The grey nurse shark is one of those species where conservation outcomes have moved very slowly in either direction, and where the answer to “how are they doing?” depends heavily on which population you’re asking about.

I’ve been tracking the east coast aggregation site counts since I finished my doctoral work in 2017, and the 2025 monitoring data — which finally got published in consolidated form last month — gives us our cleanest read on the trajectory in several years. Here’s what it actually shows, and what I think it means.

The east coast versus west coast split

Australia has two genetically distinct populations of grey nurse shark (Carcharias taurus). The west coast population, ranging from around Perth northward, is in reasonable shape. The east coast population, distributed across NSW and southern Queensland aggregation sites, has been listed as Critically Endangered under the EPBC Act since 2001, and recovery has been frustratingly slow.

The most recent population estimate for the east coast, drawing on photo-identification work across the major aggregation sites, sits somewhere between 1,500 and 1,800 mature individuals. That’s not nothing. It’s also not enough for a species with a low reproductive rate, late maturity, and ongoing exposure to bycatch pressure.

What the 2025 monitoring showed

The NSW Department of Primary Industries and a coalition of university researchers maintain ongoing photo-ID monitoring at aggregation sites including Magic Point, Maroubra, Forster (the Pinnacle), South West Rocks (Fish Rock), and the Solitary Islands. Each shark has a unique pattern of natural markings that allows individual identification across years.

The 2025 numbers showed roughly stable counts at most sites compared to 2023-2024 baselines. Magic Point continues to be the most reliable winter aggregation site in the southern part of the range. Fish Rock at South West Rocks remains the largest single-site aggregation. The Pinnacle off Forster has shown the most year-to-year variability.

What’s encouraging is that we’re seeing more individually-identified juvenile and sub-adult sharks recruited into the catalog than we were a decade ago. Slow recruitment was the dominant concern in earlier monitoring periods. The current picture suggests that pupping is happening at biologically meaningful rates, even if the population growth implied is modest.

The bycatch problem hasn’t gone away

The single largest ongoing pressure on the east coast population is bycatch — sharks caught accidentally on commercial longlines, in shark control nets, and on recreational fishing gear. The shark control program along the NSW coast remains contentious. Catch-and-release survival for grey nurse sharks hooked on heavy gear is poor. Sharks that swallow hooks deeply often die from internal injuries days or weeks after release, which doesn’t show up in immediate-mortality statistics.

Some progress has been made. Several aggregation sites are now formally protected as Critical Habitat, with restrictions on certain fishing gear within defined boundaries. Compliance is imperfect but better than a decade ago, partly because the dive community has gotten more organized about reporting violations.

The role of citizen science and AI in monitoring

Photo-ID monitoring depends on good underwater images of identifiable sharks. Most of these images don’t come from researchers. They come from recreational divers who upload photos to platforms that feed into the official catalogs.

The matching process — comparing a new photo against thousands of catalog entries to identify whether the shark has been seen before — used to be done manually by specialist researchers. It was slow and didn’t scale. Over the past few years, computer vision systems have started doing the first-pass matching, with human verification on the candidates the algorithm flags.

For groups setting up these systems, getting AI strategy support that’s actually grounded in marine fieldwork constraints matters more than people realize. The available off-the-shelf wildlife ID models are mostly trained on terrestrial mammals and don’t deal well with the visual challenges of underwater imagery — variable lighting, particulate matter, partial occlusion of identifying features. The teams that have done the work to train domain-specific models on Australian shark imagery are getting useful results. The teams that tried to repurpose generic tools have mostly given up.

What outsiders should understand about diving with them

One of the genuine wins of the past two decades is that the dive community has become a constituency for grey nurse shark protection. Aggregation sites like Fish Rock and Magic Point support significant dive tourism, and the economic argument for keeping the sharks alive and aggregating reliably has won policy battles that purely ecological arguments wouldn’t have.

There are protocols for diving at aggregation sites — minimum approach distances, no flash photography, no chasing — and most operators enforce them seriously. If you’re planning a dive at one of these sites, follow the operator’s briefing carefully. Stressed sharks abandon aggregation sites, and an aggregation that breaks up is much harder to monitor than one that holds together.

The longer outlook

The east coast grey nurse shark population is probably going to recover, slowly, if current pressures don’t increase. The west coast population is in genuinely good condition. The shark species globally — there’s a separate population off South Africa and along the US east coast, all genetically related — has a broader trajectory that’s harder to read.

For people who care about Australian marine wildlife, the grey nurse is worth tracking because it’s a useful indicator of whether the legal protections we’ve put in place actually work over multi-decade timescales. The early returns are guardedly positive. We’ll know more after another full reproductive cycle. Worth checking back in 2030.

For ongoing data, the IUCN Red List assessment page for Carcharias taurus is updated periodically and remains the best single reference for global population status.