FINE STRUCTURE LAYERING IN RADIATION RECRYSTALLIZED SNOW

Edward E. Adams and David J. Walters

Department of Civil Engineering

Montana State University Bozeman, Montana USA

ABSTRACT: Radiation recrystallization of snow is a near-surface metamorphism process that results in the development of a layer of faceted crystals. When such a layer is subsequently buried by additional snow accumulation a stratigraphic construction that is structurally unstable may result. Previous research has revealed the environmental conditions that result in the formation of radiation recrystallized layers. However, the dynamic process of microstructural recrystallization is not well understood. With the aid of X-ray computed tomography (micro-CT) and optical microscopes, a detailed time history of the recrystalli-zation process has been recorded in sifted, sintered snow subjected to appropriate laboratory-simulated meteorological conditions. Recrystallization has been observed to develop thin layers of unique micro-structural morphology within the radiation recrystallized zone. The top surface layer consists of long slen-der needle crystals connected in a cross-hatch pattern. In addition, a fine layer that consists of small chains of facets oriented parallel to the temperature gradient, which is similar to depth hoar, but on a generally smaller scale, has been identified. Mechanical testing indicates that this weak layer is likely the primary failure stratum. The development of radiation recrystallization is likely the result of vapor flux in-ducing metamorphism within the snow and deposition at the snow surface.

KEYWORDS: snow, radiation recrystallization, kinetic growth, metamorphism, vapor pressure, gradient

1. INTRODUCTION

Snow is a porous, granular material composed of an aggregate of ice particles bonded together. In the natural environment its microstructure experi-ences nearly continuous metamorphism. Varying environmental conditions generally lead to the de-velopment of a stratified snowpack. One such near-surface layer, which can develop quite rapid-ly is termed radiation-recrystallized snow (LaChapelle 1970; LaChapelle & Armstrong 1977; Birkeland 1998; Morstad et al. 2007; McCabe et al. 2008; Slaughter et al. 2009; Slaughter & Adams 2010; Slaughter et al. 2011; Colbeck 1989). This microstructure, when buried by subse-quent snowfall, can result in a weak layer, which can be of concern with regard to avalanche stabil-ity (Walters et al. 2010; Walters & Adams 2014).

Key to the development of radiation recrystalliza-tion is that the radiative absorptivity of ice is wave-length dependent. Broadly speaking, in the visible (short-wave) spectrum ice is quite transparent - although not perfectly so. Radiative energy is ab-

sorbed and converted to thermal energy, attenuat-ing with depth. In the infrared (long-wave), it is nearly a black body. Since snow is composed of ice the process of radiation recrystallization is driven its material property. Under clear sky condi-tions, solar energy induces subsurface heating, while long-wave energy is emitted to the cold sky, cooling the surface. This produces near surface temperature gradients, which can cause near sur-face metamorphism termed radiation recrystalliza-tion.

In this presentation we examine the case of a sample of sifted snow that was allowed to sinter for two weeks prior to being subjected to condi-tions conducive to radiation recrystallization.

2. METHODS

The procedure was carried out in the Montana State University Subzero Science and Engineering Research Facility. The environmental chamber used for the development of the radiation recrys-tallization is configured such that the room tem-perature can be controlled independently from the ceiling. In the center of the ceiling a metal halide luminary provides a simulated solar source to illu-minate a sample.

Snow used for the case presented was produced in another cold laboratory, and then stored at -15C

* Corresponding author address: Edward E. Adams, Department of Civil Engineer-ing, Montana State University, Bozeman, MT 59717; tel: 1-406-994-6122; fax: 406-994-6105; email: eda@ce.montana.edu

Proceedings, International Snow Science Workshop, Banff, 2014

29

for two weeks prior to subjecting it to conditions conducive to the development of radiation recrys-tallization (Slaughter et al. 2009).

The initial snow was composed of dendritic crys-tals sifted through a 0.7 mm mesh into an insulat-ed container 0.7 m x 07 m on a side and 0.5 m deep, with an open top. During the sintering phase the top was covered, but it was exposed to the laboratory environment for the radiation recrystalli-zation phase. An X-ray computed tomography (micro-CT) image of the snow sample at the be-ginning of the radiation phase is displayed in Fig-ure 1.

