[Presenter] This conference will now be recorded.

[Deana Crouser] Good morning, everyone, and welcome to another EcoFOCI seminar series. I'm Deana Crouser, co-lead of the seminar series with Heather Tabisola. This seminar is part of NOAA's EcoFOCI biannual seminar series focused on the ecosystems of the North Pacific Ocean, Bering Sea, and US Arctic to improve understanding of ecosystem dynamics and applications of that understanding to the management of living marine resources. Since 1986, the seminar has provided an opportunity for research scientists and practitioners to meet, present, and provoke conversation on subjects pertaining to fisheries oceanography, or regional issues in Alaska's marine ecosystems. Visit the EcoFOCI web page for more information at www.ecofoci.noaa.gov. We sincerely thank you for joining us today as we continue our all virtual series. Look for our speaker lineup via the OneNOAA seminar series and the NOAA PMEL calendar of events. Did you miss a seminar? Catch up on PMEL's YouTube page. It takes a few weeks to get posted, but all seminars will be posted. I ask that you please check that your microphones are muted and that you're not using video. During the talk, please feel free to type your questions into the chat and we'll be monitoring the questions and we'll address them at the end of the talk. Today, I am pleased to introduce Peggy Sullivan. Peggy is a research scientist at EcoFOCI, an affiliation with the University of Washington's Cooperative Institute for Climate, Ocean, and Ecosystem Studies, CICOES. Sea ice in the Bering Sea and Chukchi Seas have been a long term topic of interest and focus of her work, where both seasonal and multi-year changes and marginal ice affect the water column and ecosystems. And with that, let's begin.

