Innovations in the Marine Economy - Innovation in Data

You're listening to "From the Sea Up," a podcast from Island Institute. I'm Galen Kirk. In this season of "From the Sea Up," we're focusing on businesses and individuals who are innovating in Maine's blue economy and marine industries. In this episode, we'll be diving into all things data, exploring how two businesses in Maine, Marine Solar Technologies and Ocean Data Network, are working to fill the gaps in Marine ecosystem data. Marine Solar Technology is developing new ways to capture data, describing land runoff and ocean pollutants with inshore water quality monitoring buoys. Ocean Data Network, a company based in Portland and operating all over the world, utilizes an existing resource, the fishermen themselves, to collect highly sought-after data from both inshore and offshore fishing grounds. The data collected and managed by Ocean Data Network can influence everything from weather predictions to fish stock assessments. The solutions pursued by these two businesses are not only innovative in their own right, but the data they collect can drive more innovation within the blue economy. As technology advances, data has become such a common word that it feels almost meaningless, but data is extremely important. It's defined as a collection of facts, statistics, and information that is used as a basis for reasoning, discussion, or calculation. Collecting raw data allows you to capture a snapshot of the complexities of the world that can then be analyzed and turned into more meaningful information. Analyzing data leads to the discovery of trends and patterns, improved ability to make predictions, and more informed decision-making. Without data, we can observe what's happening in the world around us, but we can't determine what's causing it, or how it's changing. Data allows us to quantify or represent what we observe, so we can learn more about the world and its systems, by comparing a single observation with the observations of others, or ones made over longer periods of time. And while technological advancements have made it possible to collect more data than ever before, the marine world still poses unique challenges to data collection. You either have to get boats and researchers out on the water to collect data that is expensive, time-consuming, and weather dependent, or you have to deploy a device that can collect the data you want in the area that you want. And that device has to be able to operate in and withstand salt water, currents, storms, and possible interactions with boats and wildlife. Not to mention, you have to be able to retrieve the device and download the data collected, or the device itself has to be capable of sending the data back to you remotely. I spoke with Matt Tarpi of Marine Solar Technologies to learn more about how he's working to address some of these challenges. Matt's two startup companies, Marine Solar Technologies and Maine Electric Boat Company, are based on the Sako River in Bitterford, Maine, at Sako Bay Marine. My name is Matthew Tarpi. I am one of the co-founders and CEO of Marine Solar Technologies. You would think it would be simple to monitor water quality closer to the shore, but financial, logistical, and jurisdictional constraints make it easier to deploy monitoring technology in offshore waters, despite the increased distance. Water currents become more of an issue the closer to shore you get. Tides need to be factored in, and there's more recreational and commercial boat traffic. What we saw in water monitoring technology, specifically autonomous water monitoring technology, is that it was mostly offshore, and there was what we call coastal neglect, whereas we think that those localized spots are where the most important data actually is. The offshore stuff tells us there's a problem. We want to find out what the problem is and where it's coming from. Near shore monitoring does exist in Maine, but the sites and their capacity are limited. In 2016, Laura Sewell of Bates College Coastal Center and Caitlin Cleaver of Hurricane Island created the Northeastern Coastal Station Alliance, or Nesca, to unite small field stations along the coast to collaborate and collect environmental data. Starting with 10 small field stations, volunteers and researchers deployed intertidal loggers to collect hourly temperature data. While this is a step in the right direction, there are still gaps in near shore data to be filled. To overcome the barriers to near shore water monitoring, Matt and the team at Marine Solar Technologies have developed their own water monitoring buoy. Their design aims to make monitoring water quality more sustainable and logistically feasible, allowing monitoring to become more widespread and increasing the volume of data collected. We realized we had to start from scratch, design a customized computer board that would allow the communication protocols that we need to be able to communicate with all sorts of different kinds of sensor probes. From there, it was building a buoy shell that accomplished our goals of being light and very deployable. Another part of this is most water monitoring buoys you need a boat with a crane on it to deploy. R's is about 23 odd pounds, 32 inches, so it's something that most people, if they have an original inflatable, they're going to be able to deploy or move themselves. It's, you know, a cylindrical yellow has to be yellow for data buoys. It's relatively flat with some solar panels on it, which, you know, runs the whole thing. Again, it's really small, it's rotationally molded, so it's very light. It's something where one of our goals here is to make it so that it's going to be pretty