Session 17: Sinking Particles and Remineralization (2)

The following content is provided under a creative common license. Your support will help MIT Open Courseware continue to offer high-quality educational resources for free. To make a donation or view additional materials from hundreds of MIT courses, visit MIT Open Courseware at ocw. mmit. edu. Okay, so where we were on Tuesday is we were talking about the sinking of organic matter and the remmineralization in the water column. So if you remember if we plotted versus depth and plotted flux of organic matter, it tended to drop off fairly steeply and then in the thermocline and then was rather flat through the um through the deep water. And we talked about putting out sediment traps in the water column. These would be little sediment traps um to collect material. And we were talking about a couple of different hypotheses that people had for what controlled some of the processes involved in this decrease in flux as you went down the water column. And so in order to support a change in flux with depth, you have to have either you know transformation of large particles, large sinking particles into small particles that would then get respired. or you needed some process that would just respire those particles directly. Now, we talked a little bit about what those particles are, and I think I don't know if I ever listed them down specifically, but for example, the this flux of particles because of Stoke's law and the viscosity of water, the flux of particles, the particles themselves that are sinking have to be relatively large. Remember when we talked about Stokes law, we were talking about things that were, you know, greater than 100 microns or somewhere in that range. And so it could be, you know, large dead uh cells. It could be aggregates of cells. Aggregates. And in fact uh it's not uncommon for datom blooms um at the end of the bloom at the termination of the bloom uh often there'll be a aggregation of the cells of the dead cells and then they will or dead and sometimes living sink out as an algaal mat and this uh can be observed in the deep sea by collecting basically the organic uh mush on the bottom of the sediment layer and it will have a high chlorophyll content uh suggesting that it's very fresh recent material that has recently rained down onto the surface. It can also be z plankton carcasses. I'm not going to get carcass right, but carcass. I'll end up spelling caracus or something. Um, or zoplankton fecal pellets. Um, and these are things like for the zop plankton that form pico pellets, things like copapods and then spps which are um which are gelatinous filter feeders uh that tend to filter out a lot of the small particles and midsized particles out of the water column uh and then they end up producing these very tightly compacted um fecal pellets that tend to sink very rapidly. the last sort of um uh large um component of that of the sinking flux is what's often called marine snow. Um and these are just organic aggregates um that come from a variety of different sources. One of them could be um some organisms give off mucus. There are a variety of organisms that have organic feeding structures.

For example, the um appendic gallerians which are another form of filter feeder and they shed this they have this sort of organic web that they use for filtering and they shed that on a periodic basis and then that can go into these marine snow or marine aggregates. Um and then there's also the pos the possibility for um a spontaneous formation where you have either particle or DOM particle interactions. that could lead to the formation of a of an organic aggregate as these aggregates grow. So let's say you've got a some marine aggregate here. As they grow, remember increased size leads to increased sinking speed. And one of the things these aggregates can do is actually collect smaller particles by essentially just as the aggregate sinking, it catches up with suspended material or smaller material that isn't sinking as fast. And these can be or inorganic. So you know for example you know cockaliths or dust etc. And so these marine aggregates, they may start off from a particular source, for example, a a feeding web from an epicular, but then it will accumulate other material as it sinks down the water column because it basically just sweeps out in front of it um many of the smaller particles um that are that it falls through. Now in addition to sediment traps, sediment traps give you flux, you know, and that's usually organic carbon, maybe inorganic carbon. uh silica or opal and the aigenous component. Um and it gives you some sense of the composition. So you can look in the sediment traps and look at the material and you can pick out individual carcasses, fecal pellets, um, and sometimes dead cells, you know, forams, things that are large enough, but often there's a lot of just organic mush and the thought is that a lot of that is this marine snow. Uh people have gone out and actually um gone into the water column and free doing um scuba diving in the upper water column and then also submarine work to try to look at these aggregates as they're actually sinking because once they get into the sediment traps, they just become this organic mush at the bottom. And it's very hard to actually figure out what their original size was or any anything characteristic about them other than other than their bulk composition. But the one of the problems with sediment traps is all the sediment trap tells you is net dlux dz at best. the net change in flux with depth. But there could be a lot of interactions along the way where you could have say large particles going to small particles going back to small particles being

