The image illustrating AI technologies and plastic pollution was created by AI (DALL-E).

As our oceans and rivers increasingly bear the brunt of relentless plastic pollution, a new wave of hope rises, powered by the synergy of artificial intelligence (AI) and citizen science.

The webinar “Leveraging AI for Plastic Pollution Detection and Monitoring” (January 19, 2024) highlighted the focus on using artificial intelligence as a key tool in identifying and addressing the issue of plastic pollution in the environment. It unveiled this remarkable alliance, showcasing innovative approaches and solutions to one of the most pressing environmental challenges of our time through the expertise of visionaries and practitioners who are pioneering the use of AI in detecting and monitoring plastic pollution.

By integrating the precision of technology with the power of community engagement, AI and Citizen Science are creating a new front in the fight against plastic pollution, opening new avenues for conservation and sustainability.

Harnessing AI for Precision and Scale

Artificial intelligence is being harnessed to enhance precision and scale in marine research in various ways, particularly in the detection and monitoring of plastic pollution. The advent of AI in image recognition surpassing human capabilities, marked a pivotal turn in marine research. AI techniques like Object Detection and Segmenting allow for precise identification of various elements in marine environments, including plastics, amidst complex undersea imagery.

There are several compelling examples to illustrate AI’s efficacy. Tools like You Only Look Once (Yolo) is an openly available object detection system that can recognize items on the seabed, such as fish and misplaced plastics. Segment Anything is another tool demonstrating AI’s capability to distinguish and segment images, like identifying a bottle in complex underwater scenes. These examples underscore AI’s transformative power in enhancing data analysis for marine conservation. Moreover, there are practical applications of AI in the field showcasing how established AI techniques are readily applicable in diverse scenarios, for example companies already using AI for detecting underwater equipment.

The intersection of AI with traditional marine biology offers a promising avenue for bolstering environmental efforts, particularly in combating marine pollution. The webinar served not only to highlight AI’s potential in environmental research, but also as a clarion call for its broader adoption in tackling marine pollution challenges.

Innovations in Action

Several companies and projects have already started to take advantage of the practical application of AI in combating plastic pollution. Mapping oceanic plastic using AI for example by focusing on the application of unmanned aerial vehicles (UAVs) and camera systems, has pioneered the integration of these technologies with satellite data. This emphasizes AI’s role in transforming observations into predictions, a crucial aspect in environmental research. By utilizing advanced AI models, the Ocean Cleanup has been successful in detecting and mapping plastic litter in vast ocean areas. This innovative method not only enhances the accuracy of plastic detection but also aids in understanding its distribution and movement, providing critical data for cleanup operations.

Shifting focus to riverine environments, the intricacies of AI-based hotspot mapping and quantification of plastic flux in rivers becomes crucial. The operational challenges and methodologies employed to gather data underscore AI’s role in accurately detecting and quantifying plastic in dynamic river systems. This highlights the necessity of developing tailored AI models that can adapt to the unique characteristics of riverine environments. There is potential of AI in providing vital information for environmental strategies aimed at riverine plastic pollution.

Advanced technology, combined with rigorous research, can lead to actionable solutions in environmental conservation. However, the necessity of robust data acquisition strategies and the integration of AI with other technologies is important to maximize effectiveness in environmental monitoring.

The Role of Citizen Science

Citizen Science emerges as a critical component in environmental monitoring, particularly in complementing AI-driven efforts. By involving communities in data collection and analysis, Citizen Science amplifies the reach and effectiveness of scientific research. These collaborative efforts not only enhance the data pool for AI models but also foster a deeper public understanding of environmental challenges.

Satellite data and AI can be used to detect and monitor marine debris. The integration of these technologies not only assists in identifying pollution hotspots but also plays a crucial role in understanding and predicting the movements of plastic debris over time. Donal Hill (Plastic-I) highlighted their project’s commitment to providing actionable insights for cleanup operations and policy-making, thereby bridging the gap between problem identification and effective solutions.

It is important to underscore the significance of engaging the public in environmental monitoring efforts. Somes projects leverage Citizen Science to gather and analyze data, using AI to enhance the efficiency and accuracy of plastic pollution mapping. The transformative impact of public participation in environmental research demonstrates how community involvement can amplify the reach and effectiveness of scientific endeavors.