unobtrusive to boaters. It's something where, if you bump into it, it's just going to move out of the way. It's not going to really cause any damage, but essentially it's going to look for all intensive purposes unless you're very, very close to it, just like a small yellow mooring, almost just a flat mooring. Along with growing the regions being monitored and increasing the scale of data being collected, marine solar technologies is also interested in expanding the types of data being collected in near-shore waters. We're constantly adding things, but right now it's your simple stuff. Like water temperature, pH, oxygen reduction potential, dissolved oxygen, salinity, conductivity. And while Matt and marine solar have designed a prototype that can withstand the marine environment, provide itself with power and collect various metrics, there's still a lot of work to be done to figure out how to get the data from the monitoring buoy into the hands of stakeholders and folks who can use it to inform their own businesses. At these beginning stages, we have what I call a very, very strong prototype. We're so far through testing, it's not everything we wanted to do. We want to keep evolving, we want to keep developing. I always say, and it frustrates a lot of people, that I view this as a failure when I view it as finished, because we're going to need to have a change in conversation about what metrics we need. That being said, we do want pretty straightforward pilot projects at the beginning. We're at a stage where we need to get these deployed, so we can start both using the data and understanding how this data can be beneficial to all stakeholders, whether it be aquaculture or municipalities or education. Our immediate focus is in areas that are going to have good access to cell signal. So we're kind of taking out one of the variables while we're doing kind of the first few pilot projects. Deploying these monitoring buoys in areas that have good cell signal enables the data to be transmitted in real time, using cell service instead of requiring someone to regularly physically download the data from the buoy and upload it to a database. The buoy records the data and sends it off like you would a text message. Improved cell service or using other networks to transmit this data, such as broadband internet, cuts out multiple costly and uncertain steps and getting data from the buoy to the folks who need it. There's also less risk of months of data being lost if a buoy itself is lost before the stored data can be retrieved. We thought it was really important to be able to transmit the data in real time so that, yes, we're tracking long-term trends, but if there's something wrong, there's something that if there's some, you know, if this buoy is in a river, say, and there's something significantly different in the water chemistry, we know something's gotten into the water that shouldn't be there. This real time transmission would allow the monitoring buoys to alert businesses or homeowners or sea farmers to time sensitive pollutants or run off. For now, Marine Solar is still developing their own platform to manage the data these buoys collect. Over the last six months, they've been working on building a platform that incorporates machine learning to find difficult to test for metrics. This platform would allow Marine Solar technologies to expand the information and analytics that they're able to provide about coastal water chemistry. One of our goals in the very near future is to build out our own platform because making the data digestible and shareable is one of the most important parts of this. I think for far too long we've been kind of afraid of those answers, and my opinion is that we've run out of time to be afraid of data. And it's just, it's gotten to the point where we've seen the impact of what's going on with our climate, and the only solution in my opinion now is to find the data, analyze the data, and then create solutions based on that information rather than just kind of some vague concepts. I met with some of the team of Ocean Data Network to learn about their work in fisheries based ocean observation. Cooper Van Branken, I'm the founder and CEO of Ocean Data Network. Of Jack Carroll, I do operations in technical development for Odien. I've been working for Cooper for three years or so. Ocean Data Network collects data with fishermen partners through a collaborative process. The company's offices are on Union Worf in Portland, Maine, but their data collection spans the globe. And so really kind of bringing together a different inspiration both from mechanical engineering and then fishing industry, putting that together, and kind of being a technical bridge and also cultural bridge working with the fishing industry to collect oceanographic data, which is so crucial for a wide range of applications and everything like that. Cooper Van Branken and Jack Carroll were both involved in the fishing industry before Ocean Data Network. I grew up with Jack on Peaks Island right out there in Casco Bay and kind of grew up on the water alongside the water and was initially an engineer and then got really interested in fisheries and also decided I didn't really care about robots or airplanes, which is what I was working on. And got interested in fisheries, so then came back and fished out of Stonington for a little bit and then Alaska and really interested in those issues and then went to grad school. And initially kind of set out to do more mechanical engineering for fishing to increase really both the ecological but also economic sustainability and make fishing more of a win-win. But in grad school learned that yes I can make a better