[Automated voice] This conference will now be recorded.

[Emily Lemagie] All right good morning everybody and welcome to another EcoFOCI Seminar Series. I'm Emily Lemagie. I'm the co-lead of the seminar with Deana Crouser.

This seminar is part of NOAA EcoFOCI's bi-annual seminar series, focused on the ecosystems of the North Pacific Ocean, Bering Sea and US Arctic to improve our understanding of ecosystem dynamics and application of that understanding to the management of living marine resources. Since 1986, this seminar has provided an opportunity for research scientists and practitioners to meet, present, and provoke conversation on subjects pertaining to physical and fisheries oceanography or regional issues in Alaska's marine ecosystem.

You can visit the EcoFOCI web page for more information. We thank you for joining us today. You can please make sure to mute your microphones and that you're not using your video during the talks. You are welcome to type your questions into the chat and we will come back to those at the end of the talk.

And our first speaker today is Phyllis Stabeno. She's a physical oceanographer at the Pacific Marine Environmental Lab. She conducts research into the impacts of climate change on high latitude marine ecosystems.

[Phyllis Stabeno] Thank you, Emily.

[Emily Lemagie] You're good.

[Phyllis Stabeno] Ok.

Maybe.

[Deana Crouser] It might have to be turned off.

I couldn't get it to work last time. Why don't you do it, Emily?

[Phyllis Stabeno] It's not working still.
Pushing all of them.

[Emily Lemagie] There you go!

[Phyllis Stabeno] So good morning everybody. I'd like to talk about the history of the M2 mooring and some of the other long-term moorings in the Bering Sea. This M2 has been out for almost 30 years and why it may not be a team of thousands it certainly was a large number of people over these years that have been able to maintain it. Funding has come from a variety of sources besides our typical ones. AUS, NSF, the Arctic Research Program and even the Poly Commission provided money over two years.

So EcoFOCI began in 1984 and we began observations in the Gulf of Alaska, moving north into the Bering Sea. And finally into the Chukchi. As I say, my interest is looking at the moorings. At the moment there's five long-time term moorings in the Bering Sea.

Next Slide.

M2 in the south, which was deployed like I say, for the 29th year. M5, sorry, M4 and M5 and M8 and you began three years ago, a mooring north of Saint Lawrence. So this is the time series, this temperature time series at M2. And just, you know kind of a short history of it in 1993 we were inspired by the TOGA-TAO array for long-term moorings, in the Bering Sea. The next year we begged, borrowed, and stole money. We didn't steal, we just begged and borrowed money. And we deployed the first mooring in, from the Freeman's winter cruise in 1995. That was not one of our great successes. The ice was far, fairly far north. We gambled that the ice would not come down and within six weeks the ice had come down in the mooring.

Fortunately we found it later. It's not the picture up at the top, but ice does what mooring surface what you see in the center there. Basically rips everything off the top. But we began to put another mooring out in April, May of that year and that began this time series, at M2. So the first year we did deploy in the winter. And we then formed the rhythm of deploying a surface mooring in April, May, and a subsurface winter mooring in September, October.

The surface mooring had measurements at about every 3 meters of temperature. And the upper 30 meters or so, and 5 meters below that. And scattered within that was temperature, I'm sorry, salinity fluorescence, etc. for the NAVCP. So one of our big concerns when we began M2, we had looked at how many fishing fleets, boats were around it. And M2 is a very popular place. But it's where we wanted to deploy for ice. And so we named the mooring after Peggy Dyson who maintained a radio channel up there, talking to all the fishing fleets. And she would say please don't pull up my mooring.

And I think that was very successful for us. Over the years we had two times, a fishing boat, yanked up or pulled up, part at the mooring. So those are the next two gaps in the data, and then of course Covid. We had Covid on a ship. We couldn't do the cruise, etc. And then between that and the sea ice. So this is the time series temperature from zero to about 70 meters over all the years. The warmest temperatures occurred in 2019. They were almost 16 degrees on the surface.

