Don ANDERSON 3/28/18 New Applications of Autonomous Biosensors
Well. Thank you thank, you very much for that Jim and Dennison. Thanks. To all of you for coming here. Listen, to a, northerner. Talk to you a little bit about a problem that I know as many of you are quite. Familiar with I. Am. Going to right, away though, try to do to. Clarify, some, terminology. If. We talk about red. Tides down, here this is an image that I think, would, sort. Of resonate with many of you this is a picture of the types of red, tides you get down here this, one's in the the Gulf the. Gulf Coast of Florida you. Can see the water is brown. And discolored. Because of this this organism, we're going to talk about and. Problems. Would Eve might might have troubles walking on the beach because of aerosols, it's over it there may well be dead fish and so forth and this is what people, down here, call, a red tide if you went up north and this. You. Took a look at this body of water the, local, papers, would, say the red tide strikes, Cape Cod again and yet, if you look at it the water is not read, the. There's. No problem, breathing the air. But. If. You looked very closely at this sign way over here it, would tell you warning do not eat the shellfish because if, we, harvested, some clams right from here, the, day I took this picture just. One. Or two of, the clams would kill any one of us in this room they're, that toxic, that poisonous. Even. Though the water is blue, and. Seemingly. Harmless, so. That's. Why one. Of many reasons why that many. Scientists, don't really use the red tide term unless you I use, it in titles of my seminars, because it brings people out they know what I'm talking about but if you're talking with scientists. We use this term harmful. Algal blooms or H a B, or halves because. It's a much bigger term, and I'm going to give you some, slides I'm. Going to spend a little too much time on this in the beginning but after a lot of discussions, today I realized. That I want to cover this, general. Topic of halves and red tides a little more broadly to get you a feel, for just how diverse, the phenomena, are then, we're going to go back and study in detail some of the issues, we're using, in the northeast but, I want to give you the broad perspective first, now, if we went back to your bloom that I showed you and looked, it under looked at on the microscope you'd see this a lot of individual, cells these are tiny little, plants. Microscopic. Plants, that photosynthesize. This. Would be Karenia, brevis the organism, that causes the problems here this would be the version of the, water, in my neck of the woods in, this case, this. Little golden ball is the organism, that causes us most trouble, it's, called Alexandrian, now I show you this image for two reasons one is to to, first give you the feeling of how. Alexandrian, looks. In the field but also to realize to point out that that, this. Is, the assemblage, in which these organisms typically, typically, occur there they're rarely the only thing in the water and the, reason that's important, is for those of us trying to study them to measure them to learn, their dynamics, to learn their toxicity, all. Of, this is basically, noise in the system we care about what this cell is doing how many there are how toxic they are but, all of this other these, other organisms, are there, as well it's like me trying to do. A study here and figure out what's, the pulse, of all the people who are from, Massachusetts.
In This room very very, hard. To do with it with such an assemblage, and that's what we, have to do in this field that's where some of these the, challenges, with this bio sensor technology. Comes in so, let's let's go back to the well. That's right I forgot for those of you who, don't. Like looking at the water this is the same thing on land, it's, you, imagine each, one of these is a different species they're, blooming, we talked about algae blooming, the same way and there's. Many many many different species here, at the same time and, imagine. That the only one you he wanted to study because it was poisonous somehow was this orange one and and, that is your challenge. So. The basics, of this organisms. Life, start, here, every. One of these cells that we're talking about starts, as a single cell that then divides, the. The as they grow they don't get big like a tree they they grow and then they divide and then, they grow and they divide again so, that's why you go from two to four to six to eight cells and that is the, population. That typically can, change the color of the water or can make things dangerous. Now. To give you a few examples if, you get, an organism, like this one it may look spectacular, but. It's actually harmless, in this particular, location offshore. This, is a non toxic, organism, in New Zealand this is true color there's nothing fake about this it really does look like tomato, soup but. If all of this biomass. As we call it gets, blown into Shore and gets. Real close and starts to break down and decay it uses up oxygen and, when, that happens, it basically kills, everything. That's, there in those shallow waters, so you have one. Effect of non-toxic. ABS that. Kill. Things B just because of oxygen, depletion, but. Most, of the problems we deal with are, things like this this is again the Florida red, tide the water is is, typically, reddish, and brown now you see a lot of dead fish floating. In there these. Are fish that have died from a very potent, neurotoxin. That is produced. By and released, by that red tide, organism. And that kills the fish so, this is not an unusual sight. For some of you I'm sure millions, and millions of dead. Fish along. The the coast. So, these, toxins, can also now kill fish now. Here's a very interesting example, this is where I stuck in a few slides and I know I'll regret it timewise but it's fascinating, story. The. Farmed. Fish are also killed, by some, of these organisms, it's, actually some of them are being killed by a different, toxic, mechanism, but I won't go into that but you what you can see here is that very recently in 2016. And chilly. Almost. A billion dollars, worth of salmon, died in just, a few weeks because of a red tide down there and I. I was down there and I just came back from there again you. Can see these huge farms, they have many many many of them they're massive, in scale and a good half, or more of the, salmon, that you might eat will come from these, Chilean, farms, and what, you see though is that the fish were dying from one of these blooms just so. Many dead fish and I. Just, will show you a few pictures because, the fishermen, were the farmers were so handcuffed. By so many dead fish that they dumped them out at sea they got permission from the government to dump them at sea and, what. Happened. Was. That another. Red tide a different, species appeared. Shut. Down the shellfish, this time making the shellfish poisonous, and it.
