The “settled science” is that simply looking at the O18/O16 ratio tells you what the temperature has been. (I’m leaving off the “sigma” in front of the isotope names as it’s a PITA to look up the Unicode for it and type all the escapes to get it printed – Oh for a keyboard with dual Latin / Greek and with all those strange fiddly bits Europeans use to decorate their letters ;-)
At any rate, we just accept that the ratio tells the temperature, unadorned.
The question is “by how much” is it recording non-temperature things?
That, IMHO, is one giant Dig Here! One I can’t do right now (reality constraints – I need to work more on getting money and less on brain food). But I can still “admire the problem” and see if other folks have an interesting point of view on it.
First up, what brought this on?
Musing on Rain
I was pondering the recent rains. The globe has gotten a whole lot more wet lately. Flooding all over the “near equatorial” region. Decent snow pack “up north”.
The places with low rainfall tend to be those where the local geography interacts with the local (typically cold) offshore waters and the increased “loopy jet stream” (‘deeper’ Rossby Waves). Places like California.
California has a long history of cyclical droughts. Another “dig here” would be to see how closely they map to other natural cycles and how that package maps to hot / cold cycling. In the Midwest there is a long history of some areas having significant drought during different times. Some areas are ‘perverse’ in that they get more drought during hot times. Others get more drought during cold times (when you would expect the air to carry less water). The Rossby Waves tend to ‘stick’ for a little bit in certain configurations (likely due to the mountains and differential enthalpy heat available over oceans vs the Rocky Mountains). I would speculate that this might be the cause of relative enrichment of rainfall on places, like the South East of the USA, that are often under a “loop” headed north from the Gulf of Mexico water supply; while in California such a “loop” heads north from the Desert South of us or is fed air from a relatively cold off shore current – so little to supply rain. We get rain from two major sources. Storms that come down off the North Pacific / Gulf of Alaska where they loaded up on water from the warm Japan Current or, during times of a “Flat Jet Stream” from the “Pineapple Express” winds that occasionally will rip in off the middle Pacific Ocean.
So, IMHO, that’s why we often have different responses of California and related areas vs “Back East” to generalized “warming” or “cooling” of the planet. Or, more precisely, to the alternate “flat” vs “loopy” configurations of the jet stream. (The jet stream shape / Rossby waves likely depend in turn, IMHO, on Solar UV output changing atmospheric “height” via stratospheric and nearby layer heating and Ozone formation). But back at oxygen…
So we have these histories of “hot” and “cold” that show up as excess rains in some places and excess droughts in others. They also show up in Isotope Ratios. O18 vs O16.
Or do they? That’s the big question… One I can’t answer yet, but one on which a whole host of other presumptions about past climate rest.
Some time back I looked at the Carbon C12 / C13 ratio usage and found that we had more “speculation passing as data” than “well proven facts” about sources and sinks of carbon and their relative isotope ratios. (We just learned about “gut rocks” in fish being ‘pooed’ into the deeps, and we don’t know the isotope ratio of past already burned oil, natural gas, and coal; where each deposit has its own particular range of ratios, BTW… Oh, and there are gigatons of bacteria all eating anything that comes along in the sea or soil, many of which consume oil and methane and put it into the biosphere. We don’t know how much. So even the notion that oil and gas derived C is ‘non-biological’ is wrong on the face of it. It needs a number. One we don’t have.
That all left me wondering if there might be similar issues with the Oxygen Ratios.
In particular, the question:
“Does the Oxygen Isotope ratio tells use more about COLD or about RAIN?”
We already saw that large quantities of Silicate dissolves into the ocean, adding its oxygen to the mix. Similarly ancient carbonates dissolve as do sulphates and a host of other mineral sources of Oxygen. Erosion is a strong influence on what enters the ocean, and how much is ‘from living things vs. not’ is ‘up for grabs’.
The basic process in making ice core oxygen is that O18 is heavier, so doesn’t evaporate as well. From the start, more stays in the ocean, less gets into the clouds. Then we have rain. O18 condenses out first, leaving more O16. So much that some hurricanes can become substantially depleted in O18 before they are done as the water is ‘recycled’ a bit from cloud to ocean surface back to cloud.
As an air mass moves from a warm region to a cold region, water vapor condenses and is removed as precipitation. The precipitation removes H2-18O, leaving progressively more H2-16O-rich water vapor. This distillation process causes precipitation to have lower 18O/16O as the temperature decreases. Additional factors can affect the efficiency of the distillation, such as the direct precipitation of ice crystals, rather than liquid water, at low temperatures.
