15:44:35 Alright, so thanks, nice to have.
15:44:41 I guess I've met most of you at this point, I'm Alex better off I'm at Clark University which is this tiny place just outside of Boston, and Wooster, so I started there and they said that you have to be your own postdoc, so today I'm being my postdoc
15:44:53 and showing you work in progress, sometime next week I'll show you stuff that you know I actually have some results with so today it's about hopes and ambitions and techniques.
15:45:03 And next week is about results.
15:45:06 So the whole sort of motivation of this talk today is, we've been seeing plots that look like this an awful lot of these radar stratified metabolic communities.
15:45:17 And, you know, we've made a big deal about their ubiquity and their stability across many types of ecosystems presumably across geological time. I did my PhD thesis basically trying to interpret these things in dramatically.
15:45:30 It's their strong confidence that these things are persistent and strong.
15:45:33 So, within all the discussions we've had so far about, you know, making model communities and Model Model, you know, environments, it's, it's interesting to think about, you know, if we just start with the phone monster in all of its complexity.
15:45:50 To what extent can we understand the, the dynamics by which these metabolic strata arrives. Now their ubiquity is suggestive that the structures we see here and the dynamics shape them are somehow more general than the particular bacterial communities
15:46:05 that populate.
15:46:07 So the simplest thing that you can do is just it's just, you know, do a time lapse photo of how do these, how do these strata appear so you know if you so we go down Woods Hole we get a big pile for mode we stick it into a Winogradsky calm, and we just
15:46:21 take photos of it once an hour for for 14 days.
15:46:25 And we just want to see how these different layers appear. So the sort of naive hypothesis is that you know you you look at all these coverage structures and you take photos of them and they stole just sort of blend out, and sharpen out of the initial
15:46:38 mixture.
15:46:39 So we're going to be mostly looking at the dynamics of this white band here which are called sulfur oxidizing bacteria so they're burning probably hydrogen sulfide with oxygen, and they need to position themselves into competing gradients.
15:46:53 So here's what these dynamics actually look like. So that's a one millimeter scale. So we can see this sort of wave of software oxidizing bacteria moving upwards.
15:47:04 It becomes more intense more bright, it sharpens and sort of suggestive of, you know, a front coming to some sort of stable point. If we look at it now know depending on how much you believe that you can start seeing it for a little bit.
15:47:19 And there's maybe some sort of a timescale associated with the period of oscillation here would be something on the order of 10s of hours.
15:47:26 If we keep watching it, now it's going to rip itself in half.
15:47:30 Now we have two parallel fronts.
15:47:35 Then do that again.
15:47:38 Then maybe again if you sort of squint your eyes. So I have no good explanation for this you know you can Sherlock Holmes it and then make up some nonsense that probably sent innocent people to the gallows and Sherlock Holmes is times.
15:47:51 But you know, my basic point here is, there are curious dynamics happening here are, in many ways, you know, more complicated than we would not easily gas, but they still sort of seem low dimensional they're sort of timescales emerging their skills emerging
15:48:12 much of what we see here, you know, sort of is reminiscent of low dimensional systems. Yes.
15:48:12 So given that these are sulfur oxidizing bacteria that are watching. Should I assume that the other layers have already been established we just can't see them because that's where the gradient of the hydrogen sulfide already exists.
15:48:22 Okay, so here's the whole point of this project, where I can make movies like this and it look interesting, but it's entirely impossible to know anything definitely right because there are so many different things changing at the same time, I don't have
15:48:34 any clear idea of what's happening. I see so what I'm going to try to do in the next little bit is just design an experiment where I can at least follow one thing.
15:48:41 And if I can at least follow one thing maybe there's hope that I can follow two things. So, the answer right now is I have no idea, but I mean I did my honest best to sort of load the system without staring stuff up too much.
15:48:53 And I also want to say that you know that that propagation of the front and even this bifurcation you can repeat that, that that's a repeatable effectively saw many many times, I see.
15:49:03 All right, so the sort of hope for this whole project this sort of hope that the formation and dynamics that shape the strata might be more general than, then you know the peculiarities of any particular microbial community.
15:49:21 So the hope is that while we have extraordinary genomic diversity extraordinary phenotypic diversity. You know, we've got some strong constraints and posed by conservation of mass.
15:49:32 So, all of life certified 90% of sales are made up of carbon, hydrogen, nitrogen, oxygen, phosphorus or sulfur, that sort of 98% of the atoms in you.
15:49:42 And let these, these nutrients get passed back and forth between organisms with different types of metabolisms, and some of these structures that we're seeing here are reflections of that.