Figure 1 CT image of the top 1 cm at the initial

stage of the snow sample. The image is of a prismatic rectangular sample with 5 mm x 5 mm cross-section. The orientation of this image is looking at a corner, so that the left and right sides of the image are at the corners and the thinnest part as seen from this perspective.

During the radiation recrystallization phase the container of snow was centered beneath the lumi-nary and cold ceiling. Cold laboratory conditions were held approximately constant for the duration; with a room temperature of -15 C, ceiling tempera-

ture of -48 C, simulated solar radiation at the snow surface of 706 W m-2, relative humidity of 70% and calm air conditions.

The concept behind the experimental setup is to mimic natural environmental conditions under which radiation recrystallized snow has been ob-served to develop. The luminary provides the inci-dent solar radiation source, while the cold ceiling represents the cold clear sky, which affords a ra-diation coupling between the snow surface tem-perature and the ceiling.

The example of morphology presented here is part of a series of mechanical tests being carried out on radiation recrystallization metamorphism. In the test presented here shear tests were carried out on similar snow types prior and subsequent to ra-diation metamorphism in a manner as presented in Walters et al. (2010) and Walters & Adams (2014). In addition and relevant to this presenta-tion, samples of morphology were taken during the metamorphic progression. In an adjacent cold la-boratory disaggregated grains were examined un-der an optical microscope, as would be typical in a field setting. In addition, micro-CT scans were tak-en that allow for in-situ examination. In this test the morphology was examined at 0, 3, 7, 10 and 12 hours.

3. RESULTS

Starting from an initial essentially isothermal snow temperature, the imposed conditions prompted an alteration of the thermal conditions of the sample. A vertical array of thermocouples was placed in the upper 7 cm of the sample on a 1 cm interval. The surface temperature was measured using a non-contact infrared thermometer. Temperature profile results are presented in Figure 2.

The influence of the solar warming and long-wave surface cooling is evidenced throughout the pro-file. At the 3-hour measurement a temperature gradient of 191 C/m was recorded. This transi-tioned to -5 C cm-1, -110 C cm-1 and -179 C m-1 at hours 6, 10 and 12 respectively. Based on the Clausius-Clapeyron equation calculated for ice over a flat surface the vapor pressure gradient for this final value would be 9 kPa m-1.

In the course of the 12-hour period there was a significant change in the near-surface morphology, which was readily observable. Although there was a significant temperature gradient driving thermal energy from the surface downward during the ear-ly stage, no apparent alteration of the snow mor-phology from the initial form was observed.

Proceedings, International Snow Science Workshop, Banff, 2014

30

Figure 2 Temperature profile variation in the upper

7 cm of the sample over 12 hours; follow-ing an initial isothermal temperature of -15 C. The hours coincide with the time of morphology sampling. By 10 hours an av-erage temperature gradient in the top 1 cm of 110 C/m had been established, reaching 179 C/m by the end of the test.

Some evidence of long slender needle-like crys-tals began to become apparent on the surface by the 7-hour observation (Figure 3). There may have been some grain coarsening just below the sur-face.

Growth of the surface needles was well developed by hour 10. What also became apparent by this stage was the manifestation of faceted partial cup crystals that had developed just below the needle layer (Figure 4). Distinct hexagonal and rectangu-lar shapes were manifest in these cups. At this time the thickness of the needle layer was ~0.9 mm and the faceted layer ~2 mm. Thicknesses of the individual layers are based on subjective ob-servation of morphology obtained from the micro-CT images. Below this layer the metamorphism was not obvious, although the transition between these layers is not abrupt.

Morphologic evolution continued until the conclu-sion of the study, at 12 hours (Figure 5). The nee-dle layer developed to a thickness of ~1.25 mm and the faceted cup crystal layer to ~2.5 mm. This faceted cup crystal layer developed columnar structures, with the overall morphology similar that often observed in depth hoar (e.g. Staron et al. 2014). This columnar structure is emphasized in the inset in Figure 5 and in an optical image in Figure 6.