So if we also look to the far north is M8. It also had an interesting history. We decided we wanted to deploy in the north and so we put a test mooring out because we weren't sure how well it would go. And we wanted to measure at 15 meters. We lost that mooring. The ice age, some later work that Peggy Sullivan led, showed that ice heels at M8 can be 22 meters. And that, so in the end we decided to design the mooning at M8 to go from the bottom to about 20 meters. And gambled with it. We've been very successful up there, take a look at it.

The fishing fleet isn't up there and we've successfully done it. For the last about, 19 years. Our problem of course this last year was Dyson broke down, and they did not turn around the mooring. So we'll see what happens to that. But you can just looking at it, you can see the same patterns. You know M8 shows the nice deepening in the fall. And the sending of heat to the bottom and then the cooling as ice comes over.

Next slide. If we look at this, it's easier to see the same patterns, when you look at the two anomalies. So you see the warm periods that since 2014, and both sides see the colder things from 2006 to 2013. And we began in 2005 at M8 and there's indications that that was slightly warmer also. So what we're seeing you know when you look at M8 you think, oh you see warming. Maybe not. And then that is, we need to do a bit more work over there see what kind of data we can find, etc. Next slide. So there's a lot of ways of looking at this data. You know the depth contours are better.

So the top panel for both M8 and M2, are the depth. Integrated temperature from each of those mooring sites, and we subtracted off the annual signal. And so you're now looking at it; the anomalies. Once again at M2 you see the warm period from 2000 to 2005. The cold period from 2006 to 2013 and then the change for 2014. The same pattern, since 2005 occurs at M8. So you're seeing the whole shelf change together. It's not...we alI...I thought that the south was more separate from the north. They are connected, the timing and this all depends on ice as I'll discuss later. But there's a very great similarity of what happens across the ship.

Next slide. So let's take a quick look at M8 and you say, okay, why is temperature important? One of the importance is for the Al Hermann's model. He uses the moorings all the time to tune etc. his model. But the other is just how things function. The center panel is a percent ice cover in a 25 km box, square km box, around the mooring site.

The bottom one is top and bottom temperatures. And you know this turns out to be pretty important. Next slide.

For instance pollock have a preference of waters above 1 degree, 0 degrees somewhere in there a little bit. This line here is
about at minus .5, and this is what you see.

2005 had some above, the part above the line, shows the duration of temperature. Above the, this is minus .5. Similar for zero, etc. And what you, see beginning in 2014 is a marked increase in the duration of that. That goes with what we've seen for Walleye pollock distributions. This warmer bottom temperature provide a corridor for them to go north. If you look at the C2 or the Chukchi moorings, they also have warmer temperatures on the bottom in the winter. So it's a habitat that they can go into. Historically those temperatures were minus 1.6 for most of the summer. And you get a little bit of warming in the fall, and then it would cool very rapidly.

Basically had no warm autumn temperatures but what we've seen since 2014 is a marked increase in the duration of this warmer temperatures on the bottom. Which makes it open to many species. It also causes problems in all likelihood for crab, snowcrab, who don't you know who prefer the colder temperatures. And so these are what these long-term measurements tell you. It's how things change and what the patterns are. And yeah this is just from M8. And yeah every time I look at M8 I say, oh I wish I'd begun in 2000. To have a longer time series. But I think that's inevitably everybody's wish for the beginning of these long time series.

Next slide. So what causes all of this stuff? It's sea ice. The presence of sea ice. This is a plot of the aerial ice extent on Eastern Bering Sea Shelf, for three months: March, April, and May. And the most extensive sea ice, in April, occurred in 2012. The whole Bering Sea Shelf was covered. Just a few years later in 2018 there was virtually no sea ice. The shelf. And shown in the lower right is the aerial map of maximum sea ice, at '18, and maximum.