15:49:52 So you know if you shine visible light on it you can generate a little carbon cycle with cyanobacteria splitting water to turn to turn into sugar plus occidental then you have some people doing that in reverse.
15:50:02 You can do the same thing with infrared bite, instead of breaking in electron off of h2o you break it off with HTTPS, so just do one step down the periodic table, we can form a cycle there.
15:50:13 And, and you can also do what's called came up with the water trophy where instead of taking your energy from a photon to do that reconstruction. You can also do it with with chemical reaction so basically if you're burning stuff with oxygen you're getting
15:50:26 more energy out of that more delta G out of that then you are when you're burning it out of sulfate. So there's a buck to be made, burning the waste products have one was the rack and so the author, and that's what we see these people doing the bottom
15:50:38 right there burning oh two with HQs or HS minus dependent on the pH, and you can drive a cycle here.
15:50:46 Sorry, I think I heard my echo and I thought it was a question. Alright, cool. So the sort of. So, you know, it within these within these metabolic within these microbial mats we've got, you know, if we have a whole thing in a jar we've got a closed cycle,
15:51:02 which means that you can't, you know, turn all of your in organic carbon into organic carbon, right, all these cycles. If you average over sufficiently long amount of time, these cycles need to balance that all these cycles are a couple of to one another
15:51:16 by Redux chemistry, right, the thermodynamics of how much energy it takes to turn co2 into sugar or to burn sugar with so for all these are strongly constrained by just thermodynamics self.
15:51:31 Additionally, all these things are living in in pour water in, in, you know, without any without any global flow going through it so everything stagnant.
15:51:41 And so it's typically diffusion that's moving stuff around.
15:51:44 So that's, that's a strong constraint you know qualitatively it means that if you. So if you take that cycle there in the middle and you double it. You can imagine that you're doubling the purple software bacteria you get twice as much of that stuff going
15:51:55 on.
15:51:56 Since that's takes up more space, you're going to push this off rack oxidizing bacteria further down. And so these different cycles are literally competing with each other force base and the rights and so maybe if you put.
15:52:09 Oh, and then finally, all of these chemical cycles are strongly related to the composition of the bacteria themselves. So if you look at the Stoke geometry of living organisms.
15:52:18 They don't vary by great amounts, so the best example of these are what are called their Redfield ratios, the ratio of carbon, nitrogen, phosphorus, and typically it's about 106 carbons to 16 phosphorus is to one, sorry 16 Nigerians to one phosphorus.
15:52:33 There's some variability in this between different organisms But broadly speaking, it's consistent.
15:52:39 So, you know, it would be interesting to put these four components into a simple model, and just try to understand, you know, what are the low dimensional dynamics, you get out just composed by diffusion energetics transport and Stoke geometry.
15:52:54 You know I'm telling this to you and that probably means that I tried it myself. I don't know how to deal with death.
15:53:00 And I don't know how to deal with carrying capacity, that's that's what limited me, and I'm happy to talk more about that at a point if you're interested.
15:53:09 But anyway, well what we really want to do here is just look at this stuff happen. So what I'm going to do is I need to focus on just one of their cycles the academic part of the cycle.
15:53:19 So basically you can have photo synthesizers that split water and use that with you to make sure there's an option, and then you have things like us that go in reverse direction.
15:53:29 So if you'd like you've got one group of organisms that are moving the system away from chemical equilibrium, and the other group that are moving the system towards chemical equilibrium and dissipating that energy as he.
15:53:39 So what we're doing is we're going to start off with a very thin layer of mud. The the thickness of the virus module is a border the penetration depth of light, so light penetrates the entire thickness of the material once we project it.
15:53:53 And we're going and everything is within the penetration depth of options. So option penetrates the entire system.
15:53:58 So the sort of hope with that is that we're at least limited in the importance of, you know, the nitrogen, the D nitric fires all of the sulfur metabolisms all of the metal metabolisms, you know, Miss antigens I'm pretty confident hurricane here.
15:54:13 Maybe they're mixer I'm living in micro environments between Porter's, I don't know, but there are at least I'm trying to tamp that down a fair amount.
15:54:19 their mixer I'm living in micro environments between porters, I don't know but they're at least I'm trying to tamp them down a fair amount. Now what we want to do was be able to track the dynamics of this carbon cycle, and we're going to do that by tracking oxygen.
15:54:26 So what I've got here is you can see this red, you know, so it's a red piece of my of our, we take ruthenium die, and we dissolve on chloroform Nixon polystyrene and coat that over my of our buyer.
