He–Pb double dating of detrital zircons from the Ganges and Indus Rivers: Implication for quantifying sediment recycling and provenance studies

He–Pb double dating of detrital zircons is more reliable than conventional U–Pb dating for tracing the source of detritus in sediments and can be used to constrain the percentage of recycled material in sediments. Conventional U–Pb dating can be used to constrain the provenance of sediments if the U–Pb zircon age pattern for potential source regions is known but can only be used to trace the source of individual zircons if they are first-cycle grains. The advantages of He–Pb double dating are demonstrated using examples from the Indus and Ganges rivers, and previously published data from the Navajo sandstone. Conventional U–Pb dating can unambiguously identify only 2.5% of the Ganges zircons, and 18% of the Indus zircons as coming from the Himalayan Mountains or Tibet Plateau and only 23% of the Navajo zircons as coming from the Appalachian Mountains. The correct figure, as determined from double dating, is over 95% from the Himalayan Mountains or Tibet Plateau in the case of the Indus and Ganges rivers and at least 70% from the Appalachian Mountains in the case of the Navajo Sandstone. This result casts doubt on the reliability of the U–Pb method when used in the absence of other techniques, such as He dating, to identify the true provenance of sediments, as opposed to the ultimate source of the zircons. Double dating also shows that at least 60% of the Indus and 70% of the Ganges and Navajo sandstone zircons have been recycled from earlier sediments. Exhumation rates, estimated from the He dates, reveal that ~75% of the Indus and Ganges zircons were derived from areas where the exhumation rate exceeds 1.5 km/Myr. These rates are higher and more varied than those calculated from detrital muscovites. These results imply that ~75% of the eroded material in the Himalayan Mountains is derived from areas of anomalously high erosion where the short-term exhumation rate exceeds the long-term average. D 2005 Elsevier B.V. All rights reserved.