Recently, my paper, “Continental scale variation in 17O-excess of metoric waters in the United States,” was published in Geochimica et Cosmochimica Acta, in the same issue (Sep. 1) as another Δ17O paper by Di Rocco and Pack.
Our paper reports the triple-oxygen isotope data from tap waters in the U.S. and provides the first look at the variation in 17O-excess (or Δ17O), the deviation from an expected relationship in 17O/16O and 18O/16O ratios, of meteoric waters on a continental-scale. Our results show that there is a significant amount of variation in 17O-excess of waters from low and mid-latitudes, which had not been recognized before. I hope that our work can help contribute to the growing number of studies that use mass-dependent triple oxygen isotope variation in waters (including ice) and rocks to explore hydrological variations today and in the past.
I am usually asked what additional information 17O-excess (or Δ17O) can give.
Traditionally, the 18O/16O ratios of waters are widely used as measures of paleoclimate because they are sensitive to environmental parameters and because they are preserved in various geologic materials, including ice cores, tree rings, minerals and rocks. Nevertheless, there is much information that is not uniquely recorded by the 18O/16O ratios. For example, increases in 18O/16O ratios in surface water (measured directly or inferred from 18O/16O ratios of a mineral that formed from that water) could be due to the increased influence of evaporation, a decrease in elevation, or a change in moisture source for precipitation (as outlined in the figure below). Therefore, we need some additional proxies to tease apart the effects of equilibrium processes (e.g., condensation) and kinetic processes (e.g., evaporation) on the isotopic composition of water.
Scenarios of interpreting difference in 18O/16O ratios (δ18O) in waters A and B
Combining 2H/1H and 18O/16O ratios (i.e., δD and δ18O, respectively) into the parameter “deuterium excess” can provide additional information on kinetic fractionation processes during water evaporation. However, the utility of deuterium excess is limited in geologic studies because it is difficult to conduct measurements of δ18O and δD on a single geologic material; most minerals do not contain hydrogen as a structural component. The third stable isotope of oxygen, 17O, provides an opportunity to overcome this difficulty of characterizing the effects of kinetic fractionation using just oxygen isotopes alone, from a single geologic material.
In the past, variations in 17O/16O and 18O/16O ratios have been observed in meteorites and in ozone, and they are used as indicators of mass independent fractionations such that the enrichment of 17O is equal to or even larger than the enrichment of 18O. For Earth surface processes, the isotopic ratio 17O/16O was assumed to carry no additional information to 18O/16O, because processes associated with phase changes or chemical reactions usually fractionate the oxygen isotopes in a mass-dependent way, such that the enrichment in 17O is approximately 0.52 times of that in 18O. The mass-dependent assumption is justified when the analytical precision of δ17O or δ18O is no better than ±0.1‰. However, recent precise analyses of 17O/16O ratios in meteoric waters have shown that there are small but measurable variations in the δ17O - δ18O relationship in meteoric waters and seawater, leaf waters, and terrestrial rocks and minerals attributable to variations in mass dependent fractionation which are governed by kinetic effects such as diffusion processes during water evaporation (see the schematic figure below).
Schematics of how Δ17O could be used to tease apart the effect of condensation vs. evaporation on the isotopic composition of water.
We wish to help fill a gap with this paper.
In order to deduce meaningful environmental information from 17O-excess variations in ancient geologic materials, we must first understand how the distribution of 17O-excess in modern meteoric waters relates to modern climate.
The study of triple oxygen isotope variations in the atmosphere and ice cores from high latitudes has progressed significantly in the last ten years, although under the radar of the communities studying continental-scale hydrological cycles particularly in mid-latitudes. Not many datasets of 17O-excess from meteoric waters in mid-latitudes had been reported, and there were still gaps in our knowledge about the potential mechanisms driving the triple-oxygen isotope variation on a continental scale. These gaps must be filled before we can extend triple oxygen isotope applications into paleoclimate records.
A predicted map of 17O-excess in U. S. waters
In the paper, we presented a dataset on the spatial distribution of 17 O-excess of tap water from the continental U.S. (see the map below). In general, 17O-excess values of tap water increase with increased latitudes. Re-evaporation of precipitation and mixing of moisture sources have significant influences on the 17O-excess values. The lack of a latitudinal relationship in the central region of the U.S. is likely due to the multiple moisture sources that contribute to tap waters in this region.
Map of interpolated 17O-excess values of the continental U.S.
The average 17O-excess value of tap waters from the U.S. is lower than the reported global average value in meteoric waters, but similar to the average 17O-excess value of existing data from precipitation at low and mid-latitudes. When we put all the published 17O-excess data of meteoric waters in the literature together (see figure below), we found that there are signals to work with, which can be used to help test global circulation models that incorporate 17O-excess.
Compilation of published 17O-excess vs. δ18O data of meteoric waters. The cited data are from Luz and Barkan (2010) except as noted in the legend.
The interpolated map of 17O-excess variation among U. S. tap waters in our paper can also be used as an estimate of 17O-excess values of precipitation in the U.S., which has the potential to provide a framework for understanding the range of 17O-excess values that should be expected over large geographic regions.
Many thanks to all my co-authors and colleagues.
It was a very great and impressive experience working with Naomi Levin and Lesley Chesson on this project! I also would like to express my gratitude to all those who helped develop this study and improve the manuscript, including Boaz Luz, Andy Schauer, Ben Passey, Amaelle Landais, Saleh Sati, Ben Zatchik, Brad Erkkila, Jim Ehleringer, Thure Cerling, Laurence Yeung, one anonymous reviewer, and countless others who help collect the water samples!
P.S. Our paper got published on May 25, which is the same day when Ben Passey got great news in his academic career. Let’s just keep the memory of the day on a mug.