troll door but ultimately who cares, there are more systemic issues and really the data and the ecosystem and kind of climate effects on fisheries are the most pressing issues and we really need to understand what's happening to fisheries. The catch all term for that is kind of ecosystem based approach to fisheries management but a key problem with that is that there isn't actually any data where the fish are. And so using satellite sea surface temperature for example, well that's clearly not going to work very well for COD that could be 100 vathams down on the bottom and lobster and things like that, there's a big disconnect. What's the obvious solution to get the data where the fish are? Well stick some sensors on the gear that are going to cast them. The method of ocean data networks data collection is straightforward. Cooper, Jack and the team at ODN collaborate with fishermen, fishing organizations and industry leaders to place sensors on fishing gear. But executing that mission is pretty complicated. Both Jack and Cooper travel around the world to introduce fishermen to the data sensors and help install the systems. Everywhere from Japan to Ghana. Here's Jack Carroll. So I travel around the world installing these systems on boats from a canoe all the way to a factory trawler. So there's a receiving unit on the vessel whether it's wired in or solar powered and that collects GPS and then the sensors connected to the gear or whatever it is, a trap or a trawler or a line or whatever they're doing. And then when the sensor goes below the surface, the Bluetooth, it senses the pressure change and then the Bluetooth, it shuts off, goes down and collects that temperature depth profile or cylinder or whatever you're capturing. When it comes back out of the water, it senses that pressure change again. Hey, I'm in the atmosphere. It turns the Bluetooth signal back on and that relays to the receiving unit. And then the time stamp on the sensor stitches with the time stamp on the GPS and then you get the location of the fishing event. And then that sends to our cloud either by cellular through like a little modem or we can connect to their own vessel, Wi-Fi network if they have one. So it's a really elegant way to collect data in a really cost effective way because you're limiting the platform and the deployment cost of working with the research vessel or like we were saying robots or something like that, this is all. And again, fully automatic. Once we install it, they don't have to do anything for years. So that's why it's such a good idea. That elegant data collection, as Jack calls it, also makes this method really accessible to the fishermen who partner with ODN. Once installed, the sensor automatically transmits data back to ocean data network. Fishermen don't need to push a button or connect the sensor to Wi-Fi or a cell signal manually. Fishing occurs in these near shore areas where it's really hard to observe the ocean because it's like complex butthymetry, strong currents that make robots hard operate there. So offshore, there's really good data coverage. There's this program called the ARGO float program. It's probably the biggest ocean data collection network in the world. And it works great out in the open ocean, but near shore, it can't really operate. ARGO is an international program that deploys fleets of robotic floats that drift with ocean currents between the surface and midwater level. These floats offer a robust data set, but they can't access the nooks, crannies, or the depths of ocean data networks sensors. So in those near shore areas where the most important for economic and ecological climate research, et cetera, where all the stakeholders are, where we live, there's a lot less data. So we're right there leveraging all that fishing activity to bring more data into a really data sparse part of the ocean, which is kind of counterintuitive to what most people would think, but it's kind of the reality of the oceans today. So we are not a sensor company. We are a oceanographic data collection platform being, you know, going along for fishing vessels. And we are a data management company. And so data management is hugely undervalued, but is absolutely crucial. And if you don't manage your data right, who cares? It's useless. Ocean Data Network has a portal on their website where stakeholders and fishermen can, with a username and password, access the data that's collected from the sensors. On the portal, you can see up to date data from fishing vessels all around the world. Here's the boat. This is the boat that has the DO sensor we've been trialing. He's also dragging right now. This was. What's the DO sensor? This is all DOxygen. Oh, okay. Sorry. Too many acronyms that I have. If there's so many. So this guy's like that where you know, here we are. Oh, look, he got hung out. That's a little. Yeah, something. Or maybe he's turning around or something. But that's a way. So in terms of like a great temperature gradient, you know, it's 56 of the surface and it's 47 down low. And so like this guy, he uses this data himself to avoid cod because there's like, by catch quota, you can only catch so much cod. So he's figured out, if I stay above this temperature or get below here, I'll catch more cod or less. And so he can stay out there fishing for whatever he does, quote, or for longer. This is a great example of how Ocean Data Network is enabling fishermen to fish smarter. And how accurate data can contribute to better ecosystem-based fisheries management. How this data is used varies greatly from fishermen to fishermen and industry to industry. But one important component of the data is that when it's sold to researchers or