Due to the intense precipitation that occurs in hurricanes, the H2-18O is exhausted relative to the H2-16O, resulting in relatively low 18O/16O ratios. The subsequent uptake of hurricane rainfall in trees, creates a record of the passing of hurricanes that can be used to create a historical record in the absence of human records.
So one immediate “issue” is that we might be seeing the result of “rain / huricane fractionation” of the isotopes in Ice Cores in places like Antarctica and Greenland. Just HOW MUCH was there hurricane like isotope fractionation before the water that ended up in that ice was deposited? We don’t know. We could speculate that it would be similar to today. That generally the effect would be similar. But in the case of a climate history, we need a size not a general speculation. How big were the hurricanes in 250,000 BC? Did they follow the same tracks? Was the isotopic fractionation the same? (In degree AND in location).
Now, notice one other bit from that quote: “Additional factors can affect the efficiency of the distillation, such as the direct precipitation of ice crystals, rather than liquid water, at low temperatures”. Oh Dear. Now we need to know how much fell as snow, hail, sleet vs rain. When did the average storm, headed to deposit snow on the ice sheets, shift from making rain to making solids? Off shore? On shore? Can the ice sheet EVER record the ocean isotope ratio OR the atmospheric ratio? We know that major storms fractionate the ratios (vis the use of O16 depletion to mark the local passage of hurricanes) so how can we say it does NOT just show the “average of hurricanes”?
Makes it look like we just did a direct correlation of present temperatures vs Deuterium, called it causality, and then extrapolated. (Yes, I ought to ‘dig in’ to the prior original paper, but as noted, I’ve got real work to do or I start getting skinny ;-) so that will have to go to someone else to prove / disprove.) Deuterium acts very much like O18, only being even heavier, deposits even more strongly. Differential distillation is how we make “heavy water”.
From the article, it looks like Deuterium was used as the proof, then O16/O18 calibrated off of it (but that, too, needs a ‘dig here’ to demonstrate).
3311 m suggests that the Vostok climate records may be disturbed below 3311 m. Thus, discussion of the new data set is limited to the upper 3310 m of the ice core. Petit et al. (1999) reported an ice recovery rate of 85% or higher and a measurement accuracy of ± 0.5°/°° Surface Mean Ocean Water (SMOW). The temperature estimates are based on both experimental and theoretical arguments. One of the fundamental arguments used in deriving this temperature record is that the deuterium content distribution is well documented over East Antarctica and over a large range of temperatures (-20° to -55° C); thus, there is a linear relationship between the average annual surface temperature and the snow deuterium content. The slope of this δD/surface temperature relationship was found by Jouzel et al. (1993, 1996) and Petit et al. (1999) to be 9°/°° per °C. Further details on the methodology are presented in Jouzel et al. (1987), Lorius et al. (1985), and Petit et al. (1999).
One is also left to ponder what can “disturb” a nice stable Antarctic Ice Record below 3311 m but that’s for some other day…
The major point I’m making here is a simple one: Extrapolation of current isotope ratios to past temperatures assumes that the weather and storm patterns then are as they are now and so result in similar “slope” of the deposition curve (and that there is NO ‘fractionation’ happening ‘up wind’ as the storms approach). Has that been proven? Is it ALSO true of Greenland?
In essence, we are presuming past weather was the same as now to show that past climate was different. I find that a bit odd.
It might well be that we can show the “Hurricane Effect” is so narrow and short span as to be something that can be ignored; but near as I can tell that has been left as a “loose end” and is just assumed.
Bugs Bugs, Glorious Bugs
But Wait, There’s MORE…
It turns out that, the more I look, the more I find life itself is involved in the whole isotope game. For carbon we found it acting near sub-ocean vents of methane and volcanic vents, sucking in the ancient carbon as fast as it could and putting it into the food chain. The argument that the carbon ratio of fossil fuels is “ancient” while that of life is “new” fails on that point. Also on the point that deep in the rocks of the earth we find bacteria. More life exists inside the crust than on top of it. Life that eats a variety of things, including rock itself. Releasing C and S and other elements from those rocks. And if those rocks have Oxygen being released, too? Hmmmm….
So for both carbon and oxygen the question of “What is life doing?” matters. Are we “fixing gases” into solids? (Like those “Fish gut rocks” were Carbonate – CO3 – is precipitated. With what preference for isotope ratios?) or are we liberating it from rocks? (As bacteria break down Sulphate – SO4 – and release it). Again, we just don’t know. Life is known to have isotopic preferential reaction rates. So what is the relative rate of biological fixation vs release of the isotopes? Unknown.