15:54:47 That is really hydrophobic or ruthenium die is reversible. It's fluorescent where essence is reversible quenched biology. So if you've got some amount of dying some amount of of blue light shining on it.
15:54:52 It glows red.
15:54:54 The more option, there is around the bus the the fluorescence there is. So, there's a fair amount of work that went into this just make uniform off toads and make sure that they're reasonably well separated from all the complicated chemistry happening
15:55:09 in our, in our mind.
15:55:12 Cool.
15:55:13 So, and then what we're going to do is just go to literally, we're just going to pick up mud from the beach, and we're going to try to generate these, these, you know, stratified communities.
15:55:25 Somehow, we're going to try to watch the dynamics by which the carbon cycle comes to comes to a studies.
15:55:30 Oh I also want to mention that is a really nice way. Really nice reason to think about the, you know, dynamics of an ecosystem as a reflection of the dynamics of the carbon cycle.
15:55:41 If we're talking about it. The know dynamics by which are, you know, the genetic diversity or the proteome or what transcriptome or whatever you want.
15:55:56 If we think about how that's that's evolving, it doesn't have any clear obvious steady state conservation of mass gives you that you know the system has to evolve towards steady state of carbon cycle.
15:56:05 Right.
15:56:05 There's at least something that we're converging towards.
15:56:10 So we can start just by looking at time so this is work that Boris and I work started with this was last fall, we, these are mascot it from Carpinteria two of you have seen these maps the rest of you haven't, and the simplest thing you can do is just
15:56:25 slide it in there and just want to watch the, the time evolution of these. So this is the oxygen relative to the equilibrium concentration of oxygen mode.
15:56:35 So they started off in equilibrium, initially get this burst of, you know, you're decreasing the concentration of oxygen.
15:56:54 variability between these different experiments, because I'm using bud.
15:56:56 So these little oscillations are me turning on and off the bite at 90 minute intervals. So when the light is turned off oxygen is going down. When I turn on the light auction is going up.
15:57:07 So you can see, so I need to do a better job in a moment, but you can get some sort of qualitative ideas for how the header trust and audit grocer evolving with respect for one another.
15:57:17 If you just, you know, do the dumbest thing, and say this amplitude is a reflection of otter trophy.
15:57:25 The distance between one and your lowest point is, that's a measure of how much oxygen is being taken up in this little thin film that difference is balanced by a diffuse of flux from the atmosphere into you so it's a reflection of your metabolic.
15:57:41 So if you plot your amplitude on the vertical axis as a function of your distance from the top.
15:57:49 It kind of looks like a spiral Spain sank. So you know, sort of go around in a circle, you've got some fast dynamics there, reach some sort of a fixed point which sort of slowly evolved after that,
15:58:01 there's good reason, as we've been discussing with Boris at this that you could imagine this as, you know, we mix up all the mode, we kill a bunch of cells we're bringing anaerobic bacteria putting them in contact with the atmosphere, they all explode
15:58:14 and die there's there are washing the bodies of their breath or their brothers, they quickly devour each other you get this phase of exponential growth with fast oxygen uptake.
15:58:25 And then after a while, you know, they go into stationary state and the metabolic rate decreases a little bit and so you stabilize to a lower, lower point.
15:58:33 That's a suggestion. And now what we're doing is we're collecting material from the beginning of the voltage trough and the rebound we're trying to extract DNA out of that.
15:58:41 That's the goal. It's moving forward step by step, but I've still got much covered.
15:58:50 For sake of time, I'll just say that you know it's qualitatively similar if you increase the thickness of the month.
15:58:56 So the real interest of all of this experiment is that not only do we have access to the timescales over which option in this format, we can look at its distribution space.
15:59:06 So here's, here's a image of the fluorescence, so this is the fluorescence and then I convert fluorescence to oxygen concentration.
15:59:16 This is that time is equal to zero.
15:59:19 This dashed line represents the, the circumference of a spot that I projected onto the surface of the mud. And when I project a spot of light there photosynthesis going to be possible and presumably the concentration of oxygen is going to increase.
15:59:34 This is light off conditions.
15:59:36 And so, if you look here on the left, the two important things or this dotted line, that shows you the edge of the spot.
15:59:44 This song of mine shows you the equilibrium concentration of oxygen so this is what the concentration of oxygen would be if there were no bacteria consuming option.
15:59:53 So the difference between this and this gives you the metabolic gives you the diffuse of flux into the mud, which gives you the metabolic rate locally.
16:00:00 That's what it looks like with right off conditions.
