Do Marine Reserves Help Snapper Fishing? What the Evidence Actually Shows

The argument is appealing. Close a patch of coast to fishing. Wait. Watch the snapper population recover. Watch fish — bigger, more numerous — spill out of the reserve into surrounding water where they can be caught. Net result: a snapper fishery that’s better than the one before the closure, plus a working ecosystem refuge.

It is half right. The first part — fish recover dramatically inside no-take reserves — is among the best-documented findings in New Zealand marine science. The second part — the spillover that compensates surrounding anglers for the closure — is genuinely contested, methodologically fraught, and a fair distance from the slogan version. This piece walks through what the evidence actually shows, what is still in dispute, and what it means for the Revitalising the Gulf debate.

This is a companion piece to our Hauraki Gulf snapper recovery story and our SNA1 allocation explorer. Closing water to all fishing is — in effect — another way of cutting catch, and the case for it stands or falls on whether the spillover claim survives scrutiny. For an interactive map of every NZ marine reserve plus the Sea Change proposed sites, see our marine reserves explorer.

What Leigh actually shows

The Cape Rodney to Okakari Point Marine Reserve — most people call it Leigh, or Goat Island — was New Zealand’s first no-take marine reserve, established in 1975. Five kilometres of coastline north of Auckland. By the time the closure went in, the inshore snapper population in the area had been heavily fished for decades.

Within twenty years of closure, monitoring inside the reserve was showing snapper densities an order of magnitude above adjacent fished water. Willis, Millar and Babcock (2003) measured legal-sized snapper biomass inside Leigh at roughly 14 times the density observed at matched fished sites. Mean snapper length inside the reserve has been consistently around 30 to 40 percent greater than outside. The big-fish effect — bigger spawners, more eggs, longer-lived individuals — is the strongest single signal in the Leigh dataset.

The follow-on ecosystem effect at Leigh is just as striking. Big snapper eat sea urchins (kina). Suppressed urchin populations let kelp forest re-establish. Shears and Babcock (2003) documented the trophic cascade in detail: the percentage of reef bottom dominated by kelp inside the reserve climbed from a low base to substantial cover within a couple of decades, while adjacent unprotected reefs remained stuck in kina-barren state. The reserve was never designed to do this; it dropped out of the data once a top predator was allowed to recover.

This is the part that’s not seriously disputed. No-take marine reserves dramatically restore fish populations and trigger cascading habitat recovery, in places that have already been fished hard. The case studies are robust, the mechanism is understood, and the effect has been replicated across multiple New Zealand reserves and many international ones.

The spillover question

The harder question is what happens at the boundary. If a reserve has a wall of snapper inside, and fished water outside, do enough fish move across the boundary to compensate the displaced anglers? This is where the evidence gets thinner and the methodology gets harder.

Kelly, Scott and MacDiarmid (2002) measured the spatial gradient of snapper abundance away from Leigh’s boundary. They found a detectable spillover signal — fish density was elevated for about 700 to 1,000 metres outside the reserve relative to sites further away. The effect was real but spatially narrow. For an angler fishing 5 kilometres away from the reserve boundary, there was no measurable benefit.

This is the standard finding for snapper-style species across the global marine-reserve literature. Spillover exists. It is detectable. It is also small and localised compared with the loss of access inside the reserve. The reserve is, on balance, a transfer of fish from anglers to non-anglers, with a modest fringe benefit to anglers immediately at the boundary.

Larval export — adults inside the reserve produce eggs that drift on currents and recruit to the broader fishery — is a separate, larger-scale mechanism that’s much harder to measure. There are theoretical reasons to expect it matters at population scale, especially for stocks like SNA1 where the Leigh reserve sits within the major nursery zone. But quantifying the contribution of a 5-kilometre reserve to the recruitment of a stock that ranges from East Cape to Cape Reinga is an open research problem.

Why the science is contested

Even the strong signals at Leigh come with methodological caveats that the research community takes seriously.

