Abstract

Widespread silicate melts occur in several areas (5 km x 3 km) of the Atacama Desert near the Pica oasis, northern Chile. Although first interpreted as the result of meteoric fireballs (Blanco and Tomlinson, 2013; Perroud et al., 2016), a later study concluded that they resulted from intense grass fires (Roperch et al., 2017). Here we provide geologic evidence that these glasses must have been generated by a series of super-Tunguska-like fireballs reaching the surface during the late Pleistocene or early Holocene. At Núñez and Chipana localities, clusters of glasses occur in patches (up to 10 m x 20 m) are scattered over a range of lithologic facies (paleo-wetland and alluvial over-bank deposits of Pleistocene-Holocene age), with occasional distal clusters. In some cases, glasses trapped matted paleo-grasses below. Based on locations and depth below the glasses, however, most of these grasses had been diagenetically altered before being trapped in the glass. Individual glasses can exceed 30 cm across and 10 cm thick but are surrounded by numerous other fragments, occasionally oriented in a common direction. Many examples exhibit twisting, shearing, rolling, and folding (in some cases more than twice) before being fully quenched resulting in mixed components, including un-melted pockets or seams of clay. Several samples exhibit clusters of “fingers” indicative of injection at an angle into underlying wet sediments after partial quenching. Such morphologies indicate dynamic emplacement after initial melting. Coupled with microanalysis (see Harris et al, this volume), we concur with the original interpretation that Pica glasses formed from several near-surface meteoric fireballs. They are unlike, however, previously reported Pliocene/Miocene glasses in Argentina that contain entrained compositions and/or clasts from depth (i.e., solid-body contact). Theoretical models (e.g., Boslough and Crawford, 2008) reveal that large fireballs in the lower atmosphere retain significant momentum, similar to the process proposed for radar “splotches” on Venus (Schultz, 1992). Radiative and convective heating (along with the partly vaporized projectile and plasma) preferentially fuses fine-grained materials followed by intense winds, thereby accounting for dynamic emplacement and limited ballistic transport.