16:00:03 If I then turn on the light. You can see that within the spot option rises precipitously far outside of the spot, there's no bite we're still below the equilibrium constant.
16:00:14 There's this slow turn over here, I did my absolute best to try to focus light, get knife sharp edges. Of course I'm projecting this this spot on to a porous medium, so there's lateral scattering.
16:00:28 And a lot of these dynamics here.
16:00:31 A lot of that turnovers gets the penetration depth of light ladder.
16:00:36 So I can take an image every five minutes, and do this for about a month.
16:00:41 So here's what it looks like on day zero, we turn on the light, we get uniform oxygen within the spot uniform outside as well.
16:00:52 Now we're going to turn off the light that falls down, back down to that flat level.
16:00:59 We now go out I forget how many days.
16:01:02 Okay day 12.
16:01:04 You can say is that we get more oxygen produced auction is rising higher outside so presumably there are more photosynthetic microbes there's more options being produced per cubic centimeter.
16:01:21 Then, at nightfall, we fall down to a deeper level. These are 90 minute cycles. And so, that is an extremely terrible thing to do to bacteria, right, because I'm really messing with their circadian rhythm.
16:01:33 It's really convenient for the oxygen measurements.
16:01:38 Because what basically I want to do is in a moment I need to go through the entire inverse procedure to figure out what to invert the rate of actual production consumption.
16:01:50 You can see that if I'm only measuring the concentration that I'm so that I've got one equation, and two one notes are in function production rate of oxygen consumption.
16:02:01 So you need to. So if you want there's two independently, you need to compare a light on and light off conditions.
16:02:01 And so if you're doing that you want to make those measurements close enough together in time that you know the population is needs change dramatically between one of the other, for I think the more responsible thing to do here is do a 12 hour flight
16:02:10 cycle. And then, you know, at the end of every bite cycle do an immediate comparison. I didn't know what time resolution I wanted when I set this up. So we picked 90 minutes because it's sort of partway between the diffuse of timescale for how long it
16:02:26 takes the ingredients to reach equilibrium vertically. And, you know, I looked up a table and they told me a bunch of doubling times that I have no reason to believe, but I have chosen geometric mean between the diffuse of timescale, and some numbers
16:02:42 I looked up arbitrarily, and chose that as my on off site.
16:02:48 One every couple of days. Yeah.
16:02:54 Okay.
16:02:59 Do I know that they have circadian rhythms.
16:03:02 So I'm putting mud in here. And I know that some sign of bacteria have circadian rhythms.
16:03:10 So presumably some of what I stuck in here is doing it.
16:03:25 I guess I showed you this whole video did everybody see that nice ring up here. Okay, well then I won't show it again, when you pop it open you do actually see that right, so we're starting to see you know some sort of interesting dynamics of know the
16:03:31 the system, reaching some sort of interesting steady state which, again, this is a talk about hopes and aspirations, it's not a talk about results.
16:03:40 But maybe that's understandable.
16:03:43 Can you just clarify one thing about your inversion, your inverse problem procedure where you were inferring rates. Yeah.
16:03:51 Do you have to make the assumption like we do that the rate in the light phase of the uptake is unchanged, relative to the dark phase yeah so I need to make that assumption.
16:04:01 So basically you can write down the diffusion equation. and so the curvature of the oxygen concentration is the difference between the rate of function production and consumption.
16:04:09 So, just like you do I turn on the light, get b minus k i turn off the light and get k. And then I deserve. And so that's what that's why I'm using my 90 minute cycle so that the populations don't have too much time to change and adapt between my two
16:04:26 measurements. But that's just sort of the limitation up against right now.
16:04:31 All right, but right now I'm measuring see and what I really want to do is measure the local rate of oxygen production, be so the dentist so the density of oxygen sources of the density of auctions things.
16:04:46 This is sort of a difficult inverse problem, because you'll notice that what I'm doing is I'm measuring see here. And if I measure see if I take two derivatives of a noisy signal all I have left is noise.
16:04:59 So we're going to have to be a little bit careful to do this in verse.
16:05:14 The first thing we have is we can exploit a separation of length scales. This entire ecosystem is quasi two dimensional. The thickness, I shouldn't point on my screen because nobody can see that. So the thickness of this thing is a little less than a millimeter, and I've got a
16:05:18 one inch diameter spot. So the first thing we're going to do is turn this three dimensional diffusion equation into a quasi two dimensional one.
16:05:26 So now we've got oxygen changing as a result of three processes. The first is oxygen defuses laterally through the sacrament.
16:05:36 This is equivalent rates exponentially with the atmosphere, this you just totally passive.