Triggered, A true, social. Protest, a social. Uprising, and this is a this, is a real picture, as is. This one basically. They barricaded, all. Of the ferries they barricaded, highways. For three weeks they burn tires in the street this looks like it could be a riot in one of our inner cities this, was in Chile as a result, of fish. Kills, caused, by, one. Of these types of algae so it's, just a striking. Example of, how, this, phenomena. Can affect, society. Now, let's get back to the impacts, a number, of these, problems. That we call Habs show. Up in shellfish, this looks like a nice, plate. Of muscles but mussels and many other these bivalves will filter, the water filter. Algae from the water is food and then, accumulate. The toxins, to the point where, they. Can kill you or beat you can get very very sick so you've, got a shellfish. Poisoning, you've got seabird. Mortality. Sea lions. Whale. Mortalities. Many many of these monster, animals, now are we're, finding out or being killed by this toxins, produced by these tiny tiny. Organisms. And then. This, just to show you how broad the term is these are pictures I got years ago from Brian the point right here showing. Overgrowth. Of seaweed. Here this co diem here. In the region and and I think that's Brian standing, in up. To his waist in this seaweed, seaweed is algae also so this also comes under that term harmful, algal bloom but, just to show, you people. Are asking me what I was just China, why was I there well this, is one of the reasons I was there in 2008. They had a seaweed, bloom, that. Has now continued, every year since then that you will not believe in scale, this was at the Olympics this is one of our wind, surfers or sailors, trying. To sail in. This incredible, green, tide and it's. It. Was so, massive, that the government, had to bring in 10,000. Soldiers many of which you see here a lot, of people from the city of Qingdao, and, I'll. Just show you some. Pictures, to. Show you the scale of this. Phenomenon. This is just local citizens, trying to help clear it off the beach a hundred million, tons of algae, washed up in that part of China and it's something near that has happened every year since then that is an amazing. Story we can talk about later why it's happening, believe, it or not every time you eat sushi, and you, eat that little nori that wraps around it that, ultimately. Is the cause, of this, Chinese greed type that I don't have time to go into it so I'll, just leave you with that teaser but. You can see the incredible and this is the same ova that I saw today as. We, were going around looking at your aquaculture, facility, this is being grown and and can, be eaten and they're trying to figure out ways to do it but in this, quantity, all they did was throw it away so, just to wrap up all these this background I want, to make sure you realize that there's also fresh water equivalents, of what, I've shown you in the ocean this is a picture of Lake Erie and a bloom of, an organism, called Microcystis.
That You can actually see, from space, then. For many of you you might remember this. 2014. When, basically Toledo, Ohio shut. Down its drinking, water supply, for, several days because. Of the toxins, produced by this organism, and the plant was not ready to to. To remove them and so the people there went out went, without water for a number of days, because. Of it so. This. Is just a list of a number of haves and their impacts, and I highlighted. In yellow all, the ones that occur down, here, in, Florida, in one way or another so, you this, is really an amazing state for haves it's it's for I've got a lot of good colleagues working here for good reason because you have many, of these poisoning, syndromes, you've got at least three of them you have a fish poisoning, syndrome. That that is something that largely, associated, with reef fish the large reef fish especially. Down further in the in the keys and then brown. Tides you know you're having here in the Indian River Lagoon and, and so, forth so it's a big diverse problem. And of course those, of us studying it are always trying to find better ways to study it better ways to manage it and that then gets, me to the main part of this talk and if. I wanted to talk about new technologies. For have research there are many many many I've, just listed a couple here there are things like new. Methods for detecting the toxins, new quick, methods, like pregnancy kits for using this over much better analytical, methods methods. To even control, the blooms, to try to suppress, them and there's, many many other technologies. That we're working on but, what. I'm going to focus on are these two new sensors, to, try to look at to count and measure, toxins. In situ, it means in place like within the water as well. As numerical. Modeling for forecasts. And predicting, these blooms and these, two sort of I have to show. Them because they fit together the some, of the rationale, for doing this work is because we want to do that so. Now. We're heading up to the north although. You do have this problem here in the Indian River Lagoon this, is paralytic, shellfish, poisoning or. PSP. It's that disorder, as I showed you that could kill you if you eat shellfish, that