• Chris McLellan

E32: Can AI help to mitigate the effects of catastrophic oil spills?

Updated: May 15

Ask AI check-in host Carolyne Pelletier meets the Life Cycle Management Lab at UBC Okanagan


Click here to subscribe to our podcast or search "Ask AI Podcast" wherever you get your shows. Special thanks to Microsoft Canada and Cinchy, the Dataware platform for sponsoring the 4th series of the Ask AI podcast.

When oils rigs or machinery malfunction or break, thousands of tons of oil can seep into the environment. The impact on environments and habitats can be catastrophic: oil spills can kill plants and animals, disturb salinity/pH levels, and pollute air and water, and disrupt thousands of livlihoods.


Any spill is disastrous — ecologically, economically and socially. The Exxon Valdez disaster in 1989 killed an estimated 250,000 seabirds, hundreds of otters, seals and eagles, and two dozen killer whales. Oil vapours are toxic and contaminate seafood, harming public health and the local economy and residues linger for decades. Large spills, such as from the Tasman Spirit, which ran aground off Karachi in 2003, or from the Prestige that sank in 2002 off Galicia, Spain, caused billions of dollars in damages. Clean-ups are estimated to cost more than US$20,000 per tonne of oil spilt.


While the overall incident numbers are on the decline, we remain one earthquake, extreme weather event, or human error away from the next devastating oil spill. There are also several ticking time bombs (notably in California and the Red Sea) that we should be dealling with far more practively.


Canadian oil traffic


Transport Canada estimates that there are approximately 20,000 oil tanker movements off the coasts of Canada each year. Of these, approximately 17,000 (85%) are on the Atlantic coast.


On Canada's ecologically vital West Coast, oil tankers currently represent about 2% of total ship traffic visiting the Port of Vancouver (out of 250 total vessels per month, about 5 are tankers). However, the Government of Canada’s recent approval of the Trans Mountain Expansion Project is expected to increase the number of tankers visiting the Port of Vancouver from it's current average of 5 to around 34 per month.


In this scenario, oil tankers would represent about 14% of total ship traffic.


Team Check-In: Life Cycle Management Laboratory


Established in 2011, the Life Cycle Management Laboratory (LCML) is a research unit nestled in the Okanagan Campus of the University of British Columbia that focuses on life cycle studies related to urban development, energy systems, construction and asset management, water systems, and industrial products & processes.


The research team is composed of 20+ graduate students and post-doctoral fellows and its resources include more than 30 state-of-the-art software and databases, and the relevant hardware to perform thorough life cycle assessments.


Tune in to this Check-In as host Carolyne Pelletier checks in with Saeed Mohammadiun and Guanji Hu of the UBC Life Cycle Management Lab to ask the question:


How can AI help to mitigate catastrophic oil spills?


Research links


Please check out the these research papers published by the Life Cycle Management Laboratory:


Intelligent computational techniques in marine oil spill management: A critical review


Optimization of integrated fuzzy decision tree and regression models for selection of oil spill response method in the Arctic


Special mention: This project has received $2.8 million financial support from Fisheries and Oceans Canada.


Share this Team Check-In

Share YouTube video

Share LinkedIn post

Share Twitter post

Share Facebook post


Subscribe to podcast

Click here to subscribe to our podcast and newsletter


Visitors can also check out our recent Team Check-In with the DICE Team at Alberta Innovates.

Request Team Check-In


If you have news about an investment or project that you’d like us to feature on our podcast, YouTube channel, and newsletter, please feel free to request a Team Check-In.


Talk to our bot


This episode and hundreds of AI resources are available from the Ask AI chatbot. Special thanks to chatbot partner Ada for providing the amazing technology!


Episode credits


Sponsors: Microsoft Canada's Free AI Business School and Cinchy, the Dataware platform that eliminates integration Chatbot partner: Ada Support Series Producer: Chris McLellan Episode Host: Carolyne Pelletier Series Editor: James Fajardo Original music: Mike LeTourneau

About Ask AI

Ask AI is a volunteer project exploring Canada's fast-emerging artificial intelligence sector. Our projects include thought leader interviews, team check-ins, and live events. To learn more about sponsorships, guest blogs, and volunteering, please contact us or ask our chatbot.


