96 SEDIMENTARY ROCKS/Evaporites hydrothermal brines, which nevertheless carry large amounts of dissolved materials Hardie calculated that modern seawater could be modified by the addition of only just over 3% Salton Sea brine into water that would not precipitate evaporites containing magnesium sulphate Evaporites in Non-Rift Basins Hardie’s hydrothermal brine explanation is not convincing for some evaporite sequences because they occur in non-rift basins Kendall provided an alternative explanation for some of these evaporites Desiccation of large evaporite basins produces large and deep depressions This induces a hydrodynamic drive, which causes subsurface waters to migrate into the basins, where they evaporate Dolomitization of limestones by migrating formation waters with seawater-like compositions would release calcium and generate calcium chloride waters Basin desiccation thus provides both the drive that allows formation waters to enter the basin and a mechanism to generate waters with more calcium than sulphate Middle Devonian evaporites in western Canada provide evidence to support this desiccation–drive model Where calcium-rich formation waters entered the Devonian evaporite basin, spring-associated carbonates were precipitated, and there was mixing with seawaterderived brines that had already precipitated gypsum The addition of spring-water calcium stripped the remaining sulphate from the basin brine, precipitated anomalous concentrations of calcium sulphate far into the evaporite basin, and led to the formation of a sulphate-depleted brine, which may have caused later potash salts deficient in magnesium sulphate to form in the basin Halite-saturated brines, refluxing beneath large evaporite basins, react with all types of sediments (not just limestones) by exchanging sodium for calcium, to generate calcium chloride brines The main problem in understanding how these deep dense brines could form potash salts is to explain how the brines are transported to the surface This could occur at times of basin inversion, by heating (creating buoyant hydrothermal brines) or by evaporative draw into a later evaporite basin Past Composition of Seawater A more exciting alternative that explains the seemingly anomalous compositions of most ancient potash salts is that seawater compositions were substantially different in the past Secular variations in the distributions of magnesium sulphate and evaporites deficient in magnesium sulphate are in phase with better-known variations in the mineralogy of ancient marine carbonates These secular variations can be attributed to changes in the major-ion composition of seawater over time A model that explains how seawater can change over time was used to predict periods when aragonite and evaporites containing magnesium sulphate are dominant, and episodes when calcite and evaporites deficient in magnesium sulphate are favoured Seawater chemistry is controlled by steady-state mixing of river water and mid-ocean ridge hydrothermal brines (coupled with calcium carbonate and silica precipitation) Relatively small changes in mid-ocean ridge fluxes cause significant changes in magnesium : calcium, sodium : potassium, and chloride : sulphate ratios in seawater Such changes are believed to be responsible for variations in the primary mineralogies of marine evaporites and carbonates Variations in mid-ocean ridge flux correspond to variations in the production rate of oceanic crust (seafloor spreading rate), and this can be estimated using various proxies, such as areas of ocean floor of different ages, the global sea-level curve, and granite-pluton emplacement rates Predictions of the variation in past seawater chemistry produced by variation in mid-ocean-ridge flux rates are in agreement with the known age distribution of primary marine carbonate and evaporite mineralogies The coherence of the datasets strongly suggests that past variations in evaporite and carbonate mineralogy were largely caused by secular variations in seawater chemistry The idea that varying seawater chemistry can explain potash salt composition has been challenged Holland et al predict similar changes of seawater composition but of much smaller magnitude They argue that an apparent near constancy of the level of potassium in seawater during the Phanerozoic (demonstrated by the compositions of brines trapped in ancient halites) supports this view: Hardie’s model predicts significant changes in the sodium : potassium ratio Instead, Holland et al suggest that changes in past evaporite mineralogy are due to differences in the extent to which dolomitization of carbonate sediments occurred before or during seawater evaporation During times of rapid seafloor spreading, sea-levels are higher and large carbonate platforms are more abundant Changes in seawater chemistry (caused by increased mid-ocean ridge flux) coupled with increases in the extent of dolomitization of carbonate platforms are believed to be responsible for the formation of potash deposits that are impoverished in magnesium sulphate This explanation resembles, in part, that suggested earlier by Kendall