are heavily. Contaminated. In. Our, region up here which hole is right there this. Is called the Gulf of Maine this, is then the connect, Canada, and the United States, and everywhere, in purple, you can see areas, where there are essentially annually, recurrent, outbreaks of this. Paralytic, shellfish poisoning so, it's quite, a, prevalent. Problem throughout that region and this, is then the region we've been been, studying, within the u.s. portion of that, now, it's, also important, to understand, a little bit about the life history, of this, organism, this, is the cell I showed you earlier and we call it now Alexandrian, cat. Analysis. Is also Alexandrian. Cats nella but. It's more like a seed think of it that way a seed, of a plant and this. Is the life history so let's start with this seed which we call a cyst, which. Germinates. Produces. A cell that then divides and makes two and as I showed you those two make like one, makes two this one makes two we get four and, then the bloom begins and you get a large. Number of cells just, through that process that we call a sexual, division just. Regular, binary division but, then at some point and it's typically when these cells run out of nutrients. They, form gametes, which, fuse together to make a zygote which, then produces that cyst and this, is the, life history, of the organism, and it's hugely important, in trying to explain and predict. What, happens with these outbreaks, and I put this slide here just to show you that one. Of the organisms, that you have in the Indian River Lagoon and it's also over on the west side of Florida. Is called pirate dating Bahaman see it's a similar class of same class as the organisms, I've showed you so far it. Has these. Are the vegetative, cells here it makes chains like this but. It also has a cyst stage just, like the Alexandrian, cyst that can germinate and cause. And liberate blooms so a lot of what I'm saying about alexandria, minute cyst does, have an analogy here and a lot of the technologies, the approaches, we're, taking can be use here and it's actually a lot of the discussions, I had here today, are about possible, collaborations.
Where We can can, share some of these technologies, so. Up, in the Gulf of Maine we have two types of blooms, as we call it a valleys and reom you, can see here's Cape Cod and here's Boston, and here's Nova Scotia you, can see this this coloring is this these are the cells of Alexandrian, that we've counted on cruises. That stopped at every one of these see, these little lines and samples, and we map the cells that we've counted you, can see the scale of this bloom is incredible, it covers hundreds, and hundreds of miles, it's. All moving along this coast and around Georges. Bank that's. A major type. Of the balloon we get up there but, this you can see is much smaller look at this road look at this building you can see this is a what's, called a salt pond out on Cape Cod and, this, is another place where we do a lot of our work it's. Also the, reason we do is because it, blooms essentially, every year with, Alexandrian, and part, of the reason it does is this very narrow and shallow is, shallow Inlet channel sort of keeps the cells in here the cysts are in here they germinate they, bloom and most of them stay there without being washed out but I will I will, come back to that, so. How. Do we study these well the first thing that I have to go through if I'm going to talk about sensors, are modeling. Oh you're gonna be bored by it no I hope not because. Modeling. Is really, really important, in every, part of our life when you think of and I'll come back to this weather forecasting. They've, had to do modeling of the atmosphere, that's the reason, they can do forecast. They have to be able to model the movement of the fluid air but. We have to model the movement of the fluid ocean as well as, the organisms. That are in it and we, use these models. For looking backwards, in time for looking at what's going on right now and for looking forward. So. How. Do we do that well we start with a, hydrodynamic. Model we'd, call this a physical model because, it's just the physics, of this system this shows you salinity, up in that region this is the Gulfstream you see very salty see. Little Eddie's and filaments, spinning off of it here it, through time the, blue shows fresh water coming down from the Arctic the melting melting. Glaciers. And rivers. And so forth up here these are rivers entering, up here the st. John and the Bay of Fundy the, Penobscot the, Androscoggin. So. This, physics, though is very well modeled, very well characterized. In, these hydrodynamic. Models but that's in a sense the easy part the hard part is here we, have to put a biological. Sub, model. Into. That, hydrodynamic. Model so. That we can say well what, to our organism. And so. I cannot, possibly go through all this but what I'll say is that we, start with a map of the cysts, of Alexandrian we go up and run cruises, and count in the sediments, all, of the cysts, that are there and we get maps like this that show what we would call a seedbed here. And a, seedbed here. This would be almost like if you went out in your yard and could somehow map out. Weeds. Weeds, like like some.