[Peggy Sullivan] Okay, thanks, Deana. And thanks, Heather. Here we go. I am Peggy Sullivan, in the lab, but not everyone knows my evil alias Margaret, the banks all know that. And that's what's on my papers, so. I'm gonna talk about the seasonal ice in the Chukchi Sea in the context of nine years of ice draft data that we collected at various sites around the Chukchi Sea. Oops, but I think I'll start with a little bit of history of ice-related topics in EcoFOCI or before EcoFOCI. When I came on board quite a while ago, there existed CTD casts in the Chukchi and Beaufort Sea in the '70s and early '80s. And at that point, we weren't doing any work up there, so we don't really know what the projects were up there, but the data exist and they're kinda cool. And at the time, we were working in the Gulf Alaska and Bering Sea, in the Aleutian Islands. In 2006, we got funding and we're working through NPRB and the BSIERP program and we set ourselves up to collect ice core samples. So over three years we collected 30 samples, and there's a picture of them there and you can kinda see snow on top of one and coloration and banding in one. We sliced them up and we basically made profile data out of that for salinity temperature and nutrients. It was pretty interesting working on the ice as well. And the picture below shows one of the first trips onto the ice from a ship during the first cruise of that. And then this is the second or third year. Ned doesn't look too unhappy having fallen into the ice [chuckles]. But I think he's on this call, so we got him back. In 2019, we started working with BOEM, CHAOZ, ArcWEST, CHAOZ-X, and the DBO program, along with usual NOAA stuff. And we put out, set up a series of mooring locations in the Chukchi Sea. And it was in 2010 then that we set up the ice profilers at a number of stations in the Chukchi Sea and then that work continued as we started into the Arctic IERP program in 2017. 2018, actually 2018 to '19 was the last year we collected any ice profiler data mostly due to funding and pandemic issues. So a little description of the instrument that we're using. It's called an IPS5. It's an upward-looking sonar device. The photo to the right is the casing that contains the instrumentation and the memory card. And it's manufactured by ASL Environmental Sciences. The fuzzy picture in the middle is their cartoon that shows how these things are deployed. They're pretty close to bottom. And ours are about 10 meters off the bottom. The little, the trans, yeah, the signal that's going up there is basically the trans, sorry, the instrument up to the bottom of the ice edge. And we're getting range data that gets turned into draft data. And ice draft is from the bottom surface of the ice up to the surface of the water. So it's not to the top of the ice, but it's to the surface of the water. And that's usually considered to be 80, 85, 90% of the ice that is on the ocean. The very wordy table there are the specs for the instrument. And if anybody wants that information, I could get it to them. But right now, I'm moving on. Here's a map of the area. There were originally seven stations that we were going to use. And these stations were also being used by other projects for biophysical and biological moorings in ADCP. C8, we deployed up, if you look in the center, the map at the very top, we deployed once and we had a failed deployment so got zero data out of the whole thing. So basically, we have data from six moorings and we have focused on C1, C2, and C3 that we call the Icy Cape Line. And so if you look at the table to the right, C2 has the most data every year that we've had it out, and that's followed by C1, the next largest amount of data C3, and then five and six doesn't have, don't have much at all, and I haven't spent too much time on them yet but that's coming. And I also wanna mention that in 2015, we put one of these ice profilers at M8 in the Bering Sea so a data set exists for that, but it's not part of this talk. All right, let's look at the ice and just get a sense of what these data look like and what the shapes mean. This is an ice data record that I get out of the ice profiler or once it's processed. So this is ice draft. The blue is data that are taken every one second for the full duration. So it's really quite a load of data. The line on top, it's a red line with yellow dots, is daily averages of the same data. Here's a sense of what you're actually looking at in these shapes. So to the left are waves and to the far right are also waves, but the left is fall, so you're basically following summer into starting the freeze up. The center of the plot is all ice, once the ice has, or has been established and is there in the winter in the seasonal ice zone. And there's a shape that's labeled wave-ice transition. That is just that, the tall things are waves and in between the ice has kind of dominated that time period. And then more ice is labeled to the right. So there's an arrow in that bottom plot to the left that shows where I've taken a bit of time and expanded the plot. And this is during waves before the freeze up has started. This is 10 minutes worth of data, 600 data points. And there are about 10 waves per minute. So I'm looking at my notes. 10 wavelengths per minute, so the wavelength period is about six seconds and it's typical for surface waves. This particular plot shows two meter maximum amplitude in the waves. So the next expansion of the data, you see the down arrow in the bottom plot in the center, is right smack in the middle of ice cover. And this is pretty typical what we see also. So even though the one second data at the bottom looks really dense and deeper than that, once it's expanded out, the ice depths are, or the draft is zero to two meters with a lot of exceptions to that rule, a lot of deeper things that come through. This is a 24-minute period with 1,440 data points. And this is in February. And the last expansion I did on this data sample is that odd peak down, down around May 20th, if you look on the bottom plot and the arrow is pointing to that. One might think, if you look at the data right away, that that would be a problem data section and maybe a mistake, but it isn't. You expand it out and it looks like the plot above. And this is really something typical that we see that you have pretty obvious wave forms that start and then get stopped or are interrupted by a keel with quite a number of data points. So overall there's 2,400 data points in this whole plot, but the keel comprises 265 of those points. And it's a little under five minutes and 40 minutes worth of data. Okay, now I want to show the comparison that we've done with ice concentration satellite data. So the previous data we looked at was from 2017 to '18. And the years that I'm referring to are ice years and they start in the fall and end late summer. So this is 2016 and '17, a year earlier than the ones that we saw in the previous slides. And so to the left of most of these plots it's fall and to the right it's summer. That's pretty typical all through