used in oceanographic studies, it is all anonymous. And nobody else has access. I guess there are a few boats who want their data and want credit for their data and want publicly. But by default, everything is totally anonymous. And you can't use this for tracking, fishing, or stuff like that. That's key to the whole thing's privacy and building trust with industry. It's one of the most important parts of their company is also us coming from fishing industry background and growing up around fishing. It's kind of, you know, it's fishermen first, always, or Fisher first, always. And then, you know, you have to gain their trust. And then you can work from there. Ocean Data Network installs sensors on fishing gear all over the world. They have sensors on boats in Japan, Tanzania, and Peru, in the UK, Norway, and Alaska. Sometimes they're approached by commercial fishing fleets or associations who buy the sensors and data as a client. Sometimes they work with institutions or universities to install sensors in remote locations. The ability to be flexible is one of the greatest strengths of ODN. You know, the diversity of what fishing looks like around the world is pretty astounding. And we love that and being able to, but, you know, that also is kind of dictates a little bit how it can work. And so, you know, these canoes in Ghana, for example, like they're doing this because they have a stipend, but actually the folks in Tanzania, they have a stipend that they're psyched about the data as well. It's a very small stipend and it's, you know, the data is worth so much more than, you know, so it's a win win for everybody. But then on very large scale ones and then even, you know, medium scale, innovative Alaskan, especially personers, they're often paying us to implement this and British Columbia as well. You know, it's the fishing industry taking the lead on this and implementing themselves. You know, it's that full spectrum of, you know, okay, we need to kick in a stipend to make this work to get the data for the science and then it's, you know, sophisticated, big business of, you know, the fishing industry. This data is important to the fishing industry, to have solid specific data from fishing grounds, but it's also important to other oceanographic and climate focused clients all over the world. And then it was a little bit of a side thought is like, oh, hey, does anybody else need this data? It turns out, yes, all kinds of folks need the data from hurricane forecasting to climate type things. And surprisingly, the exact places where fishing happens, that is exactly where we are missing data. So it's kind of this perfect Yin and Yang missing piece to fill in a key kind of blind spot in our data picture of the oceans. Filling in the blind spots of ocean data is useful for climate monitoring and weather forecasting. The primary data that these sensors gather is ocean temperature and sometimes salinity. And it turns out, having subsurface data on ocean temperatures is incredibly important. Our bread and butter and really the most mature and widely deployed that we have is simply temperature profiling. So what that means is the temperature down to the fishing depth and then accompanying the pressure, which is how you get the depth. And so with that, you're able to see, oh, we've got a warmer surface layer and then you've got a thermocline and then a colder layer. That's a pretty standard setup. And that is the most important ocean parameter, this physical essential ocean variable. And so that's, you know, you can use that to fingerprint different water masses. The one that we're getting a lot of requests for now that accompanies that quite well is the salinity. So with temperature and salinity and then the depth or pressure, you can get the density of the water. And so with the density of the water, you can then fully resolve underwater current density driven current exchanges and all these different things. And in the middle of the Atlantic Ocean, you can fingerprint a water mass coming from the Mediterranean. So it's kind of the whole picture of the ocean physics. And that has many different markets and it's essential for all kinds of different things. Ocean data network is trialing some other sensors for additional data collection. Jack mentioned dissolved oxygen. They've just started to collect some of that data on some trial sensors. But as Cooper pointed out to me, fishing hits these sensors pretty hard. They might be deployed on a troll net or a dragger or inside of a lobster trap. So they're in the business of putting out robust sensors that can take a beating. When we meet with ODN on Union Wharf, Jack Carroll is about to install a surface weather station on a boat in Portland Harbor. The company's office is right above a marine shop. They're in a great location to test out new prototypes with local fishermen. There's added value to stitching together a picture, subsurface, sea surface, surface weather and this co-located data streams, particularly for linking ocean conditions to weather conditions. And those coupled ocean atmosphere models, like that's really the future for an accurate 15 day weather forecast and things like that. So that's something I'm excited about kind of unlocking on a bigger scale. Those 15 day weather forecasts can be especially relevant in coastal communities that experience hurricanes or extreme weather systems. That data can be life saving. In situations when having a few more days of accurate weather means there's more time for evacuations or disaster preparedness. I was simply putting things are changing and they're changing fast and so any data as this happens will be valuable in the future to evaluate