Next Slide. The difference between these two is 700,000 square kilometers of ice cover. That is bigger than the area of California and Oregon together.

And most... or most of the eastern states. This is a huge area, where there was. The ice extent, advance and retreat is among the largest in the world. And the variability shown here, is also immense. You can see historically there's very little to nothing [indistinct]. Next slide. 

So, let's look at these time series and add ice into it. That's the kind of purplish color, you can see, you know just looking at the ice. M8 looks like you had periods of mainly 100% ice cover and then beginning once again in 2014 you see some changes. The bottom one is M2.

M2 was chosen because it always, almost always had ice at it, as we did the analysis. You can see now there's long periods where there's virtually no ice at M2. 

Next slide. So here's the comparison you get, you know, the similarity, the extent of ice, excuse me. Are there shown in the bottom is the average ice extent in this 25 km.

M8 is extensive to it's variable, maximum ice extent, and that in late March. So this was, I just did this, I thought it was really interesting. This is the anomaly of ice and it shows how sharply things have changed. Once again 2014 appears to be the shift year for us, when you look at the average temperatures, you can see how they agree with the presence of ice and the absence of ice. Years with a lot of ice, you get cold temperatures; years without ice, you get warm temperatures.

Okay. So, so that's what we've done, as we said these were measurements. You know, early on every 3 meters of temperature at the top. Every 5 meters in the bottom. And this is the type of mooring design we had. There's a Met Package. You have T, S currents, chlorophyll fluorescence and then once in a while we have nitrate and oxygen and PAR.

And Catherine Berchok always put one of her passive acoustics forms near this. And if you look at fluoresence that's the type of time series you would get. So we partnered with the ITAE program and we changed the upper part here. We now have a Prawler here, that goes up and down. That is the type of data you get in the upper 50 meters. A number of people have begun to do papers on this major advance in technology. Including Jens, Noel, Calvin Mordy and I think Haley is also going to discuss a bit of this in the next presentation.

Thank you.

[Applause]

[Emily Lamagie] All right. We have some time for questions. I don't see any in the chat yet, so we can start in the room.
Shaun?

[Shaun] In the ice anomaly plot that you showed, it my eye wants to trace a very periodic wave in that. Is that representative of maybe like, some multi-fail signal that might be in there? PDO or some other long-term trend? Or is it still, or am I just making stuff up?

[Phyllis Stabeno] Yeah. It's too short. I think one of the reasons I showed M8 and M2, what you see in M2 and the indications you see in M2 are different than what you see at M8. And yeah there are 10 years difference in length. And 30 years provides a greater insight of variability in that type of thing. Certainly in the southern shelf there appeared there was a five, six year cycle for a while.

But that looks like it's disappeared. I think the next 5 years, 10 years, are going to be fascinating. Is it going to be a general warming with the cycle on top of it? Or is it going to get warm enough in the south that you just don't have ice in the south again because of the Gulf. And I've seen both predictions and I think it's, what you want to believe at this point. On how it is. My suspicion is you could have ice in a given year.

For instance, this year we had ice at M2, but it was just at M2. You went 10 kilometers further, [indistinct] and there was no ice. The ice only made it to the 70 meter ice of that. Which is interesting the patterns of weather said it should have been more extensive. So Wei and I are doing some work on trying to understand how warmer temperatures affect the advance of ice and, her with the ROMS model and me with just an analytic thing. And so we're trying to move forward, have a better understanding of the southern Bering Sea.

Wei.

[Wei] Yes, just this pattern, that's super interesting. Do you have any comments on why 2017 it kind of started in the sub-surface?

[Phyllis Stabeno] I will steal something from Jens. And what Jens has seen is this appears to be larger phytoplankton cells. And it I just saw the plots earlier this week or maybe it was Friday and that's it because when I first saw that I thought oh crap. Everything is upside down, but you know all the results showed this. And so Jens has done some really nice work looking at this and it looks like they're larger cells and so that meant they fell down.