Dandelion, Or something like that, before, they would actually pop so you would know there's a seed bed here, and some, here but not so much in between, that. Is the potential, the sort of area, of the. The quantity of cells that might might, actually start, a bloom then. You need to know when do they bloom and, that's. What this shows we've done studies this is all data from the lab that, shows there's certain times of the year where, these cells will germinate and, here. This next year and here's the next year here's a next year and there's lots of times where. They either germinate, not at all or germinate, very poorly, and, if any of you are gardeners you know that's true for a number of different kinds of seeds but. I won't get into that, right now once, you've got them germinating, you need to know how fast they're going to germinate and that's what's shown here you, have to once they've, germinated and, are swimming around we have to know how fast. They grow as a function, of salinity. And temperature and. Light and we put all of that in this model give them a little bit of swimming behavior and I'm skipping, by years and years work but the, result, is. Here. Here's a model of the. Year 2005. We, took a cyst map in 2004. And, we have actually the way the currents, and the rainfall and all the conditions, happened in 2005, you can see this is going through now, it's up into June July, then. It's going to start over again in March so you can see that the bloom is beginning, in the areas of those cysts, seed beds that I told you that it's being carried. Down the coast by these coastal currents of worth it's going around, Georges. Bank down here even to Martha's Vineyard and and then, tuck it and this, model is actually a very good, representation. Of what truly happened, this was a year. That the state of Massachusetts. Had a 50mm, in dollar loss to its shellfish. Industry alone, because, of this outbreak, so. We've got a model what, can we do with it well we do many things but the one I wanted to highlight here is this. One weekly, forecast so we do something, a little bit like the 7 or 10 day forecast for, weather that, you get, when. You good want to go golfing or something, else but. How do we do that well we use, that model we. Run it with, the conditions, that are happening in that year and then we get a forecast, that runs out four or five days forward, and we extend, the model with that weather forecast, and so, myself, and a lot of, interested. Scientists, and managers in the regions will then get a message. Like this from, the people who are doing. This. This. Modeling, and it gives you this link and if, you click on it we, jump out to this webpage and. There. Is, not. Quite the wet though a radar, map that you get on the weather these days but a map of the Alexandrian, bloom, and you can see here this is seven. Eight nine ten June so so, we're. We're. Showing what's happening, in June I don't even remember what year this is I think it's 2016, or 2017 and. Managers. Can look at this and say well I'm worried.
About What's happening here you can see all the cells are offshore, they're, not being blown in or if a storm comes in they come but we we, give them this this this information every week but, there's a real problem if, we looked at the beginning of this this starts, in March, or there abouts and, we. Start running it let's. See if we can see where it starts we, start running it forward, and we. Can't, correct. It we if it started if something, is wrong. Whatever. Whatever reason, it something. Happened that wasn't captured by the model the, model just keeps propagating forward, so it's more like me saying would, you trust me if I said see I missed. It but we're in May already. If. Would. You trust me if I said I'm gonna give you a weather forecast, for what's going to happen in in, June then, I'll give it to you right now you wouldn't you wouldn't believe me and you shouldn't so, what. Is it though that. Allows the. Weather, forecasts. To be. Back. Here. So. Accurate, how can, we improve, our forecasts. And the answer, is, to take a lesson from the weather service these, are their models, very similar, to, what I've shown you for the models of that. We have of storms. Moving through the ocean these are the same sort of computer. Models but why are, they accurate well, this is why because all over, the. Region all over the country they have meteorological. Stations, like, these here that are collecting, data constantly, they are putting weather, balloons up that are giving you this three-dimensional. Do feeling. For what's going on in the atmosphere and all of that data is taken to the web by the weather service it's assimilated. Into those models they adjust them and then they run them forward again and those models are constantly, being, corrected. And then run, forward so, that's. What we are not able to do so far with. Our hab, models, so, how. Can we do how, can we improve things well what we need are in situ, sensors, we need sensors that equivalent, of those meteorological. Stations, but. This time for Alexandrian, cells and toxins, that. Can provide, this real-time data, that can be assimilated into, the model thereby, improving, forecasts, and that's, where we. Have been working on two of these, two. Instruments. This one's called the ESP, this, is called the eye FCB this is I'll go through each one tell. You what they are the first one ESP, is called the environmental, sample, processor, or ESP.