The initial physically measured density of the sample was 155 kg m3. This initial density was

synchronized with the CT images and all threshold setting kept the same for subsequent images. Cal-culations were made of the density through time using micro-CT images for the region in which the cup crystal chains developed. This yielded values of 136 kg m-3, 124 kg m-3, 114 kg m-3 and 126 kg m -3. These values are based a small subset of the entire cross-sectional areas.

Figure 3 CT image of a representative sample at hour 7. Orientation and scale is as de-scribed in Figure 1.

Another metric that was considered is what is termed the “structure separation” in the Bruck-er/Skyscan software manual. The parameter is applied to the pore space and represents the di-ameter of the largest sphere that would be entirely bounded by the pore surfaces. Values calculated in the region in which the faceted cup crystal chains developed were 0.35 mm, 0.37 mm, 0.33 mm, 0.44mm and 0.52 mm respectively for each of the sampling times.

0 1 2 3 4 5 6 7

-11 -9 -7 -5 -3 -1

Dep

th c

m

Temperature C

3 hours 6 hours

10 hours 12 hours

Figure 4 CT image after 10 hours. Orientation and

scale is as described in Figure 1.

4. CONCLUSIONS

Results from one laboratory trial on the morpho-logical development of radiation recrystallization are presented. The example presented here is taken from a study that is aimed primarily at exam-ination of mechanical properties. Results present-ed for this single trial are representative of the morphologic development observed for all trials in the mechanical study run under similar environ-mental conditions using comparably produced snow samples, but including different sintering times. A relevant observation from these mechani-cal tests indicate that failure occurs in the faceted cup chain layer.

It is clear that in the process of radiation recrystal-lization, under at least the prescribed conditions presented, two distinct morphologic structures de-velop simultaneously. This results in three distinct stratigraphic layers within the top 5 -10 mm of the snow surface, but predominantly in the top 5 mm.

We hypothesize that the two different morphologi-cal developments are both driven by the large temperature gradients. One process occurs due to recrystallization within the snow, the other due to

water vapor exiting from the pore space being de-posited onto grains at the snow surface.

The faceted and cup crystal chain growth within the snowcover is driven by mechanisms that de-scribe the development of depth hoar. That is, sublimation from the upper surface of a warmer grain; diffusion of water vapor across the pore and condensation onto the bottom region of a colder grain (Akitaya 1974).

Figure 5 CT image of snow sample at 12 hours.

Orientation and scale of the primary image is as described in Figure 1. The inset on the left is an image of the needle layer viewed from the top. This layer is ~1.25 mm thick. The right inset shows an isolat-ed cup crystal chain attached to a needle. The height of this chain is ~2.5 mm.

In the second process, large vapor pressure gra-dients will drive vapor from pores that are open to the atmosphere directly from the snow into the overlaying air. This would supply sufficient vapor to be supersaturated relative to surface grains that are cooling through long-wave coupling with the cold sky (ceiling). This process is in agreement with a recent study on the Antarctic plateau by

Proceedings, International Snow Science Workshop, Banff, 2014

32

Gallet et al. (2014) in the growth of what they termed sublimation crystals.

A schematic representation of the presumed sim-ultaneous processes leading to the development of the near surface fine scale stratigraphy due to radiation recrystallization is presented in Figure 7 Hypothesized schematic of the dual processes occurring during radiation recrystallization meta-morphism.

Natural snowcovers in general are well stratified. This stratification, which is a response to varying atmospheric forcing, is important with regard to structural integrity. Here we present a particularly fine scale stratigraphic structure that results pri-marily due to snow and atmosphere radiation ex-change. Two distinct morphologic structures are observed to develop simultaneously at the snow surface - atmosphere interface.

Figure 6 Optical images of disaggregated grains

illustrating the two morphology structures observed during radiation recrystallization. (a) needle crystals, (b) faceted crystal chain and (c) Both needle and faceted-cup crystals; note that these are disaggregate displays and do not represent natural con-nection of the crystal types. Grid scale in these images is 1mm.

Figure 7 Hypothesized schematic of the dual pro-

cesses occurring during radiation recrys-tallization metamorphism. The red arrow is taken to represent the metmorphism oc-curing within the snow in a process similar to that which occurs in depth hoar. The green arrow represents the condensation onto the snowcover surface of water vapor that has is source as the snowcover itself.

