It was a feel-good story that, for once, seemed to justify the introduction of an exotic species.

When a delegation of Winnemem Wintu Indians arrived in Canterbury from the US in March, they wanted to take home eggs from chinook salmon, introduced from Northern California to the South Island a century ago.

Tribal salmon runs had declined when a dam blocked rivers in the 1940s, so New Zealand had become a chinook sanctuary. A feel-good story no doubt. But also a lesson in the importance of hanging on to what you've got.

When it comes to fish, New Zealand is often held up as an exemplar of what happens when you get things right. Adoption of the Quota Management System, introduced by a Labour Government in 1986, was a milestone in the fishing industry.

Instead of treating fish as a food source to be exploited without restraint, it married conservation and economics by leasing the perpetual right to harvest quotas, set by government, within our 4.4 million sq km Exclusive Economic Zone (EEZ), the world's fourth largest. The system gives the industry a vested interest in maintaining stocks and ensuring that no one violates the compact. Defaulters face stiff penalties, including the loss of leases and confiscation of vessels.

"The QMS is imperfect," says Wayne McNee, chief executive of the Ministry of Fisheries (MoF). "But it certainly puts incentives in the right place, as opposed to any other system I've seen for managing fisheries to protect the commercial resource into the future."

Last year Fisheries Minister Phil Heatley trumpeted a report in the US journal Science that praised New Zealand's fishing record. "We can justifiably be proud that, along with Alaska, we are regarded internationally as leading the world in terms of management success, by our efforts to put management interventions in place before drastic measures are needed to conserve, restore and rebuild our marine resources," Heatley said.

Nonetheless, he warned, New Zealand could not afford to rest on its laurels. Heatley cites our hoki fishery as "a good example of rigorous management delivering sustainable stocks into the future."

As the world's demand for fish grows, fishing already ranks as one of our largest export industries. While our 440,000 tonne annual catch last year was tiny compared to the 92 million tonne world harvest (in 2006), we have a reputation for producing a high-value product.

Last year the total value of our seafood exports was $1.4 billion, out of a total quota value of $4 billion, says lobby group the Seafood Industry Council.

Crayfish, hoki and squid were the top export earners. In 2007 the Ministry of Fisheries calculated that there were 1644 quota owners, 1277 fishing vessels and 7155 full-time staff in the industry (the Seafood Industry Council quotes almost 27,000 people employed by the industry, directly and indirectly).

But are there weaknesses in our management system that might put this asset at risk? For while the ministry and the industry laud the QMS as sustainable - and imitated by Alaska and Iceland - environmentalists see flaws.

In February, Charles Clover, environmental correspondent for the London Telegraph, told the Herald, "the closer you look at [the QMS], the more holes there are in it". While the QMS managed 628 fish stocks, detailed information was only available on 117 stocks or sub-stocks. Clover, whose 2004 book The End of the Line inspired a 2009 documentary of the same name that slammed industrial fishing, warned that this ignorance was "the No 1 hole" in our system.

"Of the assessed stocks," he said, "32 per cent are defined as depleted or collapsed, which is more than in America."

And in March, John McKoy, chief fisheries scientist at the National Institute of Water and Atmospheric Research, told the Dominion Post that stock research was "highly susceptible to influence," and often based on guesswork.

Despite Heatley's enthusiasm for hoki, which garnered an eco-certification from the Marine Stewardship Council (set up by corporate and green interests) the British supermarket chain Waitrose banned hoki sales because of the damage done to the seafloor by bottom trawling. And then there is orange roughy, a species heavily fished by Kiwi boats. The Ministry of Fisheries has closed three of eight orange roughy fisheries.

McNee concedes the QMS is not perfect and could do with improvement. "It is not aligned perfectly by any means with environmental interests. It doesn't take into account necessarily the impact on non-target species. So marine mammals, seabirds and other by-catch need protections."

Greenpeace NZ argues that the QMS is unsustainable. Citing a 2007 Forest and Bird assessment of 75 commercial fisheries, Greenpeace says none have a management plan, 26 are over-fished or have experienced a substantial decline in stocks, 51 cause habitat damage, 42 kill significant numbers of seabirds, 45 kill significant numbers of marine mammals, 58 cause adverse ecological effects, and 64 catch too many non-target fish. Moreover, industry litigation has often overturned government efforts to regulate catches. The most common fishing technique, trawling, has a large by-catch.

This grim toll could harm foreign sales if retailers and consumers boycott fish from New Zealand. Fish are swimming on to the global conservation radar, joining such iconic species as whales and tigers.