Developed, By one. Of my former students, and now the head, of the Monterey, Bay Aquarium Research, Institute Chris, :, you, see the instrument here it's about the size of a trash can and here. It's being deployed in one form but over, here is what's important, is that this. Technology. It's like a laboratory, in a can and I'll show you how it works in a second but what it can do is measure both, the the abundance. Of species. Of hap species, as well as their, toxins, it, measures them, autonomously. Robotic. Li inside. A can and so, I'll be quiet, for a minute life, and the, search for it is perhaps, our greatest adventure. On. Earth, the most abundant forms, of life are, those that we cannot see with the naked eye. Three, kilometers below, the ocean surface, microscopic. Life-forms, flourish in the liquid seeping, from the Earth's interior. What. Are these organisms, how, did they get here and what. Impact do they have on our environment. Each. Year, massive. Blooms of algae render, parts of our ocean of toxic, to humans, and Wildlife, imagine. If we had an automated, detection, Network warning. Us were harmful my. Present. As. We. Further industrialized, the ocean what, is the effect of microorganisms. In our harbours in off our coast is there. A better way to monitor, how we are impacting, the planet. Fish. Farms, are especially vulnerable to outbreaks of harmful algae and disease what, if we were instantly notified, of changes to the health of the waters around, aquaculture, areas. The. Environmental, sample processor, is an automated laboratory able. To detect microorganisms. In real time consider. The mysteries, this biological. And chemical sensor, is beginning, to address our. Ocean, houses, life-forms, we know almost nothing about where. Do they travel how, is humankind, impacting, their environment, and how are they in return, impacting. Ours the. ESP. Provides, us with a model, for how we might monitor, the health of our ocean in the future, and seek, out new forms, of life in, the darkest, depths of our world and. These breaches, of violence. So. I wanted. To, show. You in that last one and I couldn't stop it in time that one of the sponsors. Of that research was NASA and they helped a lot in putting that video, together as you can imagine because at the very end you can see what one of the reasons they were sponsoring, it is that they would, like to see this very same, technology, this, robotic. Laboratory. In a can technology, head. Out to other planets, for the search for life and so they're helping fund, research. In our own oceans looking, down the bottom and under all these hostile environments, to. Help advance, that technology, but we have been able to apply it to these these halves now, how do we do that we've, had to develop a very, robust, mooring, system because of the currents and the tides and, so forth in, our region it these see this is the ESP, here connected, to the bottom very. Strong. Surface. Expression, of that here's of the cable, that stretches, two and a half times its length and still. Can conduct information. Up, and down here's, a few more pictures. Of. It's. A big operation of a number of our ESPs on a ship being, deployed you, can see that there's the ESP, and it's canister, here, batteries. There, here's this surface, buoy and to. Give you some data here's. An example from 2014. Where, we deployed. Three, ESPs. Along. The coast up here in New Hampshire and Western Maine, and. The ESP, s were put very close to, the, mountain stations, where the state of Maine. Tests. The shellfish, flesh for the toxins, that the, PSP, toxins, we, wanted to see how what kind of information we got right near those stations, basically. We had a six-week deployment. We had ESP measurements. One a day once per day and we had measurements, of the, shellfish. Toxicity, by the state every, week and this is the kind of data that we get from those instruments over, here is the cell. Abundance. Over here is the the time and over here is the measurement of toxicity, and the shellfish that is this red. Line, here once a week this, line shows you the level at which it becomes dangerous, to eat shellfish, that's a threshold or an action limit so, here is the ESP, data you, can see how patchy the blooms are as they go by that that, Mord instrument, that, their cells there there's none as there is on but gradually, we get more and more cells, you, see that in general it, follows, this pattern that's showing up in the shellfish, that there's low toxicity, here, higher toxicity.
There You also see some interesting times. When we, get get rather big jumps, in the cell abundance, that, may, not have been picked up very much in this weekly. Data. From the state here's another ESP, deployed a little further up the coast again. You, see the, state data that talks of shellfish becoming, more. Toxic, the. Quarantine level, is actually. The quarantine level, in this case is way up there here's, the Alexandrian. Abundance, following. Some general pattern and we give these I've just just two examples of, many we, give this data on real-time basis, to the managers so they can know, what's happening. But, we it, also occurred to me that that, these are useful but how can we make them even more useful, to the resource managers, and one of the answers. There is that they this is the currency they talk in it's the toxin. Measurement. This is the currency that the instrument is giving them which is cell abundance, they're not the same thing they're. Related, but they're not the same thing so. What we've done is developed something and don't don't go bleary-eyed here this is really very simple what we've done is make the ESP. Into, a robotic, mussel we're, saying okay instead, of counting so we'll count your cells but let's use those cell, counts, and convert. Them to shellfish, toxicity, as if that ESP, were a mussel and this, is a formula that it, takes the toxin, at one that the next time point and says it's equal to how much toxin, we had at our last time point plus, an uptake, number, minus. A loss number, and this is the, uptake numbers just some constant, times how many cells, we. Measure, with the ESP, and this, is a constant. Times how much tox when we had at the previous time points real real, simple we fit this, model to, the data from the ESP, and the state. Agencies, and, this is the kind of data results. We get now you see this says toxin, over. Here on the ESP, side and it's. Toxin, well, it's toxin for both and. This. Is the state data these circles, and now. We see, the ESP data that we are hitting those, state results, perfectly. Almost, but, we are also seeing all the dynamics, in-between and as I said between, this point and this point look there, was a number, of days where our robotic. Muscle would have said there were dangerous, levels of toxin, that the state was not picking up with its weekly sampling that, a sensor, out there every day would, have we. Look at the other example, I gave you again. An extremely. Good fit. To, the data, but. Also these kind of Peaks, that we're not necessarily seeing in the the, state data and, we've. Done this for a number of different years and this is a rather remarkable, result where we've got, our model toxicity. In the ESP, plotted, versus what's observed in, the state, shellfish, records and you see this very, good straight line extremely. Strong. And statistically. Significant. Regression, so, we now have the opportunity, to use these instruments, for. A number of things one of which is to actually try, to give estimates of shellfish. Toxicity, without the state having to do all of those kind of tests, so, one, vision of the future for, us in monitoring, these problems, is to put an array of ESP. S along the coast in key locations, so. They, can provide this data the cell count data that can be assimilated into, the models like I showed you to make our forecast, models better but, they can also give, us estimates, of shellfish toxicity. At, these near shore monitoring, stations, to complement, what the states. Are doing so that's a that's a big big, plus and a big advance so. Moving quickly to the second. Instrument it's. Called the IFC. B or the imaging flow saito bond and this, is basically a submersible, microscope. It's a it, goes underwater and, it takes pictures of cells, very high resolution, pictures so here's another movie. This one I have to narrate so we're going to open, up the cover of the IFC, be so, you can see it it's, got. This. Laser. Over here you'll see in a second if we bring drawn a sample from the outside, and then, we push it down through this flow cell and as, it goes through the flow cell it, gets hit these particles. And the cells are in single file they get hit with a laser that's a piece of dirt it just gives off somewhat thing called side scatter if it hits if it's an Allen's with algal cell that gives off chlorophyll, fluorescence, which is a red color and that, triggers, a camera, which, takes a picture of that particle and that sounds complicated but, this instrument, can do 10 images, a second, so one 1000, I just took 10 images of cells like this, one and.
