Patterned ground, which occurs principally in polar, subpolar, and alpine regions, is broadly classified into sorted and nonsorted varieties of circles, nets, polygons, steps, and stripes. This descriptive classification and the associated terminology eliminate confusion resulting from the many overlapping and synonymous terms in the literature.
The origin of patterned ground is far from satisfactorily explained. Hypotheses are reviewed and summarized according to dominant processes as follows: (1) ejection of stones from fines by multigelation (often-repeated freezing and thawing), (2) mass heaving, (3) local differential heaving, (4) cryostatic movement (movement by frost-generated hydrostatic pressure), (5) circulation due to ice thrusting, (6) frost wedging, (7) absorption of water by colloids, (8) weathering, (9) contraction due to drying, (10) contraction due to low temperature, (11) contraction due to thawing, (12) convection due to temperature-controlled density differences, (13) convection due to moisture-controlled density differences, (14) movement due to moisture-controlled changes in intergranular pressure, (IS) differential thawing and eluviation, (16) vibration, (17) artesian flow, (18) rillwork (for stripes), (19) solifluction in combination with one or more of the above processes (for stripes).
Conclusions regarding origin are that: (1) the origin of most forms of patterned ground is uncertain; (2) patterned ground is polygenetic; (3) some forms may be combination products in a continuous system having different processes as end members; (4) climatic and terrain interpretation of patterned ground, both active and “fossil”, is limited by lack of reliable data about formative processes.
With respect to future research, it is apparent that: (1) laboratory experiments, including cold-room studies specifically dealing with patterned ground, are urgently required; (2) excavations rather than surface observations should be stressed in the field; (3) physicists, pedologists, plant ecologists, and engineers versed in soil mechanics have much to contribute to patterned-ground research, and joint work between them and geologists should produce particularly valuable results.