Episode transcript


Please note: this transcript was generated by an artificial intelligence and some typos are invevitable:


Carolyne: Hi, this is Carolyne Pelletier your host for ask AI team check-ins. So we came across this article discussing the use of AI and other computational techniques for Marine oil spill management.


And so we caught up with Dr. Guangji, who postdoc research fellow and Saeed PhD student, both at the school of engineering at UBC Okanagan to talk more about the recent work.

Carolyne: Hello, everyone. This is Carolyne Pelletier from the ask AI podcast. Today. We're doing a tin check-in from sunny Montreal. I'm very happy to introduce our guests today. I'll give you the honors for, to make an intro.


Guangji: Hello, Carolyne. It's my pleasure to be here. I'm Dr. Guangji Hu I'm a postdoc research fellow at a school of engineering University of British Columbia, Okanagan, I'm an environmental engineer and my research mainly focus hazardous waste management, environmental risk assessment, remediation of a contaminated environment. And of course Marine or your sphere response.


Saeed: Hello, Carlene. Hi everybody. My name is Saeed Mohammadiun. I am a PhD student at a university of British Columbia Okanagan campus. Currently I am working on actually development of some AI based and soft computing based decision-making tools for a million oldest in management. And I'm very glad to be here.


Carolyne: Great. Thank you so much for joining us today. So I'd love to hear what are you currently working on?


Guangji: So we're currently working on a Marine oil spill response related project, founded by Marty partner research initiative with fisheries and oceans, Canada. And a site is developing.

His PhD says is based on this project. It's the market partner research initiative provides 45.5 million to support a wide range of research projects related to Marine or your spear response. It's a fundamental goal. Research initiative is to comprehensively enhance the preparedness and the response capability for any potential or your spear extant in Canadian ward.


Carolyne: Great. So what specific project are you working on right now?


Guangji: So the research team at a UBC Okanagan has been collaborating with researchers at other universities and the oil spill response industries on one project, this project for cause the canteen of oily wastewater and development of decision support tools for the management of oil waste generated from all your spear response operations.


Our specific task in this project is to develop. AI software computing based tools to help complex decision-making in all your spear response management. This task is important because the success and effectiveness of our response greatly rely on how efficiently the information such as oils, leak location or your property.


Whereas or see conditions and response resources such as equipment manpower vessels can be used and how optimally the decisions and actions can be made.


Carolyne: So say what's new with the project?


Saeed: well, let me start with a short introduction to oil spill management and the significant role of computational techniques in. Largely spills would have catastrophic consequences in terms of environmental economic and social impacts. As an example, one infamous accident in north America was the Deepwater horizon spill happened in 2010 in the Gulf of Mexico.


You know, this incident was an environmental disaster and cows more than 2 billion us dollars of economical. Therefore minimizing the negative impacts of oil spill is absolutely necessary. And this can be done by actually applying an effective in order to speed management. This management system consists of multiple, I would say consecutive action.


It usually starts with oil spill detection and monitoring then continues with. It's efficient response selection and obviously it's implementation. And then it also covers all your base management because usually we expect huge amounts of all your waste generated from a response operation intelligent techniques, such as AI can be applied to any of this action to facilitate timely decision-making and also increased efficacy of responsible.


Carolyne: And say, can you tell me why this matters?


Saeed: actually an effective Marinol, a spin management system can be developed based on these computational techniques. Such a comprehensive system can consider both a proactive and reactive actually practices to prevent an incident from happening and also mitigate the adverse impacts of an oil spill.


Let's consider the previously mentioned actions in Marin or Lisbon management. For example, we can time to detect an oil spill event, even in highly remote areas, by the application of, for example, deep learning techniques on some data that can be obtained from PSAT or hyperspectral imagery, for example, using satellite or area.