What. You'll see in a second is that we, can't humans are not fast enough to identify them so we have to teach the computer to it and so computer. Machine, learning, then is, what's neck is what's needed here's. Just a few statistics over here 8 to 10 images per second, 600,000. Images in a day, basically. We can't humans can't deal with that we usually joke, at a university. That that. That's. Typically what we have graduate, students do but even, graduate students can't can't, can't do this this is asking, too much so. This. Is why. We have to use automated, species, classification just, like on TV or in the airport where they're doing that facial recognition it's. The same thing being applied to cells here it's not you, know it's not rocket science so here. Are. This. Is what we call a dashboard of output from that I FCB, it shows all the big cell types here, and smaller ones this is actually an Alexandrian. Bloom all in this as, you'll see these are big alexandrine cells smaller ones that we know are gametes, these. Are 10, Tenon's and other organisms that are large these are diatoms, there's. Many many different kinds. Of organisms, here and one. Of the ways these IFC, bees are being used this ones in the Gulf Coast near Texas but it could easily be used in Florida. As well is. For early warning they, have put three es bees out along the coast on piers, that draw, water up from below and you can see they've had early warning of seven have events, so the instrument sits there. And is just taking samples three samples, an hour 24/7. And then, you, can see that it picks up this organism. Which causes, a syndrome, called d SP and. It. It at this point they alerted the managers, went out and tested and they found things where toxic. And they've done that multiple times this is cranny of the species you have a lot of problems with again, being, picked up here the instrument basically, has, thresholds. Set in it that if you get to a certain point send, out an email and text message to, managers, so the scientists, saying you've got a bloom happening, in this location so, it's being used for early warning very. Very effectively, but. I know some, people here might be interested, also in climate. Change and, changes, in pollution can you clean up the Indian River so, I stuck. In a few slides here about, that. What we might call trend, analysis, let's look over long periods, of time what, can this instrument, give us so, here's an example of. Think. Of this as cell number. It says bio volume but it's it's the same thing over, six, months of time off of where, Woods Hole is and. This. Is a measure of the total number of micro. Plankton in the water but the blue or this one class of algae we call diatoms, and you can see these six species are, the ones that are largely responsible for. Those, patterns and that looks interesting like, okay we have something going on but, if we actually take, it apart we, see that this, organism. Blooms. In January, this one blooms benching February, in March this, one between March and April you, can see how, different all, of these are and. That. Pattern. That I showed you in the beginning is. Really. The, result of all of these different species, that are called and this instrument, sitting out there unattended. For six months sending all this information, back to someone's computer at the lab everything's, being done it's an incredible, way to monitor, and this is what I've been telling people here, that I think that. You should have some of these in the Indian River that's. A great way and this is where I'm looking that maybe we can help help. Make that happen so, back, to long-term. Monitoring. This shows, a number of years and this and this is a one, type of organism, a little little blue-green alga, called Seneca caucus, these are diatoms, again buddy this, just goes through 2009. But imagine we have this data through, right, now more, than 10 years 12 years worth of data and this, shows you through, these months the of, the year the inter-annual, variability, in, just that one group, it's an incredible, data set to look at climate change or changes, in pollution you don't have some. Technician, who's the one who's been counting your cells leave and, go take a new job and you have someone else come and try to learn how to count so that the data are questionable, from year to year this is being done the, same the, same software. Year, after year after year so. Back, to our our. Use of the IFC, B we this is where we've been using it thus far in these small little salt, ponds that I mentioned, and, surprisingly.