mediated by um either some of these zol plankton. And so we need some other way of looking at the dynamics in the subsurface of these particles. And we're going to turn back to radionucleides because for two reasons. one is they're quite useful in two its chemistry class. So this will look very similar at the outset to what we were discussing with thorium 234 for the upper water column except we're going to choose different isotopes of thorium that have time scales that are more relevant to the thermocline and deep ocean. So in particular, if you remember when we were doing thorium 234, we had uranium 238 decayed to thorium 234. But that thorium 234 will decay to uranium 234, which then decays to thorium 230. So we were using this in the upper water column because thorium 234's halflife is short. It's approximately do you guys remember about 24 days. the halflife of thorium 230 is more like 75,000 years and where what we saw before was that in the deep water or actually below the euphotic zone thorium 234 um was in secular equilibrium with uranium 238. 8. We're now going to look at deviations and in the surface waters, thorium 234 had a deficit with respect to uranium 238. We're going to do the same thing where we're now going to look at the deficit of thorium 230 with respect to uranium 234. Uranium has a long um we talked about this for the upper water column. Uranium has a fairly long resonance time in the ocean um because there aren't a lot of large sources and sinks. So we can approximate the uranium concentration as a function of salinity and we did that for uranium 23 238. The activity of 234 uranium is slightly higher than the activity of 238 uranium. And Meg mentioned this when she was doing her lectures on weathering. Does anybody remember what the cause of that? Yeah, the in this process of the decay of uranium 238, it's releasing alpha particles and it's actually breaking up and damaging the mineral lattice um the crystal lattice that the uranium was originally in. And so you get a little bit more leeching out of the uh uranium 234 because of that damage. And so there's a slight in seawater a slight elevation of uranium 234. um some of the original work on the thorium 230 or one of the seminal papers is a paper by bacon and Anderson 1982 and we'll see if we can track the reference down for you. So what you have is something very similar to what we saw before, which is you have uranium 234 that's going to decay into thorium 230 into a dissolve phase. That dissolve phase has two paths. It can either decay itself through radioactive decay or it can get absorbed onto a particulate phase that would then sink out of the system being carried with the particles themselves.

So the first variant of this we're going to look at is an irreversible scavenging model. So in irreversible scavenging there's one kinetic constant and that's the forward scavenging of dissolve thorium onto particulate thorium and once it gets on there it never comes back off. So the equations you would write out would be for an equation for dissolved thorium. So the rate of change of dissolve thorium with time and we're going to be interested in the steadystate solutions of that. And so that's going to equal the activity of U234. So that's the production term. And then we would have these two loss terms, one due to decay and one due to scavenging and they would be multiplied by the dissolve thorium concentration. Similarly, you would have an equation for the um particulate thorium and it steady state the source term would be coming from the scavenging of the dissolve phase. There would be decay on the on of the particulate thorium. And then there'd be a loss term where the loss term is going to be the vertical gradient of the particulate thorium and the sinking velocity. So this would be something like that would look like this is if let's say you had a layer and uh or a layer bounded by say two surfaces. So you have in inside the layer you have formation of particulate thorium and you would have two fluxes of particulate thorium. One at the top of the interface, one at the bottom of the interface. flux is going to equal W, the sinking velocity, times the thorium concentration, right? And so you would have a flux in and a flux out. And so that the if you have net formation of particulate thorium in that box, the flux out has to exceed the flux in. Right? So if you have if the formation is greater than zero, then flux out has to be greater than flux in. And the way of looking at that is to look at the um the way of thinking about that is by looking at the gradient you're actually taking this depth range delta Z and looking at the rate of change of that flux assuming that W is constant over that small interval. And so this term basically gives you the divergence of the flux in and flux out term. And at steady state if there was just production those two would have to equal each other. Does that make sense to you guys? Does that make sense? Yeah. Question. If flux out is greater than flux in, I always sort of think that the particulate thorium concentration would be going down. It just and you have formation greater than zero. I'm not sure what that means. — Okay, the so let's look back at this equation. If let's ignore for the moment the decay on the on of the particulate thorium. This would say that the formation term K1 thorium dissolved.

