Apr 2, 2019
The technological revolution in fishing
Rapid and sometimes radical advances in technology around the globe have opened up a new era of opportunities for commerce and environmental protection. In particular, advances in artificial intelligence, smart sensors and digital networks are revolutionizing industries ranging from medicine to manufacturing and beyond.
These exponential technologies — ones that improve at an exponential rather than a linear rate — are revolutionizing how we approach and solve problems today.
Sensors: New, low-cost sensor technology has proliferated in recent years, allowing detection, visualization and measurement of a wide variety of variables at a fraction of previous costs. Phone camera sensors have gone from gimmickry to greatness in just over a decade, and low-cost sensors can now be placed nearly anywhere to measure nearly everything.
Networks: Advances in networking technologies, such as the development of low-power wide-area networks, allow data to be shared more cheaply and quickly over longer distances. And companies such as OneWorld and SpaceX are planning to launch thousands of micro satellites that would dramatically increase existing worldwide, high speed and low latency connectivity — allowing any device with satellite connectivity to rapidly transmit and receive data.
Artificial intelligence: Recent advances in AI, in particular machine learning and the neural network-based subfield of deep learning, have led to rapid improvement in automated pattern recognition. This can be applied to visual, audio and industrial sensors to convert data into insights. Use cases range from identifying a person’s face to predicting when a piece of machinery is likely to break.
These technologies enable a wide range of new applications and are ripe for accelerated adoption. While their deployment has penetrated many parts of our economy, the fishing sector is still in the digital dark ages. But that is about to change as exponential technologies are poised to help solve one of the most pressing social and environmental challenges of our time: overfishing.
Technological solutions to watch
Many of the most exciting advances have nothing to do with fishing or even maritime industry, but could nevertheless have a dramatic impact. For example, the cost of micro-electro mechanical sensors — which are found in everything from cell phones to automobile engines and also have many maritime applications — dropped by 70 percent over the past decade. Inexpensive mechanical sensors could spur a revolution in fisheries monitoring, essential for good management but which currently is the exception rather than the rule. And new sensors such as solid state LIDAR, which is being developed for the autonomous car industry, could revolutionize the ability to map the undersea world, revealing sensitive habitats that need protection.
At the same time, both active (sonar) and passive (microphone based) acoustic sensors continue to improve in performance and decrease in cost, opening the door to new applications. For example, Environmental Defense Fund is working with scientists at Cornell and Stonybrook to test the efficacy of using active acoustics to estimate sardine abundance in the Philippines. They are also developing a method for using sensitive underwater microphones to detect and deter illegal fishing.
Metamaterials — materials engineered to have properties not found in nature — are being harnessed by start-up companies like Kymeta to enable low-cost satellite connectivity that can easily be incorporated into ships. And vessel tracking systems by companies like Pelagic Data Systems sidestep more costly satellite systems in favor of a tiny solar powered box that is easily installed on small vessels and can automatically transfer data to a server when the vessel comes within range of a cell phone tower.
Incorporating this data into a program of sustainable fishery management can improve both profits and the sustainability of our oceans. Underwater cameras can dramatically improve stock assessments so that managers know both which and how much fish can be sustainably caught. Cheap cameras and micromechanical sensors on deck can improve catch monitoring, helping to achieve compliance with these improved regulations, and inexpensive GPS trackers can improve compliance with restrictions aimed at protecting these habitats.
The data collected by the proliferation of networked sensors would rapidly overwhelm human analysts. Fortunately, machine learning can help. Research at the Alaska Fisheries Science Center and University of Washington have developed computer vision algorithms that are currently up to 94 percent accurate, and improving, at identifying species of fish aboard fishing vessels. And new systems based on commercial machine learning products — like smart-recognition remote video monitoring software from Camio — can provide descriptions of what is being captured in the video (e.g., a package on the porch, deer in the front yard, etc.). The Nature Conservancy piloted Camio’s technology on fishing boat cameras to automatically identify video segments containing various fishing activities, reducing the amount of review time required from hours to minutes. The pilot demonstrated that machine learning can amplify the impact of a single fishery reviewer, allowing them to go from monitoring a couple to a couple hundred boats per day, thus reducing the cost of compliance for fishermen and the cost of enforcement for regulators.
A powerful vision for the future of fishing
Sustainable management of global fisheries depends upon robust and scalable systems for collecting accurate, timely and consistent data. As the old adage goes: you can’t manage what you don’t measure. While some of the world’s fisheries use sophisticated technology to ensure compliance with catch limits, avoid killing marine wildlife and achieve other management goals, they constitute only a tiny fraction of the world’s fisheries and vessels. Most fisheries have inadequate monitoring systems, or are not monitored at all. Even in some of the world’s most advanced fisheries, any data collected onboard (usually by human observers) takes weeks or months to be processed and used.
The result is that the global fishing industry is plagued by illegal and unsustainable practices, human rights abuses, fraud, predatory business relationships and inadequate science — without the entrepreneurial flexibility and resilience needed to respond to major challenges like climate change that are already re-shaping the abundance and distribution of fish around the globe.
The lack of data and transformative technology also hurts fishermen by leaving them isolated at sea and shore, and cut off from oceanographic and market information that could boost their profits. It leads to a lack of trust among scientists, regulators and industry that undermines effective management. This cascade of market failures results in high costs, low revenues, waste and plundered resources.
EDF’s Smart Boat Initiative addresses the costs
The stakes are huge, not just for the oceans, but also for those whose lives depend upon them. Globally, over 1.5 billion people rely on fish for nutrition and for their livelihoods. However, the majority of fisheries worldwide are seeing production level off or decrease in recent years. Research by EDF and leading scientists estimates that every year the world loses $53 billion in revenues from mismanagement of global marine fisheries. This is a trend that can and must be reversed.
Enter EDF’s Smart Boat Initiative, our program designed to address the major information gaps that hinder the improvement and successful uptake of good fisheries management systems and practices. We envision networked fisheries that will leverage multiple technologies to extend a ”digital nervous system” across our oceans. Our plan for transforming global fisheries depends on equipping fishing vessels around the world with an increased ability to collect, share and use data by leveraging the latest developments in sensor, network, artificial intelligence and other technologies, effectively turning these vessels into smart boats.
With more than four million fishing boats — large industrial vessels and small skiffs alike — plying the world’s oceans at any given moment, we see enormous potential to create fleets of smart boats that can generate an unprecedented new level of information about fisheries and the oceans. This information can help improve decision-making, expand market opportunities and improve the quality of life for fishermen at sea. It’s also good news for your dinner plate.
Of course, technology on its own is not enough. It must be responsive to the unique needs of fishing communities and ecosystems, validated by scientific assessment and utilized within a well-designed regulatory framework. EDF is committed to making that happen, and will work to ensure that these new technologies are incorporated into a regime that addresses the needs of all stakeholders. We believe that smart boats, and the networked fisheries they enable, will transform fisheries management. They would allow managers to cost-effectively monitor fisheries, empower fishermen to increase their profits sustainably and give communities the power to control their own food security, all while helping to protect our marine ecosystems. Not a bad catch all around.
We are entering a new era of environmental innovation that is driving better alignment between technology and environmental goals — and results. #FourthWave
