 Our next presentation is by SEP Hakebo of the Environmental Defense Fund. EDF will talk to their smart pass framework. Hello, everybody. My name is SEP Hakebo with Environmental Defense Fund. I will share my screen to get my presentation started. Today, my colleague Chris Cusack and I will be presenting on SmartPass, a technology for measuring fishing effort with cameras and AI. We work on EDF's Oceans Technology Solutions Team, a group of researchers working with fishermen, tech providers, and fishery managers to bolster pilot's policies and people that advance technology solutions, fisheries issues around the world. To understand SmartPass, it's helpful to first highlight that a core component of fisheries management is understanding the total catch. Surveys, interviews, and other methods help us understand catch for boat. But one of the greatest challenges is measuring total effort in a timely, accurate, and cost-efficient way. Our solution is to enable others through technology. The SmartPass framework is made up of cameras mounted at a coastal pass, port, or river mouth. SmartPass includes a platform to assist with video review and AI training. And it also includes AI to maximize efficiency, including identifying the type of vessels, in our case, recreational or small-scale fishers, and count the number of trips, aka measure effort. Now, we'd like to show you a little bit more about this approach. More accurate and timely data is essential for good, sustainable fisheries management. We're in the middle of a digital revolution. We're trying to use the latest new and emerging technologies to solve the world's fishery challenges. SmartPass technology really has potential to shine a light on small-boat fisheries globally. The first time I went down to the Gulf, it fished for red snappers, boy, tell you why. It wasn't just the fish that got hooked, I got hooked. Why isn't it important? I mean, it's the resource that I enjoy harvesting. It's the resource I enjoy going to the restaurant and having. It's a resource that plays an important part in the ecosystem of the Gulf. Recreational fisheries in some parts of the country account for more than the commercial catch. That's especially true in the Gulf of Mexico with the red snapper fishery, but we don't really have a good handle on how many boats are participating. What we're doing is we're putting a camera at a natural pass or a pinch point where vessels have to funnel through to get to the ocean, recording that effort and using artificial intelligence to figure out how many vessels and of what type go through that pass. We have implemented two systems in the United States. It does take time to build the algorithms, to gain confidence in them, and then to get a level of reliability out of them. But when you get there, these efficiencies could really be a game changer. The more accurate we can be, it benefits everyone. We're getting a count of the number of boats going out, getting an average catch per boat when we interview them dockside, and then taking those two elements and being able to expand it to the total catch as well as the total number of anglers that went fishing. Everybody thinks the ocean's so vast. If the science tells us, we can harvest X number of this fish and still build the fishery back up, that's what we should be looking at. Getting more regions to adopt this technology, we really push the needle on fisheries monitoring and recreational fisheries, but also small-scale fisheries. Blue Simigrap Fishers in Lampung is very important. The value is really high. It's number three, the biggest export from Indonesia. The urgent thing that need to be solved is the sustainability of the Blue Simigrap itself. Everyone can catch Blue Simigrap. If you have money, you can buy a boat. If you have net, you can go fishing and then catch Blue Simigrap. At the moment, there are no exact number on the fishers in Lampung. All over the world, small-scale fisheries account for approximately half of all fisheries catch, and we just don't know how much is being caught. We need to know about how much is the effort. Imagine like four to 10 hours a day to sit in front of the poor counting the peso. If you have the camera, you just record the footage and then assign AI to identify the pesos. We compare that data with the data we're collecting with monitors in the fishery to estimate the amount of catch. We've come to realize that fisheries are not inexhaustible. We need to understand what's happening under the ocean. With data, with that knowledge, we will find ways to ensure the sustainability of the ocean ecosystem in the future. Now, in our pilot in Oregon, we're building off innovations at the Department of Wildlife and Fisheries already implemented. In 2007, they switched from human observers in the field to observers reviewing footage from shore-based cameras. And this increased coverage to all daylight hours, and it greatly reduced the costs from the previous methods. Adding smart pass and AI to augment the process was the next logical step. As we continue to develop this tech, the cost will go down. But as of now, the tech provider we're working with, C-Vision AI, offers the cameras for approximately $5,000, including the monitors at the location. And the review platform is between $100 and $300 US, depending on the upgrades you have for. The AI they've developed is open source and available to anyone interested in building up. In our Oregon pilot, we saw great results from the vessel detection algorithm seen here in Table 1 and the classifier algorithm seen here in Table 2. These results are further detailed in our white paper, which I'll share with everyone. And we have strong performance metrics from our Indonesia and Oregon projects. And we want to bring this approach to small scale and recreational fisheries globally. Our partners continue to refine the tech, but we're also at this inflection point of considering the next barriers to scaling. So we'd like to discuss with you a couple questions. As for barriers, the first is that the current market size appears small right now. And this may not be enough to incentivize tech companies to invest in developing a scalable platform. So our question here are there other opportunities in REC and small scale fisheries around the globe? What about opportunities to apply this approach to other environmental monitoring needs? Now the second barrier is that most small scale fisheries don't use effort and total catch estimates to conduct stock assessments or as a management control. We feel this is partly because these metrics have been not been available, either expensive or it was just too difficult to tackle. So how do we start to disrupt how we think about managing small scale fisheries? For more information, feel free to email me. Here's my address. We also have a lot of information on our SmartPass webpage. That's where you can find the white paper as well as the SmartPass video. Thanks so much for your time today. Thank you very much, Steph and Chris. As you know, we've broken up this forum into a number of different sessions and this morning session, we were lucky enough to have Alan Steele who spoke to a similar question around the blue swimmer crab fishery in Indonesia and where they're actually following individual consignments of crab coming out of the fishery and