(Some characters in the original are shown as question marks. I think they are likely a Greek letter and probably a sigma. Another ‘dig here’ if only a small one)
, markedly decreased d18O values of NO3 � and SO4 2� in forest floor water suggest that microbial processing occurred in organic soil horizons. Similarly low d18O values of NO3 � and SO4 2� were observed in forest floor and mineral soil leachates, groundwater, and streams. Over the winter observation period, most of the NO3 � and SO4 2� in stream water was from a watershed-derived source, whereas atmospheric contributions were relatively minor. Despite differences in soil water NO3 � concentrations between watersheds, the isotopic composition of NO3 � (d15N-NO3 � , d18O-NO3 � ) was similar, and indicated that in both watersheds most of the NO3 � was produced by nitrification in the forest soils. Although there was likely some contribution of SO4 2� from microbial oxidation of carbon-bonded sulfur, most of the stream water SO4 2� appeared to be derived from weathering of S- containing bedrock or parent material. The decreased d18O values of NO3 � and SO4 2� in upper soil horizons indicate that atmospheric deposition of N and S was not directly linked with stream water losses, even during winter and spring snowmelt. Citation: Campbell, J. L., M. J. Mitchell, and B. Mayer (2006), Isotopic assessment of NO3 � and SO4 2� mobility during winter in two
The point here is very simple. Two things are confounding the isotope ratios:
Life is playing with isotopes.
Erosion of soils and decomposition of rocks matters.
So we have to ask: How much is the isotope ratio showing the COLD and how much is it showing the RAIN / Erosion?
How much is LIFE changing those ratios in the surface runoff that takes 3000 years to reach the ocean bottom?
Does covering 1/4 of the planet with a glacier and causing a semi-desert in the rest introduce some kind of systematic bias?
Is there a hysteresis shown in the ‘entering’ vs ‘exiting’ an ice age glacial? (And what might that tell us).
I’d hoped to make this longer and more detailed, with a better set of conclusions, but as I’m not a paid researcher on the Government Dole getting $Million Dollar wet kisses from the Elite Socialists who run the NGO handouts for politically correct results process, I can only do what can be done ‘for free’ (and for truth…)
So we have to end here at questions.
The general point is that the notion that we know the past was COLD vs WET rests on assumptions. Many of them. Assumptions about the role of erosion. Assumptions about the impact of life and how much of it there is at any one time. Assumptions about weather in the past.
This might not matter so much except that as we’ve take a turn to the cold/wet side, the Warmistas are all hollering that it is hotter than ever and that the snow falling in Europe is because it’s hotter in the arctic. (A particularly stupid kind of thinking, IMHO, but ‘it is what it is’ to quote Paul, my Mechanic… a particularly bright fellow, btw. So put down “It is what it is. Paul The Mechanic” as one of your truth touchstones…)
So lets just speculate for a moment. We look at an ice core and find it has very high O16 in it (so the air must be very depleted in O18, so it must be “very cold”) and we find the ice thicker (so it must be snowing alot, so times are colder)… that’s per the present ‘received wisdom’… But what if the reality of those past ice cores is that the thick snows and depleted O18 are due to LESS arctic ice, so the storms cycle the water a bit more depleting O18 and then dump a load of snow?
What if the past “cold” is really measuring past rain, erosion, and level of life?
In essence, what if the past “very cold” times were exactly as they are now, but we just got screwed up on the details of how the isotopes report things?
Are we using “different rulers” to measure PAST “cold events” vs present cold events, so getting different results?
So that’s the question. I’ll leave you with some links and one or two other minor observations.
Has a load of interesting links / citations at the bottom
That wiki on oxygen isotope ratios includes:
Limestone is deposited from the calcite shells of microorganisms. Calcite, or calcium carbonate, chemical formula CaCO3, is formed from water, H2O, and carbon dioxide, CO2, dissolved in the water. The carbon dioxide provides two of the oxygen atoms in the calcite. The calcium must rob the third from the water. The isotope ratio in the calcite is therefore the same, after compensation, as the ratio in the water from which the microorganisms of a given layer extracted the material of the shell. The microorganism most frequently referenced is foraminifera.
Which has an important implication. As various limestones get subducted and volcanically turned back into lava and CO2 gas, the oxygen isotope ratio released into the air will depend on what it was millions of years before. Similarly, ancient limestone erosion and weathering will put “non-biologic” carbonate into the environment (both the C and O isotopes) As the Pacific Ring Of Fire is one of the major subduction zones on the planet “this matters” as things will be ‘time synchronous’ for much of the subducting rock. We need to put a number on it to know “how important”. An irrelevant fractional percent, or not? (My guess would be irrelevant, but that’s a guess…) “What is the total volcanic O production and with what isotope ratios? With how much dependent on prior depositions?” needs to be asked, and answered.