16:05:42 How do you just sort of the you know the, what would it be it would be the thickness of the settlement if I divide by the diffusion coefficient.
16:05:52 Then we got the same b minus k, we pick up some pretty factors for everything because we're specifically measuring the concentration of oxygen at the bottom of the mat, not the average concentration.
16:06:02 So this is the see that we measure, not the representative see it's the same.
16:06:08 Alright.
16:06:10 So as I've already discussed.
16:06:12 You know, we've got one equation and two unknowns, so we have to look at the difference between bnk to get these two independent by any case so okay so that's what we do.
16:06:26 So, the important and much more interesting question than that is what's the spatial resolution that we have with this inverse method. If I have two sources right beside one another, how close can they be before I can no longer tell the difference between
16:06:39 the two.
16:06:40 And this is where the quasi two dimensionality of the system really becomes helpful. So we've had exponential damping, with the atmosphere.
16:06:49 So physically what's going on here is if I've got some thin layer of mud. If oxygen is produced at some given point, it's going to diffuse a distance laterally of order the thickness, the sediment thickness before it escapes into the atmosphere diffusion
16:07:04 in the atmosphere is fast. And so once it gets into the atmosphere, it's pretty well mixed. So, opera so sources and sings only influence one another within sort of you know one sediment thickness from you.
16:07:17 So, so there's a finite atmosphere on top of my layer of mud. and then the entire thing, so I've got a few different versions of this experiment.
16:07:28 What were either the entire thing is in a piece of milled aluminum or it's two acrylic sheets separated by an O ring and clamped down to make sure that it's a closed ecosystem, but yes we're still got we've still got interactions of oxygen to and from
16:07:42 the atmosphere,
16:07:45 this exponential damping really simplifies her life, so you can work out the greens functions of this. And, and so the Greens function is basically decaying exponential.
16:07:55 This is really convenient for inverse method, because what we can do now is projects the measured option concentration on to the greens functions of this equation.
16:08:05 And because the greens functions decay exponentially we're projecting onto a basis it's almost orthogonal.
16:08:10 So, so every measurement we take as long as they're separated by about one sediment thickness from one another, they're they're basically independent so we've got good spatial resolution down to the thickness of the setup.
16:08:25 Alright so now. Now let's do it. So now I'm going to show you the same information that I showed you before, of the spot arising, except now i'm going to be showing you with the so blue here is the rate of auction production rate is the rate of oxygen
16:08:40 consumption.
16:08:43 And we can watch them change so initially be increases quickly, then it drops off.
16:08:50 And it's going to then it grows up and stabilizes again and out crashes out again.
16:08:57 The time period of for that oscillation is about nine days.
16:09:05 So you know here here's the results of one experiment, blue here is a total integrated option flux averaged over a day night cycle. The right here is the rate of oxygen consumption.
16:09:17 So these two things are going to a quick break these curves need to move together.
16:09:20 Right, so I'm still far from equilibrium, I'm still not equilibrium. I'm still far from steady state.
16:09:26 If you'll forgive some mindless curve fitting that red curve looks an awful lot like logistic growth. If you do that, you get a doubling time have about four days.
16:09:37 The, and then as I discussed about the oscillation was a nine day period, the flood, if you look at the fluctuations about the logistic growth there substantially correlated with the oscillations, the have the option producers know it's a 48% correlation
16:09:52 Now it's a 48% correlation coefficient, which means that it, you know, of substantial importance. No, may it may be the, it's of substantial importance it's certainly not the only thing going on.
16:10:04 And the big surprises that it's still not an echo liberty and after, after all this time.
16:10:09 So, where we are right now is that we've got, I think a good method to track rates of option production consumption in quasi two dimensional microbial mats, you know it's it's a lot of work to keep the stable, over, over the timescale of weeks.
16:10:27 First epic and in a test.
16:10:29 But, you know, we're gradually making some progress and sort of, you know, initially we see some dynamics that sort of looks for charmingly simple. It seems like we could
16:10:43 at least give a suggestion for why we have these spirals sinks why we get these time periods of observation where these skills are coming from that they're, they're seeing a lot of questions that look like eigenvalues are changing signs or something like
16:10:56 that. I think that there are tractable problems within us.
16:11:02 I also just want to note that one of the big challenges of working with, with these thin systems is that if you want to sample anything from them. You need to pop them open and then you kill it.
16:11:14 So the whole problem now is setup five identical communities let them all evolve open them one at a time.