So that would be formation of the particulate thorium, right? That's the absorption of the dissolved thorium onto the particles has to equal the gradient times the sinking velocity. And so if let's say we start off with a flux at some point and then you're producing more particulate thorium. The next point should have a larger flux. And then again, If I looked at the gradient of flux, which would just be DZ of W thorium P. Um, and then we're going to approximate that by saying w is constant, you get that term. And so this is actually positive because it's increasing as you increase Z. And that's just balancing it's it's balancing the formation. So if you have if you're forming it in the water column, it has to go someplace. The decay term isn't going to be big enough because it's not going to hang around for very long because it's going to get it's going to be scavenged onto particles that sink out. And if it's scavenging onto sinking particles, then the sinking flux of thorium has to actually increase with depth to basically balance the formation in the water column. Is that so to answer that or does that answer your question? Scott, you're forming particulate thorium at the same rate throughout the water column, correct? — Yes. The assumption — falls out. So, as you go down deeper, there's more and more falling out because there's more above. — The the assumption if we went all the way back up here was that you have a fairly uniform um source due to uranium 234 decay. So, you're forming thorium, you're forming thorium 230 throughout the water column at a fairly uniform rate, barring small deviations in salinity, but we'll ignore that. Um, and the thorium is getting scavenged fairly quickly onto the sinking particles. And so, essentially the depth is integrating all of the thorium. the flux at any particular depth is integrating all of the production that has occurred above that depth. So the if you see if I trying to see I have another way of that I explained it in the notes and let me go back and we'll try that and see if that um if that helps — quick clarification. — Yeah. — And let's say we have formation form greater than zero. You're talking about formation of dissolved thorium. — Well, here I was talking about the formation of there there's two formations, right? There's the there's the production of dissolved thorium from uranium decay and that can go two pathways. It can either decay itself or it can get scavenged onto thorium. When it gets scavenged on thorium you can think of it as the formation of particulate. It gets scavenged onto particles. You can think of that as the formation of particulate thorium. And so once it gets scavenged onto there, you have this net flow of thorium onto the particles. So you have a net production term at steady state. Something has to balance that. And that's the sinking out flux is balancing that. Let me maybe let me try showing this in equations and see if that makes sense. Um, so if I take this equation up

here and assume steady state, which I did, and then rearrange it, I would get that the activity of uranium 234 would equal K1 thorium dissolved. So this is the scavenging term. plus the decay term um we're going to make the assumption that for thorium 230 30 that K1 is much greater than lambda. Basically the scavenging is going to dominate over decay in the water column. And this would then allow us to approximate, you know, this can be written as thorium dissolved equals the activity of uranium 234 over K1 + lambda. With this approximation um that collapses down and we get that the thorium dissolved concentration is approximately equal to the activity of uranium 234 over the scavenging coefficient. — Question about that. — Sure. So generally we used to always pick things where the scavenging rate was sort of on the order of your time scale of the life or whatever because otherwise you're sort of like you know using a meter stick to measure the distance to the moon or something. You know what I mean? It's like — not an appropriate measurement. Um the it this isn't perhaps ideal because what you'll see is that the um amount of thorium that's left in the water column is very small relative to the uranium activity. Um we'll show that in a minute. The problem is that there's not another isotope there's not another set of isotope systematics that is better and easily measured. So you know you take what you got and in this case the radionucleides just don't line up as well with the processes. Um that said, you can still as long as you can make precise measurements of thorium, you can use this. Um but what you'll find is that thorium concentrations are, you know, orders of magnitude lower than the original parent. So it's quite a ways out of secular equilibrium. I don't know if I have a plot of that. Well, you you'll see that in the homework um the homework that goes online tonight. So, what does that say? If thorium if thorium dissolved is approximately equal to the uranium activity over K1. If we were to plot that with depth, we know that uranium is semicon is basically a conserved species in seawater. It's only going to vary with salinity. So the thorium dissolved concentration is going to be approximately a constant with depth assuming that K1 is constant. And we'll take that as you know we'll say we're we're dealing with a homogeneous water column for the moment. Now if we go back to the original equation for particulate thorium right we can write out similarly the and solve for um and solve this equation. Now, we now know what thorium dissolved is, right? We have an approximation for that. So

I'm going to plug that in. And so from the dethorium particulate DT T equation, we're going to get that the activity of U234 is going to equal lambda thorium particle. So this is now the um the formation of dissolve thorium from uranium 234. And remember we approximated thorium dissolve time K1 is equal to the uranium activity. So we're going to plug that in there. We have gamma. We have the loss rate of particulate thorium by radioactive decay. And then we have W dorium P DZ. We're going to make the assumption now that the activity of uranium 234 is much greater than the activity of thorium on particles. Remember gamma times thorium the decay constant times the number of atoms just gives you the activity. And then that will allow us to solve for u 234 is approximately equal to this is an assumption. Solve for vertical gradient of the particulate thorium. So the partic the thorium is a is equal to a constant times W. If we assume W's constant, that says that the thorium particulate concentration is just going to be equal to a U234 over W * depth. Right? So if we go back to our diagram up here, this would then be the this would be the dissolve curve. particulate curve. So you'd expect to see uniform thorium uh concentration and increasing um particle concentrations of thorium. And since flux equals W thorium P, um the flux of thorium should also increase linearly with depth. Now that's the prediction. This is the prediction for the irreversible model. The problem is that when people went out and actually made observations by either pumping putting out sediment traps and measuring the flux or pumping large volumes of seawater at depth to see how much thorium was on the particles. What they observed was that thorium dissolved actually increased with depth and so did thorium particulate. So it violated the there must be something wrong with the irreversible model because it didn't work. Bacon and Anderson formulated a slightly more complicated model, one where they now allow reversible scavenging. So the uranium still goes to the dissolved, but then you can have dissolved thorium going onto the particulate phase and then thorium coming off the particulate phase going back into the dissolve pool.