this talk. So in each of these three segments, well, first, there's the map that's repeated here, the Icy Cape Line C1, 2 and 3, and to the right of these plots it's labeled C1, 2, and 3 to give you some bearing on where these data were collected. So ice concentration was pulled for each of those locations and that of the pairings, that's the bottom, it's labeled there with the green arrow. So the red box is generalizing when the transition zone is from summer waves to winter ice. And especially in fall, I compare fall to spring when I look at the whole record. And spring is a really chaotic time, that is a little hard to pin down of exactly when it happens when the transition happens. But fall is a little better defined. The transition doesn't happen on a dime; it takes time. But it's better defined over years with a lot of variability when the transition starts and when it is complete and ice is finally covering the area. But the ice concentration data agree quite well with the ice drafted data that we've got. And I also wanna point out that in the yellow box I'm generally framing the winter months, January through March, the three-month period where we persistently see this signal of ice looking a little deeper and having more volume. And this at, wait, yeah, at C2 we've looked at this through the whole data record and the statistics show that, on a daily mean, the ice does show a deepening from January, then to February, then to March. So of the ice year, it's the most reasonable time that there would be full ice cover as opposed to kinda chaotic breaking and waves intruding and things like that. All right, jumping, well, actually back to the transition topic. I've used this graphic in a couple of posters over time and it is also a generalization of me going through the data, this is all at C2, going through the data and trying to figure out, well, let me say the colors first. The blue are waves. That's summer. That's summer ending. The red is what I've chosen as the transition zone between the summer waves and the winter ice. And the light blue designates that winter is here and ice is here and they're the dominant medium. So the red are the transition zones. And I went through the data and kinda eyeballed when the waves weren't more dominant when the ice was more dominant, when waves really weren't breaking things up very much anymore. And then based on some of those rules, it shows when the transition started and when the transition ended and waves were mostly gone. No guarantee of that, but they're mostly gone. Anyway, if you kinda eyeball the red areas that are the transition zones, it was interesting to look at especially 2012 to 2017 of the trend of the transition zones and the ice formation trending later in the fall. And this really agrees with ice concentration data. Okay, now I'm onto ice keels. So ice keels are also interesting in the data signature. We're kinda classifying deep ones as greater than 20 meters. And they happen regularly, but not with a lot of frequency. This is a scatter plot, showing all the keels that are greater than 20 over all the years at C2. So 2011 to 2019. Kinda interesting to note the variability. Many of the points occur in April and May, but there's definitely quite a bit of activity March through early June. And then if you look at maybe 2013, it has five points, but two of them are January and February, and two of them are June and July. So that is fairly anomalous with the rest of the keels that we've come up with. 2014 has no keels. And 2016 in particular has a really large keel count, so much bigger than all the rest so that kind of skews how many keels are in what months. But April and May are the dominant months for these to exist. Okay, moving onto, I'm looking at MODIS True Color images. On the left is an image that also shows the Icy Cape Line. It gives you a good sense, well, it actually gives you a really good sense of what this area looks at at different times of the year. Along the coast there, there's fast ice that you can kinda see, I don't know if you can see my mouse, but shorefast ice along here, but often there's a coastal polynya that pulls away here, so the darker color is the polynya or what the, at the place where the polynya would form. So that definitely impacts the data signal that you'd see at C1 because it's influenced by that polynya action. This plot, so the blue or turquoise line down through the middle of this plot is the time that this satellite image was taken. And I plotted the data plus or minus, well, it's, yeah, it's one day plus or minus 12 hours around the time of the picture, just to show what the data or what the ice signal looks like next to the visual of the satellite image. And it's pretty interesting. C1 is pretty flat, as you might expect with little tendrils coming down, but they're not very much to write home about, they're not very big. C2 really has quite a bit of ice signal down to its 17, 18 meters. And if you look at the picture, it's sitting right on an ice flow. We don't know, plus or minus, whether that ice flow has moved very much or how it's moved. And before the time of the photo you have the deepest keel. You also see some signal that is small and so it definitely designates, suggests movement of the ice flow or the instrument picking up data at the edges of the ice flow at times 'cause it's a moving system. C3 looks like it's in a lot more ice, but if you look carefully at it, the ice looks pretty broken up. And the plot here reflects that. So you have clusters of ice and clusters, or sequences of time that are close to zero that are very thin ice. Okay, so this is still the MODIS True Color images. And this is from 2012. I went through those pictures and found a 20-day period that was pretty visible, and that's a little unusual. So I made a little movie out of it. You might take note that those, that this is skewed a little bit and the north indication is up in the upper right corner. But those three markers, there are C1, 2, and 3 from right to left. And I'm gonna, play the movie, probably run through it a couple of times, but we start on March 14th, 2012. So a lot of ice cover, a lot of cracks in the ice, but you get to watch what it does. Yeah, so big breakup. All right, we're playing it again. So it's kind of interesting to see the different scenarios that each of the instruments would look up and see in this scenario as all this ice breakage is occurring. So definitely C1 sees a lot more variability. C3 out there is kind of outside the riffraff for a while anyway. Okay, enough movies. So based on that movie, or from that movie, these are four stills from the movie. So the movie was March 14 to April 2nd. This is March 29 to through April 1st. I keep movin'. And these are the stills. The orange markers are C2. I put blue markers on a couple of ice flows just so your eyes can follow where the ice flows are moving, which is interesting. It's because the whole system is moving and changing and it definitely gives you an idea what the instrument is looking at, moving over--