how things have changed and what it's doing. In a world that is teaming with venture capital and startup businesses, ocean data network have differentiated themselves by approaching their data management with scientific rigor and with a focus on local and global consortiums. I'm really really proud of how we've navigated that and what we've built and kind of the rigor in which we've approached this by doing this in an incredibly scientifically rigorous way and really kind of co-designing and co-developing these solutions with the scientific community. So we are a private startup but then we are also leading the fishing vessel Ocean Observing Network or FEON for short, which was recently endorsed as an emerging network within the global ocean observing system, which is a UN level coordinating body. It's kind of the highest possible ocean data accreditation coordination around the world and that is a level that no other ocean data startup has gotten to it all and I think is absolutely crucial to building this and scaling this in a way that's not going to crash and burn. ODN's work with the fishing vessel ocean observing network is highly collaborative between scientists, fishermen and nations. That's part of the work that feels a little different from other entrepreneurial ventures and startups. ODN is working with other data management and ocean observing businesses and networks in order to get more accurate and comprehensive ocean data. ODN stands on the shoulders of some pioneers of ocean observation, including Michaela Martinelli, one of the researchers behind the FOOSE, the Fishery and Oceanography Observing System in Italy, and Jim Manning in Massachusetts, who founded the EMOLT project, which stands for Environmental Monitors on Lobster Traps. George Maynard is now taking over the project and is doing a great job at NOAA Northeast Fishery Science Center. That is also a Gulf of Maine Lobster Foundation, an air and peltier, and then there is CFRF, which is a slightly different commercial fisherman's research foundation down in Rhode Island. They're running a program. So around here it's a great group of us all kind of working on this together. You can't forget the fishermen themselves, and then particularly in this region of New England. There's a lot of New England and the Middle Atlantic. There's a lot of history here, and it's been great to kind of jump in with these folks and then work to make this all better and more streamlined and more efficient. Providing this kind of baseline understanding of what's actually happening under the surface is like, okay, we got to start somewhere, and this is absolutely crucial for understanding the biology, the ecosystem impacts all of that. That's kind of what we're providing, and a lot of that is this really fascinating intersection between fisheries science and oceanography is like better forecast of what's happening at the bottom to better stock advice for the future. And that stock advice can be, we need a higher quota or we need a lower quota, but better anticipating it and particularly not leaving quota unutilized, which is what happens a lot of that. Now because of the uncertainty in a lot of the data streams and the management of traditional single stock fisheries management, we're leaving quota that could be fish sustainably. And that's disenfranchising communities and, you know, leading to a lack of trust and, you know, cod collapse, you know, there's all these, you know, this the strongest example probably of anywhere in the world of like a failure of science and industry and, you know, kind of that lack of understanding. This is where the data hits close to home. The collapse of the cod industry in the Gulf of Maine touched many fishermen here and paved the way for the lobster industry that now teeters on the edge as lobsters move north to find colder temperatures at the bottom. Ocean Data Network works closely with the Maine Fisherman's Association based in Brunswick to deploy sensors on fishing boats in the Gulf of Maine, and they work with the American lobster research program at the University of Maine to contribute data for lobster stock assessments. For Cooper van Franken, Jack Carroll, and the team at Ocean Data Network, there is a ceaseless opportunity to expand their data set to reach more fishermen across the world. We're being less and less shy about how bold we want to kind of take this and that, you know, our goals are, you know, within the next 10 years have tens of thousands of fishing vessels operating around the world doing this. And the current global ocean observing system, everybody put together, it's only 8,000 platforms. So, you know, we want to kind of dramatic or as a magnitude, a paradigm shift in understanding the co-solutions and, you know, understanding the regions that are most relevant to society and the blue economy. And it's, you know, it's achievable because this going along for the ride is so cost effective and is a win-win for everybody, so we're giving our hell, you know. Thank you for listening to From the Sea Up. This episode was produced by me, Gaelin Koch, and production assistant Olivia Jolly. Nicole Wolfe takes the photos that accompany this episode. Special thanks to Matthew Tarpi and Marine Solar Technologies, and to Cooper Van Franken, Jack Carroll, and Ocean Data Network. Most of the music in this episode is by Q-shop. You can hear more of their tunes at www.cuehyphenshop.com. From the Sea Up is made possible by the Fund for Main Islands, the Repartnership between Island Institute, College of the Atlantic, Main Sea Grant, and the First Coast. Here are past episodes and for more information, visit www.islandinstitute.org/podcast.