And then you have sinking and you also have decay. You also have decay of this but it's very small. We can similarly then write out a set of equations again for thorium dissolved. Assume steady state. It's going to look very similar to the last one except for the addition of one term. So you have production by radioactive decay symbolized by the activity of U234. You have a loss term of scavenging a dissolved thorium onto the particles. You have radioactive decay and then you have a source term due to the release of thorium off of the particulate phase. That's a k minus one. Often you'll see rates written as K1 and K minus one. So it K1 is the forward reaction for one. K minus one is the backward reaction rate for one for equation one. Then we'd have the same sort of thing for the particulate phase where you'd have K1 thorium dissolve plus W dorium P DZ minus K minus one thorium P minus lambda thorium B and if we make the same assumptions as we made earlier right that uh we'll say K1 and K minus one are much greater than lambda then you get that thorium dissolved is approximately equal to a U23 234 plus K -1 thorium particulate over K1. So before we got a term that it was the dissolved thorium depended solely on the activity of uranium 234. Now with reversible dissociation we find that it also depends upon the flux. So this is a production term because it's radioactive decay. for dissolved thorium because it's a release of thorium. And so uh we actually need to account for both of those. Um the let's see the model for dissolved thorium is actually somewhat more complicated but we can still approximate it as it you write out a very long equation. But what you find is when you actually scale things properly is that the um that the particulate thorium still scales approximately as U234 activity over the sinking velocity which is what we found for the reversible the irreversible model and that's simply because um the end the end reaction or the end uh the the main sink of thorium production from uranium decay still is this sinking particle is the sinking thorium flux on the particles. So it has to get there. there somehow. And so it eventually all does still pile up in the particles and sink out. And the nice thing is this now is much more consistent. This kind of simple reversible model is much more consistent with the observations in that you have uh dissolved thorium uh increasing linearly with depth and that increase has to do with the um it it's basically scaling because there's more particulate thorium at depth that's releasing more dissolved thorium uh at depth from this term. the I had mentioned earlier this

assumption that the that you know the scavenging rates were much greater than lambda the halflife was order you know 75,000 years the particle resonance time based on estimates of thorium in the um in the dissolve phase and in the particular phase is only about 40 years. And so very little of the thorium actually hangs around long enough to get to decay in the water column. Almost all of it is scavenged onto particles on some decadal time scale um and then ends up sinking up to the bottom. Now in addition to um being useful for estimating scavenging rates. So these are scavenging rates um and tell you a lot about particularly this whole solution that you know the irreversible model didn't work. There needs to be some way for exchanging thorium on and off of particles. Now the next step would be to if you could actually start to differentiate between small particles and large particles. you might be able to um begin to distinguish some of the models that we talked about earlier uh between um whether the flux is simply of organic matter is simply sinking down and decaying or whether that sinking flux might get turned into smaller particles and then um aggregate back to larger particles. sink down smaller particles, large particles. This solution where you see this increase in thorium with depth says that there that it's likely that at least thorium must be coming off of the large particles that are sinking down, right? Because that's what's transporting additional thorium down to depth. Uh it's likely that organic matter is also coming off and on as well associated with these reversible processes. So the thorium data and some of the other radionucleides are starting to elucidate what the interactions are between the different pools of small particles and large particles. Um but the conclusions aren't yet you know sort of robust enough um to be completely convincing in part because these measurements are fairly hard and there aren't there isn't enough data yet. um or at least they it's on the borderline of having enough data to really make um to make a good case. There's one other thing I wanted to um say about the thorium before we move to protectinium. Oh, that's right. Um the other thing is you can use thorium 230 as a deep sediment trap calibration. Um because if we go back up, remember the or actually you can look here that the this thor this particulate thorium flux is increasing as we go down the water column. And if you since you know the activity of uranium 234 and you know the depth over which you're integrating sorry I'd forgotten that Z I think when I first wrote this out. Um this is the same as the irreversible case where it's the integrated over Z. Um you might not know W specifically. the thorium particulate concentration specifically, but the flux is constrained by the total amount of activity of uranium 234 and the depth. And so if you have a sediment trap where you can measure how much thorium 230 ends up in your trap for a particular time, that would be the flux of sinking of the sinking out flux of thorium 230.

