#01 2024

Digital Twins of the Ocean – the future of ocean management

By Giles Fearon, Operational Ocean Modeller, Egagasini Node, NRF-SAEON

Imagine an interactive user interface which allows decision makers to leverage the best available ocean observation and model data to address complex questions around the management of our oceans. You are imagining a digital twin of the ocean. 

I was lucky enough to attend the second International Digital Twins of the Ocean (DITTO) Summit held in Xiamen, China between 9 and 12 November 2023. Some 450 participants from 19 nations and 118 institutions descended on Xiamen to present recent developments in digital twin science and technology from around the world, discuss how the field might evolve in the future, and how we might collaborate to ensure a shared vision is realised.

But what is a digital twin of the ocean exactly?  

Isn’t an ocean model just a digital version of the ocean, and therefore a digital twin of the ocean? Once observations of the ocean are digitised, can they be considered a digital twin of the ocean? These are the questions I had in my own mind on my way to China, which I rather sheepishly admitted to another one of the attendees I met on the plane. I realised that I was not alone in my uncertainty around the concept.

Through the conference, an idea which crystallised in my mind, is that a key feature of a true digital twin of the ocean is that the relationship between the user and the tool must be interactive. It is not enough for ocean observations or ocean model output to be visualised in an easily accessible way. While this is indeed important, the user must in addition be able to run different management or response scenarios on top of the data. Only then can the cutting-edge understanding of ocean science be leveraged to truly inform decision-making.

Attendees of the second International Digital Twins of the Ocean Summit held in Xiamen, China

Giles Fearon presenting an overview of the Ocean and Coastal Information Management System on behalf of Marjolaine Krug, the OCIMS programme manager

The gap between engineering and science  

The concept of a digital twin is not new. Digital twins are routinely used in engineering and technology to design the world around us. In principle, we create a digital version of something in the physical world, real or imagined, which gives us the ability to test different configurations and assess which version we would like to build in the real world.

This is more easily done for a physical object, like a car, than it is for the ocean. For a start, it is impossible for us to observe the entire ocean, like it is for a car. In-situ ocean observations are expensive and provide us with only localised information in space and time. Satellite observations of the ocean are powerful, but only provide information at the surface, ignoring the complexity of processes at depth. Ocean models can provide us with data everywhere and all the time but will always be approximations of the real thing and need to be informed by the best observations we have.

Despite these challenges, digital twins of the ocean are in fact commonplace in the coastal and ocean engineering community. For example, engineers routinely build models of the coastal ocean to test different configurations of a port layout. They imagine scenarios of a port layout and build a digital twin of the port to test things like whether the entrance channel will fill up with sand too quickly, and whether waves inside the port will be small enough for operations to take place smoothly.

A key goal of the DITTO community is to connect the world of engineering and technology to the world of research. This concept resonates particularly strongly with me, as I spent the first nine years of my working life as a port and coastal engineer, making the jump to ocean modelling in the service of research relatively late in life. I have personally seen the chasm between the worlds of engineering and research, which DITTO aims to bridge.

Scientists are at the forefront of what we know about the ocean and the dynamics of how it changes over time, but do not typically engage on issues around how a proposed human activity, such as an aquaculture farm or ship bunkering activities, might impact and be impacted by the ocean environment. This task is typically left to the engineers, who have the difficult task of answering these tough questions on small budgets and tight timelines. This is often done in isolation from the ocean science community, who may then be prone to criticize the approaches taken by engineers.

Presenting on SAEON’s SOMISANA ocean modelling initiative

Discussing the intricacies of modelling coral reef atolls with Keston Smith from Woods Hole Oceanographic Institution

What are we doing in South Africa to bridge this gap?  

In South Africa, we are currently establishing the building blocks for our own digital twins of the ocean through the Ocean and Coastal Information Management System (OCIMS), led by the Department of Forestry, Fisheries and the Environment (DFFE) in partnership with the Department of Science and Innovation (DSI). This is done by leveraging ocean observations and model data maintained by the Marine Information Management System (MIMS), to provide various decision-support tools (DeSTs) which are co-designed with stakeholders.

If you are confused by this sea of acronyms, you are not alone. Earlier this year, a participant at one of our stakeholder engagements put forward the following analogy:

DFFE is building a sushi bar. At the centre of the sushi bar are the ingredients, which need to be cleaned and easy to work with (MIMS). The chefs then ask potential customers what they would like to see on the menu (DeSTs, co-designed with stakeholders) and go about using the ingredients to create these dishes (OCIMS).

What does success look like?

In my eyes, OCIMS will be viewed as a success when ocean observations and ocean model development are no longer viewed as purely esoteric research enterprises. Rather, the need to fund these activities will be viewed as an essential component of how our oceans are managed. To achieve this, the OCIMS decision-support tools must become an integral part of the management of our ocean space, such that they cannot be done without.

The continued and sustained acquisition of ocean data and development of ocean models can only have positive effects in the private sector. Engineering consultants, who typically need to rely on data provided by international institutes and/or local data paid for by their clients, will find themselves in an ever increasingly data-rich environment. A better shared understanding of our coastal and ocean environment means that uncertainties are reduced, which will in turn reduce the risks associated with private investment in the blue economy.

The need for funding of accurate atmospheric data and forecasts is not questioned, as they are already essential for the management of various terrestrial activities which are central to our economy. I sincerely hope OCIMS will achieve this success for our oceans.