CHEMOSTRATIGRAPHIES OF GRANDE RONDE BASALT (GRB) AND OTHER FORMATIONS OF THE COLUMBIA RIVER BASALT (CRB) GROUP

Over the past half-century, the chemostratigraphies of Grande Ronde Basalt (GRB) and other formations of the Columbia River Basalt (CRB) Group have been determined without consideration of the effects of low-T chemical water-rock interaction, or chemical alteration, on lava compositions. Alteration has occurred in two distinct environments: oxic near-surface settings and anoxic groundwaters, the latter of which has received little consideration. Chemical alteration differs significantly between these environments in primary mineral and glass reactivity and redox-sensitive-element solubility (e.g., Fe). In both environments, the principal effect of alteration is removal of soluble cations by dissolution, resulting in decreased rock mass, increased porosity and decreased density. Mass changes are determined from a parent:sample immobile element ratio in which parent compositions are referenced to Pl + Cpx fractionation trends. Changes to lava chemistry from alteration often exceed changes from magmatic processes among petrogenetically related magmas. Evaluation of anoxic-altered GRB Sentinel Bluffs Member and oxic-altered Wanapum Basalt basalt of Dodge lavas shows that soluble cations in reactive phases decrease linearly with decreasing sample mass, indicating steady state dissolution. Mineral reactivity differs markedly, however, between anoxic and oxic environments. Plagioclase and glass are unreactive or weakly reactive in anoxic environments but clearly are reactive in oxic environments. Such differences in mineral reactivity, as well as in Fe solubility, result in marked differences in the composition of material removed by dissolution, such that altered compositions are not proportional to primary compositions. Thus, volatile-free normalization does not yield eruptive compositions but, instead, yields compositions proportional to altered compositions. Immobile element abundances are also modified by alteration but are readily recovered as a function of sample mass change (Sawlan, 2018, https://doi.org/10.1130/GES01188.1). The prior inattention to chemical alteration of CRB lavas has led to ambiguous chemostratigraphies, proposal of improbable emplacement processes, and has clouded identification of petrogenetic processes in CRB magma generation and evolution.