The question now is how do they have enough light to exist in the whole, you know from 30 to 70 meters. And that I think is an interesting question.

[Emily Lamagie] Yeah, the bloom was before the Prawler was deployed, that year. You see the spring bloom
in the winter [inaudible].

[Phyllis Stabeno] Yeah, but that high production at the bottom is strange.

[Participant] At what rate does the Prawler go up and down?

[Phyllis Stabeno] It climbs up with wave action. And it goes up and down very rapidly. We keep one hour of data.

[Participant] Okay.

[Phyllis Stabeno] I suspect you could have more than one hour of data but that's it so it climbs up and it falls down. And it climbs up and it falls down. And it goes up and down as fast basically as it can climb.

[Participant] Okay. Thank you.

[Deana Crouser] Ok.

[Clapping] Thank you!

[Emily Lemagie] And also about a meter every
eight seconds, profiling on descent.

Great. Thank you Phyllis for your time.
[Applause]

And the next speaker is Haley Cynar. She's a PhD student at Oregon State University in the College of Earth, Ocean, and Atmospheric Sciences with a concentration on Ocean Ecology and Biogeochemistry. Her research focuses on measures of ocean productivity using dissolved gases and isotopes as tracers of biological productivity.

Alright.

[Haley Cynar] Great! Thank you, Emily and Deana. Can you hear me and see my screen all right?

[Emily Lamagie] Yes.

[Haley Cynar] Okay. So today I'll be talking about net community production rates by oxygen-nitrogen gas ratios at M2, in the southeastern Bering Sea from 2021.

And this is work that wouldn't be possible without all of these co-authors. So to start off, just a brief overview of primary productivity. This is the rate of production of organic matter by phytoplankton. And this phytoplankton production fuels the ecosystem which is then consumed by zooplankton and fish, crabs, and whales. And generally primary productivity is an estimate of the food available to the ecosystem, as like an upper limit.

And in this diagram we see phytoplankton here, in the surface uptaking carbon and producing organic carbon. And then some of that organic carbon is respired by those phytoplankton themselves. They get grazed upon by zooplankton, and there's also zooplankton respiration. That's kind of in this upper box. And then where that organic carbon ends up depends on the ecosystem and everything else that's happening, as well. So next I'm going to go over a few metrics of primary productivity

that are used in this observational study, in the Bering Sea. So first we have gross primary productivity or GPP, which is the total rate of production. For example, if you measured all the organic carbon produced in this equation or all the oxygen produced. There's also net primary productivity (NPP) which is that total production minus the autotrophic respiration by phytoplankton themselves. And this is, sorry, and then we have Net Community Productivity (NCP) which is that gross production minus all the respiration, both by autotrophs and heterotrophs.

And NCP is kind of an estimate of like the ocean metabolism, where it's the balance between photosynthesis and respiration. Where like, if you just measured the net oxygen produced for example and there are a lot of ways to estimate each of these. I'm just going to go over the basis of the methods that are shown in this talk.

So for gross primary productivity, this was estimated by diel changes in oxygen in the mixed layer from Jens. And then net primary productivity (NPP) this is what people will be pretty familiar with, as commonly measured by getting a measure of carbon uptake. Either from like C13 or C14. But here this was estimated using a vertically generalized production model which is based on chlorophyll, sea surface temperature, and PAR.

And then Net Community Productivity (NCP) here we're using the net biological oxygen super saturation from an oxygen nitrogen ratio. So that's kind of getting at the net oxygen produced here. And we have to account for the physical oxygen changes as well. So getting more into that I'm just going to go over oxygen saturation. Oxygen solubility is primarily a function of temperature. And here in this figure, this gray line here is shown as oxygen solubility concentration, with temperature. And then all of these points are observations from the surface oceans.