Look at the history of past ocean isotopes. Rock deposited “then” will have a different ratio than “now”. If we are suddenly starting to decompose a different set of rocks, with a different ratio, then there will be biases introduced into the releases that end up in the snows.
Look at that for a minute. We have a very long term increase in O18. Near the middle, the Antarctic thaws and that moves it all of 1 point. What moved it the other 4 points from “0” to “4”? Now, from the text, we find that the “zero” is the present value:
This figure shows climate change over the last 65 million years. The data are based on a compilation of oxygen isotope measurements (δ18O) on benthic foraminifera by Zachos et al. (2001) which reflect a combination of local temperature changes in their environment and changes in the isotopic composition of sea water associated with the growth and retreat of continental ice sheets.
Because it is related to both factors, it is not possible to uniquely tie these measurements to temperature without additional constraints. For the most recent data, an approximate relationship to temperature can be made by observing that the oxygen isotope measurements of Lisiecki and Raymo (2005) are tightly correlated to temperature changes at Vostok as established by Petit et al. (1999). Present day is indicated as 0. For the oldest part of the record, when temperatures were much warmer than today, it is possible to estimate temperature changes in the polar oceans (where these measurements were made) based on the observation that no significant ice sheets existed and hence all fluctuation in (δ18O) must result from local temperature changes (as reported by Zachos et al.).
Now contemplate that for a moment. Today is ZERO. But when Antarctica was melted and gone (so presumably it was warmer than today) the number was about 1 3/4 or 1.75. I.e. close to 2 out of 4.
So… how do we get 1/2 of the present value when it was warm enough to melt Antarctica (while today Antarctic ice is growing)?
How can WE be twice as hot as then? Yet adding Ice to Antarctica?
Something is wrong in the O16 / O18 ratio thesis. What, exactly, is a “dig here”… as is what does it mean…
My speculation is that it is confounded by rain and life and rain driven erosion.
In talking about this graph:
On geologic time scales, the largest shift in oxygen isotope ratios is due to the slow radiogenic evolution of the mantle. A variety of proposals exist for dealing with this, and are subject to a variety of systematic biases, but the most common approach is simply to suppress long-term trends in the record. This approach was applied in this case by subtracting a quadratic polynomial fit to the short-term averages. As a result, it is not possible to draw any conclusion about very long-term (>200 Myr) changes in temperatures from this data alone. However, it is usually believed that temperatures during the present cold period and during the Cretaceous thermal maximum are not greatly different from cold and hot periods during most of the rest the Phanerozoic. However, recently this has been disputed by Royer et al. (2004), who suggest that the highs and lows in the early part of the Phanerozoic were both significantly warmer than their recent counterparts.
In other words, we just fudge the data based on a “wiggle fit” and then say “warm was warm about like now and cold was cold”.
Somehow, the more I “dig here” the less satisfied I am about what the Oxygen Isotope ratios have to say. In short time periods they are subject to local effects like hurricanes. In long time periods, erosion (and thus a dependence on rainfall rates) along with “radiogenic evolution” matters. Perhaps cosmic rays, too… So exactly when in that continuum of time scales is it an accurate thermometer?
How do we know that the present large snow falls are not the entry signature to a new glaciation? We are showing O ratios that say it is as warm as when there was no Antarctic Ice Sheet, yet the Antarctic ice is growing, and both Europe and New Zealand (and other S.H. locations) are getting increased snows. Could it not be that the increased precipitation rate is what is reflected in the ice cores, and THAT is the signature event about which we ought to care? The added snowfalls, not the absolute temperature?
Is all the hollering about “warm in the Arctic” getting it ‘exactly wrong’ due to the past “cold low O18” snows being from “warmer water cycling depleting O18 prior to snowing”? We could check the snows falling now, but near as I can tell that is not being done.
I don’t see oxygen ratios as a proven clean thermometer.
Present snowfall is more important than folks think as an indicator of “cold coming”.
More rapid water cycling in heavier rains most likely confounds Oxygen isotope history.
It all needs a big review and needs to be re-proven from the first assumption on up before we use oxygen ratios as a reliable thermometer. Right now it says we are too warm to have an Antarctic Ice Cap and I find that hard to believe… and the “wiggle fit” correction for that just as troubling.