New Zealand's environment watchdog wants greater scrutiny of a computer model used to calculate farm run-off into waterways – and increasingly being used to regulate pollution.

Overseer is a widely-used model that takes nutrients on a farm, models how they are used by plants and animals, and then estimates how they leave and in what form.

By modelling nutrient flows, Overseer could provide a farmer with estimates of what nutrients were lacking and could be supplemented through fertiliser to maintain growth and production.

But in more recent times, regional councils have been also drawing on the model to help set limits around nutrients such as nitrogen.


This threw up thorny questions around who owned and funded Overseer, and whether its data was reliable enough to support regulations to ease farming's pressure on our freshwater estate.

In an-depth report on the issue today, Parliamentary Commissioner for the Environment Simon Upton said that, if the Government wanted Overseer used as a regulatory tool, it had to be a transparent one.

"To help build confidence there needs to be more transparency around how the model operates," he said.

"Any model operates with a measure of uncertainty. That's normal. The question is whether the level of uncertainty is an acceptable one."

Upton pointed to the limited amount of research trials that had been conducted to calibrate the model to regional conditions.

As an empirical model, Overseer tried to generate results that match real-world conditions, including soil, climate and management systems.

But some parts of New Zealand had not been covered by trials, meaning uncertainty in some regions may be greater than in others.

A comprehensive range of regionally specific calibration trials would support wider use of the model, Upton said.

He called for a comprehensive evaluation to ensure the model was independently peer-reviewed, along with more openness about how the model worked, and aligning its funding and ownership with the transparency needed for it to be used in regulation.

Upton also noted that greater certainty meant farmers and councils could be more confident about its results.

"It will take time to improve Overseer and provide transparency around how it operates," he said.

"In the meantime, regional councils can continue to use it but they need to be aware of its limitations."

Federated Farmers backed Upton's calls.

"Even for types of farming systems and geography within models that have been calibrated, actual results for nitrogen losses can be 25 to 30 percent off the mark - in either direction," the lobby group's environment spokesperson Chris Allen said.

Outside these calibration ranges, results could be up to 50 per cent inaccurate.

"The significant inherent inaccuracies in the Overseer model means that is very unfair when the model is used to regulate farming activity central to farmers' livelihoods, and even more importantly to mount prosecutions," he said.

"How would an ordinary New Zealander feel if he or she was prosecuted for speeding in their car based on the model they are driving and when the radar or speed camera had an accuracy range of plus or minus 50 per cent?"

Federated Farmers agreed Overseer needed more Government investment, third-party peer review, and greater transparency.

The group particularly argued Overseer should not be used to assign absolute limits to discharges on farm activities, that can then be traded.

"It is more useful for guidance about relative change or 'direction of travel' in terms of reducing nutrient losses, or a comparison of changes to farming systems rather than assigning absolute numbers," Allen said.

He added other shortcomings also needed to be recognised.

Overseer modelled nutrients lost from the farming system – but not what happened to the nutrients after that, nor what happened to the surrounding environment.

Freshwater scientists also saw a case for greater oversight.

Waikato University's Professor Troy Baisden pointed out the report's conclusion that Overseer had achieved a safe monopoly on regulatory use.

"Essentially, Overseer is the best model we have, because it is the only model we have," he said.

"As a result, there are reasons to recommend the Government address issues of openness, enabling the science community to do more to check and improve Overseer."

Professor Richard McDowell, who was one of the reviewers of the report and also a contributor to the model's development, believed the best use of Overseer would be connecting it to farm environment plans.

"Farm environment plans should fall under the same national guidance recommended by the commissioner for Overseer, to ensure plans are effective," he said.

"Currently there is potential for significant variation in quality between 16 regional councils."

Beyond this, McDowell said, Overseer needed to evolve so it could map and target critical areas of nutrient loss within a farm, while measuring the impacts of day-to-day decisions.

"The next step would see Overseer and other models used to help farmers and growers identify the most suitable land uses for areas most prone to nutrient loss."

Environment Minister David Parker described the report as a "critically important" contribution to the Government's Essential Freshwater programme.

Parker agreed that if it was to be used more widely as a regulatory tool, then more investment in it would be needed, so that it better accounted for different land types and farm practices.

The Government had already put $5m into Overseer in the 2018 Budget.

Motu Economic and Public Policy Research's Dr Suzi Kerr said creating a strong version of Overseer to estimate greenhouse gases was significantly easier task - although one still in progress.