You can compare that to what you expect based on um based on the amount of decay of uranium 234 in the overlying water column. And if those don't match, then probably something's wrong with your sediment trap. You know, there's been some hydrodnamic effect and you've either overcolcted or undercollected. Similarly, this is often and commonly used in sediment work um to look at the um the accumulation at the surf in the surface layer of thorium 230. basically the flux of sediment of thorium 230 that's making it into the sediments um and comparing that with the overlying activity of uranium 234 to see if there's winnowing or convergence of sediments if you're in a spot where there's bottom currents that are resuspending sediments and then depositing it all uh in one spot. It's sort of like, you know, when the in a snowstorm when the wind blows and then all the snow piles up on the side of your garage or house and there's no snow out on the street. Well, the same thing happens with sediments. And this gives you an idea or a calibration point for saying whether there's been excess uh deposition of sediments or a deficit of sediments. You can use the thorium 230 flux estimates and the water column to get at that and calibrate your sediment data as well. So surface sediments thorium 230 that I want to go over and that has to do with comparing thorium 230 with another isotope system namely protectinium. So thorium 230 has a halflife we said of about 70 it was 70 it's either 72 or 75 now I'm forgetting 72,000 protactinium 231 one has a halflife of 32,000 years. So they both have fairly long half lives. And protek 23 protectinium 231 the source of it is uranium 235. So all of what we've said about you know uranium being sta you know stable and conservative in seawater also holds for protectinium. So just as how you can use you know your uranium estimates to come up with a source flux of protectinium or thorium. You can do the same thing for protectinium. So it has a known seawater source. Now the difference is that thorium is more particle reactive than protectinium. And so the idea is to is that you know the production ratio of thorium to or excuse me I'll turn that around of protectinium to thorium in the water column. The thorium is assumed to be removed locally. So all the thorium that forms in the water column sinks out right underneath where it's formed. The protectinium eventually gets scavenged, but it can move laterally before scavenging occurs. So, protectinium, it's not scavenged as quickly. So, it might invect a bit. Remember we were saying that the um the scavenging time scale for thorium was about 40 years.

Protactinium might be more like a couple hundred. And so it can actually in the deep water it can move uh it can move and the what you often see. So this is about 40 years of scavenging time. This is more like a 100 to 200 years a scavenging time. And if you were to plot make a little schematic of your ocean, so you have thorium and protectinium being formed in the water column. Some of it's going to sink and be scavenged right there. And some of the protectinium is actually going to advect over and get scavenged someplace else. And in particular, it's going to get scavenged in regions like the margins where you have high productivity and high organic flux. So regions such as the margins, if there's just more organic flux sinking down, you have more surface area under which to scavenge. So this would be a region of high scavenging and this would be low scavenging. So if you were to look in the sediments or in the sediment traps, what you'd find is that out here you would have low protectinium to thorium ratios and here you would have high protectinium to thorium ratios. Um with the argument that all the thorium is constant because it just depends upon water depth. But that some of the protectinium that was formed out here that should have fallen here had a long enough time that it ended up in this high scavenging environment and it got trapped and removed in the margins. So there's been considerable work um to understand this in modern day as a modern-day proxy for scavenging. Um but you can also use it in a paleo sense because these isotopes have a fairly long halflife. So you can go back through the glacial interglacial cycles and see did scavenging change? Did my after you've decay corrected them, did my accumulation of protectinium to thorium in the open ocean or margins change during glacial cycles? which might tell you something about how productivity changed there relative to modern conditions. Okay. So, any questions on uranium isotopes, thorium isotopes? Cuz we're going to change gears a little bit. Nope. Any questions up at MIT? Okay. so far we've been talking about um we've been talking a lot about sinking particles. We're now going to go back remember so this is say the euphotic zone and there were two pathways we talked about for export. there was the particular organic matter pathway, but there's also this dissolved organic matter pathway. And so we want to go back and talk a little bit about dissolved organic matter um and how that impacts export and what the fate of that dissolved organic matter is. So I'm just going to flip the page. I have a clean page. Oops. Okay. So, dissolved organic matter. Um, we've talked a lot that we don't know

