Many geological processes may create clastic wedges and polygons, with a surficial appearance similar to frost-created features, like wetting and drying, (non-freezing) thermal contraction, in gilgai, sedimentary compaction, gravitational loading, small scale tectonics, flexure over an uneven surface, volume change during cementation, in sheeting joints, and almost any volume change in sediments, and it is not always clear cut how to interpret different features (e.g., Butrym et al., 1964; Everett, 2006; Van Vliet-Lanoë et al., 2004; Van Vliet-Lanoë, 2005; Dixon, 2009; Van Loon, 2009; Superson et al., 2010; Robinson et al., 2017; Molén, 2023b). Such processes have on occasions been misinterpreted as permafrost features (e.g., Butrym et al., 1964; Eyles and Clark, 1985). As permafrost undergoes degradation, significant changes occur that impact the formation and appearance of these fissures, necessitating a nuanced understanding of how to differentiate between those found in non-frozen sediment and those present in perennially-frozen sediment (e.g., French, 2018). But, the papers referred to by Soreghan et al. (2023) do not document such detailed data, and therefore the data presented here is also more general.

True ice wedges are commonly 1) V-shaped vertically 2) arranged in polygonal patterns, 3) filled from above (and display vertical lamination), 4) associated with deformations in the flanking sediments, and 5) display vertically standing stones, if stones are present (Butrym et al., 1964). Soreghan et al. (2023) acknowledge that their interpretation of their documented polygonal network of sand wedges is controversial, but they defend their interpretation as of probably periglacial origin, referring to studies by Sweet and Soreghan (2008). Except for the above five criteria, which were only in part documented, other features were documented by Soreghan (references as of above) which are at odds with a periglacial interpretation of the sand wedges.

The features (Soreghan et al., 2022, Soreghan et al., 2023; Sweet and Soreghan, 2008) refer to are incipient (Figure 1), in general outside of the range of freeze and thaw polygons, and present in superimposed horizons. In Quaternary periglacial areas, polygon diameters may be between 1 and 59 m, wedge depth 0.25–80 m, and wedge width 0.1–10 m, or more (Eyles and Clark, 1985; Murton, 2013; Bertran, 2022), while those documented by Sweet and Soreghan (2008) were 15–78 cm in diameter, 13–61 cm in depth, and 3–22 cm in width. But polygons and ice- and sand wedges do of course always start as small and incipient. A longer or more severe climatic deterioration discussed but not suggested in this area by Soreghan et al. (2022), would induce structures more similar to those close to Pleistocene alpine glaciers or inland ice caps. Furthermore, smaller wedge polygons often occur in finer material (Murton, 2013; Andrieux et al., 2016; Bertran, 2022), and not in coarse sandstone and granule conglomerate as is the case for the structures documented by Sweet and Soreghan (2008). Areas of smaller polygons, more similar in size to those documented by Sweet and Soreghan (2008), may be a subdivision within larger polygons (Bertran et al., 2014; Andrieux et al., 2016; Bertran, 2022). However, the structures documented by Sweet and Soreghan (2008) are not a subdivision of larger polygons.

Except for the size, Quaternary examples are still different in appearance compared to those documented by Sweet and Soreghan (2008), i.e., the latter do not display clear deformations in flanking sediments nor vertical lamination - and only in one locality the sediments are interpreted to be eolian which may not preserve evidence of vertical lamination. Laminae in the host sediment (where present) “commonly are truncated and rarely bent upward at wedge walls” (p. 198, Sweet and Soreghan, 2008), and there is only one presented example of upward bent wedge walls. Quaternary polygons and wedges commonly display both vertical and horizontal lamination, embedded strata of organic material, and commonly upturned and downturned laminae in the wedge walls (Murton, 2013; Andrieux et al., 2016; Bertran et al., 2018; Wolfe et al., 2018; Bertran, 2022). Undeformed strata, which are the most prevalent appearance in the areas studied by Sweet and Soreghan (2008), would be rare and difficult to explain if present in permafrost areas (Murton, 2013), but Sweet and Soreghan (2008), p. 201) refer to special conditions “… that the hosting stratum was massive and/or lacked high water content” in their complete research area. Also, at least some of their documented “frozen-ground” fractures are tapering off upwards (Sweet and Soreghan, 2008; Soreghan et al., 2014), similar to structures produced by soft sediment tectonics (Butrym et al., 1964). There is no undulation or slumping in the sediments next to the wedges, which is a common feature of frozen ground (Sweet and Soreghan, 2008).

To summarize, features of wedges and polygons mentioned by Soreghan et al. (2022), are:

* Thin, not widening much upwards (Sweet and Soreghan 2008, Figures 10A–B; Soreghan et al., 2022; Figure 5C).

The SEM work on surface microtextures by Soreghan et al. (2022), Soreghan et al. (2023) uses the correct definitions, but selectively use only parts of a classification scheme, and do not refer to the combination of surface microtextures necessary for a correct identification. Apparently single surface microtextures, whatever size or number of ocurrences on single grains, appear to have been recorded as a basis for their interpretations, not using overall descriptions of the surfaces and statistics (Mahaney, 2002; Molén, 2014). Single occurrences of small scale surface microtextures commonly have little value and may be misidentified. The problems with this were clearly outlined by Molén (2014) referring to operator variance that sometimes differed between 0% and 100% detection on the same samples, even though the researchers still had the same interpretations of the depositional environment. (Soreghan et al., 2022) and other papers by the same group, e.g., Sweet and Soreghan, 2008) commonly mark surface microtextures on only a few grains, but the entire grain surfaces show, e.g., solitary fractures or those typical of weathered grains released from bedrock and no abrasion, i.e., these grains had to have been missed by any nearby glaciers because they acquired no irregular abrasion. They showed typical non-glaciogenic grains but proposed (probable) nearby glaciers (e.g., Sweet and Soreghan, 2010; Keiser et al., 2015). They provided no evidence of regular (fluvial) or irregular (glacial) abrasion (e.g., different kinds of “edge rounding”) and therefore misidentified grains that commonly originate from release from bedrock or simple fracturing without providing evidence of glaciation or (glacio-) fluvial abrasion (e.g., Sweet and Brannan, 2016; Soreghan et al., 2014; Soreghan et al., 2022; Smith et al., 2018). A typical glaciogenic quartz sand grain is shown in Figure 2.

The assertion by Soreghan et al. (2023), that for the method by Molén (2014) to be relevant it “requires observation of the entire grain surface” goes against their own observations where they actually have pictured “entire” grain surfaces (i.e., what is shown by SEM), and also against the work of most published SEM work on surface microtextures (e.g., Mahaney, 2002; Molén, 2014; Molén, 2017; Molén, 2023b; Molén and Smit, 2022). Their assertion, shown from their microphotographs, that apparent ocurrences of only single small scale surface microtextures have to be documented, and commonly minute evidence of abrasion, goes against the work by Mahaney (2002), Molén (2014) and others (e.g., Ma et al., 2024). Soreghan et al. (2023) also missed the Molén (2014) reference in their reference section.