Government discarded expert advice not to use 'potentially misleading' data about dolphin threat

Thursday, 20 June 2019

WWF NZ researcher Amanda Leathers explains why Hector's and Māui dolphin are on the brink of extinction. (First published May 9, 2019)

A major finding from a government proposal to increase the protection of endangered dolphins continues to be fiercely disputed by some scientists, one of whom said it was plainly wrong and amounted to misinformation.

It has prompted questions about why the data was included in a public discussion document, contrary to advice from the government's own panel of international experts, which warned the figures 'could be seriously misleading'.

A member of that expert panel said he found it 'hard to believe' the data in dispute was accurate, but praised the government agencies involved and said the overall process had been fair and constructive. Fisheries New Zealand, which jointly authored the document, said it had confidence in its risk modelling, and the document was the first step in a wider process and further research was necessary.

Massey vet clinic pathologist Wendi Roe does a post mortem on an adult female Hector dolphin, found on a beach in Hokitika.

The dispute concerns new data laying out the major threats to Māui and Hector's dolphins, two sub-species only found in New Zealand. Māui dolphins are considered the world's most endangered dolphin, with an estimated adult population of 63. Hector's dolphins number around 15,700, and are deemed 'nationally endangered'.

* Three Hector's dolphins killed in net off Canterbury coast

* United States adds Māui's and Hector's dolphins to endangered species act

A Maui dolphin surveyed off Manukau Harbour in 2017.

* Ministry for Primary Industries denies claims of Maui's dolphin death cover-up**

A 40-page discussion document, jointly produced by Fisheries New Zealand and the Department of Conservation (DOC), included data showing the major threat to both sub-species was toxoplasmosis, a parasitic disease spread through cat feces which can infect dolphins when washed into the ocean through stormwater run-off.

It contradicted conventional wisdom that bycatch from commercial fisheries was the primary threat.

The difference in the scale of the two threats was stark: For every 0.12 Māui dolphins killed in a net annually, 1.9 would be killed by the disease, the figures show. For every 58 Hector's dolphins killed annually in a net, 334 were killed by the disease (all figures were averaged estimates).

The figures suggest the disease kills 16 times more Māui dolphins and six times more Hector's dolphins than commercial fishers.

The data in dispute, as presented in the discussion document.

A New Zealand marine mammal expert said the numbers were plainly wrong.

'This is as close as you can come, in my view, to lying without actually lying,' said the University of Otago's Professor Steve Dawson, a marine biologist specialising in Hector's dolphins. 'I think it's extremely misleading.'

The expert panel had itself cautioned against using the figures, in one of its 37 recommendations to the government. If the toxoplasmosis numbers were accurate, it said, it would mean the population of Hector's Dolphins was 'in rapid free-fall towards extinction'.

'[W]e are not convinced that it is appropriate for the toxoplasmosis necropsy data to receive the full modelling treatment: the uncertainties and potential biases in these data are too large,' the panel wrote.

The concern was in the way the data was collected: It came from necropsies carried out on 31 dead dolphins found washed up on beaches, in which a cause of death could be established.

Nine of those dolphins had died from toxoplasmosis.

Dawson pointed out that using dead dolphins as a representative sample of the living population was flawed.

He likened it to estimating the risk of pneumonia to humans: Many people died of pneumonia, but nearly all who did were either very old, very young, or chronically ill. Just because a dolphin died of toxoplasmosis did not mean the disease posed an equal threat to every living dolphin: 'The question of what carries you off in the end isn't the real issue,' he said.

A tribute recognising all of the Māui and Hector's dolphins who had died, held in Auckland in 2018.

The panel had pointed this out, too, and said assuming the dead dolphins were representative of the living population, 'or even a rough approximation' of it, was problematic.

'There is no reason to believe that beach-cast carcasses (and particularly such a small sample of these) are representative of deaths of all kinds throughout the dolphin population,' it said.

If the government chose to present the toxoplasmosis data, it should do so while detailing its strengths, weaknesses and biases.

Whether the government did this is debateable. A footnote beneath the data refers to uncertainties, but with little detail.

A broader explanation is included in a 120-page supporting document on the Fisheries New Zealand website, which says there is 'substantial uncertainty around the estimated number of toxoplasmosis-related deaths'.

A Fisheries New Zealand spokesman said the uncertainties in the modelling were made clear 'both in the discussion document and the longer technical document on [the agency's] website'.

A Hector's Dolphin with its calf in Akaroa harbour.

He said the risk modelling was the first step in a wider process, and the agency had made it clear more research was needed to establish the risk of toxoplasmosis to the dolphin populations.

The panel also recommended greater clarity about observer coverage and how that factored into calculations around bycatch rates.

Bycatch data comes from government observers on fishing boats, who record the deaths of non-target species.

Observer coverage varies around the country; in parts of the Māui dolphin habitat, coverage is 100 per cent, but in parts of the South Island, coverage can be as low as two or three per cent.

The panel said it should be abundantly clear that observer coverage is low in some places.

'We are particularly concerned about making it transparent to readers that many areas have little to no observer coverage and even in those that do, observer coverage is low and often from some time ago,' the panel wrote.

This was not reflected in the public discussion document, which makes no reference to observer coverage.

The Fisheries New Zealand spokesman said low levels of observer coverage were factored into its modelling, and the explanation of the factors underpinning the modelling was included in the appendices of a supporting technical document on its website.

The panel experts contacted by Stuff said they were still absorbing the proposals, and would be able to provide a considered response at a later date.

One of the experts, Mike Lonergan of Dundee University, said he felt the process - including the workshop in which he and other experts produced their recommendations - had been fair and transparent, particularly given the complicated issues involved.

'My one sentence summary of the workshop would be: We know bycatch can be a serious problem, and have quite a good idea of how to limit it, but things are frighteningly unclear for toxoplasmosis,' he said.

'Personally, I find it hard to believe toxoplasmosis is as serious a problem as the data suggest but I know I have no real basis for that feeling.'

He said the workshop had been constructive and focussed on the science.

'I think it was right that we were asked for recommendations, and the decisions are to be made by the New Zealand government after consulting wider society,' he said.

'Despite the strong and differing views that were clearly expressed there was very little animosity and a general acceptance that everyone was honestly engaging with a difficult problem.'

Professor Dawson, however, said he did not believe the document was an accurate representation of the science.

He pointed to his own decades-long research, which found a significant increase in the survival rate of Hector's dolphins in the Banks Peninsula after a marine mammal sanctuary was declared, which halted commercial fishing.

If toxoplasmosis was the major cause of dolphin deaths, eliminating commercial fishing would not have led to such an increase, he said.

'If you asked me if toxoplasmosis in the South Island could be nearly six times as bad as commercial set-netting, I'd say not on your life,' he said.

'I mean, come on, it's just a ridiculous suggestion – If that were true, there would be dead dolphins all over our beaches.'