So you see that these points generally fall pretty close to that line, that would be estimated if oxygen was at saturation. But they're not quite there and the difference from that saturation tells the story. So for example this point is at around 110% saturation. So it's like 10% super saturated. If you look at the difference, in concentration there from that line. So that could be due to physical drivers. For example if the water warmed rapidly, that would decrease the saturation concentration and make it supersaturated before they kind of had time to equilibrate. Or from bubble processes which could also supersaturate oxygen. But it could also be from biology

through that photosynthesis equation and producing oxygen. So how do we figure out what's biological and what's physical? We can separate the biological oxygen portion of that, from physical using an abiotic gas tracer, along with oxygen.

For example, this is commonly done with oxygen argon ratios, because argon is a really good, it's very similar to oxygen in physical properties. So we can kind of take out the physically caused oxygen changes, with that. But argon is a little tricker to measure. So here we're using oxygen/nitrogen ratios and we get those from a Gas Tension Device which is shown here that measures the total dissolved gas pressure.

And when we use this alongside an oxygen sensor or an optode, we can get that oxygen/nitrogen ratio. Now nitrogen doesn't respond quite so similarly to oxygen in terms of physical forcing. So it's less perfect than argon but we can kind of adjust for those differences if we know the water mass history. So in this talk I'm actually going to be showing data from oxygen/nitrogen prime, which was developed by Robert Isett as a way to kind of like convert nitrogen into responding like argon.

Based on what we know happened in the water mass, if it warmed or if there were high winds that kind of thing. And this is pretty simple to deploy and requires minimal maintenance, which is great, in terms of getting more data with it. This was tested in a shipboard comparison in 2019, in this same region. And it worked pretty well. So here we're using this oxygen/nitrogen ratio to estimate net community productivity rates using a simple model, in the surface mixed layer.

So where was this deployed? This was deployed at M2 as Phyllis was talking about, in the southeast Bering Sea. And this area of the shelf is highly productive. It's a foraging area for seabirds and marine mammals and is also home to commercially important fisheries. And M2 has been shown to be pretty representative of this area of the shelf. So, this Gas Tension Device or GTD was deployed at 6 meters on M2 in 2021, and the mooring stayed out until February of 2022 just before ice cover.

So I'm going to be showing a few Prawler profiles showing kind of an overview of the water column 
over time from 2021, and this Prawler is located pretty much right next to M2. So we have depth on this axis and then temperature where we see that it's pretty well mixed in cold at the start of the season. And then it becomes more stratified and warm into spring. Becomes very strongly stratified through the summer period. And a lot warmer at surface and then here you see the mixed layer deepening and the water column becoming pretty well mixed, by late fall.

Okay next we have chlorophyll. This scale is attenuated, so the actual values in the bloom are much higher than five but this is kind of to show other patterns as well. So here we see the spring bloom with high chlorophyll concentrations in the surface. Then after the bloom you see much lower concentrations through summer. Since a lot of the nutrients have been used up in that spring bloom. And then at the same time that that mixed layer is deepening. And there are some strong wind events we see evidence of the fall bloom, with greater chlorophyll concentrations in that period as well.

And then oxygen percent saturation, is shown in this bottom panel where we see high oxygen supersaturation in that bloom period, in the spring. And then, percents that are a lot closer to saturation through the summer. Some of those mixing events caused undersaturation from mixing of deep water at that point. And then you can't, it's a lot harder to tell if there's a fall bloom in this oxygen data, as it's shown. But we do see like a change from undersaturated after some of those mixing events, back to closer to saturation. So some of this could be from biology, producing at the same time as this chlorophyll concentration is elevated.

So next I'm going to go into some productivity metrics. And just to start out, here we have chlorophyll concentration shown on the right axis in green. These are both 5-day moving averages. And then we have net primary productivity shown, on this axis in millimoles of oxygen per meter cubed day. And this one was based on that, that model that includes chlorophyll. So you'll see that they have pretty similar patterns for this data set. And that's because chlorophyll goes into the NPP calculations. So now we're getting rid of chlorophyll and just sticking with these three productivity estimates.

