An Isotopic Study of Lake Effect Snow

I’ve written a few blog posts about what can be done with isotopes from precipitation, and how that might assist us in understanding how to interpret isotopic data collected from ancient rocks and fossils. (Look here and here.) As I live here in western New York state, close to Lake Ontario, I frequently have opportunities to further study how the isotopes from precipitation (in this case Lake Effect snow) are related to the isotopes of the water that originally evaporated to make the clouds that do all the snowing.

Right now, we’re looking at a Lake Effect snow event that’s due to start sometime tomorrow, so I’m throwing together is quick and fun isotopic study that I’ll share with you when the data come in. I’ll describe it here.

As review, let’s think about isotopes in water. First, what do I mean by isotopes? The term worries people, because they immediately think of radioactive isotopes and OMG, we’re gonna die! No, it’s not like that. The word isotope just refers to the fact that some atoms of the same element are heavier or lighter than the others.

Water is composed of hydrogen and oxygen (H2O). Hydrogen comes in two types. Most of it has a mass (think of it as weight) of 1. Some of it has a mass of 2. (The hydrogen with a mass of 2 is called deuterium. It’s one of the few isotopes that has its own name.) So water is mostly made with hydrogen atoms of mass 1, but some water has hydrogen of mass 2. The water with the mass 2 hydrogen is heavier than the water with the mass 1 hydrogen.

Similarly, oxygen comes in two important isotopes. The most common form of oxygen has a mass of 16. A more rare (but not radioactive) form of oxygen has a mass of 18. Either type of oxygen can be in a water molecule, but the water with the mass 18 oxygen is heavier.

With mass spectrometry, we can measure water to see how much of it has the heavier hydrogen and the heavier oxygen. This is what I do for a living.

To get any kind of precipitation (rain or snow), water must first evaporate to make a vapor mass in the atmosphere. You can think of this as just making a cloud or a storm. In the case of Lake Effect precipitation, the water that’s evaporating is the lake itself. When the water evaporates, the lighter water evaporates more than the heavier water because, well, it’s lighter. So the cloud that you get from evaporation is isotopically lighter than the lake it evaporated from.

We measure ‘lighter’ or ‘heavier’ with isotopes using what we call ‘delta notation.’ The numbers we get are given in ‘permil’ (‰) even though they’re not a concentration. What’s important is that more positive delta values means that there’s more of the ‘heavy’ element. More negative values means there’s more of the ‘light’ element. So, if the lake has a delta value of -1‰, then the cloud should have a more negative value, like -3‰. When a cloud rains or snows, the heavier elements fall out first, because they’re heavier. If the cloud has an isotopic value of -3‰, the snow should have a more positive value, like -2‰.

The change between lake and cloud, or between cloud and snow, is called fractionation, and is controlled in part by temperature. (This means that the numbers I just gave you are completely made up.) The fractionation is also different for hydrogen and oxygen, and we measure these separately. (Hydrogen and oxygen isotopes in water do tend to vary together, but it can get pretty complex.)

As a cloud rains, it loses its heavy isotopes. If we take a cloud or storm (or say a hurricane) and take it from its water source (a lake or the ocean) and move it over land, this fractionation will go on. If no more water vapor is added, then the cloud gradually gets isotopically lighter. This means that the precipitation will also get lighter (but will always be heavier than the cloud). This process is called ‘Rayleigh Distillation,’ and is an important assumption in isotope geochemistry. Luckily, it has been shown to be a good model.

All right, let’s get back to Lake Effect snow. We’re looking at a Lake Effect event that is expected to start sometime tomorrow. We can get snow bands off of the lake that make great stripes of snow across the landscape.

What Lake Effect snow from Lake Ontario teach us?

We know that the snow will be forming from water evaporated off of Lake Ontario, so it will be useful to know the isotopic values of that water as a baseline. We have no way of measuring it isotopic values of the water vapor (the cloud) but we can find out the air temperature close to the lake surface and calculate the the isotopic value should be.

Then, we can measure the isotopic value of the snow that falls. We can collect snow that falls right at the lake (that which first forms from the freshly evaporated water) and we can look at snow that falls some distance away. We can make predictions about what patterns we might see.