And bad news travels faster in the internet era. Our decision not to support efforts to list Atlantic bluefin tuna as endangered, even as we took 27 per cent more southern bluefin tuna, didn't win us any brownie points. Nor did it look good when New Zealand boats, seeking Antarctic toothfish, were implicated in the decline of Ross Sea orca, which eat toothfish.

Industrial fishing is usually viewed as a post-World War II phenomenon, and for decades fleets bristling with high-tech detection gear and armed with huge bottom trawls and hooked lines many kilometres long, plundered fish as an inexhaustible resource. The 1992 collapse of the Grand Banks cod fishery, off eastern Canada, and concerns about over-fishing, toxic algae and bacteria, pollution and climate change have failed to check the slaughter.

But a steady drip of scientific papers sent out disturbing signals. In 2003 Boris Worm and Ransom Myers, writing in the British journal Nature, warned that 90 per cent of big predator species had vanished since 1950. "Industrial fishing has scoured the global ocean," they said. "There is no blue frontier left." By 2006 Worm was warning that, unless the world's fishermen were made to limit their catch, "100 per cent of [fished] species will collapse by 2048, or around that."

Worm drew fire because two of the 15 stocks he cited had not crashed as precipitously as he claimed. But, like the furore generated by climate change deniers over the false claim that Himalayan glaciers would vanish by 2035, the criticism missed the wood for the trees: market-driven systems were fast squandering vital food supplies.

Fortunately, as the Winnemem Wintu quest for chinook salmon indicates, New Zealand may have an edge as a sanctuary for fish species. Our EEZ has 10 fish management areas. "What people don't realise is how diverse marine systems are," explains Pamela Mace, the MoF's chief scientist.

Bill Ballantine, a biologist with the University of Auckland's Leigh Marine Laboratory, says "New Zealand has 16,000 marine species that we know about, but there are probably at least as many we haven't discovered yet." After 50 years studying sea life, during which knowledge has quadrupled, he is most impressed by "how ignorant we are".

The MoF's aim for fisheries and aquaculture, set out in the report Fisheries 2030, is to balance profit with conservation. "We don't expect there will be greater catches," says McNee. "Catches go up and down depending on the state of the fishery. It's about working with the [industry] so that they can get more value from the catch they're getting."

This involves a regulatory review, relaxing rules that "constrain" the industry. For example, the MoF is open to industry improvements on bird-scaring devices that achieve the same goal for less cost.

"We've still got a lot of regulations that interfere with the operation of the QMS," says Nici Gibbs, policy manager with the Seafood Industry Council. The council wants to free up money to invest in market research and product development.

But isn't deregulation a threat to conservation? "Actually, there is no tension between commercial imperatives and sustainability," Gibbs insists. "The main thing about the QMS is that it is a perpetual property right: the value of today's asset is the worth of all the future assets consolidated into today. So it immediately has a long-term aspect, which provides property owners with an interest in stock sustainability. And each quota is proportional. It goes up and down and all owners feel the effects." She says this creates a "strong interest" in sustaining stocks.

However, true sustainability will demand more than fine-tuning rules.

Science has a major role to play. Under the user-pays system, the industry funds much of the research behind the QMS. It may need to fund far more: the NZ Biodiversity Strategy, published in 2002, admitted that New Zealand lacked enough knowledge to know whether we were managing our marine environment sustainably.

Fisheries' science budgets have shrunk by over half since the mid-1990s, even as the industry targeted more fish stocks. "Less is known about more and more," warned McKoy, who said industry-funded research focused on profitable species, a trend that risked missing warnings elsewhere.

Gibbs says science is "increasingly collaborative" and that the industry's role is to "act as a watchdog on the MoF's science processes. Making sure the science is robust."

While "at risk" species are checked, MoF science focuses on the most valuable commercial species. "We try and monitor everything in the system," says the ministry's Mace. "But some of it isn't monitored quite as intensely." It is a selective strategy that makes it hard to gauge the impact of climate change, for example.

Given the MoF's tiny science budget, work on climate change is mostly limited to keeping a watching brief on overseas research.

Targeting fish without full scientific knowledge can have ruinous consequences, as we discovered after hammering orange roughy, which take 20 to 32 years to reach sexual maturity, and can live to 100 years. Bottom trawling devastated the species, its habitat and caught other species.

Besides assessing a stock's commercial value, it helps to know about fish lifecycles and how they interact with ecosystems. Should we set quotas lower to allow for biological variation within target and by-catch species? Crucially, how much should we err on the side of caution when setting quotas, even if this reduces profits? This question goes to the heart of the sustainability debate.

Sustainability is a loaded word. Does it mean a sustainable economy, a sustainable ecosystem, or a combination of both? Getting it right is critical. Fisheries 2030 emphasises sustainability and the MoF is adopting eco-management: looking at fish populations in the context of complex and interwoven ecosystems which shelter target stocks.