Where we have GPP in red, which is based on that diel change in oxygen. And then we have NPP in purple and NCP, in blue, which is based on that oxygen/nitrogen prime ratio. So, and then these are also 5-day averages. So kind of initial impressions we expect that gross primary productivity is greater than net, which is greater than net community production. That's kind of based on the definitions and we do see that that holds true for the most part. Despite these being three different methods that are based on different data.

Let's see, so kind of focusing on the spring bloom, to start out with. We see that all of these metrics have pretty similar patterns, with the peak in the bloom. There are a couple things to note about that. One is this like dip in net community productivity. And this is, in part because there's a mixed layer depth, conversion that goes into this, to get it into volume units shown here.

Which correspond to these two which don't directly have mixed layer depth in those calculations. Another thing you might note is that these gross primary productivity and NPP rates, kind of drop off around the same time.

When the bloom is coming to an end and a lot of the nutrients have been used up. Whereas the NCP rates have a little bit of lag, and this is based on the way that we estimate NCP with a steady-state model where like there's a lot of oxygen produced during this bloom, and then all the oxygen doesn't just disappear. Once like the production slows down it takes time for that to be ventilated, by air-sea gas exchange. So that's kind of just an effect of the time scale of the measurements.

[indistinct] spring bloom we see much lower production for the summer period. And then with GPP and NPP we see kind of a small peak around that fall period, corresponding to the elevated chlorophyll from the Prawler and a fall bloom. The NCP data here isn't the way that we estimate, NCP has some assumptions in it that aren't really valid during this time. So I'm still working on getting a better estimate for NCP in this period.

So compared to previous years we're just looking at the spring bloom here so this table on the top is an excerpt from Jens' paper where he calculated rates of GPP, NPP, as well as NCP, for a variety of years. For each like bloom period and these rates are in milligrams of carbon per meter cubed day. And then below here I have the estimates from the study from 2021, for the bloom period, with the same units milligrams of carbon per meter cubed day. So and then I should mention that GPP and NPP are calculated based on the same methods from this paper versus the study whereas NCP is measured differently here.

Where the 2021 data is based on that O2, N2 Prime, from that gas tension device. So general takeaways here, we're always going to expect some interannual variability and some of these were really warm years. But we see relatively similar values in GPP and NPP, as well as NCP. For these years they're kind of in line with what we expect. Also shown here are the ratios of net primary productivity to gross, and NCP to gross primary productivity. Which are also relatively similar to those from that previous study, where for example like this value of 0.5 for net community production to gross primary productivity, basically means that half of the production from this balloon period, is available

like for consumption or for lateral transport out of the system or export to the seafloor. That kind of thing. And this is pretty consistent with what we expect from a bloom. Whereas after the bloom period when nutrients are a lot more limiting, we expect this to be much lower with like more recycling in that surface portion. And then just a quick comparison to another method. So if you sum up, all the production based on NCP rates here, for basically the month of May or from when this was deployed, until June 1st, we get around three moles of carbon per meter squared that was produced.

And if you estimate how much carbon you would expect based on the nitrate, in the surface, which Noel Pelland calculated, based on this paper you get around 3.66 moles of carbon per meter squared. So those are roughly in the same ballpark for how much carbon is available to be or, sorry for, like the carbon available based on the nutrients that allow that production.

So kind of to sum up overall this data provides a more complete picture of the seasonal patterns, for what's happening in this region of the Bering Sea. And also contributes to like our understanding of both interannual variability and those seasonal patterns.

So next steps for this, I think this is a neat method that really enhances our like, the, it makes it so we can get a lot more productivity estimates, in places where, ships can't spend that much time.

So, to kind of enhance these measurements a GTD and optode have been installed on the Sikuliaq since late 2020, continuously collecting data wherever they go. This Gas Tension Device was deployed again this year in 2023, at M2, and then there's also potential for more autonomous deployments, for example, on saildrones in the future.