the chemical characterization of DOM as well as we would like. It's a very heterogeneous um set of organic compounds many of which are not produced um many of which were not what was originally produced that the DOM has undergone chemical reactions and cross-linking and so it's a some fraction of it is a large geopolymer um and so it it's hard to characterize size other than in some of the bulk uh um the bulk characteristics which we talked a little bit about. We'll talk a little bit more today. Um, and we talked the other day about its measurement using things like UV oxidation or high temperature combustion where you're basically just burning or oxidizing the dissolved organic matter, turning it into CO2 and then analyzing the CO2. If you go out in the water column and look in the upper thermocline, so this would be the upper thousand meters, and look at the dissolved organic matter as a function of depth. By the time you get down to about 1,000 mters, you're going to have a fairly uniform background concentration of about 40 micro moles per kilogram. And as you head up to the surface, in many places you'll actually see uh increases with surface values ranging between say 60 and maybe 90 micro moles per kilogram. And those are micro moles of carbon. These are in places like the subtropics and convergence zones. Subtropics are regions of convergence. Remember the water uh the waters being upwell in the subpolar and the equator. And so the water in the subtropics uh has been at the surface for a fair length of time. And if this organic matter is pool is kind of building up, it would be transported into the middle of the subtropics and and um so it's had some time to build up concentrations. If you look in the polar waters, particularly in the southern ocean, um they tend to have lower dissolved organic carbon concentrations, not that much above this deep background. So that gives you some sense of the profile with depth. We could also look for example along the path of deep water formation. So if we started in the North Atlantic, Southern Ocean, North Pacific, now we expect the nutrients concentrations to do what? As you go from the deep north Atlantic. So this would be deep to go from the deep north Atlantic to the deep north Pacific. Nutrients will — increase. Oxygen is going to — decrease. Um, some of the early data suggested that there was a slight decrease of a dissolved organic matter, dissolved organic carbon from something like maybe 48 to 34. Um, the more recent data kind of looks more like that. And so it's not completely clear that there is a gradient. There may be um but that still needs to be worked out. And in fact, right now they're trying to collect enough dissolved organic carbon samples on some of the new uh global transexs to be able to really clarify what the deep water concentrations of DOC are and whether we can map them out uh as well as we can map out nutrients and things like that. Now, these concentrations are actually fairly high. um they're not as high obviously as dissolved inorganic carbon, right? Remember dissolved inorganic carbon DIC

was something like approximately 2,000 micro moles per kilogram. But 40 is a lot of organic matter and 60 to 90 is a huge amount of organic matter. Particularly in the subtropics where the um amount of biomass or the amount of particulate organic matter may be um you know one or two orders of magnitude smaller. So there's a lot more DOC than there is PC in the ocean. And the global estimates are that there's something around 700 pedigogs carbon as DOC where there was only about say three pedagrams carbon as PO. What else do we know about the dissolved organic pool? Well, we can make measurements of DO. Do is something like 4 to 6 micro moles per kilogram which would give us C to N ratio of 10 to 15. Um, does anybody remember what typical red field ratio is for carbon to nitrogen ratios? — Carbon to nitrogen, not carbon to phosphorus. — 106 to 16. So 6. 6. So by comparison, Redfield would be about 6. 6. So the material is carbonri or nitrogen poor. Um if you were to go down the list of carbohydrate, protein, DNA, lipid, um you know, you'd be looking at things like carbohydrates. you know, just based on C to N ratios, you might think that, well, maybe the source material for DOC looks more like carbohydrates than it would say proteins because the CDN ratio of carbohydrates is fairly high. It's carbonri material where amino acids have a lot of nitrogen in them. — Um, question. — Sure. Um backing up to that uh the deep water graph you say DOC are you meaning interchangeable with DOM? Yeah, these are this was in carbon. If I go back up to this plot, these concentration this was actually these were the amount of DOM that's the amount of carbon in DOM. So these measurements were based on carbon measurements. These are carbon measurements. And now I'm saying well what are the other components? How much nitrogen and how much phosphorus is in that organic matter? — Is that okay? Yeah. Along the trajectory from North Atlantic to North Pacific. — Yeah. — Um I guess I would think that more organic matters either sinking or being — dropped and it would increase. So why is it decreasing not increasing? And then if DOC has a long resonance time, it's pretty stable. — So it's not being destroyed by other processes. Why would it decrease rather than increase? Well, that's a good question. If it's decreasing, that means there has to be a net lo there. If it truly was decreasing, and what it may be, you know, from some of the data I've seen is that what it might end up being once we have more data is there was a sharp decrease right in the North Atlantic and then there it might be relatively flat and then a little bit of a decrease right in the North Pacific. The original paper didn't have enough data to well describe this curve and then there might be some scatter but what would be if okay so let's say there's a net loss so you mentioned there might be a source due to decay of organ or decaying organic particles right so along this trajectory particles are sinking down and some of that is being those particles are being destroyed and forming new dissolved organic matter. What would be a sink?

