Hydrogen and Oxygen isotopes in sea water and marine sediments

We will turn our attention now to the isotopes of hydrogen and oxygen contained in water and in sediments precipitated from the water column. We'll learn what the concentration of these isotopes tells us about the volume of glacier ice and the temperature of various parts of the ocean. "Normal" water is H2l6O, but other isotopes of hydrogen or oxygen can substitute for the most prevalent isotopes, leading to various forms of heavy water, notably HD16O and H28O.

^ Light Water ^ Heavy Water

Natural Air Composition

Figure 1.5: Sketch showing how the growth of ice sheets on land affects the isotopic composition of ocean water. The water vapor which evaporates from the ocean is enriched in lighter forms of water, and becomes more isotopically light as the heavy forms of water preferentially rain or snow out before the remainder is deposited on the glacier. This process systematically transfers isotopically light water to the glacier, leaving the ocean isotopically heavy.

Figure 1.5: Sketch showing how the growth of ice sheets on land affects the isotopic composition of ocean water. The water vapor which evaporates from the ocean is enriched in lighter forms of water, and becomes more isotopically light as the heavy forms of water preferentially rain or snow out before the remainder is deposited on the glacier. This process systematically transfers isotopically light water to the glacier, leaving the ocean isotopically heavy.

At any given temperature light molecules, on average, travel with higher velocity than heavier molecules. This implies that during evaporation, the lighter isotopic forms of water evaporate more readily than the heavier forms, leading the vapor to be enriched in light water and depleted in heavy water relative to the liquid reservoir. Furthermore, when water vapor condenses into liquid or ice, the heavier forms condense more readily because the slower moving molecules can more easily stick together without bouncing off each other. In consequence, the rainfall is enriched in heavy forms relative to the vapor in the air, while the vapor left behind in the air is further enriched in the light forms and further depleted in the heavy forms. This is a form of distillation, very similar to the process by which one makes brandy from wine (or moonshine from fermented corn mash). Alcohol is more volatile than water, so the vapor in contact with heated wine is enriched in alcohol relative to the liquid; if part of the vapor cools and condenses, the water condenses out preferentially, leaving a potent essence at the end of the still if the remaining vapor is then condensed separately.

The way the distillation process affects the isotopic composition of sea water is sketched in Fig. 1.5. Let's suppose that the ocean starts with ó18O of zero relative to VSMOW. Water will evaporate into the air until the air becomes saturated with water vapor (a concept that will be made precise in Chapter 2). Since heavy molecules evaporate less readily than light molecules, the water vapor will be depleted in 18O relative to the ocean - in other words, it will be isotopically light. More precisely, the ratio of 18O to 16O for the water vapor in the air will be less than the ratio of the original ocean water, leading to a negative S18O for the vapor. The amount of depletion depends weakly on temperature. At 273K, the water vapor

S18O

is shifted by -11.7 relative to the ocean. At 290K the shift is -10.1%oand at 350K the shift is -6.0%o5. This reduction in isotopic contrast between reservoirs as temperature increases is typical of almost all isotopic fractionation problems. Because the amount of water stored in the form of water vapor in the atmosphere is utterly dwarfed by the amount of water in the ocean, the selective removal of light isotopes makes the ocean only very, very slightly isotopically heavy. But what if we removed the atmosphere's water vapor, sequestered it in a glacier on land, and repeated the process many times over until a substantial fraction of the ocean water had been transformed into an isotopically light glacier? In that case, the systematic removal of large volumes of isotopically light water from the ocean would leave the ocean water isotopically heavy by a significant amount - it would have a significantly positive S18O. Thus, the degree of to which the ocean is enriched in isotopically heavy forms of water tells us how much ice has built up on land. As ice volume becomes greater, the S18O (or similarly, SD) of the remaining ocean water becomes more positive. As an example, let's suppose we build an ice sheet by removing 200m depth of water from an ocean with a mean depth of 4km, assuming the glacier to be built from vapor with S18O = -10%o. If Sj is the S18O of the ice and So is that of the ocean water, then conservation of molecules implies that 200Sj + (4000 — 200)So = 0, if the ocean started with S18O = 0. From this we conclude So = .526%o.

In fact, for the reasons sketched in Fig. 1.5, the water that eventually snows out to form glaciers is much more isotopically light than the -10 value one might expect from just looking at the vapor in equilibrium with ocean water. The initially evaporated water vapor may have S18O = -10 , but on the way to the cold polar regions, some of that water will rain out back into the ocean, and the condensed water is isotopically heavy relative to the vapor, since heavy species condense more readily. That means that each time some atmospheric water vapor is lost to rainfall or snowfall back into the ocean, the vapor left behind becomes lighter. The precise extent of the additional lightening by the time the snow eventually falls out on a glacier depends on the amount of water lost on the way, which is in turn a function of the temperature difference between where the water was picked up from the ocean and where it was dropped on the glacier. Over the past 100,000 years, the S18O of the snow falling on Greenland has varied from -42 in the coldest times to values around -35%otoday. Antarctic ice is somewhat isotopically lighter than Greenland ice, and has S18O values ranging from -40 to -55 depending on location and age of the ice. Because of the additional fractionation on the way to the pole, the formation of the glacier would leave the ocean much more enriched in heavy isotopes than our previous estimate suggested. For glaciers having isotopic compositions comparable to the present ones, removing 200m of ocean to build glaciers would leave the ocean enriched by about +2%o, rather than a mere .526%o. The preceding discussion also shows that in order to translate the S value of the ocean into an ice volume, one needs some estimate of the isotopic composition of the glaciers being formed. For the present glaciers, this can be determined by drilling into the ice, but for past ice that no longer exists one must rely on modelling.

So, if we could go back in time and grab a bucket of sea water, measuring its isotopic composition would tell us the volume of ice on the Earth, and this would tell us much about how cold the planet was. Wouldn't it be awfully nice if there were some way to do that?

5In reality, the isotopic composition of water vapor in the atmosphere just above the ocean deviates somewhat from the equilibrium value, because the water vapor is in a dynamic balance between evaporation from the ocean and mixing of dry air into the layer from aloft. It is only when the air is saturated with water vapor that the equilibrium fractionation applies exactly.

Figure 1.6: The shell of a departed benthic foram (Cibicidoides Robertsonianus. The specimen is about one half millimeter in diameter.

Was this article helpful?

0 0

Post a comment