Global Chemical Erosion over the Last 250 My: Variations Due to Changes in Paleogeography, Paleoclimate, and Paleogeology

We utilize predictions of runoff from two series of GENESIS (version 1.02) climate model experiments to calculate chemical erosion rates for 12 time slices that span the Mesozoic and Cenozoic. A set of ‘‘control’’ experiments where geography is altered according to published paleogeographic reconstructions and atmospheric pCO2 is held fixed at the present-day value was designed to elucidate climate sensitivity to geography alone. A second series of experiments, where the (elevated) atmospheric CO2 level for each time slice was adapted from Berner (1991), was executed to determine the additional climate sensitivity to this parameter. By holding other climate forcing factors (for example, vegetation) constant throughout the sequence of experiments we evaluate the effects of systematic/coherent paleogeographic changes on runoff and temperature, and thus on global rates of chemical weathering. By using empirical relationships between runoff and bicarbonate fluxes for different rock types and maps of paleogeology we calculate global bicarbonate fluxes, taking into account spatial variations in lithology and hydrology. The climate model predicts that many regions experienced dramatic changes in runoff (for example, from wet to very arid) since the Early Triassic. In general, the supercontinent (Pangean) paleogeographic regime was driest, times of dispersed continents (Tethyan regime) wettest, and the modern geography intermediate. Total bicarbonate fluxes, as well as those from silicate mineral weathering only, closely parallel these trends. We find that spatial variations in lithology account for little variation in the total or silicate chemical erosion rates. In contrast, changes in hydrology due to differences in paleogeography are significant and are the main clear trend in our results. Here we also add an Arrhenius-type temperature dependency that modifies the runoff-determined fluxes for each rock type. High activation energies for the weathering reactions substantially increase absolute values, but we find that the ratio between the flux for a particular time slice and the Present Day remains very similar for different temperature dependencies, following the paleogeographic-controlled trend in runoff. Our calculations suggest a weaker-than-expected CO2-climate weathering feedback. The reasonable atmospheric pCO2 variations specified for the climatemodel simulations do not lead to climatic effects that support large changes in the chemical erosion rate, compared to those generated by changing paleogeography. In general, however, we find that silicate weathering rates are similar to outgassing rates of volcanic and metamorphic CO2. Times of supercontinental stasis represent low outgassing but also high aridity due to extreme continentality and thus low chemical erosion fluxes. In contrast, times of continental dispersion represent high outgassing as well as high runoff (and fluxes) due to increased proximity to moisture sources. Where a mismatch occurs, particularly in the case of the early to mid-Cretaceous, we infer higher CO2 levels than those used in our GCM simulations to balance the carbon cycle. * Center for Climatic Research, University of Wisconsin-Madison, Madison, Wisconsin 53706 ** Department of Geological Engineering and Sciences, Michigan Technological University, Houghton, Michigan 49931 *** Department of Earth and Planetary Sciences, University of New Mexico, Albuquerque, New Mexico 87131 **** Department of Geosciences and Earth System Science Center, Pennsylvania State University, University Park, Pennsylvania 16802 [AMERICAN JOURNAL OF SCIENCE, VOL. 299, SEPT./OCT./NOV., 1999, P. 611–651]