have millions and millions of new codes developed to follow hundreds and hundreds of fishermen. I'm just wondering if you could throw your mind deep into the future. Do you foresee your framework linking your type of measures with the collaborative development of others and how do you believe that it's possible to bring the real value chain vision of what's happening to allow the most important question of all, predictability. Predictability on how the stock's doing, and the visibility of the people who are important to these types of fisheries, be they fishers or traders or consumers. Yeah, thanks, Kim, for that question. It's a good one. I think a key note to take home from our SmartPass presentation and the framework is that it's just one enabling component. So SmartPass has that ability to be modular. We wanted to design something that was open source with our partners on the tech side so that it could have broad applicability around the world. For instance, in recreational fisheries here in the U.S., but also small scale fisheries around the world. And I think another key component is that the SmartPass framework is not a holistic solution. It's one piece and helps truly identify that effort component. So you still need a better understanding of what's being caught per vessel. Again, whether that's a recreational vessel or a small scale, maybe blue swimming crab vessel. So, you know, as we're in these earlier phases, we also want to make sure that we're mindful of drilling down into really unique cases, but developing the capabilities more broadly so that they can be paired with, for instance, kind of this Q code approach that you mentioned. And the other thing is there's the capability to implement the SmartPass framework as either a baseline study to help give managers in other countries an idea maybe over a year or a two-year study, or as our partners with the Oregon Department of Fish and Wildlife have done to really set up a continuous census, basically. So measuring during all daylight hours, every day of the year, which can also start to give you an idea of those long-term shifts in effort. Recreational fisheries we're seeing on the rise throughout the U.S. and a lot of other parts of the world. We're going to start to see a lot more changes in fishing patterns in the face of climate change. In some places that will be an uptick. In some places there'll be a lot more disturbances and that'll be a decrease. And so really starting to have this a combination of data coming in from, for instance, the Q code approach that you mentioned, but also the SmartPass approach can give managers a much better idea of in real time how those shifts are taking place. And with that you can start to make much more real-time including if you've got an incredibly busy day that you weren't expecting that SmartPass can let you know within the first two hours of the day if 10 recreational vessels have gone out or in some places here in the Gulf we see 400 or 500 recreational vessels going out in the morning. So you can start to ramp up your dockside intercepts by the afternoon. You can start to ramp up on the water enforcement. So I think thinking about all these approaches again as modular but also compatible is a really important way of approaching these issues. Yeah, thanks very much. My mind just thinking now safety at sea there's so many different areas that this type of data could feed into and making your technological innovations open source for other people that's fantastic too in the scheme of sharing ideas. Matt, could you come in with any question please? Thanks, Seth for your presentation. Thanks so much. I'm just interested in issues of privacy really and you know dealing with stakeholders and saying we're going to watch you every day going to work. What's your strategy to cope with and deal with empathy with their perceptions of being filmed constantly? Yeah Matt, it's a great question and it's one that we thought about before even ramping up any of our pilots. I think a key piece is stakeholder engagement and being completely transparent about the goals of this work. You know we've seen I work in the Gulf of Mexico on the US side and there's a few states that approach recreational monitoring with mandatory reporting. You know some folks see that also as an infringement of fishing rights or privacy and so I feel like this is an approach that takes some of the onus off of the individual fisherman to report and starts to make that process more efficient can allow those anglers to focus on the fishing whether it's recreational or commercial really stakeholder engagement from the early stages is a critical component of this and there's no doubt there's going to be folks even with that stakeholder engagement that are going to feel like this isn't appropriate and so we also wanted to make sure that there's there is the right amount of security around the data that we're not monitoring individual vessels we're not recognizing vessel numbers we're not tracking how long those vessels are out it's really just a holistic identification of total effort from the day from the week from the year and perhaps there'll be a time that comes in the future where IOT will give us an opportunity just to produce raw data on the spot rather than collecting all the video. I think that's possibly in 5-10 years time when small-edge computing devices have got the power to actually have the algorithms inside them themselves and I think that's maybe have you envisaged that sort of technology in the future in the future because I think it would be much nicer to tell a stakeholder it doesn't produce any image data whatsoever it just counts folks and we just know how many folks there were of this type. Yeah Matt that's a great point for a lot of reasons we've thought about being able to compute this on the edge and not even have to upload any sort of images I mean the videos that we're taking we're trying to get as low resolution as we can so that we can transmit via cellular or eventually transmit via satellite internet but I think another critical question in terms of where we are right now is this is a pretty niche area I think everybody on this forum today can probably recognize that Fisheries monitoring a lot of these applications can be pretty niche but there a lot of these technologies have been applied in numerous other industries around the world and so we're thinking about how can how can you also present that business opportunity to tech providers there's some great programs here in the US like the small business innovation and research grants that can help develop that ecosystem of innovation from the private and the public sector but how can you also think about different applications for instance the smart pass that aren't just focused on maybe fisheries but are focused more broadly on environmental monitoring what other opportunities are around the world in fisheries but also beyond fisheries to use these technologies well thank you very much Matt and Seth and thank you very much for the presentation in fact a large amount of money gets spent in shopping centres to track your movements and a lot of that code is being transferred around to develop other so it's very interesting to see where we're going to get good payoffs and where we might offer opportunities to other types of things okay