The devil is in the details, says Daniel Pauly, a marine biologist and global fisheries expert at the University of British Columbia, in Vancouver. "You cannot embed fisheries management in an ecosystem unless you assess the effect of fishing on an entire ecosystem. Fisheries management is only half the story. You must protect the environment."

The MoF, says McNee, is developing plans to think of EEZ fisheries as "a whole", a consultation process that considers, for instance, how different stocks interact. He says awareness of fishing's ecological impacts is growing. Having a sustainable fishery is "a price of market access".

The industry believes it has already captured the environmental high ground with what are known as Benthic Protection Areas, or BPAs - areas where bottom trawling and dredging have been banned. They cover 32 per cent of our EEZ to protect seafloor - or benthic - habitats.

"More than any other natural resource-based industry in New Zealand, our economic destiny is inextricably linked to our environmental practices," Owen Symmans, the Seafood Industry Council's chief executive, wrote in the Herald last November. He says BPAs were driven by the industry and wants policy set for all economic activity, including mining or oil exploration, in the EEZ.

"Different people have different views about what an ecosystem approach to fisheries management means," says Mace. "In my view what it means, and this is essentially what the ministry is doing, is that we have a quota system that requires getting TACs [total allowable catches by all fishers] and changing them as needed." Other restrictions can include fishing methods, the size of fish caught and when and where fishing can take place.

The MoF also does multi-species surveys. A hoki study on the Chatham Rise records over 500 species, including commercial species such as ling and hake.

"You can track if the abundance of animals is going up or down, or staying stable, from the point of view of individual species," says Mace. "We can also see if the ecosystem is changed." Nonetheless, she agrees that TACs are plagued by "high uncertainty" due to limited data. "Dealing with uncertainty is a really big part of fishery science," says Mace. "Essentially, you need to take a precautionary approach." She says the trend is to fish the lower levels of quota estimates and believes our stocks are "in really good shape" compared to the rest of the world.

This merely shows how bad things are elsewhere, retorts Greenpeace. Fisheries are managed at "the absolute extreme end", says Karli Thomas, Greenpeace NZ's ocean campaigner. It is a risky policy that doesn't insure against shocks. "What we advocate - and it can be done with the QMS - is an ecosystem approach." The precautionary principle, she says, spares more fish as insurance against shocks like disease or climate change. It is a holistic viewpoint, embedding market-based economics in sustainable ecosystems.

"The problem is that catching fish next week is quite different than the ecological sustainability of the marine environment," says Raewyn Peart, senior policy analyst with the Environmental Defence Society and author of Managing the Marine Environment. "We know so little about how many fish there are. It's not as if we can go and count them." She believes the EEZ needs a new legal framework, akin to the Resource Management Act, to gauge ecological impacts.

The Fisheries Act addresses utilisation. "It's there to promote the fisheries industry, to manage it in a way that's acceptable," says Peart, who cites the rapid decline of sea lions - accidentally caught by boats fishing for squid - and set nets killing Hector's dolphins as "flow-on effects" that are poorly managed by the QMS. "Modern marine reserve laws are there for biodiversity protection. But we don't have that. The Fisheries Act doesn't protect the whole ecosystem."

Ballantine, who helped create the marine reserve at Goat Island near Leigh, believes no-access zones are the solution. He suggests marine reserves in fish-rich "diversity hotspots" need to cover at least 10 per cent of our waters to promote science, 20 per cent for conservation, and 30 per cent if we want to sustain our commercial fishing industry.

Greenpeace suggests 40 per cent be set aside as part of a global network. Reserves, which benefit adjacent areas via the "spill over" of aquatic life, are catching on elsewhere: California has designated 18 per cent of its waters as no-catch zones.

Such visions meet a tepid reception in the fishing industry. Both the MoF and the Seafood Industry Council say reserves may protect biodiversity but the QMS is the best fisheries management tool. But why not have both? Given the gaps in scientific knowledge of our EEZ, including the possibility of finding unknown species, this is like clear-felling rainforest without searching for beneficial plants.

A nascent overseas legal movement regards such devastation as "ecocide" and wants legal consequences for offenders. When I suggest the industry needs to get in front of the issues and press for more reserves, Gibbs points to BPAs. It's all about risk assessment. Certain activities cause risks to biodiversity. Others don't. "So why ban the activities that don't cause risk to biodiversity?"

Mace says science "errs on the side of caution". But how far should science err? There are some nasty threats out there.

Ultimately, the debate comes down to short-term versus long-term visions. Between cashflow pressures and ecological concerns. Wild fish stocks are only as sustainable as their habitats. Fisheries 2030 pushed out the boat. We have a way to go.