Frost heave:
monitoring with strain gauge transducers
Automated monitoring of frost heave on sorted stripes in the Swiss Alps.
Norikazu
Matsuoka, Institute
of Geoscience, University of Tsukuba, Ibaraki 305-8571, Japan.
Background
Frost heave is a fundamental periglacial process dominating
over soil surfaces in cold regions. On periglacial slopes, frost heave is
followed by mass movements including frost creep, gelifluction, debris flow and
active layer detachment slides. The repetition of frost heave and thaw
consolidation also produces periglacial features such as solifluction lobes,
sorted patterned ground and earth hummocks. Field monitoring of frost heave (see
figure above) allows us to understand these geomorphological features and to
estimate the potential frost creep (PFC), which is given by PFC = H tan θ,
where H is the heave amount and θ is the slope angle.
A number of attempts have been undertaken on measuring frost heave. The
remoteness of the research sites with periglacial features, however, have
hindered long-term, continuous measurements with a highly instrumented
monitoring system, because such locations are normally visited only once to
several times per year and electricity is not available.
Field measurements during the 1960’s addressed the annual heave amount of heaved targets (Washburn, 1967; Benedict, 1970), but the amount tended to underestimate the actual value because the targets may have failed to record partial settlement during thawing. Measurements during the 1970’s-1980’s used “the bedstead” as an anchored frame on which a mechanical recorder is fixed (Fahey, 1973; Smith, 1987; Matsuoka and Moriwaki, 1992). A soil heave curve was drawn on a rotating sheet. This technique allowed continuous recording of heaving, but low resolution (at most 0.1 mm) and frequent mechanical troubles prevented acquisition of long-term, high-quality data.
Since 1990’s, electrical sensors connected to data loggers have allowed high resolution, continuous recording of frost heave, and this system requires only annual maintenance (e.g. Matsuoka, 1996; Matsuoka et al., 1997; Hallet, 1998). This report introduces an example of such electric measurements used on high mountains.
Techniques
The figure below illustrates an instrumentation in the field. Soil heave with reference to a fixed point is recorded with a sensor attached to a bedstead anchored in subsurface permafrost or bedrock. An angle iron frame is commonly used as a bedstead. Upheaving of the angle iron legs is avoided by inserting the base of legs well into the bedrock or permafrost, with the aid of an anti-heaving material (e.g. timber or concrete) attached at the base. Where such a subsurface hard bed (bedrock or permafrost) is absent, the base of the bedstead has to be installed deeply in sediments, at least below the seasonal frost depth. Where a deep snow cover (> 2 m) is expected, the frame has to be reinforced to resist snow pressure, for instance, using a three-dimensional frame.
Fig. 2. Field monitoring of differential heave on sorted pattern ground.
A displacement transducer senses soil heave. The capacity of the transducer should exceed the expected maximum heave amount. The use of a single transducer allows recording of 1D (vertical) soil movement, while a combination of two intersecting transducers permits recording of 2D displacement (cf. laboratory techniques by Harris et al., 1997). Matsuoka (1996) and Matsuoka et al. (1997, 2003) use a commercial strain-gauge type transducer (DT-100A) manufactured by Kyowa Electronics, Japan. This sensor can record displacement up to 10 cm at a resolution of 0.04 mm. Since strain gauges show some temperature dependency (usually negligible compared with heave amounts), calibration may be required. A plate (a few cm in diameter) is attached at the base to sense planar heave. Subsurface heave can be measured by installing the base at a depth below the surface. A spring in the sensor enables elastic response to settlement of the heaved surface. However, the spring may in turn exert pressure to the soil. For example, the DT-100A sensor presses the ground with a 450 gf force, which often results in slight penetration of the basal plate into the soil softened during thawing (e.g. Matsuoka et al., 1997). Waterproofing is also necessary, where snow meltwater is significant.
The strain-gauge type sensor requires a date logger having strain inputs. The strain readings are translated to heave amounts using a correction factor. The detailed process of heave and settlement can be evaluated by a concurrent measurement of soil temperatures at various depths (Fig. 2). Diurnal frost heave cycles are detected by recording at 1 to 3-h intervals. Shorter intervals are required to detect detailed differential heave, for instance, between fine and coarse stripes of patterned ground.
Examples of field monitoring
1. Timing and amount of heave in the Japanese Alps (Matsuoka, 1996).
Fig. 3. Soil heave, temperature and moisture during the autumn freeze-thaw period of 1991 on a debris slope (2880 m ASL), southern Japanese Alps.