[Woman] Hey.

[Man] You all right?

I'm hearing somebody talking.

[Man] I just wanted to--

The middle frame is a plot that coincides timewise with the pictures. So it's four days of the ice draft data. And then the--

[Woman] No, I have the adherence word thing on my calendar.

Oh, you do?

Yeah.

[Man] Oh, do you wanna--

[Woman] But I should probably RSVP--

The two lines connect down to the bottom and it's an expansion of a shorter term out of that plot just to expand and show the ice keel and the roughness factor of that and how much data are included in that keel. I meant to mention too that in each of the image frames, the vertical black line is the time of day that the satellite picture was taken. All right, we're jumping to a comparison of the C1 and the C4 station. So as I had said before, I mostly worked on the C1, 2, and 3 data, the Icy Cape Line. C4, as shown up in the map, is closer to the coast and C1 also, it's not close to the coast but closer than some of the other stations, so the comparison is of interest. And really my one takeaway from this so far, and I have another couple sets of data to put together to add to the story, is that C4 station seems to retain more of an ice signal a little bit later into the summer. So for C1 in 2014, it goes, ice is really a strong signal 'till about mid-July. And C4 station in 2014 is a pretty decent signal through the end of July. And then 2015, there's less of a difference but the ice, the bigger ice signal ends a little earlier at C1 than it does at C4. So I haven't really delved very far into that yet, but that's an upcoming thing to look at. All right, and then this is kinda coming up on the close, but this is box plot statistical analysis that we did at C2 for all but one of the years of data that we have. We took, or I took the data and daily data and subsetted it into four bimonthly sets. So November, December I combined, January, February, March, April, and then May, June just to see what the statistics of the ice in those groupings looked like. [paper rustles] So the box plots show the median of the daily data and they show the 25th and the 75th quartiles, the whiskers show the extreme data point value limits and the outliers are above the whiskers or the flat parts in red. The outliers were defined as values more than 1.5 times the interquartile range. Additionally, that number above the ΔD we basically calculated change in the draft so ΔD is draft, from the median of the November, December grouping of data to the median of the March, April. So you're basically getting the slope between the first grouping and the third grouping and it indicates the increase of ice over those months. And it pretty strongly supports, especially what I spoke about before, the January, February, March, winter months, where it's pretty solidly ice covered, well, not solid, but it is ice covered and it increases over those three-month periods. Yeah, that about sums it up what I'm gonna talk about today. We just finished, or I just finished a paper with Phyllis for the Arctic IERP DSR-II second special issue, the title is there and it covers these and a bit more working with the data. And then I guess I just wanted to mention, when we first started collecting these data in 2010, we were working with BOEM and the oil companies were up there pretty actively and it was interesting to get the first sets of data processed far enough so that we could see the signals in them and we were seeing these keel depths that were in the high 20 meters, like 28, 29 meters deep and that was a noteworthy thing to report back to BOEM and it was included in the Department of Interior report that spoke to hazards of drilling and putting things on the bottom up in that area. I've also worked with Catherine Berchok, that they are seeing walrus up in the Bering Sea earlier in the spring and she found a few signals at C3 of walrus soundings right after spring keel events. So that was also the topic of a poster and needs further looking at. And listing the funding agencies there. And then finally, the summary. We have nine years of data and 28 data sets over 6 sites, mostly at C2 and mostly on Icy Cape Line C1, 2, and 3. Also, noting that there is one site at M8 in the Bering Sea, but no others in the Bering Sea. There's a seasonal pattern in an ice year. The ice year is fall to late summer, that season starts with waves, it has a fall freeze up, there's a winter ice period, there's a chaotic spring melt, and then the melt goes into summer waves in the seasonal ice zone. The timing of ice development in the fall is trending later. Winter months January to March generally have more full ice cover and the ice volume increases over each of those three months. Deeper ice keels greater than 20 meters occur regularly, especially March to early June, with most occurring in April and May. And the frequency of these events vary year-to-year. And the ice draft compares well to satellite ice concentration. And with that, I'm ending. And thanks for Phyllis for the help and Ron, who's my husband, for making the movie. And thanks for coming and listening.

