Sea ice showing ice (grey), brine (white) and air (black) (Obbard et al., 2009, Fig. 1).
Second, at the
The brine volume and chemistry of sea ice are clearly temperature-dependent,
although the exact freezing pathway, with respect to phases, especially
precipitates, present at each temperature is still an area of controversy. In
the following discussion we accept the Gitterman pathway, described in Marion
et al. (1999). Key is that as seawater begins to freeze, at about
Light et al. (2003) were interested in the effects of inclusions on the
optical properties of sea ice and, in addition to conducting lab
experiments, produced a model to determine temperature-dependent equivalent
cross-sectional areas for brine pockets, tubes, gas bubbles, mirabilite
crystals and hydrohalite crystals. According to their model, the equivalent
cross-sectional area for all types of salt inclusions (brine pockets, brine
tubes, mirabilite and hydrohalite) falls between
Saturated salt solutions and solid salts would have very similar X-ray attenuation coefficients, so one cannot determine analytically the phase present in brine inclusions in the reconstructed XMT images. However, an understanding of the thermodynamics of freezing seawater allows those conducting such experiments to accurately predict what we are seeing. XMT is indeed an excellent method to investigate the distribution of brine in sea ice, and combined with an understanding of phase changes in sea ice can give a very good idea of liquid brine volumes and distribution. We will publish a lengthier examination of temperature-dependent changes in sea ice in the near future.
The preparation of this manuscript was supported by National Science Foundation (NSF) Grant PLR-1304134. Edited by: E. Wolff Reviewed by: T. Bartels-Rausch and one anonymous referee