16:11:22 I've got another system which is not closed anymore, so you can see. So this is one of the options that we have the red thing is a big gasket. The white thing is the optical installation that separating the, the community from the option detector.
16:11:36 And so we can flow in any material we want.
16:11:40 And, and do similar experiments. So,
16:11:50 this is what it looks like so you can see that over here.
16:11:54 It's bright, we've got a lot of oxygen in that corner where we're flowing water and we're we're flowing oxygenated water into it.
16:12:01 As time goes on, you can see that I'm turning on and off of Aight, I'm projecting a spot.
16:12:08 Sort of right in here.
16:12:13 And maybe I'll just fast forward a little bit so you can see it and start saying that I turn on the light, and we generate this pulse of oxygen, and then that walks through the rest of the ecosystem where it gets consumed.
16:12:26 So I kind of like this setup as a mechanism to be continuously sampling from it, I would love to do stable isotope work with this right everything I've been talking about so far just trying to track down that carbon cycle.
16:12:40 Know, with the inflow, I could give it a spike of, of, you know, see 13 labeled, whatever or, you know, I majored in compound or I could start to tease apart how the different aspects of the carbon cycle, some of which I can measure, you know, our projected
16:12:56 onto all the other cycles we were talking about the difficulty as you can see is that pretty quickly bubbles born and break the experiment, so that's that's the experimental channel bubbles are the inevitable destruction of every experiment.
16:13:15 Alright so, that is all hope and aspiration. And so, to end on something even more ambitious.
16:13:24 So I think set a show the the plot on the left.
16:13:39 Alright, cool.
16:13:41 I guess I was so clear that nobody has any questions or maybe everybody just wanted beer. So, last slide and maybe I'll take a little bit of little bit of time on this.
16:13:53 You think you don't know what what food is synthetic bacteria are actually. Yeah. How would you know if that four day replication time is roughly what you would expect for the amount of the light intensity and the, so the four day doubling time was that
16:14:14 was for the auction think not the source, so that the auction think is presumably some variant or average over some large number of header traffic streams.
16:14:21 See, on that slide Did you have an estimate for the science. So the scientist plus the alga plus whatever else is producing oxygen. I'm not going to fit that thing to a logistic.
16:14:34 So I'll just report that it's a 90 day period and say that I don't know what's going on. Other than that.
16:14:41 Yeah, that's what I felt I was going to ask, because that sign is don't divide it when it's dark and you're making a dark every 90 minutes and yeah it's tops DNA.
16:14:51 Oh that's good guests that was not a Hoffman wouldn't even divided all if you
16:14:58 mean fast enough yeah exactly but it doesn't seem that fast either but
16:15:07 an unrelated question actually the ring formation in your desk is that something to do with the auto trolls wanting to be close to the boundary where the header tropes are helping them out with the oxygen production or so.
16:15:21 So, again, my short answer is I don't really know if I give you my slightly longer answer I can say a few different things.
16:15:29 So, closer to the ring that there's more space to be in the sort of shoulder of the ice intensity center in the region where light is scattering laterally.
16:15:41 So, so cyanobacteria can sort of build up within that shoulder, if they want to lower of it. You've got more great so you've had a fight gradient. So maybe you've got more types of science advisor live near the get, I don't know.
16:15:53 Alternatively within the center.
16:15:59 If so, you can also easily imagine that, that, you know, ultimately you're running out of pick your favorite nutrient phosphorus, and so you deplete the center of phosphorus there's no more life as possible there, and everybody's just living on what's
16:16:16 diffusing through the edges. I can give you the spot wouldn't give you the oscillation. maybe the, you know, if.
16:16:25 Lots of Volterra voters and obvious way to get off of it, but I don't
16:16:29 use it possible to distinguish between the eukaryotic eukaryotic photo tropes and the sign of bacteria and maybe some of the if you have if you have an Isagenix what attrition they're through spectroscopy.
16:16:42 So yeah so so this is, this is one of the projects that I'm trying to do here in the morning about doing extraction ball the photo pigments, as a function of time to get I get ideas of who's doing what.
16:16:55 So, we were doing the acetate extractions, of the, of the horror film, so that occurrence protocols that I've read.
16:17:04 You know, if you go to the different version peaks you can see, Phil ABC and all that, and you get good a good measures of the relative proportions. That might be a hint of Buddhist Buddhism, if that's a measurement you've done before that then I'd be
16:17:19 happy to chat with you because I'm still figuring.
16:17:29 Just to finish up this thought because I feel that people have no will.
16:17:37 This is a really curious spot. You can imagine how much work went into it, the fossil record on atmospheric oxygen concentration is probably not so good.
