Problem:

Evidence shows that atmospheric water vapor has been increasing globally, both over oceans and land surfaces. Long-term datasets in the Upper Midwest indicate higher amounts of water vapor in the surface layer, which is likely to have significant consequences. Understanding the recent trends in increased atmospheric water vapor is extremely complex and there is a fair amount of uncertainty related to the climate feedback processes. Generally, increased atmospheric water vapor is expected to be driven by surface warming, but changes in land use, atmospheric circulation, and human water use may also be contributing factors. In addition to being a major greenhouse gas, water vapor also has a large influence on atmospheric processes including stability, convection, cloud formation, and the development of storm systems.

The isotopic composition of atmospheric water vapor is a forensic tracer that can be used to better understand the physical and biophysical processes involved in land-atmosphere transport and recycling of water. Although there is considerable research of water isotopes in the condensed phase, little research exists on the isotopic composition of water in the vapor phase.

Goal:

Our goal is to provide new information on how the land surface and atmospheric dynamics influence the isotopic composition of water vapor within a region. By identifying the dominant forcing mechanisms and their influences on the isotopic composition of the atmosphere, we can better utilize water isotopes as tracers of global and regional climate change.

Background:

The distillation and fractionation processes of water isotopes are well known. The difference in mass between isotopes in water molecules controls fractionation. As water moves through the atmosphere and biosphere and changes phase between liquid and gas, the ratio of heavy to light isotopes change. Heavier isotopes (²H, ¹⁸O) preferentially exist in the liquid or solid phases, and lighter isotopes (¹H, 16O) preferentially exist in the gaseous phase. An evaporating body of water will preferentially evaporate the lighter isotopes because less energy is needed to do so. When water vapor condenses, the heavier isotopes will condense first because of their heavier mass. This basic principle of partitioning between phases is the basis for water isotope studies.

The classic Rayleigh Distillation Model (RDM) describes how the isotopic composition of an air mass changes as it moves from its source of moisture: As an air mass moves, it will become relatively more depleted as the heavier isotopes condense and precipitate first. There are predictable patterns of rainfall that agree with the RDM. Observed patterns of precipitation show relative depletion of air masses of the heavier isotopes at higher latitudes, at higher altitudes, and at distances farther from coasts. Air masses from different moisture sources show distinct isotopic signatures. Although this model is proven to work over relatively long time scales, it does not adequately take into account small scale processes occurring over short periods of time. A number of variables the RDM does not explicitly account for are evapotranspiration effects, boundary layer entrainment, raindrop evaporation, and others.

Craig (1961) showed that the isotopic compositions of precipitation collected from all over the world are linearly related by the Global Meteoric Water Line (GMWL):

δ²H= 8 × δ¹⁸O + 10
The slope of 8 is related to the fractionation of water as it condenses, and the y-intercept of 10 is the result of a higher enrichment of D during evaporation. The GMWL provides a useful reference point to compare local and regional precipitation against. The differences in a local meteoric water line compared with the GMWL can be interpreted based on a number of variables including land surface cover, evapotranspiration, and surface temperature. The y-intercept of the GMWL equation is known as the deuterium excess (d) parameter. The d-excess value is useful because it reflects the conditions under which evaporation of the precipitating air mass occurred. The value of d-excess increases as the relative humidity under which evaporation occurred decreases. Because the air moisture deficit is positively correlated with temperature, d-excess is positively correlated with temperature. Precipitation from mixed air masses (both advected and locally evapotranspired moisture) have higher d-excess values. The d-excess value can be used to identify the mixing of evaporated air into the atmosphere.

In combination with precipitation patterns, groundwater, surface water, plant water, and soil water are useful in this type of study because their fractionation processes are relatively well known and we should be able to trace the history and path of water movement. However, water vapor isotopes are not routinely measured.

Sampling & Measurement Techniques:

We are archiving precipitation, groundwater, plant (leaf and stem) water, soil water, and surface water. Through isotopic analysis of water at these stages as it moves through the biosphere and atmosphere, we will be able to develop an understanding on how local land-atmosphere processes influence the fractionation of water isotopes.

A TDL system (TGA200A, Campbell Scientific, Inc.) is used to measure isotopic water vapor mixing ratios and fluxes in a flux-gradient and eddy covariance mode.

Leaf, stem, and soil water are extracted using a custom vacuum line. This system is designed to fully extract water from our plant and soil samples to prevent fractionation.

Samples are loaded into the vacuum line, frozen with liquid nitrogen and then opened to the vacuum and pumped down to a pressure of ~10 millitorr (mTorr). They are then heated using a water bath, and the liquid nitrogen is used to cool a collection tube. This temperature gradient drives the water from the heated plant and soil samples to the frozen collection tubes. The samples are heated for at least 1.5 hours to ensure the plant and soil samples are completely dehydrated and all of the water is collected. The water samples are then transferred to 1.5 mL vials, wrapped in parafilm, and refrigerated.

The extracted water from plant and soil samples as well as precipitation, surface water, and groundwater samples are analyzed for isotopic composition using an off-axis laser spectroscopy system (DLT-100, Los Gatos Research). Typical precision for this system is 0.2 and 1.0 per mil for delta O18 and delta D, respectively.

Our archived precipitation is from 2006-2008 collected from our research site in Rosemount, MN. Groundwater samples have been obtained both from the Rosemount research site and from domestic wells within a certain vicinity of the research site. Surface water samples have been collected from lakes and rivers within a radius of Rosemount and the St. Paul UMN campus. Soil samples have been collected from the Rosemount site from 2006-2008. These samples have been taken ~20 cm from the surface. These data have been shared with the MIBA (Moisture Isotopes in the Biosphere and Atmosphere) program.

Our research on water vapor isotopes is a collaborative project with Dr. Xuhui Lee at Yale University and Steve Sargent at Campbell Scientific Inc. For further information please contact Dr. Tim Griffis.