(U-Th)/He dating and chemistry of diagenetic Fe- and Mn-oxides in Mesozoic sandstones of the Colorado Plateau

Diagenetic mobilization and redeposition of solutes by groundwater in sedimentary rocks commonly concentrates Fe- and Mn-oxides into concretions and other types of local cement-mineral masses. Remarkable examples come from Mesozoic sandstones of the Colorado Plateau, where ancient groundwater flow produced locally abundant pipes, cones, sheets, and spheroidal concretions defined by high concentrations of hematite, goethite, and Mn-oxides. These diagenetic oxides provide striking analogies to similar features on Mars that have also been interpreted as indicators of ancient groundwater flow. Despite their potential for understanding fluid migration, diagenesis, and Martian hydrogeology, the origin and significance of these concretionary features are still poorly understood. A fundamental limitation to our understanding is a huge lack of temporal constraints on their age of formation and possible subsequent modification. To better understand the history and compositions of these features we measured (U-Th)/He ages and trace element compositions of 75 diagenetic Fe- and Mn-oxide samples from a dozen locations in northern Arizona and southern Utah, and performed step-heating He diffusion experiments on Fe- and Mn-oxide samples. We find (U-Th)/He ages ranging from 0.2 to 25 Ma, with the majority of ages between 0.3 and 5 Ma. One sample of Mn-oxide from near Moab, Utah yields a reproducible (U-Th)/He age of 2.36 × 0.09 Ma (1σ, n=5). Small hematite-dominated concretions from SW Utah and another hematite cement from SE Utah are consistently the oldest, reaching ages of 25 Ma, near the ⁴⁰Ar/³⁹Ar step-heating plateau observed in Mn oxides by Chan et al. (2001). Other samples, including goethite-dominated Moqui marbles from the Paria Plateau, and Fe- and mixed Fe-Mn-oxides from various areas are mostly Plio-Pleistocene. The range of ages and compositions is not easily explained by a wide range of formation ages, nor by diffusive loss of He from unretentive domains. Instead, two distinct types of age-composition relationships are most consistent with post-formation uptake of U and Th to variable degrees by these mineral cements. The maximum observed ages provide a minimum age of formation (~25 Ma), whereas the minimum observed ages require that most of the U and Th was added to some of these samples at 2.4 Ma and to others as recently as ~100-200 ka. We suggest that these diagenetic oxides remain open to parent nuclide uptake long after formation, and probably to depths of a few hundred meters or less, possibly due to changes in groundwater compositions accompanying exhumation or changing surface conditions. Although this complicates simple geochronologic interpretations, it shows the potential for tracing compositional changes of diagenetic materials with time. Although cement phases are notoriously difficult to constrain except by relative age relationships, this approach may constrain the timing of the last episode of significant groundwater interaction.