Control site selection. Every patch of coast is different. Habitat, exposure, depth profile, current pattern, baseline fish density before the closure. When you compare snapper density inside Leigh to a “control” site somewhere else, you’re not comparing like with like — you’re comparing two slightly different patches of coast, one of which happens to be closed. Better designs use multiple control sites and statistical adjustment, but no design eliminates the underlying problem.

Before-After-Control-Impact (BACI) data is rare. The strongest study design — measure baseline before closure, measure for years after at both reserve and matched controls — exists for some NZ reserves (Tāwharanui, established 1981 with formal pre-closure surveys) but not for most. For Leigh, the closure happened in 1975, before systematic underwater visual census methods were standard. We are partly inferring the pre-closure baseline.

Effort displacement. When you close 5 kilometres of coast, the anglers who used to fish there don’t disappear. They move to the next-best ground. So a fishing-pressure increase shows up just outside the reserve, partially confounding any spillover-driven density increase. Some fish-density-vs-distance gradients in the literature are picking up displaced fishing pressure as much as spillover. Disentangling the two requires effort data that’s often unavailable.

Stock-wide recovery confound. SNA1 has been on a rebuild trajectory since the 1980s. Snapper densities have increased everywhere on the upper-NI east coast, not just inside reserves. Separating the reserve-specific signal from the broader stock-recovery signal requires either matched controls or very long time series, and ideally both.

Hot-spot bias. Reserves are sometimes placed in productive habitat to begin with. The big-fish-inside effect at Leigh is real but partly reflects that Leigh is on a piece of coast that was always good snapper country.

None of these caveats invalidates the basic finding. They mean the headline numbers should be treated as informed estimates rather than precision measurements.

Tāwharanui and the cleaner experiment

Tāwharanui Marine Reserve, north of Whangaparaoa, was established in 1981 (initially as a marine park, fully no-take from 2011 after legal upgrades). Critically, baseline surveys were conducted before the original closure, giving researchers a proper before-after comparison. Denny and Babcock and several follow-up studies used Tāwharanui to validate the Leigh findings against a stronger experimental design.

The headline result was that the snapper recovery and the kelp-cascade effect both replicated. After full closure in 2011, snapper density inside Tāwharanui climbed substantially within five years, with the rate of increase consistent with what Leigh showed in the late 1970s and early 80s. This is one of the closer-to-controlled-experiment data points in the NZ marine-reserve literature, and it broadly confirms the Leigh story.

Who pays, who benefits

The reserve debate is often framed as a conservation-vs-extraction trade-off, but that is too coarse. The benefits and costs both flow to specific groups.

Leigh attracts roughly 350,000 visitors a year. The dive and snorkelling industry built around Goat Island is the largest concentration of marine ecotourism in the country. Auckland University’s Leigh Marine Laboratory does world-class research on the back of the protected ecosystem next door. The visitors are a fairly specific demographic — overwhelmingly urban, often international tourists, mostly middle-class. The benefits are real but they accrue to a particular group.

The cost-bearers are different. Local commercial fishers excluded from the area when the reserve was established. Local recreational anglers, including the Mahurangi Club anglers who used to fish the same coast. Local Māori, whose customary harvesting rights were curtailed even where they were grandfathered into reserve regulations. These groups are fewer in number but more directly affected. They tend to live in the area year-round; they tend not to be the ones running the dive operations.

This is not a reason against marine reserves. It is a reason to be honest about who benefits and who carries the cost when one is established. The benefits are diffuse, urban, and ecological. The costs are concentrated, local, and economic. The political economy of reserve creation tracks this: support tends to come from cities, opposition tends to come from coastal communities adjacent to proposed sites. Dismissing the local opposition as anti-conservation misses what’s actually happening.

What this means for Revitalising the Gulf

The Hauraki Gulf Marine Park, established in 2000, is not a no-take reserve. It’s a management overlay that allows fishing within the park subject to normal SNA1 rules. The genuine no-take protection in the Gulf consists of small reserves: Leigh, Tāwharanui, Goat Island (the no-take part), Long Bay-Okura, Te Matuku on Waiheke. Together these cover a small fraction of the Gulf’s area.