1) Snow collected near the lake will be isotopically heavier than snow collected further away. Even though it’s only a few miles, Rayleigh Distillation should have some effect.

2) Over time, the snow collected at one location should not change in isotopic value, unless air temperature at the lake varies significantly. Because the cloud will be continuously replenished from Lake Ontario, I don’t expect to see any variability over time. The isotopic value of the lake water should not change consequentially. What can change is the air temperature, which will alter the fractionation of the isotopes (when it’s colder, less of the heavy water will evaporate). Also, colder temperatures could result in freezing of the lake surface, effectively moving the shoreline further into the lake.

It’s a pretty simple thing to test these predictions. I just need to collect some snow samples (and recruit other people to do the same). Specifically, I’ll be collecting every six hours, since I’ll be measuring snow depth at that time interval anyway. Collecting every twelve hours would probably be sufficient. I run a laboratory that has a liquid water isotope analyzer, so analysis will be easy. Once I’ve got the results, then it’ll be a quick write-up that everyone can benefit from here. It’ll be interesting to see how well my predictions hold.

Our water analyzer, Norm, analyzing waters from hurricane Sandy.

Our water analyzer, Norm, analyzing waters from hurricane Sandy.

If you live nearby and think you might be interested in helping out with this little project, let me know in the comments below. The more the merrier!

UPDATE 1-21-13

After waiting for 24 hours, there has not yet been any snow. But I’m assured it’s on the way!

UPDATE 1-22-13
Snow is falling. Finally.


17 thoughts on “An Isotopic Study of Lake Effect Snow

  1. I’d be happy to help Penny. I live a little over 4 miles due west of downtown Williamson. (I can provide latitude and longitude to within a few meters if you need them.) Just let me know when and how much snow you need collected. Would zip-top bags be suitable containers?

  2. Dave, That’d be great! I think if you could sample every 12 hours that’d be fine. 7am and 7pm (or as close as you can to that) would jibe with my own sampling schedule. The sampling method is to clear a patch or set out a dish somewhere, preferrably minimally affected by wind, let the snow collect, then try to get it into your container. Focus on the top inch or two of snow. Write the time on the container with a permanent marker (I actually suggest doing this before you go outside). Zip-top bags should be fine, as long as you’re confident they won’t leak. Try to make sure there’s not much air in the bag when you seal it as well. Oxygen in air can actually exchange for the oxygen in the water. Crazy, right? We don’t need a lot of water. The equivalent of three or four ice cubes would be a ton. If you have the snack-size bags and can fill them half-way, that would be more than sufficient. But if you can conveniently get more, that is good. Store the samples in your freezer, then I’ll figure out how to get them from you.
    We’ll watch this storm and see if the forecast stays the same or changes. If it starts snowing tomorrow night, then grab a sample before going to bed. We’ll sample on Monday, then see what happens from there. I’ll put updates here.

  3. You got it boss. I’ll set up a container this afternoon and if it has anything at 7:00 PM I’ll collect it, then again tomorrow morning and evening.

    As for getting it to you, I can meet you any public place in or around Williamson. Breen’s comes to mind since most people around here already know where it is.

  4. Nada in Ontario as of 7:30 this morning too, and the only white substance on the ground in Rochester is road salt.

    I hope that analyzer’s not a rental. (grin)

  5. Heh. Thankfully, we own the beast. I’m told that there is still lake effect snow in the forecast. We’ll keep vigilant.

  6. I’ve got 3 samples here. Tuesday night, Wednesday morning, and Wednesday evening. I’m not sure what time the snow stopped falling Wednesday though, so the Wednesday evening sample’s timing is uncertain.

  7. That’s fine and wonderful. My weather-friend keeps telling me that there’s more snow on the way overnight. If there is more snow in the morning, please grab a sample. If not, then don’t. 🙂 We’ll try to connect this weekend maybe to hand off the samples?

  8. I’m afraid I’m going out of town this weekend and won’t be back until Sunday evening. Some evening this week or next would be easier for me, or the following weekend.

Leave a Reply

Your email address will not be published. Required fields are marked *