Needle ice forms in porous, wet soil or sediment when the soil temperature is above freezing and the surface atmosphere temperature is below freezing. Permeability is important. Capillary action pulls water up toward the surface where the water freezes at the bases of growing ice crystals. The growing crystals are capable of lifting small particles and vegetation above the normal soil surface. This differential motion can thoroughly disaggregate and disrupt the upper several centimeters of the soil profile in temperate climates.
The images in this gallery were made in an elementary school yard where patches of bare soil are exposed. By the time I walked over the yard, the sun had melted most of the crystals except for those protected by shadows. Most of the single, larger grains on top of the crystals are coarse sand, perhaps around 2 mm (millimeters) in diameter.
All photographs in this post ©Andy R. Bobyarchick.
This is a cluster of mostly standing needle ice crystals. Some of the crystals have toppled over as they began to melt. It’s likely that these crystals grew during a single freeze and growth event. In clusters with multiple growth events, there is often an intermediate layer of soil particles embedded across the axes of the crystals.
Bundles of needle ice crystals in this cluster have fallen in different directions, probably oriented toward the least resistance or toward first melting. Some of the crystals have become detached fibers.
If you look closely, you can see that many of the needle ice individuals have longitudinal trains of bubbles. These may be air bubbles released as the crystals warm up and expand, or trapped during crystal growth. Whatever they are, the bubbles enhance the intricate internal structure of the crystals.
As though swept over by a wind, these translucent, melting crystals all tilted in the same direction, roughly toward the east. It may be that eastern members of the cluster melted earlier in the day (this image was made at 1:29 PM), creating space for these crystals to fall.
Here are some fibrous crystals, all fallen but many retaining their sandy caps.
Here is a very close view of a crystal forest with sand parapets overlooking great chasms.
Into the deep canyons you go.
Here is a streamlet of needle ice not quite frazil ice. This cluster did grow in a small gully, perhaps with a fair amount of water in the soil at the time of freezing. Several of these crystal streams have deep axial rifts. It could be that the thalwegs of these little streamlets were too saturated to allow crystal growth. It did not appear that the rift floors had been modified by crystal growth.
This is the setting of the ice palaces. Not very impressive at this scale, is it? That’s because the most beautiful complexity in geology in the field is sometimes only visible in the Underworld – that diminutive realm within our vision but not often visited.
Carter, J. R., 2013, Flowers and ribbons of ice: American Scientist, v. 101, p. 360-369.
Li, A., Matsuoka, N., and Niu, F., 2017, Frost sorting on slopes by needle ice: A laboratory simulation on the effect of slope gradient: Earth Surface Processes and Landforms, p. <xocs:firstpage xmlns:xocs=””/>. 10.1002/esp.4276
Li, A., Matsuoka, N., and Niu, F., 2018, Frost sorting on slopes by needle ice: A laboratory simulation on the effect of slope gradient: Earth Surface Processes and Landforms, p. n/a-n/a. 10.1002/esp.4276
Soons, J. M., and Greenland, D. E., 1970, Observations on the Growth of Needle Ice: Water Resources Research, v. 6, no. 2, p. 579-593. 10.1029/WR006i002p00579
Yamagishi, C., and Matsuoka, N., 2015, Laboratory frost sorting by needle ice: a pilot experiment on the effects of stone size and extent of surface stone cover: Earth Surface Processes and Landforms, v. 40, no. 4, p. 502-511. 10.1002/esp.3653