Earl Bardsley says the prospect of introducing hydro fill and spill to avoid LNG imports is worthy of further investigation

By Earl Bardsley*

In a recent presentation, Energy Minister Simeon Brown argued that importing liquified natural gas (LNG) was the best option to quickly provide some certainty of avoiding dry winter power price impacts like in 2024.

In contrast, consultation advice delivered to the Ministry for Business, Innovation and Employment (MBIE) has now been revealed as stating that there is not a strong case for LNG imports.

This raises the question of whether new renewable generation combined with existing thermal generation might be able to give the desired level of dry winter security instead.

There is one potential alternative winter security approach along these lines that could be operational in time for the 2027 winter.  The term “dry winter” is used here rather than dry year. This is because it is typically in winter when low hydro inflows are most likely for the main southern hydro storage lakes.

By way of background, a distinction is made between the causes of low winter water levels in natural lakes and controlled hydro lakes. Low water levels in natural lakes are a response to dry weather. Their water levels fall in sync with declining river inflows.

In contrast, dangerously low winter levels in hydro lakes arise when too much water has been released over the months prior.

As illustration of this, we can consider a hypothetical situation where the 2024 low hydro inflows are somehow already known in the spring of 2023.

With this advance knowledge, water would be held back in the storage lakes early on. Electricity spot prices would rise, but not to massively high levels. Additional fossil fuel power generation would make up for the reduced hydro output. The result would be having the hydro lakes much higher than the actual low levels of the 2024 winter.

In summary, dangerously low hydro lake levels and associated high power prices are not caused by dry periods. They are caused by inability to forecast those dry periods.

Of course, we will not have long-range accurate forecasting anytime soon, if ever.

Nonetheless, the way now opens for the possibility of an alternative winter energy security approach. Instead of importing LNG there is an obligation established that from 2027 the hydro lakes should always be high going into winter. This is a special case of “generator-focused winter energy obligation”, in the terminology of the current MBIE discussion document.

For example, a ruling could be set up that all the main hydro lakes must be at their maximum levels on the first day of June in every year.

Suppose this system was operating presently. Starting from the coming spring of 2026, there would need to be sufficient water held back in the hydro lakes so that they are all in the required full state on June 1, 2027. There is no waiting until some kind of signal appears that a dry 2027 winter is coming.

The advantage with this approach is that the associated burning coal and gas for power generation would be distributed over many months prior. That is, in the case of a repeat of 2024, there would not be the same risk that an intense few months would result in having insufficient coal and gas generating plant to offset the low hydro power output.

However, there is also an obvious disadvantage that comes from the reality that dry winters are less frequent than normal winters. A hydro fill that was aided by fossil fuels would therefore most often translate into subsequent spill. That is, in hindsight there would be needlessly high power prices and carbon emissions in the first half of the year, followed by wasted hydro energy.

On the positive side, the fossil fuel component of the hydro fills may only be required for about two years. The revealed advice to MBIE against the necessity of LNG imports included mention of additional new security from increased renewable generating capacity built since 2024.

This renewable build is set to continue into the future.

With renewables dominating, the hydro fill will then be achieved mostly by wind and solar substituting for hydro power. This allows the lakes to rise to the required maximum at the start of winter. Rooftop solar at scale could be part of this because the resulting reduced domestic demand is analogous to a permanent power conservation campaign.

The obligatory high winter water levels will still be required, however, to ensure energy security. This is because the water level obligation avoids any temptation to divert the needed additional renewable power away from hydro lake conservation to large energy consumers like data centres.

There remain two downsides, however, to a fill and spill model even when powered almost entirely by renewables.

While LNG importing is avoided despite the reduced current availability of local gas, the inevitable spill component nonetheless represents lost generation opportunity.

Secondly, the abundance of renewable intermittent generation will eventually dampen further wind and solar development. This is simply because the wholesale power prices would be so low that they would deter new renewable constructions.

However, an over-abundance of renewable generation is a good commercial environment for setting up a pumped storage scheme with a large energy storage capacity. The most-investigated option here is the Lake Onslow scheme. There may be other sites as well.

Unfortunately, the required large energy storage capacity means that distributing energy storage over multiple small pumped storage schemes is not realistic economically.

Any large pumped storage scheme remains some years into the future and is still to be confirmed. However, if constructed, its influence would be felt well before completion. This is because such schemes would be massive consumers of energy at times. For example, the Onslow scheme pumping at capacity would require more power than the Tiwai smelter.

With a confirmed large energy purchaser on the time horizon, further wind and solar development would be set up perhaps years in advance. If Onslow proceeds, there may be almost free power beforehand.

If a scheme like Onslow was operating, there would no longer be a need to maintain the obligatory high winter water levels in the hydro lakes. Instead, the electricity market could return to its present state with the large scheme now taking the role of providing winter security.

At this point we have almost set up an energy strategy because Lake Onslow (or equivalent) would operate similarly to the pile of coal at Huntly. That is, Huntly coal earns income for Genesis outside of dry winters simply by virtue of being available if needed via the Huntly firming option.

Likewise, a large water storage volume in a pumped storage scheme earns income outside of dry winters just by being available, analogous to an insurance premium. There would also be an additional commercial advantage of helping to meet power peak demands with fast-response generation for brief periods.

Will there be an energy journey that gets started in 2026 by introducing fill and spill to avoid LNG imports? It’s very tentative, but the prospect would at least seem worthy of some further investigation.

*Earl Bardsley is Associate Professor at the University of Waikato School of Science. He is the original proposer of the idea of pumped storage at Lake Onslow in Central Otago, as an alternative to burning coal and gas in dry years when hydro lakes are low. Bardsley spoke about the Lake Onslow idea in an episode of the Of Interest podcast in 2022.