How do you map a sea floor for kelp restoration? How do you accurately map the impact of urchin removal or monitor the return of biodiversity?

At Ocean Green partner akvaplan-niva,Pierre Priou – a researcher specialising in active acoustics – has been doing something new: adapting a technique used to map the ocean floor, to look at whether the substrate is sandy or rocky for example, to also look at kelp and turn data into a map of the water column.

‘The novelty here is that we are using a small, autonomous vessel together with a type of sonar. or multibeam echosounder, to map kelp forests,’ he explains. ‘Most of these instruments were designed to map the seafloor but here, we are not only looking at the sea floor but also what is going on above, in the water column. This has not been done a lot – and in particular not from an autonomous vessel.’

The autonomous vessel in question is an Otter, an un-crewed, multi-beam seabed mapping device, which Priou has been using to gather data for the first test site of the Ocean Green project. There, a harvester developed by Ava Ocean will remove large numbers of native urchins in order to help restore a more natural balance in the water that allows kelp forests to return.

Northern Norway has lost some 80% of its kelp forests over the last 50 years, partly due to an over abundance of native urchins, which have become prolific in the face of fators including predator overfishing and climate change.

‘The Otter was originally developed to map the sea floor,’ explains Priou. ‘What we’re doing here using it to record sea floor data, but also what's going on in the water column. That is something new. When we send sound into the ocean, we get everything that's in the water column – fish, plankton, the seabed and some noise as well,’ continues Priou, who came to the project from a background in fish and acoustics. ‘It is more complicated than this but usually, the sounds you don’t want are discarded and the data is used to map the sea floor. We want to use it to also monitor the return of kelp.’

Finding the kelp to map

Because there currently isn’t much kelp around Tromsø, where the Ocean Green project is taking place, Priou explains that he has been mapping areas that are mostly urchin barrens – but is also developing the tools and analysis methods to manage kelp in the future. ‘I'm trying to find some areas where there is some kelp, so that we can work on the methodology and have an idea of what to expect when we begin to see the restoration we expect.’

So far, the Otter has mapped – and Priou has reported on – an area around 700m by 1 km. ‘The goal is to map kelp – to create maps of the density of returning forests.’ To do this, he sent the Otter into areas that volunteers have restored by hand. ‘We send sound into the ocean and get a raw output from the instrument that covers everything in the water column. The first step is then to clean this data, to remove unwanted echoes.’

Capturing density from the echo sounder is not straightforward, he explains. ‘The way we usually evaluate density is through extremely fine-resolution imagery, where we could almost count single kelp.’ In reality though, he says this is not doable because of the simple fact that kelp moves in the water column and even an air bubble within the forest can impact the echo. The solution ‘is to take a small volume and study the echo within that, meaning that this back scatter would become representative of kelp. It requires some careful work.’

The size of the data set also poses a challenge. ‘It is a gigantic data set – probably the largest I've worked with. We’re talking about several hundreds of gigabytes.’ At the moment, that data is just point data – depth and coordinates – but Priou sees potential to develop it in a lot of different visual ways – including the possibility of 3D mapping.

Ground truthing with traditional methods

Because what Priou is doing is relatively new, it is also being backed up by the work of human divers. At the pilot area, this has been done by Camilla With Fagerli and the team from niva, another Ocean Green partner. They have been conducting ‘ground truth’ research designed to assess the accuracy of the echo data – to ‘estimate or detect the kelp forest and the biomass of the kelp when the harvest is done,’ she explains.

Having been to both the pilot site – where an overabundance of urchins means there is no longer any kelp – and a restored site, Fagerli talks about how drastically different the two areas are. ‘It’s two very different environments. In the area where we will test the harvester, we were diving with a frame that we put out every second meter for 60 meters, counting everything we saw – and it was urchins, maybe a few snails, maybe some blue mussels and nothing more. It's not very diverse. It's very heterogenic with quite a high density of sea urchins and it was a quick count.

‘The day after, we did this same thing in the small area where volunteers have restored the kelp forest. Even though it's only 18 months since the first kelp returned, it's already very different. We dived for just six frames but it took two hours to record every species we saw.’