And then back to some of the data that I showed I'm also going to be working on quantifying that fall bloom, based on NCP rates from the Gas Tension Device by incorporating the entrainment of deep water during that period when the mixed layer was deepening.

So that's all for today and like to acknowledge a number of funding sources that made this possible.

Thank you.

[Emily Lamagie] Thank you, Haley. Do we have any questions from the room? Al?

[Al] Yeah. This is great stuff. Realizing you only have a few data points for the ratios, do they in general line up with the idea of a bigger ratio of net community production to gross primary production under cold conditions? Than under warm?

[Haley Cynar] That's a good question. So the comparisons that I showed before were just for each bloom period. So time wasn't really considered and that it was just selected based on, like the two weeks on or, the like week on either side of that peak. But yeah, I would have to look into that more.

[Al] Okay, thanks.

[Emily Lamagie] Are there any questions from the chat? Or someone wants to unmute themselves and ask?

[Participant] Yeah, I have one.

[Emily Lamagie] Yeah.

[Participant] So when you were showing the Prawler data, it seemed when you had like both with temperature and all the other measurements, that you would have a bloom and then it was just, done. Like it seemed very stark. Like where like within a day like, I noticed with a lot of the temperatures it's like temperature is this temperature, then all of a sudden the next day it goes right down.

Now I don't know if that has something to do with how you smooth the data. What the way it's done or something. But especially with the bloom data it seemed to just cut right off. Like you had it and then it was over. Is there, can you comment on that?

You're muted.

[Haley Cynar] Are you talking about the, productivity estimates like showing GPP and NPP?

[Participant] No, it was the probability [indistinct] time.

[Haley Cynar] Yeah, there wasn't any smoothing done to that data that would have caused that.

[Participant] That's really interesting to me.

[Haley Cynar] Yeah, I mean one thing to look at
would be like the mixed layer depth changes.

I don't know that there was anything, like significant,
that happened towards the end of the bloom.

I think it was most likely just that the nutrients had been consumed at that point, and the like dominant phytoplankton were kind of at the end of their life cycle.

[Participant] Was just, it was really interesting to see that.

[Evan] Hi Haley, Evan.

So I think that chunk of the Bering Sea is kind of a hot spot for sedimentary denitrification and oxygen demand.

Does that do anything in terms of your mass balance or the Gas Tension Device? You know, I don't I don't really know what assumptions go into the Gas Tension Device that might be modified by those sort of extra flexes.

[Haley Cynar] Yeah that's a good question, so that would have an effect. And it depends on what time period you're looking at. For a lot of the calculations, where I'm saying that like the data is valid. That's when we're assuming there's not a lot of vertical mixing.

So a lot of so, like I would be assuming that those fluxes that are happening at the seafloor are not affecting the data at the surface. But when there is a lot of vertical mixing or entrainment of deep water that would affect those estimates.

[Evan Thank you.

[Emily Lemagie] Shaun.

[Shaun] Hey Haley. So I know that characterization of the oxygen optode is really important in order to, you know, from our own discussions. Are there any other particular measurements that you think that you would request to make this data set a little bit more useful for you in the future? Not just for our own field work but you know for the associate fieldwork on vessels like the Sikuliaq and all. Lessons learned that we could pass along, to get better data sets to work with.

[Haley Cynar] Yeah, I mean like you said the oxygen is very critical to those calculations. So I mean, I would say that the most important thing is just to get calibrated oxygen measurements more often. And like as often as possible, especially at the deployment and the recovery, to yeah, take Winkler samples and show that we can really trust the oxygen data.

As far as other sensors, I would have to think on that more. But I think that the number one would definitely just be getting oxygen data that we're confident in.

[Deana Crouser] Well, let's give Haley another round of applause because we were muted the first time. She didn't hear how loud we were.

[Laughter]

[Applause].