Provenance and depositional conditions of Cretaceous–Paleogene boundary sandstones from northeastern Mexico

We compare Late Maastrichtian siliciclastic sandstone in northeastern Mexico with those representing the Cretaceous–Paleogene (K–Pg) boundary to reveal differences in provenance and depositional conditions between background sedimentation and K–Pg sand. Lithology and compositional variations are presented for the deep-water Burgos Basin and the shallow-water La Popa Basin.The Late Maastrichtian sandstones in the Burgos Basin have sharp lower contacts, contain abundant trace fossils and are separated by meter-thick marl layers. They represent sporadic mass flows from coastal areas separated by long phases of hemipelagic sedimentation. The K–Pg sandstone layers are amalgamated, contain thin marl intercalations only in the uppermost part and trace fossils are present in the top sandstone layer only. Also this succession represents mass-flow deposits, but the sand may have been deposited during a very short period. The La Popa Basin sandstones represent deltaic sedimentation interrupted by submarine channel deposition during the K–Pg boundary transition with abundant rip-up clasts and bioclasts at the base.The sandstones of the Burgos Basin are quartz to akosic wacke dominated by quartz (> 90%) and some feldspar (< 10%) in calcite cement and matrix. Lithic fragments are rare and dominated by chert and bioclasts. Ultra-stable heavy minerals (ZTR = 50–80) and plutonic quartz grains (ca. 40% of the total quartz population) are particularly common in the K–Pg sandstones. In the Maastrichtian sandstones, metamorphic heavy minerals, particularly chlorite, clinozoisite, and tourmaline (20–50% of the heavy mineral population), and metamorphic quartz (80% of the quartz population) have higher abundances. The La Popa sandstones are subarkose to arkose and arkosic wacke and have a high abundance of feldspar (15–30%) and lithic fragments (5–20%), mainly of siltstone and carbonate. The sandstones from both basins have chemical compositions typical for recycling (Zr/Sc = 12–27 and 17–140 in the Burgos and La Popa Basin, respectively) and influences from mafic source rocks (Th/Sc = 0.4–1.1; Ti/Nb = 350–510). Therefore we suggest that all studied successions share a common provenance with transport of recycled orogenic metasedimentary components from northwestern Mexico and magmatic arc material from western Mexico. Subsequently, longshore currents mixed the detritus with limestone clasts derived from the tectonically active Sierra Madre Oriental, which probably is the cause for compositional changes in the sandstones. Due to increased sediment input from western Mexico at the K–Pg boundary, provenance changes cannot be related to the Chicxulub impact.

► Sandstones of Maastrichtian and Cretaceous–Paleogene transition in Mexican foreland. ► Sedimentology in accordance with both short-term and long-term deposition. ► K–Pg boundary sand is more recycled than in the older units. ► Provenance change can be attributed to tectonic activities in the Mexican inland.