16:17:48 So it's curious to imagine what they did to get us. So if you tell them to the methods. I tend to trust this. That's a good upper bound, we can talk about how that's done if you're interested.
16:18:03 This little bit.
16:18:03 I tend to buy that the there's a good lower bound there's a good upper bound, and there's the Wiggles, there's good reason to think that that's always correlated with the truth.
16:18:15 The big straight part in the middle, I think tells us as much as as much a reflection of our ignorance as as our now. So you could add some wiggles to that you could add a curved that and not be inconsistent with the data made this
16:18:33 this curve.
16:18:34 So okay so the way that they make this curve.
16:18:39 So, so the way that they make this curve is they've got sort of 50 different proxies for atmospheric oxygen concentration.
16:18:47 And some of these are really good proxies.
16:18:50 And some of these are totally mediocre proxies. So we've got some good proxies back here, and then we swerved dramatically to avoid some mediocre proxy.
16:19:02 So, this is a good bound, how high up or, you know, you're now.
16:19:09 Drawing a straight line to avoid some things that you are somewhat confident about.
16:19:18 No, it's not an upper bound it's sort of an intermediate bound.
16:19:23 So well we've got so you can find some minerals that only form a very low oxygen concentrations, you can find some minerals that farm at higher than this concentration, you can imagine that there are some uncertainties about where others are.
16:19:35 And then if you find there's minerals that gives you some idea of what the oxygen concentration in the atmosphere was there, if that's globally representative is unclear.
16:19:45 So, this whole region is sort of
16:19:50 not unreasonable gifts.
16:19:55 This is a plot that maybe most of you haven't seen that I'm just showing it here in the hopes that some of you will become more excited about it.
16:20:05 So the basic idea here is, this is the rate so the the vertical scales are complicated.
16:20:14 Intuitively, it's just the ratio of C 13 to total carp.
16:20:18 So the router the fractional abundance of ice topically heavy stable carp.
16:20:25 So, a neat thing about Cisco, is that it tends to favor carbon 12, as opposed to carbon 13.
16:20:31 So if you pick up a block of limestone.
16:20:34 You can dissolve out all the carbon from that all the inorganic carbon, and you can measure the, the isotopic ratio, and that tells you the isotopic that tells you the isotopic weight of the, of the inorganic carbon.
16:20:51 So, if that if in times where that goes up, that means that things are taking right carbon out of, out of the ocean, and so going up on this part is generally correlated with more photosynthesis.
16:21:04 Going or more primary production going down on this part is generally correlated with more hetero trophy. These are not the only effects. But this is a primary effect, or the interesting thing to look about look at here is, you've got isotopic, the state,
16:21:20 you've got stabilized topic carbon going from, you know, as early as we can measure up to the origin of photosynthesis. So this state is pretty good.
16:21:31 And then we're auction goes up again is with the origin of of multicellular life. And you can see that after each one of these occurrences the carbon cycle goes totally out of whack.
16:21:45 It's curious to imagine that as we're introducing this new accident or as we increase its, its availability.
16:21:55 Does that destabilize the carbon cycle as a whole. And what we're seeing here, are you know some, no reorganizations of a carbon cycle, after these major events.
16:22:06 Yes.
16:22:10 Okay. So these are distributed or for space, much like these, you'll see that I actually have data points here, so that these are different cores. These are so that this is a big combined data set from a good review, I suggest, so this paper has the better
16:22:28 answer to that.
16:22:30 But this is, you know, our best faith.
16:22:33 Look at all the rocks we can put them all on the same plot together.
16:22:37 So, These features are generally thought to be global.
16:22:51 These are difficult data sets to work with.
16:22:54 I think that they're under explored by physicists.
16:22:59 And I think that there are some good questions that you can ask your. If you find somebody who actually understands isotopic geochemistry.
16:23:08 So make friends with isotopic geochemists because they're lovely.
16:23:13 They party.
16:23:16 What is your ambitious hope. Can you say it out explicitly, you have an ambitious hope as your slide title. Oh, what is the hope that we get excited about isotopes.
16:23:26 Oh, I mean yeah yes.
16:23:29 The I'm already there, so okay sorry, that is sort of a baseline hope and expectation that if I
16:23:40 you know I started with my slide about like, you know, understanding how different elemental cycles are linked to one another, are strongly constrained by energetics dokey geometry, transport, and conservation of mass.
16:23:58 If we think about the global elemental cycles, you know, clearly things are moving around in ways that are more complicated than diffusion.