[Heather Tabisola] Thank you, Peggy. I jumped us back on mute. You did that in record time, by the way [laughs].

I did?

[Heather Tabisola] You did. Yeah, no, it's really fun to hear all of the effort that you've been putting into this. You didn't explain to people what a pain it is to pull this data, but somebody who has had nothing to do with your project but has heard the pains. It's really nice to see this. And Ned also says, "Wow, that is a fabulous data set." And Jessica, "Fantastic talk. Thanks, Peggy." So if folks have questions, please feel free to put them in the chat. And if you'd like to ask it yourself, just let me know that you have a question so we can call on you. And I was actually, Peggy, I was gonna ask. Knowing that there are trends in ice declining and thickness and all of that, you said, the paper you referenced at the very beginning, there's sort of the known, like you see about 85% of the ice to the surface and then there'd be this percentage over it. How old is that paper? Do you think that has changed at all? Would that change the data set in any way knowing that there's this historical sort of decline of ice?

[Peggy Sullivan] I don't think that, no, I don't think that has changed.

[Heather Tabisola] Okay.

[Peggy Sullivan] That ratio has not, definitely the, as we all know, the integrity of the ice and the thickness and such has changed and it has been hard to quantify. But no, that ratio probably does not change.

[Heather Tabisola] Okay. Ryan has a question. Good morning, Ryan. "You mentioned fall ice formation trending later. Is there a pattern for spring formation?"

[Peggy Sullivan] Well, spring would be the demise of the ice, but it definitely is a way more chaotic and longer, usually a longer process where ice can be gone and then come back, waves come in, break things up, then you'll have much less ice, more open space, but things can freeze up on the surface. So it goes back and forth for a while and it's a lot less easy to quantify. Yeah, and it really ranges a long time. I mean, I like that January, February, and March grouping that I've made because things can occur as early as late March into April, that really break up the ice, but sometimes not until mid-May. And sometimes everything is gone at the end of May, early June. But sometimes you get something coming through in late July or August. So it really kinda goes on and on, where, how do you mark the end of the ice season when you still have ice coming through or showing in the data set. Was that much of an answer?

[Heather Tabisola] See if Ryan answers [laughs]. He says yes, with an explanation. Let's see, I see Calvin responding to Jim earlier. [clears throat] What other questions do folks have? Noel--

[Shaun] Hi, Heather. Can I ask one?

[Heather Tabisola] Oh yeah, let me just read Noel's comment and then I'll come back to you, this is you, Shaun, right? Noel had said, "Great talk. Follow on to Ryan's question. Do we know why fall formation is trending later?"

