JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 108, NO. E4, 8032, doi:10.1029/2002JE001864, 2003

2. Rock Glacier Definitions

[3] Because of the difficulties of applying a unique model for rock glacier formation and flow mechanisms (discussed in the next section) it is desirable that a nongenetic definition and terminology be used. This is helpful on Earth and essential for defining Martian analogs. Some confusion has been, and still is, associated with adherence to a genetic definition for terrestrial rock glaciers. A definition based on morphology, which does not assume a particular formational model, appears to be a reasonable starting point. Genetic findings can then be added to discussions about individual cases. Current generalizations are, unfortunately, likely to mislead as much as clarify.

[4] A suitable descriptive definition of a rock glacier [Potter, 1972; Washburn, 1979] is “a tongue-like or lobate body, usually of angular boulders, that resembles a small glacier, generally occurs in high mountainous terrain and usually has ridges, furrows, and sometimes lobes on its surface, and has a steep front at the angle of repose.” Definitions sometimes also include a mention of the low flow rates [Martin and Whalley, 1987], which are typically <1 m a^-1.

Figure 1a. Jumbo Rock Glacier, Wrangell Mountains, Alaska. Note the two phases of formation distinguished by shades of gray and also the lack of lobes from the valley sides.

Figure 1b. Aerial Photograph showing typical rock glaciers in the Wrangell Mountains, Alaska; a has a single lobe, and b has a second lobe which appears to have advanced on top of the first as in the case of Jumbo Rock glacier (Image: US Geological Survey).

Figure 2. Debris-ice components in mountainous area with suggested terminology for the main features and alternative terms in parentheses. After Martin and Whalley [1987], Humlum [1988], and Hamilton and Whalley [1995].

[5] Definition by visual analogy is especially helpful for the interpretation of photographs, especially MOC, so the form itself is best illustrated with a photograph (Figures 1a and 1b). Figure 2 (from Hamilton and Whalley [1995] and Martin and Whalley [1987], after Humlum [1988]) places rock glaciers in the context of mountain landforms. The diagram also shows other debris related forms, discussed below, and provides some of the alternative names found in the literature [Martin and Whalley, 1987]. Capps [1910] and Wahrhaftig and Cox [1959] described the Alaskan landforms seen in Figures 1a, 1b, and 2. We follow Capps' description and term this a rock glacier sensu stricto (but usually dropping this Latin tag).

Figure 3. Aerial photograph of protalus lobes, Axel Heiberg Island, Canadian Arctic (Image: Canadian Department of Energy, Mines and Resources; Ottawa).

Figure 4a. Protalus rampart (a) and poorly-developed protalus lobes (b) (compare to Figure 3) below cliffs (Mynydd Ddu, Wales). The area between the ridges and cliffs would have held a substantial body of ice which, in this case, was probably not supplied with a continuous supply of rock debris.

Figure 4b. Active production of protalus rampart below small “glacierette” (Pic Campbell, Pyrenees, France). This is similar to the construction of a terminal moraine by a glacier.

Figure 4c. Active glacier with protalus lobe-like feature in front. Note that, although the form is rather different from that in Figures 1a, 1b, and 5a, it is only some 3 km from the feature shown in Figure 5a and is also thought to contain a glacier ice core as seen by the “thermokarst” lakes set in the debris cover.

Figure 4d. A probably inactive rock glacier but with features similar to Figure 4c (Olympus Range, New Zealand). A “spoon-shaped depression” occurs between the headwall talus and the main rock glacier (compare Figure 5a, where the glacier still exists).

[6] Several authors have subsequently modified the terminology and even used “rock glacier” as indicative of a rather different feature (e.g., Figure 3) and illustrated as a “protalus lobe” in Figures 1a and 1b. Thus Outcalt and Benedict [1965] introduced the term “valley-side rock glacier” after a study of features in Colorado. The terms “valley floor rock glacier” or “tongue-shaped rock glacier” [Wahrhaftig and Cox, 1959] were used for the sensu stricto feature. Unfortunately, later discussants have often omitted the “valley side” or “lobate” part and thus “rock glacier” has been used for both features regardless of topographic form. Some authors have indeed suggested differences in origin of these two forms as well as them being in a continuum of landscape elements (see, e.g., Whalley and Martin [1992] and Giardino and Vitek [1988] for discussion). If a genetic connotation is implied in the use of either term, still further confusion arises when discussing genesis. The term “protalus lobe” was introduced as a nongenetic term [Hamilton and Whalley, 1995] acknowledging that there might be differences of ice origin as well as topography. The diagram (Figure 2) summarizes these terms. Figure 3 illustrates distinctive protalus lobes and Figures 4a, 4b, 4c, and 4d show possibly related features but which might also be “protalus ramparts” as well as a rock glacier with convoluted transverse ridges. Martin and Whalley [1987] and Hamilton and Whalley [1995] engage in further discussion. The possible origins of these features are reviewed in the next section.

[7] The discussion of possible rock glacier-like features on Mars has been similarly confused, not only because of failure to distinguish between the two types of rock glacier but also because the rock glacier form may be derived in perhaps three possible ways.

Citation: Whalley, W. B., and F. Azizi, Rock glaciers and protalus landforms: Analogous forms and ice sources on Earth and Mars, J. Geophys. Res., 108(E4), 8032, doi:10.1029/2002JE001864, 2003.