Upfreezing by Frost-pull as a thermal ratchet

Not sue what happened but this post should have appeared before the following post discussing stone and rock circles.

It would appear that the process responsible for the gradual upward movement of rocks, and other inclusions of greater than average grain size, is fairly well understood. The process of “frost-pull” seems to account for the phenomenon known as “upfreezing”, whereby stones, boulders, fence posts and even it seems coffins, are over a period of time extruded from the earth to end up on the surface, see for example Figure 4.13 in Washburn (1979). Because upfreezing represents a form of thermal ratchet, in the current sense, it might be helpful to briefly summarise what it involves. The use of the descriptor, stone, in what follows could be applied to any material objects whose size and mass are greater than the average grain size of the soil.

Upfreezing by Frost-pull

Figure 5 reproduces in somewhat different form Figure 4.13 from Washburn (1979), describing qualitatively the process known as frost-pull. By many this is considered to be the dominant driving force underlying the process of upfreezing, whereby, over a period of time, larger than average objects are expelled upward from the earth. It certainly appears to represent the most compelling explanation. In Figure 5(a) a typical annual temperature cycle is used to indicate when each of the time-sequenced sections is taken through a typical rock at depth, d, and having a vertical thickness of t. These sections are taken at the eight indicative times shown in Figure 5(b). At time (1), sometime in early autumn, the average ground surface temperature will have dropped below zero and frost will be just starting to form at ground surface level. Over the next few weeks the temperatures at ground level will continue to drop until the lower frost line (LFL) reaches the top of the stone. During this period the ground surface will experience frost heave as indicated at time (2). When the LFL reaches the depth of d at the top of the stone the ice will affect a bond with the top of the effectively rigid stone. This means that the increment of frost heave, , occurring over the time it takes for the LFL to further aggrade to the level d+t at the bottom of the stone, corresponding with time (3), will also be experienced by the stone. Over this period of frost aggradation a cavity of depth  will have opened up beneath the stone. With the stone now firmly bonded into the frozen ground the remaining heave, occurring over the time it takes for the LFL to reach its lowest level, at time (4), will also be experienced by the stone. Some of the cryostatic expansion occurring during the aggrading of the LFL below the stone may be absorbed by a partial closing of the cavity beneath the stone. On account of the time lag associated with the conduction to this LFL the lowest point for the LFL will correspond with a time somewhat after minimum surface temperature is reached.

                                      

As surface temperatures begin to rise in late winter, indicated by section (5), surface thawing will commence. As the upper frost level (UFL) lowers to reach the bottom of the stone, through section (6) and (7), the previous heave +, experienced when the LFL aggraded through this same depth, will at the surface be largely recovered. During this period the stone will remain unmoved as a result of its base still being bonded to the relatively rigid lower frozen ground. Only when the complete thaw has been achieved will the stone move with the rest of the now unfrozen soil to recover that part of the heave experienced when the LFL aggraded from the bottom of the stone to its mid-winter lowest level. With any remaining cavity beneath the stone likely to be filled with slump material during the thaw beneath it, the stone will now be left at an elevation roughly above its starting point at the end of the previous summer. Each annual cycle of frost will result in the stone being pushed a little nearer to the surface, with the rate of its rise being directly proportional to its vertical height t. Similar processes would be anticipated to occur for shorter period frost-thaw cycles. Whatever the dominant driving periodicity the stone will eventually be expelled from the soil to lie at the surface.

 For this phenomena the ratchet results from the interaction of the volume increase occurring when ground water experiences a phase change to ice and expelled pore water pressure beneath the LFL inducing back-fill into the cavity created by the upward heave of the stone relative to its underlying soil. Its occurrence also depends upon the differential bond between the stone and frozen ground compared with that between the stone and unfrozen ground.  While the cause may differ slightly it is suggested here that at least some of the horizontal movements of stones to form stone circles, polygons, nets etc may result from a closely related form of thermal ratchet.