Diurnal frost heave activity was monitored on debris slopes subject to deep seasonal freezing (1-2 m) in the southern Japanese Alps. Figure 3 displays results on a 30° slope during an autumn freeze-thaw period. A comparison between data on the heave amount, frost depth and soil moisture shows frequent heaving up to 3 cm with shallow freezing (< 10 cm). Not all freezing events result in heaving. Moisture supply controls heave amounts more critically than the diurnal frost depth. Despite nocturnal sub-freezing temperatures soil desiccation does not allow heaving after continuous sunny days. In contrast, the heave amount reaches a maximum when soil freezing immediately follows precipitation (rain or snow) and then decreases with soil desiccation.
Frequent cycling of heave and settlement in the topsoil is responsible for shallow but rapid soil creep with a surface velocity in places exceeding 50 cm/a (Matsuoka, 1998). Such a movement eventually produces small-scale periglacial features including thin stone-banked lobes and small sorted stripes, which are widespread on low- to middle-latitude high mountains (e.g. Bertran et al., 1995; Matsuoka, 2001).
2. Differential heave on sorted stripes in the Swiss Alps (Matsuoka et al., 2003).
Small-scale sorted patterned ground (with a diameter or spacing less than 30 cm) in which sorting occurs within the top few centimeters of soil is considered to reflect diurnal frost heave cycles. Differential heave, rather than soil convection requiring a high moisture level, is likely to be responsible for such small stripes (e.g. Werner and Hallet, 1993). However, previous field studies have rarely linked differential heave to the formation of small patterns, with a notable exception of a field experiment by Ballantyne (1996).
Fig. 4. Soil heave and temperatures in 1999-2000 on small-scale sorted stripes, Valletta site (2810 m ASL), the Swiss Alps.
An attempt has been done to monitor differential heaving between coarse and fine stripes on a limestone hill in the Swiss Alps (Fig. 1; Matsuoka et al., 2003). Here the ground undergoes both frequent diurnal heave cycles and an annual heave cycle (Fig. 4). The seasonal heave amount (ca. 1 cm) is similar for the coarse and fine stripes. Differential heave mainly accompanies diurnal freeze-thaw cycles. Figure 5 shows the amount and timing of diurnal heave sometimes differ between the two stripes. In particular, shallow freezing with needle ice formation seems to be most responsible for the differential heave amount, because needle ice occurs only on the fine stripe. In addition, the surficial coarse material delays heaving of the subsoil below the coarse stripe, causing a time lag in heaving between the two stripes. Such a time lag may enhance sorting, because the earlier heave on the fine domain favors transport of coarse debris on to the coarse border by toppling of needle ice.
Fig. 5. Soil heave and temperatures during the autumn freeze-thaw period of 2000 at Valletta site.
References
Benedict JB. 1970. Downslope soil movement in a Colorado alpine region: rates, processes, and climatic significance. Arct. Alp. Res. 2: 165-226.
Bertran P, Francou B and Texier JP. 1995. Stratified slope deposits: the stone-banked sheets and lobes model. In Steepland Geomorphology, Slaymaker O (ed). Wiley: Chichester; 147-169.
Fahey BD. 1973. An analysis of diurnal freeze-thaw and frost heave cycles in the Indian Peaks region of the Colorado Front Range. Arct. Alp. Res. 5: 269-281.
Harris C, Davies MCR and Coutard J-P. 1997. Rates and processes of periglacial solifluction: an experimental approach. Earth Surf. Process. Landform. 22: 849-868.
Hallet B. 1998. Measurement of soil motion in sorted circles, western Spitsbergen. In Proc. 7th Int. Conf. Permafrost: 415-420.
Matsuoka N. 1996. Soil moisture variability in relation to diurnal frost heaving on Japanese high mountain slopes. Permafrost Periglac. Process. 7: 139-151.
Matsuoka N. 1998. Modelling frost creep rates in an alpine environment. Permafrost Periglac. Process. 9: 397-409.
Matsuoka N. 2001. Solifluction rates, processes and landforms: a global review. Earth-Sci. Rev. 35: 107-134.
Matsuoka N and Moriwaki K. 1992. Frost heave and creep in the Sør Rondane Mountains, Antarctica. Arct. Alp. Res. 24: 271-280.
Matsuoka N, Hirakawa K, Watanabe T and Moriwaki K. 1997. Monitoring of periglacial slope processes in the Swiss Alps: the first two years of frost shattering, heave and creep. Permafrost Periglac. Process. 8: 155-177.
Matsuoka N, Abe M and Ijiri M. 2003. Differential frost heave and sorted patterned ground: field measurements and a laboratory experiment. Geomorphology (in press).
Smith DJ. 1987. Frost-heave activity in the Mount Rae area, Canadian Rocky Mountains. Arct. Alp. Res. 19: 155-166.
Washburn AL. 1967. Instrumental observations of mass-wasting in the Mesters Vig district, Northeast Greenland. Meddr. Grønland 166: 318.
Werner BT and Hallet B. 1993. Numerical simulation of self-organized stone stripes. Nature 361: 142-145.
Latest update: 8. April 2003.