Science Briefs

Isotopically Speaking: Tracing the Water Cycle

Water is the most fundamental component of the Earth's climate. Its presence in all three phases — liquid, solid and gas — defines our planet in a very profound way. Tracking the movements of water through the system — in oceans, air, clouds, rain, snow, ice, lakes, rivers, and back to the oceans — is therefore a primary concern of climatologists. Where does the water come from? Where does it go? What feedbacks are involved?

To answer these questions in computer models is relatively easy — one can "paint" different water masses different colors and then see how, for instance, the "red" water spreads around the system. But what about in the real world? Can one "paint" all of the water coming from the tropical oceans and see how it moves about? Of course not. However, it turns out that nature has provided us with something almost as good as paint... isotopes.

Simple diagram of the three hydrogen isotopes showing their numbers of protons and neutrons
The three isotopes of hydrogen, showing their protons (marked in yellow, with positive charge), neutrons (green, with no charge), and electrons (red, with negative charge)

An isotope of any particular element (such as oxygen or hydrogen) is an atom of that element that has the same chemical behavior as all the other atoms, but has a different number of neutrons in the nucleus which makes it a little heavier or lighter than the average. For instance, oxygen normally has 16 nucleons (8 protons and 8 neutrons), and is written 16O. There are also stable oxygen atoms with 17 or 18 nucleons (one or two extra neutrons) written 17O and 18O. Similarly, hydrogen comes in three flavors 1H (usually written as simply H), 2H (D, or deuterium) and 3H (T, or tritium, which is radioactive). The most abundant isotope of water is H216O, but the isotopes H218O and HD16O are also relatively common and can be easily measured.

Because isotopes differ in mass, they have slightly different physical properties such as how quickly they diffuse or evaporate. This can cause fractionation, particularly when there are changes of phase. For water in the climate system, the most important fractionations occur when water evaporates from the ocean (the evaporating water is slightly lighter, or more depleted, than the ocean water it came from) and when it condenses in clouds (the condensate is a littler heavier, or more enriched, than the water vapor). This pattern leads to a very clear pattern of progressively more depleted isotopes as you move towards the poles from the equator. Thus, water from different places or processes can have a different isotope signal which can be used as a kind of "paint" in the sense described above. This happens in polar studies where the proportion of ice melt can be tracked throught the ocean, and in hydrology where seasonal variations in isotopes can be tracked through a watershed.

There are many key questions that are relevant in atmospheric science that can also be looked at with isotopes. However, in most cases there several complicated processes happening, and so models are required to elucidate what is going on. The isotopes then provide both a inside view to what is happening in the real world, and a tough test for the models.

Climate modellers at GISS were pioneers in incorporating isotope physics in general circulation models (GCMs) back in the 1980s. With the latest generation of models and the advances in modelling clouds, ice and oceans, we have now brought that work up to date, and applied it to new sets of questions (see Schmidt et al. 2005).

In particular, we explore two questions related to how water gets into the stratosphere (where it is particularly effective as a greenhouse gas) and how water vapor and clouds behave in the upper troposhere — a key part of the water vapor feedback to climate change.

Model simulation (top) for the isotopic ratio in precipitation compared to observations (bottom) compiled by the GNIP project. Ratios are measured as a deviation from a standard, with more negative values implying less heavy isotopes than normal, postive values indicating more. The general pattern is for more depletion as you go towards the pole or away from the coast. (View as large GIF or PDF).

Previous authors have suggested that the isotopic signature in the stratosphere is somewhat anomalous and could only be explained if a significant amount of ice from convective plumes was being transported directly into the lower stratosphere (as opposed to water vapor slowly being transported up). While we are not yet able to give a definitive explanation for the phenomena, the model suggests that the stratospheric values may not be as anomalous as previously thought. The key result in the model is that the upper troposphere is not as depleted as simple one-dimensional models suggest, mainly because of the mixing of vapor and condensate in the complicated cloud physics. Whether the model is getting that right is, as yet, unknown because the observations in this regions are not yet widespread enough to test the model.

It turns out that the isotopes in this region are actually quite sensitive to quite small changes in the parameterisation of cloud effects, and so as measurements improve, the isotopes may be able to provide a constraint on the model and thus help improve the climate simulation as well.

Given the importance of the water cycle, and the pressing need to improve model representations of clouds and other "wet" processes, we anticipate that the study of isotopes will become more ubiquitous both in modelling and in observations. Already, satellite measurements are being planned using the output of this model to assess the accuracy and coverage required for useful isotopic measurements.


Aleinov, I., and G.A. Schmidt 2006. Water isotopes in the GISS ModelE land surface scheme. Global Planet. Change 51, 108-120, doi:10.1016/j.gloplacha.2005.12.010.

Schmidt, G.A., G. Hoffmann, D.T. Shindell, and Y. Hu 2005. Modelling atmospheric stable water isotopes and the potential for constraining cloud processes and stratosphere-troposphere water exchange. J. Geophys. Res. 110, D21314, doi:10.1029/2005JD005790.


Please address all inquiries about this research to Dr. Gavin Schmidt.