Mechanisms and timing of haloturbation in the northern Atacama Desert derived from a subsurface network of calcium sulphate wedges 

The presence of subsurface wedges and polygonal patterned grounds on the Earth’s surface is usually associated with cycles of cryogenic subsurface processes in periglacial environments. However, similar though calcium sulphate-dominated structures are found at numerous sites in the central Atacama Desert (N Chile), including particularly well-developed wedges in the subsurface of the Aroma alluvial fan in the Central Depression. Here, the subsurface wedges are covered by a ~20 cm thick, gypsum‑dominated surface crust, impeding the detection of the polygonal structures on the present-day Aroma fan surface. Due to high salt contents in the local alluvial fan deposits, the wedges are thought to be preliminary formed by haloturbation and may represent a hyperarid equivalent to periglacial wedge structures. The dominance of calcium sulphate phases in the vertical lamination of the wedges, accompanied by clastic minerals, is revealed by X-ray diffraction analysis. Hence, haloturbation is likely to be the key driver of wedge formation, caused by significant volumetric changes in the deposits and soil cracking induced by swelling and shrinking during calcium sulphate phase transitions. Geochronological information on subsurface wedge growth under conditions of extreme water scarcity is crucial for using these laminated wedges as an additional terrestrial palaeoclimate archive for arid to hyperarid environments in the northern Atacama Desert. Information on the processes and timing of wedge-polygon formation may also be important for interpreting wedge-polygon formation in other water-limited environments such as on Mars. Therefore, in order to unravel the mechanisms and governing environmental conditions of calcium sulphate wedge and crust formation at the Aroma site, we here present mineralogical, geochemical, and sedimentological data of wedge and crust material. In addition, our chronological investigations aimed at constraining the age of wedge growth activity by using a combination of feldspar luminescence dating and meteoric 10Be dating techniques as well as 239Pu concentration measurements. Based on a minimum age model of our luminescence dating results, wedge growth was last active at the Pleistocene-Holocene boundary. The presence of the overlying gypsum-dominated surface crust could reflect an environmental change from slightly marginally ‘wetter’ conditions to present-day hyperaridity, which ultimately inhibited wedge-polygon formation during the Holocene. However, 239Pu concentrations measured in surface crust samples indicate recent downward migration of soil fines through the crust body. Therefore, it remains an open question whether surface sediments and/or moisture can penetrate the surface crust to promote processes of wedge-polygon formation even under present hyperarid conditions, leading to wedge growth over longer time scales.