This Is Alexander, mcat nello I didn't think this is really good I'm going to work but it has the, reason I didn't think it was going to work is that these are all, Alexandrian. Cells here and we would somehow have, to teach that that. Software, how to identify and, distinguish all of these stages from the many other co-occurring things, that noise that I showed you in that other picture so, we have five different types of what we call singlet cells by themselves, these, are doublets, cells, that could be dividing. Cells, that could be be fusing. We, have cells that are infected, with a parasite, you can see here you can have have quadruplets. You, have these large large. Planets. I goats that we have here that are the result right the precursors. To our cysts and we train the software, to do that and and are getting incredible, insights, and this is part of this, is a maybe. A little, too deep for some, but I but I hope you can follow the. Excitement, of a scientist, who for years has been studying, an organism, that has this complex life history, and now. All of a sudden it's like we have got a window, we're looking into the bedroom you know we're in there saying we're a peeping toms we are seeing what's. Going on in a way we never could, before so. What do I mean this, is just, an example, of the bloom in one particular, year as as tabulated. By the IFC, beware again, it's being done involve, we've turned cell counts in the volume it's the same thing. So. Among, the things we can observe here this, is now the volume. Of the singlets, those single. Cells I showed you it, can be multiple types but any single cell of, Alexandrian, what's its volume and you can see that it goes up and down and up and down and up and down and you see this is nighttime in daytime nighttime daytime nighttime, daytime and the. First point to be made is that through this early stage of, this bloom the, volume doesn't really change very much but just when this bloom is going to peak. Cells. Start getting smaller and smaller and smaller and they get really small down here and, then what happens they start getting larger. You. Don't see this up-and-down. Pattern. Anymore it's just they're getting larger so, what's happened, is this. Is where we. Are having basically. Mass. Gametogenesis, we're, producing, huge, numbers, of gametes. During. This end of the bloom this is why the bloom ends the cells have gone through that sexual, transition, where they make gametes that, that. Be all the small cells here that fuse together to make those large zygotes, which don't divide, anymore which, then fall out as cysts and interestingly we. Get more than 90% of. The population making. These gametes in the field when, we do this in, the lab if. We get 15, or 20% in, our culture's we're happy so the sexual induction. Of these this, organism, by the way you'll see a few mistakes here this we've had to rename it it's now cat nila buts I'm not changing species, on you the, sexual induction, is far more robust, in situaion, cultures, so we're, seeing really, what's happening, now, let's, take a look at that same pattern and what, else can we do because, this these cells are getting bigger and then smaller and bigger and smaller what's happening is they're at nighttime they.
Grow And they get big and then they divide and one cell makes two and therefore they're smaller and they, get too large, and then they divide large and dubai when, that happens in this kind of phase division, you can actually calculate what we call an in situ growth wait for the first time and in in a long time it sounds like it's an easy thing we can tell how fast these, cells are, in, the field again. It sounds like we should have been able to do that but any phytoplankton. Ecologist. Will tell you how hard that, is to make that measurement but we can do that with, this instrument, with. That equation, and what it shows us is. Here here's a growth rate just think of that as how fast do they grow how fast they divide, here's temperature, 5 10 15 20 these. Are some laboratory, data where. We took in brought cells, from, that area grew them in the lab and at, 10 degrees these are the various growth rates we got at 15 we get this these, are the growth rates we're calculating in the field. At those same temperatures, from, that I FCB, data and you, can see that they're much much higher and I. Could give you other reasons to tell you to convince you that that's not an artifact. But. I won't I don't have times but believe, me we looked into it and we, do believe those are true Institute, growth rates so, what does it mean that these, cells divide two to four times faster in situ, than in culture so not. Only are they having being. A lot more having a lot more gametes a lot more sex they are also growing faster, than then, we thought. Let's. Look at how fast they're swimming another, thing we can do is follow their, swimming. Up and down you, can see this is the surface of the water here and they migrate each day they swim up to the surface and then they go down they swim up but they don't quite get to the surface and we. Can actually measure the rate of that swimming and basically. We're. Getting rates of 15 to 20 meters, per day which. As you see up top here in the, lab and in tanks and so forth we were thinking they could swim 5 to 10 meters per day so, what. We've also seen, in this is, interesting. Here remember this pattern where they're, they're. Sort of going up and down and up and down and then this. Is the migration of those cells through. Time blown. Up and what we see is that these cells are avoiding, the surface, here which is one of the reasons they would stay within this pond but, after the point where they have, become these zygotes, they, are swimming up higher, and higher in the water column, which. Means that they're up near the surface here and can, get washed out of the pond well is that true well, this is where we took the IFC be put, it inside a jet yak which is a kayak, with a jetski. Engine. To it and has writ some remote controlled we can let it go out here into this shallow, Marsh system, so, here's the pond we, did all sorts of surveys here with it brought it out there did all these surveys, looking, for what, was out there and what do we get and now you've seen this - type of dashboard before but, what you see here now is the only Alexandrian. Cells we're seeing are these big big big dark zygotes. This. These are all a different organism, is something called header a caption this is yet another organism. There so, the only cells, leaving. The pond are that late history. Stage. So, what, we're saying then is that, two. Things these. Cells swim and migrate, faster, in, sits you than predicted, and migration.
