Paleo ice-cap surfaces and extents

The distribution, equilibrium-line altitude (ELA) and timing of Pleistocene alpine glaciers are used to constrain paleoclimatic reconstructions. Attention has largely focused on the geomorphic evidence for the former presence of simple valley glaciers; paleo alpine ice caps and their outlet glaciers have proven to be more problematical. This is especially so in the remote continental interior of Asia, where the research invested in the Alps or Rocky Mountains has yet to be duplicated. Even the putative existence and size of paleo ice caps in Tibet and the Kyrgyz Tien Shan is controversial. Remote sensing offers the opportunity to assess vast tracts of land quickly, with images and co-registered digital elevation models (DEMs) offering the most information for studies of paleoglaciers. We pose several questions: (1) With what confidence can nunataks be identified remotely? (2) What insights do their physiographic characteristics offer? (3) What characteristics of the bed of a paleo ice cap can be used to identify its former presence remotely? and (4) Can the geomorphic signatures of the edges of paleo ice caps be recognized and mapped? Reconstruction of the top surface of a paleo ice cap depends on the recognition of nunataks, typically rougher at 1 m to 100 m scales than their surroundings. Nunataks in southern Siberia are commonly notched by multiple sub- horizontal bedrock terraces. These step terraces appear to originate from freeze-thaw action on the rock-ice interface during periods of stability, and presence of multiple terraces suggests stepwise lowering of ice surfaces during deglaciation. An older generation of step-terraced nunataks, distinguished by degraded and eroded terraces, delineates a larger paleo ice cap in the Sayan Range (Siberian - Mongolian border) that significantly pre-dates the last glacial maximum (LGM). Large ice caps can experience pressure melting at their base and can manifest ice streams within the ice cap. Valleys left behind differ from fluvial valleys in their width/depth profiles: the channels maintain width but get shallower near their sources. Link junction angle distributions within superimposed drainage networks are broader and distinct from those of evolved fluvial networks, and their character and statistics can be used to identify the perimeters of large paleo ice caps. (This work was carried out in part at the Jet Propulsion Laboratory of the California Institute of Technology under contract to NASA.)