CONTROLS ON THE SPATIAL VARIABILITY OF MODERN METEORIC δ~(18)O: EMPIRICAL CONSTRAINTS FROM THE WESTERN U.S. AND EAST ASIA AND IMPLICATIONS FOR STABLE ISOTOPE STUDIES

摘要

We analyze a dataset of multiannual modern precipitation and meteoric water records from the western U.S. (n = 206) and east Asia (n = 478) to (1) determine which environmental parameters correlate best with and account for spatial variability of meteoric water isotopic (δ~(18)O and δD) compositions and (2) assess the degree to which this variability is a function of physiography and climatology. Multivariate linear regression analysis including five environmental parameters [latitude, longitude, elevation, mean annual temperature (MAT), mean annual precipitation (MAP)] reveals that latitude and elevation are consistently strongly correlated with meteoric δ~(18)O distributions throughout much of the western U.S. and east Asia, but correlations with site longitude, MAT, and MAP can also be significant, depending on region. Our analysis also indicates that isotope-environment relationships are regiondependent. Isotope-elevation gradients are reduced by a factor of two or more in continental interior rainshadows (for example, Basin and Range) and high elevation continental plateaus (for example, Tibetan Plateau) in comparison to isotopeelevation gradients observed in orographic settings with a single dominant moisture source and relatively simple storm track trajectories (for example, Sierra Nevada and Himalaya). Published global and continental predictive equations for the calculation of modern meteoric water isotopic compositions that do not account for this regional variability are shown to be characterized by significant predictive uncertainties for modern δ~(18)O values (～ ±3-4‰), particularly in regions of complex moisture source interactions, continental moisture recycling and increased convective storm activity where reduced isotope-elevation gradients are also observed. We present new empirical relationships between environmental parameters and isotopic compositions that inform stable isotope-based reconstructions of climate change, timing and location of groundwater recharge, and paleoaltimetry by providing quantitative constraints on the statistical relationship between precipitation δ~(18)O and individual environmental parameters (for example, latitude, elevation, temperature) and identifying how isotopic relationships are influenced by physiography and climatology. In particular, paleoelevation estimates in regions of low-magnitude isotope-elevation gradients (Basin and Range, Tibet) are shown to be prone to significant uncertainties (in some cases > ±2 km). These relationships improve the interpretation of stable isotope proxy data, particularly in cases where the paleogeographic setting of proxy formation can be constrained.