The Sea Change Tai Timu Tai Pari plan (2017) recommended a substantial expansion of no-take and Type 2 protected zones. The Revitalising the Gulf strategy (2021) carried that forward but implementation has been slow and contentious. The Ahu Moana co-management framework is the current vehicle, with iwi partnerships shaping site selection. Several proposed sites — High Court Reef, Slipper Island Reef, Mokohinau-adjacent water — would meaningfully change the shape of Gulf fishing if implemented.

What the spillover evidence says about that programme: a network of reserves is plausibly more effective than a single big closure for stock-level effects, because larval export happens at population scale. If Revitalising the Gulf’s MPA network goes ahead, the strongest realistic outcome is reserve-internal recovery (high confidence), local kelp-cascade effects (high confidence), narrow boundary spillover for adjacent anglers (modest confidence), and stock-level recruitment contributions across the wider Gulf (uncertain). The displaced commercial and recreational catch will land elsewhere in the Gulf; that’s not gone, it just shifts.

What a good next study looks like

Several questions remain genuinely open and would benefit from new research:

• Better quantification of the spillover gradient — how far from a reserve boundary the fishing benefit extends, and how big it is, with proper effort-displacement controls.
• Stock-level larval export modelling using the recent generation of NZ oceanographic models. The 5-kilometre reserve at Leigh is small relative to SNA1, but if its export contribution is even a small percentage it would be a non-trivial input to the rebuild.
• Pre-closure baseline surveys for any new reserve proposed under Revitalising the Gulf. The Tāwharanui example shows what proper BACI data buys you scientifically, and it is cheap to do compared with not doing it.
• Long-term tracking of the displaced fishing effort — where it goes, what it catches, whether the cumulative effect on the broader fishery is positive, neutral or negative.

None of this is novel research direction. NZ marine science has the people and the tools to answer these questions. They have not been answered yet because they require sustained funding and access, not because they are conceptually hard.

The honest bottom line

No-take marine reserves restore fish populations and ecosystems inside their boundaries. That is settled science. They produce modest, narrow spillover into adjacent fished water. They are not, on the available evidence, a way to make surrounding snapper fishing dramatically better; they are a way to protect an ecosystem inside the boundary while shifting some fishing pressure elsewhere.

If you support reserves on conservation grounds — biodiversity, kelp recovery, tourism, scientific reference — the case is strong. If you support them on the grounds that they will improve your fishing — that case is weaker than it’s often presented as. Be careful which version of the argument you’re buying.

Sources and further reading

• Babcock RC, Kelly S, Shears NT, Walker JW, Willis TJ (1999). Changes in community structure in temperate marine reserves. Marine Ecology Progress Series 189: 125-134.
• Willis TJ, Millar RB, Babcock RC (2003). Protection of exploited fish in temperate regions. Journal of Applied Ecology 40: 214-227.
• Kelly S, Scott D, MacDiarmid AB (2002). The value of a spillover fishery for spiny lobsters around a marine reserve in northern New Zealand. Coastal Management 30: 153-166.
• Shears NT, Babcock RC (2003). Continuing trophic cascade effects after 25 years of no-take marine reserve protection. Marine Ecology Progress Series 246: 1-16.
• Denny CM, Babcock RC (2004). Do partial marine reserves protect reef fish assemblages? Biological Conservation 116: 119-129.
• Hauraki Gulf Marine Park Act 2000.
• Sea Change Tai Timu Tai Pari plan (2017).
• Revitalising the Gulf strategy (Department of Conservation, 2021).

Related reading

• The Hauraki Gulf snapper recovery — the wider rebuild story this piece sits inside.
• SNA1 commercial vs recreational allocation explorer — how the catch from the rebuilt stock is split, with reserves as another implicit allocation lever.
• Twenty years of Hauraki Gulf sea surface temperature — the climate variable pressing on every reserve effectiveness question.
• Best time to fish for snapper in Auckland, Coromandel and Northland — the main angler’s guide.