[Peggy Sullivan] Well, in a phrase, climate change [laughs]. Yeah, I think just warmer conditions. But yeah, don't quote me, I can't prove it yet. Yeah, it seems to go along with its trend of the same signal that's shown in the ice concentration, the satellite data.

[Heather Tabisola] Shaun, was that you asking the question or was that actually you, Noel, and I just--

[Shaun] I am the one who spoke, Shaun is the one who spoke up asking a question.

[Heather Tabisola] Okay, go ahead, Shaun.

[Sean] Yeah. Hey, Peggy. So as Heather alluded to, getting the data off of, retrieving the analysis from the ice profilers is a bear. And instrumentally, the Acoustic Doppler Profilers have an ice tracking mode, and so I'm wondering if you've investigated all, combining data sets or the other data sets, again, additional information in the region?

[Peggy Sullivan] Yeah, thanks for asking that. That's a really valid question. Yeah, I think in most cases people using these instruments add in ADCP so that they can get more information. So turn it from a time series into more of a, more information about the ice. We haven't had the resources to do that for most of our data sets. I think we did it for one and those, the ADCP data have not been combined with the ice profiler data to come up with that next iteration of the data. But for the most part, we have not done that and mostly due to resources and wait on the ships and things like that.

[Phyllis] So, Shaun, this is Phyllis. Can you hear me?

[Shaun] Yes, Phyllis.

[Phyllis] There's only a few of the ADCPs. They had bottom track in them. And so that's a comparison we made. We had some failures at the beginning and that's on the list. There's also a question about the satellites and that's something that Peggy and I are working with with a graduate student. And it's not as easy as we thought. Peggy might wanna talk about how many hairs we've pulled out over that exercise.

[Peggy Sullivan] Yeah, and that exercise is working with the ERA5 ice thickness data. So yeah, we haven't really come up with too much yet on that.

[Heather Tabisola] Ned's question, in what Phyllis is referring to, is that, "What can be done," well, he says, "The variability of the keel depth amazes him so that means the overall thickness is very variable. What can be done to compare these measurements to satellite ice thicknesses?"

[Phyllis] So do you want me to address it, Peggy?

[Peggy Sullivan] Sure, go for it.

[Phyllis] Okay, so what we started with is the ice moves basically, unless it's a big giant sheet, but when it's broken up, it moves basically at the speed of the currents. And we looked at several movies and compared it to the currents and the correlations are very good and the slope is one. So yeah, they basically move with this current. So we thought, "Okay." So the satellite goes over and it has a footprint of X. And we have second data. So you think, "Well, how much do we have to average to kinda map a flow," again, average depth of the flow which should give us a height, and so that's what we've done. And it's shocking. Sometimes you have a straight line and the data are beautiful and sometimes it's a blob. And so it's difficult. It's not straightforward, but we're still working on that, and that is a paper that we will get out at some time. But at the moment, the thesis is supposed to be done in July, I think, and so it probably won't make it into the master's thesis, but we all continue working on it with our colleagues and try and get a paper out of it once we hammer it down. At the moment, what we say is this is hard. It's not obvious how it works. And there's contamination by snow is one of the problems. And we've talked way a little bit and we're hoping to do some comparison with the models of, and see if we could get any type of relationship there because it's turning out that ice thickness is real important for how rapidly the ice retreats. And so if we could come up with better estimates of that, that would also help the modeling, which doesn't always retreat ice at the right time.

[Heather Tabisola] Thanks, Phyllis. [clears throat] Anything to add on that one, Peggy?

[Peggy Sullivan] No, I don't think so. She's covered it.

[Heather Tabisola] Well, Emily also asked a question about the coupling from satellite images to the keel data. She asked, "Is there any indication from coupling the satellite images with the ice keel data of some of the dynamics during the chaotic spring? For example, is the variability likely advection of chunks from those breakup events like we saw in the movie or would a spatial average of ice coverage be useful for interpreting that process?"