— Water upwelling out of the — Yeah. Another one would be a few hundred years to a thousand years lifetime. Maybe some of it's being removed by respiration. So ba based solely on this kind of plot you would say that well there might be a source but the net sink from respiration has to be larger than that. So as the water ages you're losing dissolved organic matter in the deep water. That's certainly what you would sort of think from this plot where you've got more in some regions more dissolved organic matter in the surface region and less at depth that as you move down the water column this dissolved organic matter has been respired. Um if this water gets ade you know some of this gets mixed down or advected down there must be some respiration going on to remove that dissolved organic matter. But that's a good question. In fact, one of the things people are looking at right now are one fraction or subset of this dissolved organic matter is the part that's colored that's chromophoric. So you can actually remember we talked a little bit when we were talking about optics how that the junk the gelb stuff in the water uh absorbs a lot of photons particularly down into the blue region. Well, you can actually measure that um and look at it in the deep water. And one of the questions they still don't know is in the deep water is respiration a bigger sink than release of say chromophoric dissolved organic matter from particles is a net source. And so they don't even know if you were to a take a parcel and age it. We think if you age a parcel even in the deep water that dissolved organic matter will go down that the that there is a net sink over time. Um but for the chro um for the chromophoric subset or subfraction we don't even know what the sign of that is yet. There's sort of conflicting data on that. So yeah that was a good question. Sorry I I looked at this plot long enough I didn't even think to put it in the right context. So that was a good question. Does that make sense to you guys? — Another one. I guess I don't understand if you said earlier that the solved organic matter is resistant to being respir. — Yeah. — You're making the case that it respires in the deep ocean and you're simultaneously making the case that it's accumulating in the surface because of ocean circulation. — Right. — Okay. being respired in both spots. Okay, it's it the question is in the surface you might have a net it's it's what's the net right in the surface you have a lot of production and so you have production of dissolved organic matter and so the production in the let's say you have let's say we go back to our subtropical gy so these are isopal surfaces so let's say you have upwell dwelling of water and net convergence and along the way you have production of DOM you would also have some respiration but let's say that respiration is slower than the production term so as you moved along that axis you might see DOM concentrations increasing as you got into the interior of the subtropical gy. But once some of this water gets pumped down into the deep, there's a lot less production. And in fact, there might be very little production. Um, depending upon what's going on with the particles, then respiration is going to take over because you've moved away from the production zone. And so you just have a net sink. And so it has to do the question is what are the relevant time scales? The time scales in the deep water are hundreds to thousands of years. Up here we're talking, you know, one to maybe 10 years for surface circulation or tens of years. And so let's say you had a slow respiration term. Even if term that was working

everywhere, you wouldn't see it up in the surface because you're only it's only working over about 10 years where if you've got a thousand or a thousand years for that slow respiration term to work in the deep water, it could lead to a slow decline of DOC. So you just have to compare the time scales. Okay. I did want to get back and cover a couple of other things today. We'll have um if if memory serves next Tuesday's lecture um we we should be on time for finishing this block of lectures. And so if there's still after I'm done today, if there's still questions on dissolved organic matter, we can come back to it on Tuesday because I think I should have enough time to finish everything. Um and then next Thursday, Bill Martin will be talking about some of the sediment uh fluxes. Okay, so we talked about the DO. The DOP is relatively small. So this would be dissolved organic phosphorus. It's only like maybe 0. 1 to 04 microar. It's not nearly as well characterized. Um most groups don't measure it. Uh most groups measure DO who do this measure DOC. Some subset of that measure DO and a smaller subset measure DOP. These would be in the surface. there's hardly any DOP at depth. Um, and this gives C2P ratios of something like 200 to 600, which is quite a bit bigger than the red field, canonical red field, which is, you know, say 116. Now, I've mentioned that the DOM is heterogeneous. Um, and it's actually a spectrum. There will be a whole spectrum of characteristics, but we kind of try to lump it into a few pools. Um, so you'll often see DOM fairly artificially partitioned by time scale. into a labile component. And this would have time scales of say hours to days. And it's going to be small. less than five micro moles per kilogram. And that's only going to be in the upper water column. And this is micro moles carbon. There's a semile labile. You can tell this is an exact term, semile labile, of days to maybe months. And that's going to be anywhere from about zero up to maybe 30 micro moles per micro moles carbon per kilogram. And this is if we go back to that plot on the previous page, this bulge of do of DOC that you saw in the subtropics, that's this semile labile stuff. It lasts long enough to build up in the surface. Um, but it's not so refractory that it can be it isn't respired before it gets into the deep water. And then there's a refractory component which is you know hundreds to thousands of years and that's this background that's about 40 micro moles per kilogram. Now you do have a net production of DOC near the surface