A Unified Approach

The future prospects and challenges in integrating AI with environmental conservation, and particularly for managing plastic pollution emphasizes the need for a unified approach, calling for collaboration among diverse stakeholders, including technologists, environmentalists, policymakers, and the public. There are several key areas for future development:

1. Interdisciplinary Collaboration: The importance of integrating AI with traditional environmental research methods was emphasized. By combining the precision and scalability of AI with the nuanced understanding of environmental science, more effective strategies for monitoring and managing plastic pollution can be developed.
2. Policy and Regulation: The role of policy in shaping the adoption and ethical use of AI in environmental monitoring was discussed. With the evolving landscape of AI technology, there’s a growing need for regulations that not only foster innovation but also ensure ethical and responsible use of AI.
3. Public Engagement and Education: The webinar underscored the significance of public awareness and education in environmental initiatives. Engaging communities and educating the public about the role of AI in environmental conservation can lead to more informed and supportive public policies.
4. Technology Accessibility: The need for making AI technology more accessible and sustainable was highlighted. For AI to have a broad impact, it must be developed in a way that allows for widespread adoption, ensuring that its benefits and reaches all sectors involved in environmental conservation.
5. Looking Ahead: The future of AI in environmental monitoring includes advancing capabilities, improving data accuracy, and expanding the scope of what AI can achieve in this field. The speakers discussed the potential of AI to not only aid in detection and monitoring but also to play a crucial role in shaping environmental policies and actions.

Navigating Challenges, Embracing Opportunities

Looking towards the future, the webinar underscored the challenges and prospects in integrating AI with environmental conservation. The speakers called for more interdisciplinary collaboration, combining AI’s precision with traditional environmental research methods to develop more effective pollution monitoring and management strategies.

The evolving landscape of AI technology also brings forth the need for ethical considerations and regulations that ensure responsible use. Public awareness and education, along with the accessibility of AI technology, are vital for a broad impact. The future of AI in environmental monitoring lies in advancing capabilities, improving data accuracy, and expanding its scope, ultimately shaping environmental policies and actions for a sustainable future.

The journey toward integrating AI in environmental conservation is fraught with challenges yet brimming with opportunities. The insights garnered from the webinar’s featured projects serve as a testament to the potential of AI to not only aid in the detection and monitoring of plastic pollution but also to play a crucial role in shaping the future of environmental policies and actions. As we venture forward, the collective efforts of technologists, environmentalists, policymakers, and the public will be paramount in harnessing the full potential of AI and Citizen Science in safeguarding our planet for future generations.

Acknowledgements

We would like to express our sincere gratitude to the expert speakers whose excellent contributions made this webinar to such an enlightening experience. First, thanks to our co-host, Prof. Morten Goodwin from the University of Agder for setting the stage with his introduction to the intricate ways in which AI is being harnessed to enhance precision and scale in marine research, particularly in the detection and monitoring of plastic pollution, and for providing interesting questions and discussions throughout the webinar.

Thank you to our expert panel; Robin de Vries from The Ocean Cleanup who presented his work on mapping floating plastic in the ocean using AI, Robrecht Moelans from VITO Remote Sensing, who presented AI’s application in riverine environments and particularly in hotspot mapping and quantification of plastic flux, Dr. Klaas Pauly, the Project Manager of The Waste Watchers Project who underscored the significance of engaging the public in environmental monitoring efforts, and Dr. Donal Hill, Co-Founder and CTO of Plastic-I, who offered a compelling case study on their project’s innovative use of satellite data and AI to detect and monitor marine debris.

The role of AI in the making of this synthesis

In keeping with the webinar’s theme, this synthesis was also written with the help of AI. First, the transcript tool otter.ai, using advanced AI and machine learning algorithms to convert speech into text, was applied to provide a detailed transcript of the webinar recording. For this text, chatGPT 4 (developed by OpenAI) was used to distill and synthesize the complex insights and discussions from the webinar transcript into an accessible and engaging narrative.

Leveraging natural language processing and understanding capabilities, it organized the content, focusing on the findings and conclusions rather than on individual contributions, to emphasize the collective knowledge and advancements in the field of AI and Citizen Science against plastic pollution. Finally, an iterative process of text generation, minor amendments and some restructuring has been applied to the suggested text.

• Read more about the webinar and speakers: Leveraging AI for Plastic Pollution Detection and Monitoring
• Watch the webinar recording here.