[Peggy Sullivan] Yeah, I think you'd have to pick your time scale for averaging carefully, but it definitely is worth looking at. We've done some averaging over time of the data. But one second is incredibly dense and somewhat hard to work with sometimes, so we've gone to minute, daily, and early on looked at monthly, which really wipes out most of the information. But yeah, I think that definitely would be useful to find the right periodicity to filter the data and do some comparison.

[Heather Tabisola] Thanks, Peggy. Any other, oh, okay [clears throat], Ryan. "Didn't hear everything you said regarding the movie period, but it looked like the near coast flows was southward during the ice breakup. Is that typical during such ice breakups, at least in that area?"

[Peggy Sullivan] I would hesitate to say that that always happens 'cause I don't think that's true, but that kind of, those kind of pictures are really typical, if one looks through the satellite data, that motion of cracking and almost pulling things south. So it must have to do with warm influx further south and melt, having some kind of pressure on the whole ice pack and then the changing seasonal conditions, increased sun and increased atmospheric heat. And I'm sure it's supported by winds because they really drive the ice up in that area.

[Heather Tabisola] Yeah.

Sorry.

[Phyllis] Yeah, it could also be the currents. The wind and the currents are highly correlated in this area. And even under the ice, you get a very, very good correlation. And--

[Peggy Sullivan] Did we lose her?

[Heather Tabisola] Every time Phyllis is talking she's clearly with Calvin, and I just see Calvin's name and it throws me off. Let's see, Phyllis-- [laughs] Maybe Calvin dropped off. There's a person, Robert Grumbine, who, and sorry, Robert, I don't know where you work or coming from, but he made a comment, said, "Some of the complexity of spring observations from the passive microwave side is that once ice starts melting, the passive microwave becomes less reliable/accurate."

[Robert] So yes, and first, thank you, Peggy for some excellent and very interesting observations. And especially if you can get 'em near real time, that would be delightful for me because I'm sitting at the Environmental Modeling Center in the National Weather Service and I'm the ice guy. And part of my duties are the sea ice observation. Hence, my comment about the passive microwave. I'm the one feeding that into the global models, the weather models, the ocean model. But the thing is in the fall the passive microwave is a nice clean observation because cold ice is pretty, pretty well-observed. Once you start seeing some melting, or the even before melt occurs changes in the snow crystal structure start mimicking open water as far as passive microwave is concerned. So what you start seeing is those effects rather than they're actually being less ice. So this adds to the whole complexity of the spring season. But one of my other hats, which isn't near real time, is the model verification for the unified forecast system and this would be excellent data for us. So I'll be emailing you in a bit for some of the time averaged observations so that we can start to see whether the model is having a reasonable evolution of the ice thicknesses. As you said, this is very important for us, initializing the model and assessing whether it's actually going to be useful in the fall and spring seasons, especially where things are changing rapidly.

[Peggy Sullivan] Yeah, thanks for the input on the microwave. It really, it's nice to hear that added layer of the story on the fall versus spring. And I look forward to communicating with you about the data.

[Heather Tabisola] Thanks for joining us, Robert, and chiming in as well. Really appreciate that. Jiaxu made a comment, and if folks have more questions, we can do probably one more question and then we'll need to wrap up. Jiaxu said, "Just a comment to Ryan and Noel's questions on fall trending later. I want to mention that the Arctic marine heatwave has a similar feature that the prolonged period is mostly extended into the fall, but not in spring. The explanation is that the stored ocean heat in summer is responsible for the fall delayed ending," and she linked a paper in there. So if you are in the chat, you might wanna check up on that.

[Peggy Sullivan] Yeah, thank you. Great.

[Heather Tabisola] Anything, Ned, do you have a question? Now that you're on video, you're gonna have to sign my form.

[Ned] Oh dear. Yeah, so, Peggy, thanks a lot. This is a great seminar, with just incredible data. So unfortunately, my question is not about your work, but maybe you can answer it anyway. So the satellites, do you have some idea what the satellite footprint is like and how often satellites go over the top of your moorings or your ice profiles?