Acknowledgements

This research was supported by the US National Sci-ence Foundation (NSF) grant No.1014997. The Mon-tana State University Subzero Science and Engineering Facility was funded by the NSF and the M.J. Murdock Charitable Trust under two separate grants.

REFERNCES

Akitaya, E., 1974. Studies on depth hoar. International Association of Hydrological Sciences …. Available at: http://itia.ntua.gr/hsj/redbooks/114/iahs_114_0000.pdf#page=54.

Birkeland, K.W., 1998. Terminology and predominant processes associated with the formation of weak layers of near-surface faceted crystals in the mountain snowpack. Arctic and Alpine Research, 30, pp.193–199. Available at: http://www.jstor.org/stable/1552134.

Colbeck, S.C., 1989. Snow-Crystal Growth with Varying Surface Temperatures and Radiation Penetration. Journal of Glaciology, 35, pp.23–29.

(a)

(b)

(c)

Proceedings, International Snow Science Workshop, Banff, 2014

33

Gallet, J.-C. et al., 2014. The growth of sublimation crystals and surface hoar on the Antarctic plateau. The Cryosphere, 8(4), pp.1205–1215. Available at: http://www.the-cryosphere.net/8/1205/2014/ [Accessed July 14, 2014].

LaChapelle, E., 1970. Principles of avalanche forecasting. Engineering and Avalanche Forecasting and Control, National Research Council, Canada, Technical(No. 98), pp.106–113.

LaChapelle, E. & Armstrong, R., 1977. Temperature patterns in an alpine snow cover and their influence on snow metamorphism. Institute of Arctic and Alpine Research. Technical Report. University of Colorado.

McCabe, D. et al., 2008. Near-surface faceting on south aspects in southwest Montana. In B. Gould & H. Steiner, eds. International Snow Snow Science Workshop. Whisler, BC Canada, pp. 147–154. Available at: http://hultsjz.issw2008.com/papers/P__8110.pdf [Accessed October 31, 2012].

Morstad, B.W., Adams, E.E. & McKittrick, L.R., 2007. Experimental and analytical study of radiation-recrystallized near-surface facets in snow. Cold Regions Science and Technology, 47(1-2), pp.90–101. Available at: http://linkinghub.elsevier.com/retrieve/pii/S0165232X06001169 [Accessed August 3, 2014].

Slaughter, A.E. et al., 2009. An investigation of radiation-recrystallization coupling laboratory and field studies. Cold Regions Science and Technology, 59(2-3), pp.126–132. Available at: http://linkinghub.elsevier.com/retrieve/pii/S0165232X09001281 [Accessed November 5, 2012].

Slaughter, A.E. et al., 2011. Field investigation of near-surface metamorphism of snow. Journal of Glaciology, 57(203), pp.441–452. Available at: http://www.ingentaconnect.com/content/igsoc/jog/2011/00000057/00000203/art00006 [Accessed October 31, 2012].

Slaughter, A.E. & Adams, E.E., 2010. Field and analytical examination of near-surface facets. In R. Johnson, ed. International Snow Science Workshop. Squaw Valley, CA USA, pp. 32–38. Available at: http://alpinix.com/Avalanche/ISSW/2010/Papers/ISSW_O-006.pdf [Accessed October 31, 2012].

Staron, P.J., Adams, E.E. & Miller, D.A., 2014. Nonequilibrium thermodynamics of kinetic metamorphism in snow. Cold Regions Science and Technology, 97, pp.60–71. Available at: http://www.sciencedirect.com/science/article/pii/S0165232X13001699 [Accessed February 20, 2014].

Walters, D.J. et al., 2010. Shear deformation of radiation recrystallized near surface facets. In R. Johnson, ed. arc.lib.montana.edu. Squaw Valley, CA USA, pp. 80–87. Available at: http://arc.lib.montana.edu/snow-science/objects/ISSW_O-015.pdf [Accessed October 31, 2012].

Walters, D.J. & Adams, E.E., 2014. Quantifying anisotropy from experimental testing of radiation recrystallized snow layers. Cold Regions Science and Technology, 97, pp.72–80. Available at: http://www.sciencedirect.com/science/article/pii/S0165232X13001511 [Accessed February 20, 2014].

Proceedings, International Snow Science Workshop, Banff, 2014

34