Behavior Varies. Among these life history stages we never knew that before so there's very this. Has very significant. Implications. In terms of the dispersal, of the species and the final little set of these slides. You're. Now familiar this, is the bloom we actually had sort of two blooms this year I won't, go through all of this but, look down here here is where we actually deployed an ESP, and and I have CB together we, use the IFC, B to watch the cells tell us what's going on and then when we thought things, were interesting, and there were cells there we would trigger the ESB and have. It measure toxin, and this. Line down here shows. You the toxicity, that's what this axis is the. Toxin, per cell of the cultures, established, from this very salt pond you, can see it's it's, in this fairly narrow range this. Red line shows. You the toxicity, that, we. Measured in the field and it, basically was, much more toxic, than. What, we thought it was and I. Won't go into it but we also have, Institute. Growth rates that are going in here that, again are faster, so what, did we learn from that. Experiment was first of all that the toxin, content, of these cells varied, more, than four fold, during. The blooms so the cells aren't equally toxic, all the time sometimes they're really toxic, sometimes, they're less toxic, but, it was higher for the vast majority of the time than what we thought that, those cells could produce so. Once again these in situ instruments, are showing, us things that, are very, different from, what, we thought was the case using our laboratory. Studies we've been doing and everybody's been doing for years we, also saw that this Institute, growth rate buried. As much as four fold over just a few days, temperature. Would get cold growth rate they'd stop growing they grow slowly warms. Up they would grow faster so we saw a lot of dynamics, in that as well okay the. Final little point here is that you've already seen the. Jet yak we now are, using a number of different mobile platforms for. The IFC be in the ESP, here. You see the jet yak I know that Harbor branch here has a wave. Rider which. Uses, wave energy, to move autonomously, through, the ocean, the, IFC B has been put on this and can be be, be again robotically, just move all over the place taking samples we, put it on board vessels, that can move rapidly. And give us maps of organisms, I show. You this because the ESP, is now in its third generation that. Down outed and Bari this. Device. Here. Looks. Here this is now the ESP, so now it's gone from the size of a trash can to the size of a soccer ball and it. Fits inside this, vehicle which is kind of cute here here it is this, waves diving, here's this little vehicle here, which, which, can dive come up to the surface send, its information, go, back down you can tell it where, to go and it's got an ESP, inside, it so this dream that I showed you what we think a, future, monitoring. Program. Should look like can change before. I showed you either, ESPs, or iri, FC, b's moored along a coast, or situate. The long coast to give you early warning, to give you data for Loomis. Model. Assimilation. And. Estimates. Of toxicity, but we, also now, can, go mobile that's. Sort of the next phase of all this so, we can have an array of moored and autonomous. ESPs and IFC B's and other instruments, providing. Real-time data, on the distribution. Toxicity. And even the physiological. Condition, of these, cells so scientists. Are supposed to dream this, is mine, and I know that pieces are there, it's. There's a lot of expenses, there's a lot of trial to go but this, is not too far-fetched believe, me so my, final. Thoughts for you I hope. You believe that. The hab problem, the red tide problem, is serious it's, a global, threat to. Health and coastal, fisheries aquaculture, tourism. Water security lots, of things and it's, not going away it's not gonna change it's. Been around for many many years it's worse now I haven't gone into that at all worse for lots of reasons, in most, parts of the world but. It's it's not gonna go away so we have to manage it, numerical. Models, of hab, dynamics. Are advancing, rapidly you have some in this region, and they. Are providing forecasts. At both weekly, and annual. Time. Skills but. In all cases the accuracy, can be greatly improved through. The use of this in situ. Biosensors. That I like that I've showed you and two. Instruments, at least in the northeastern, US right now show. Great promise that's, the IFC be in the ESP, but in addition to their use in providing data for the modeling, and forecasting. They. Are providing insights, not, possible, with traditional sampling. And modeling approaches that's what I'm an about we're looking, into the life history, of this organism we're, looking into their blooms we've got cameras. Right there watching.
It Happen it's. An amazing, new, which it's a transformational. Technology, you hear about Amazon, and others being transformational. I believe, that about this technology and in, particular, what, it's shown us is that the cells swim, faster, divide, faster, for, more gametes, and are more toxic, than, Sagitta and we thought. And. The. Bottom line is that if, we're going to understand, and forecast, these, haves here. Florida. New, England wherever. Characterization. Of critical, rates and behavior, in these. Natural populations. Basically, that's what we need and those, need to be generated whenever possible through, in situ observations. Concurrently. The bio, sensors that we're talking about also can provide early warning information. As well as cell abundance, for the. Forecast like we said many. Of the technologies, and approaches, described, here can be used on Florida hams and that's my sort. Of what I'm saying to all of you and my colleagues here at Harbor branch this. Is a wonderful opportunity for, collaboration, I think there are things, that can be done here to greatly expand, your knowledge of what's going on especially in the Indian, River Lagoon very. Very important, site perhaps and I hope. That something may come of it and we could work together and. Maybe I can come back and talk to you again but thank you for your attention. You.
2018-04-21 10:13
