Predicting the future of Aotearoa New Zealand’s coastline

Understanding the timeline of shoreline erosion with numerical models, satellite data and machine learning algorithms, is key to developing resilience.

An image of The Parade, Bucklands Beach. The road on the left and sea on the right are divided by a sea wall.
The Parade, Bucklands Beach.

Nearly two-thirds of Aotearoa, New Zealand’s population lives by the coastline. Our communities and infrastructure are inextricably intertwined with the coast, and the future of our shorelines seems to be playing on everybody’s mind, especially after the severe weather events of earlier in 2023.

Professor Giovanni Coco, from the University of Auckland School of Environment, is aware of the need to understand shoreline processes and to predict the future impacts of climatic change on our shorelines if we are to safeguard ourselves, and our coastal communities, more effectively in the future.

We’re entering a stage where we have a lot of data available and so that means we can better test models. The hope is that we can build better models, and yes, our ability to predict is improving.

Professor Giovanni Coco, Faculty of Science University of Auckland

Urban beaches

Broadly speaking, one of Giovanni’s research interests focuses on what happens at the very end of the huge body of work that is attempting to predict the impacts of climate change on the coast. This includes a variety of hazards, such as coastal flooding and erosion. One of the areas that Giovanni has become passionate about and intends to explore in detail is the future of urban beaches in Aotearoa New Zealand.

Urban beaches, such as those found in Auckland’s Waitematā and Manukau harbours and in Wellington, are beaches that the public tends to use a lot. Beaches in these areas are not exposed to the open ocean waves and are often backed by sea walls. They serve as popular recreational areas, but also as protection for public infrastructure and private properties.

While necessary as a kind of defence mechanism, the presence of sea walls on these urban beaches also restricts the natural movement of their shoreline in response to waves and rising sea levels. Because of this, as sea levels rise, Giovanni points out that these urban beaches will slowly but inevitably get narrower and narrower. The sea walls backing these beaches will also become less effective against more frequent inundation events.

The need to improve current strategies or develop alternative strategies to protect infrastructure and manage erosion in urban coastal areas like this is, therefore, becoming increasingly important.

We know our shorelines will be impacted by climatic change and are already aware of these impacts occurring. Giovanni’s job, as he says, is trying to predict when this is likely to happen.

Our changing coast

One of the ways in which Giovanni is working on these predictions is through a project funded by the Ministry of Business, Innovation, and Employment (MBIE) and led by Victoria University in Wellington called 'Our Changing Coast’. This multimillion-dollar project has multiple packages. Giovanni leads the one dealing with theimpacts of climate change on the coastline.

Giovanni and his team have developed numerical models that help to predict shoreline evolution also under the effect of sea-level rise. He and the team take predictions from global climatic models and scale those results into the local storm surge and wave climate. Using a variety of models, they have also calculated the projected wave climate and storm surge until the year 2100. This type of data is critical to assess what is likely to happen to our beaches in Aotearoa New Zealand.

Satellites and data

For ‘Our Changing Coast’, Giovanni’s team is collaborating with researchers at the University of New South Wales, who have developed techniques to extract shoreline data from satellite images.

To collect this satellite data, “First of all, you need the sky to be cloudless,” Giovanni chuckles, “which in New Zealand is a little bit of an issue.”

After that, using several images and correcting for tidal levels, researchers can see where the shoreline is.

Using this data, they then develop and test numerical models, also based on machine learning, that allow the team to develop the future of Aotearoa New Zealand’s beaches, under climatic changes.

The future of our beaches

So, what does this mean for your favourite local beach? Thankfully, we will not lose all our popular beaches. Giovanni explains that there are still a lot of beaches in Aotearoa New Zealand, where we haven’t built infrastructure too close to the dunes. Beaches like this are likely to move landward and, therefore, might get a bit smaller, but they will not disappear.

However, those urban beaches that are backed by sea walls have less room for movement. These beaches pose a unique challenge when it comes to shoreline erosion.

“It’s not a bright future for those beaches,” Giovanni says.

Predictions and projections

While predictions regarding shoreline erosion and sea level rise are valuable, Giovanni makes it clear that it is crucial to acknowledge the uncertainties associatedwith such long-term predictions.

“Every time you have a prediction, you also have to account for the uncertainties,” Giovanni explains.

Uncertainties become more and more significant the further into the future you try to predict. Giovanni points out that some scientists, therefore, wonder how significant the idea of ‘predictions actually is, and if we should even call them predictions at all.

Perhaps ‘projections’ is the alternative. “A projection,” Giovanni explains, “focuses on scenarios, on plausible futures. Whereas a prediction looks at the most probable future.”

Where to next?

Predicting shoreline erosion accurately has been challenging in the past due to the complexity of beach dynamics and sediment movement.

“There are a lot of things that we know that we don’t know,” Giovanni says.

Perhaps the key to filling those gaps is in harnessing artificial intelligence.

“I thought it was a good idea to try to explore the possibility of using artificial intelligence and machine learning algorithms to predict shoreline erosion. Right now, a PhD student is looking at what is called ‘deep learning’ to predict shoreline change, and his results are better than any existing model we compared.”

In the face of an uncertain climate future, Giovanni speaks positively about improving our ability to make predictions about the shoreline and coastal dynamics.

“We’re entering a stage where we have a lot of data available, so that means we can better test models. The hope is that we can build better models, and yes, our ability to predict is improving.”

Find out more