The application of machine learning techniques can also facilitate all this sort of response selection. There's intelligent models can be trained based on previous oil spill response data, actually historical data, and can be used to select, for example, the best management strategy for any given conditions for any future incidents.


Carolyne: and talk to me a bit about your roles in this.


Saeed: So my role in this project is helping with the management of a research works and the providing graduate students with the assistance needed.


So I think I can talk about the big picture of this project. project is more relevant to the technical part, such as application of a different machine learning and a soft computing techniques in Marine or your spear response.


Carolyne: And Dr. Guangji could you talk to me about who's involved in this project?


Guangji: There are many other participants in this market partner research initially. Funded project mainly including researchers from other Canadian universities, such as a university of Northern BC memoir university and at our house, a university, the research teams at a different universities are working on specific tasks under the same umbrella of improved or your spear response management.


University of Northern BC is working towards field scale candying technologies for oil wastewater treatment and memory university is working towards improving the canteen using demassification and advanced oxidation. Since the school of engineering at a UBC Okanagan is good at data-based computation, our work mainly focus on enhancing or your spear response performance.

So the use of artificial intelligence and a soft computing based tools and assistance.


Carolyne: Wonderful. And are there any sponsors on the project that you'd like to make?


Guangji: Yes, the sponsor of this project is Fisheries and Oceans Canada. And as this project has received $2.8 million financial. It's the national senior science advisor, Dr. Kenley and other personnel at DFO also provide their technique device to have steers the course of the research progress. We also have a technical advisory committee consisting of experts from government and the industries, including Western Canada, Marine response Corporation and Canadian coast. We have routine workshops with a technique advisors.


So out of this project to report research, progress, identify challenges, and to discuss potential solutions. So technique and advisors can bring data. And there are suggestions for our practical field operation perspectives to help us improve the developed AI based to computational. This will make our work more meaningful in terms of a reward or your spear response operations.


Carolyne: All right. And say, can you tell me what's next?


Saeed: Oh yes. Actually considerable progresses have been made in advancing computational techniques in recent years. The robustness and accuracy of this techniques have been continuously improved such as for example advances in developing effective instant segmentation and also flexible evolutionary.


The performance of computational models in this field can also be enhanced by applying this improve techniques, for example hybrid artificial intelligence and soft computing techniques can be considered to develop a more reliable and more robust models in metanoia spend management.


Carolyne: Great. And finally, how can you make a comprehensive decision for oil spill management?


Saeed: In order to obtain a comprehensive management system it sits necessary to integrate different intelligent, computational technique-based models through a multiagent system. It means we can connect, for example, numerical models.


Optimizers detection, models ascertain to assessment, all of them through a multi-agent system, they can communicate with each other and provide with each, provide each other with required data and updates in a dynamic manner. Having this perspective, all important aspects of Maryknoll speed management must be considered holistically to reach the best management strategy.


This is going to avoid the pitfall to make sure that one optimized action will not create burdens to orders. For example, a spill response operations should not be optimized separately without considering the following management of only based.


Carolyne: all right. And as we wrap this up, is there anything else that you'd like to.


Saeed: I just want to mention that we shouldn't forget that you still need more enrich data sets of real world oil spill incidents to enhance the performance and the prediction power of decision-making models besides experienced decision-makers still play a significant role in Mary NOLA speed management. However, I think it's possible to reduce the reliance on human being in the near future.


Carolyne: Wonderful. Thank you so much for joining us today and we'll be following your work closely.


Saeed: My pleasure. Thank you. It's been a pleasure to be here. Thanks for having us here.


Carolyne: Thank you so much to our guests today for joining me on this Ask AI Team Check-in.


If you're interested in learning more about their work, we've linked it to published papers in the journal of hazardous materials and knowledge based systems. So the first one is entitled intelligent computational techniques in Marine oil, spill management, a critical review, and the second optimization of integrated fuzzy decision tree and regression models for selection of oil spill response method in the.


Thank you so much.


This is Carolyne Pelletier AskAI team check-in host until next time.

209 views0 comments