and this net production gets export into the thermoc mostly into the thermocline and that's via mixing and advection. The current understanding is that in the upper thermocline, if you were to compare the respiration of this dissolved organic carbon versus the total respiration, it's going to be less than about half of total respiration. There were some hypotheses a decade or so ago, oh gosh, no, longer than that, going on two decades ago, that DOC could be the dominant source of organic matter for respiration in the upper thermocline. But the a closer examination of the data and new data suggests that that's not actually true. In addition to sort of DOC spatial gradients, we also know something about it about DOC dynamics via um radiocarbon measurements. So you can measure the radiocarbon content of dissolved organic matter. And a fair amount of this work has been done. Uh, one of the leaders on this is um is Ellen Duffle who's out at UC Irvine and uh Bower and I can't remember Bower's first name off the top of my head. Uh this is fairly new work in the last oh say 10 or 15 years. But if you were to look as a function of depth and most of their samples have come from the North Pacific. So if you plot radioarbon so you have minus a,000 which is essentially completely deadus 500. Now, does anybody remember for the pre-industrial, so this is delta C14 and Bill went over quite a bit about radiocarbon before the midterm. Does anybody remember what the pre-industrial ocean was, say for inorganic carbon, the surface levels? What was the pre-industrial atmosphere for radiocarbon? What does del C14 mean? — Oh, come on. We got three minutes left. — You don't actually want the content. You just want zero. — Yeah. — Okay. — Why zero? That — That's the definition. — Yeah. the the pre-industrial atmosphere had a del a delta C14 of approximately zero. Right? That was the standard that's used for delta C14 was set up such that the pre-industrial atmosphere had about zero. And that meant that bio freshly produced biological material in the pre-industrial period had a delta of 14C of about zero. The surface ocean was somewhat depleted. The inorganic pool was somewhat depleted relative to the atmosphere because of reservoir effects. Right? that just takes some time for carbon 14 rich 14 carbon radiocarbonri CO2 to mix into the ocean. Um and during that time you have the chance for decay. So if I were to look at the pre-industrial ocean, the dissolved inorganic carbon would have been slightly below zero in the surface and then would have decreased some with depth. So this would be DIC and it decreases in depth with depth because deep water is older than um than surface water. And so you'd expect deep DIC pools to be somewhere around minus 200 per mill.

the because of the therm the release of C-14 into the atmosphere by weapons testing. If you actually go out and look at the actual values now, they're above zero. Um, and that's because you've had an invasion of bomb C14 into the surface water. Currently, particulate organic matter all has a bomb spike. You know, if you go out and look at sediment trap material or even suspended particles, they tend to be elevated above the DIC concentrations and elevated above what we think pre-industrial levels were. This is one evident line of evidence that the suspended particles in the deep ocean have to be communicating with the large sinking particles. Because if you were to go out to the deep north Pacific and look at a suspended particle that if that had invected there from the North Atlantic or the Southern Ocean, it should be a couple hundred years old. It shouldn't have any bomb radioarbon in it. But they do. And that's another line of evidence that there's in addition to the thorium isotopes that there's exchange of information of material between the sinking particles and the suspended particles. If we go out and look at the dissolved organic matter, there's a surface increase, but the dissolved organic matter is really old relative to both the current source of particulate organic matter and the current dissolved inorganic carbon pool. So some fraction of the DOC must be really old. And in fact, if you just do calendar ages based on um radioarbon, you get that in the deep water DOC in the Atlantic is about 4,000 years old and in the Pacific it's about 6,000 years old. Now that doesn't mean remember DOC is not a heterogeneous colle it's is not a homogeneous organic material it's heterogeneous some of the DOC might actually be dead it might have no radiocarbon at all might be really refractory other components might have um some recent there might be some exchange between the PC and the DOC pool and in fact there is evidence for that in the radiocarbon data but on the whole DOC has a much lower radiocarbon and that's one of the reasons why we talk about DOC as being quite refractory. So, we'll stop there and we'll pick it up with um looking at um red field ratios and pre-formed nutrients and respiration.