[Peggy Sullivan] You mean for the images?

[Ned] No, if one wanted to go through and take all the satellite measurements in your area and compare them to all the measurements you've made with all the ice keels, I'm just trying to get an idea of how big a footprint the satellite is seeing and how often the satellite images happen, the measurements happen.

[Peggy Sullivan] Yeah, definitely daily. There are different crossings, but about daily is the frequency. And the footprint, I probably know that and I don't know it off the top of my head so I can't tell you.

[Phyllis] Yeah, so, Ned, the actual crossings over aren't that often. Peggy, I think we, looking at M2, I think we only have 20 or so. It's not a huge number depending on the month and that type of thing. And we have basically 10, 9 years of data at M2 of this type of stuff. And we're not convinced at C2, sorry. And it's not the footprint so much of the satellite. The satellite goes over an ice floe, and so it's, I mean, you can have deep keels and then you have shallow areas and the floe floats not at an instantaneous depth, but at some type of average, that we average the bottom of the keel depth and that type of thing. And we haven't succeeded in cracking that idea of how much averaging to do. And that is, I think that's our real problem that we've been having with this paper we're trying to do with Mike and Gary in Florida. And so it's the instantaneous thing of what's over you and what the height of the floe. They're not the same. You have to average. You have to get some type of average depth of the floe and that hasn't succeeded yet.

[Heather Tabisola] So Robert's chiming in. [clears throat] I just wanna read it. He said, "Up to 14 times per day for polar orbiters. And then this area is on the edge of being able to get GOES-16 observations, but those are available hourly if they cover the area." And Ned, I think we had some pretty detailed info on satellite coverage during our saildrone projects prior years that Shell had provided as well.

[Phyllis] But yeah, I think a conversation with Robert would be great.

[Heather Tabisola] So I'm just gonna read Ryan's last comment and then I think we'll wrap up. And I certainly encourage folks to, sounds like everybody will anyway, chat offline. Ryan had just said, just a comment, "The fact that the standard deviation of ice completely swamps even the daily mean is nuts." [chuckles] And Robert did post his email in there, although I'm sure if you just plugged in his name, you could find it too. But I can copy that for you, Peggy, and paste that over in your chat. And then, Peggy, any, I'm gonna close up the seminar after this, but any last thoughts or anything you wanna leave with folks today?

[Peggy Sullivan] No, not too much. It's certainly ongoing work and it's just a ton of data. So there are lots of things still to be filled in and lots of moving forward for comparison and I look forward to keep working on it. And these data haven't been uniformly distributed, so that is on the soon to be done thing as well. So if people are interested, we can get them data or it will be more publicly available soon. I think that's it.

[Heather Tabisola] All right. And I'm really looking forward to reading your paper. Somebody did ask, "Will we be able to access those data?" And I think, yeah, asking you once those are put forth, maybe reach out to Peggy in a couple, in a month, couple months for more details on that.

[Peggy Sullivan] Yeah, yeah, that sounds good.

[Heather Tabisola] And so folks, this is the last, Peggy was our last speaker of this series. So this spring, as EcoFOCI will be moving into our field season in a few short weeks through October, and we will begin seminar again in some form come the fall, it's usually in the month of November, so stay tuned for that. If you're not on our seminar list, we do post everything on the OneNOAA seminar series. We try to get those up as early as we can. And then also you can check the PMEL calendar of events once they're posted and ready to go or feel free to reach out to myself or Deana, if you're itching to either give a presentation or just know who will be in our lineup. So again, thank you everybody so much for joining us this season. Peggy, thank you for closing this out. Thank you for giving this talk. I know it's been a long road to this, but I really appreciate it. It's really fun to hear and see the work that you've been putting in.

[Peggy Sullivan] Cool, thank you. Thank you for doing all this and thanks for people comin'. Have a good one.