16:24:06 But the other three, but you know they move around due to, you know, physics at any right.
16:24:24 And can we use the instabilities that may arise in such a model to understand a closed ecosystem that is.
16:24:33 That is the ambitious hope.
16:24:41 How do I get there. I sat in front of a bunch of really clever people and wave my hands and try to smile and get you excited about it.
16:24:50 No more seriously.
16:24:53 So I set up my four equations, or you know my how to even equations my four statements of, you know, energetics transport Stoke geometry conservation of mass.
16:25:08 As I mentioned, I don't know how to deal with death.
16:25:14 And there are some nice theological experiments that.
16:25:21 So, one measurement that I got really excited about some time ago was, if I take a settlement core I measure the concentration of bacteria is a function of depth.
16:25:30 So this is under sedimentation So, the deeper you go the longer it's been buried the concentration of bacteria decay is like a power ball with some exponent last one, the exponent, isn't that nice is you can parasite the site, but you know it's within
16:25:51 bounds of environmental microbiology it's a halfway decent power law. So I've got some suggestions for, you know, the, The faster that these bacteria grow and divide.
16:26:00 Know the faster they could be killed by things like phase, the faster they could be killed by things like Canada biotics. So there's some relationship between the rate at which you grow and the rate at which you die.
16:26:12 In parallel to that there's a lot of work done on the rate at which organic carbon gets broken down. So if I start off with a bag full of leaves and measure the fraction of biomass, the fraction of that fixed herb minutes remaining after some time, that,
16:26:28 that you get widely distributed so you get the rate of decay decreases by one over time. So you get this logarithmic decay of organic carbon. So when you bury an ecosystem on one hand, you've got everything slowing down logarithmic way.
16:26:44 At the same time, your rates of birth and death are going to be sort of coupled to the rate at which you're consuming this. Is there some way to make these to stretch out to give you the sort of power loss that you see.
16:26:56 I think that's a nice test for you know this general model that I set up at the beginning that my general hand wavy speculations about all these different constraints.
16:27:06 I feel that there is something there. And I need to talk to people who are better theorists then I about how to do that without sort of putting my thumb on the scale too much.
16:27:24 So when you say you don't know how to deal with death, it's a statement which is you don't know how to parameters, the rate. But, though if you just stick some rate on there, then they just die with that right.
16:27:40 Kind of. So, another. So, okay, so I think I'm okay at building little experiments where it can track auction. I feel some confidence in my ability to stand in front of you and talk about how to do that and talk about the inverse problem.
16:27:55 Now I'm talking about models that I kind of played with for a while and maybe this is better spent you know with us drink beer together, but but the basic idea of, you know, one way to kill these things would be to say, I've got some population of, you
16:28:08 know, and many bacteria, they're taking up carbon at some diversity of rates, as time goes on, you're eating the more and more recalcitrant material so it takes you longer and longer to do that, if not one timescale for how long does it take the community
16:28:19 to accumulate one self worth and body mass that gives you a rate at which sells would divide.
16:28:27 And then you could say, given that division right you could then select a bacterium to randomly die. No, to you've randomly selected bacterium to die at a rate proportional to its growth rate so that gives you a growth and death rate, if you've got some
16:28:40 way to hide that carbon to hide some fraction of that carbon, then as time goes on, you grow fewer and fewer cells based off of how quickly they're dying so maybe that can get you some sort of expert, give you some sort of power of odd decay of cell number,
16:28:54 but now it's a whole question of how do you hide that carbon.
16:28:58 And so understanding how carbon leaks out of the carbon cycle. I think is a fundamental question that's related that I don't know how to answer the way that ecosystems die without understanding that better.
16:29:23 When you're proposing early on in your talk when you were proposing a model of, of those cup couple cycles.
16:29:33 You mentioned the red field ratio of one or six to 16 to one.
16:29:38 I don't quite know how you plan to use this ratio in those models because the way I see it, you have a basically an energy cycles which are being coupled.
16:29:49 And if everything is limited by energies and you can be first approximation ignore.
16:29:57 You know the Fluxus of nutrients to biomass so it's it's kind of somehow it will materialize. If not, then it becomes a really hopeless or, at least, very complicated problem.
16:30:09 But at this scale, I don't know if you know those models by Rosalyn Allen, which sort of also has, as far as I remember she has one cycle work to, you know, molecules are stated between oxidize it and reduce state and back.
16:30:29 And they are coupled above to something and below to something, and she even has this extra factor which sort of mimics carrying capacity which, in your case would be a space.
16:30:42 So did you try to compare it to this model, or I haven't tried that model.