Energy and the Earth

Energy and the Earth

Direct action to an emissions freefall

For more than a decade, climate science has been flagging the need to greatly reduce greenhouse gas (GHG) emissions. To avoid the possibility of serious and irreversible global warming, climate science is telling us we need to reduce emissions by some 80% by 2050.

Said like that - “80% by 2050” - the challenge seems highly improbable. It may well prove to be. Despite the best wisdom of several decades of science, global GHG-emissions continue to increase at or above trend (about 2%). The huge investment in the infrastructure that serves our current energy needs and contributes to the bulk of those emissions has invested a substantial interest in the status quo with some very powerful players, including governments. And then there is the issue of adding another 1-2 billion people to a global population that is already short on adequate energy services.

But another way of looking at the problem, year-by-year, sector-by-sector, can help to make the challenge seem a little less daunting. A focus on a yearly reduction target of 4% seems a little more manageable than the equivalent “80% by 2050”. For one thing, it allows us to assess what is working now, and what is not, and why.

With that 4% in mind it might come as something of a surprise that one sector that is doing its bit is the Australian electricity sector. And the reason is largely about direct action - if you’ll excuse the pun.

To illustrate why we can refer to the figures below, that show both the trend in demand for electricity served during the six-month winter period on the National Electricity Market (NEM) and the trend in GHG-emissions from the sector. (It doesn’t matter particularly that its the winter period, as the same is more-or-less true for any similar period). The data are from AEMO - the market operator.

The key point is that emissions from the Australian electricity generation have been in virtual freefall for five consecutive years, tracking downwards on the back of an unprecedented reduction in electricity demand.

NEM-wide electricity use (excluding Tasmania) for the six month period April through September, expressed as average electricity demand in gigawatts. Sourced from AEMO half hourly demand data.
NEM-wide GHG emissions for the six month period April through September, expressed as millions of tonnes of CO2-equivalent. Sourced from AEMO daily emissions data.

Since 2008, winter electricity demand has decreased 10% in real terms - at an annual rate of 2%. Emissions have fallen even faster - at an annual rate of 3.2%. Note that winter emissions are already below year 2000 levels. The sector is already on its way of smashing Australia’s bipartisan political target of 5% below year 2000 levels by 2020, by a whopping 6 years.

While there are a number of factors that are contributing to this dramatic emission reduction, the key underlying reason is the sensitivity of the most emissions-intensive (read coal-fired) generation to the ongoing imbalance in supply and demand.

Recent history shows that reductions in electricity demand give an emission reduction multiplier of about 160%, because the over-supply is being disproportionately absorbed by the most emission-intensive part of the electricity supply chain. In driving down demand we are also driving down emission intensity.

In effect, this nexus between demand reduction and emission intensity reduction is our free bonus. The key message is that if you want to get emissions down, then take a direct action by reducing electricity consumption and then watch the market do its thing.

Going the extra percent

Not withstanding the help provided by a high Australian dollar, low aluminium prices and the sluggish global economy, getting our electricity-sector emissions substantially into the negative has been relatively painless.

In the near-term it shouldn’t be very difficult to maintain the current rate of emission reduction, and increase it by an extra percentage point or so. There is still a significant energy efficiency dividend to play out. While we have already reduced domestic demand significantly, well targeted efficiency measures could reduce it another 20% or so without significantly impacting the amenity provided by our energy use. With domestic demand contributing around a third of total demand, that would drive a further 10% reduction in electricity-sector emissions.

Distributed generation is clearly playing its role in reducing demand for centralised electricity services. With small-scale solar rapidly approaching grid parity, there is a growing incentive for deployments that will further reduce demand for grid-based services. That should provide another 10% or so of electricity-sector emission reduction in the near- to medium-term.

Further emission reduction will require replacement of the remaining coal-fired assets with much lower emissions technologies. While there is no shortage of alternatives, such as nuclear, renewables and gas, building any new generation at scale is fraught. With falling demand, and the market screaming “get rid of supply”, there will be little appetite for any significant new supply-side investment in the near-term.

One near-term option is to make more use of existing gas generation capacity. Historically, natural gas has been used to provide peaking capacity. So there is significant potential emissions benefit in making more use of the existing gas capacity at the expense of coal.

In effect, this scenario has been played out in the US in the last few years, resulting in quite a staggering nominal GHG-emission reduction (greater than 4% in 2012) on the back of a surge in unconventional gas production. Of course, an important caveat is that the new gas production is not leading to any significant increase in fugitive emissions. (In the US, the coal-to-gas transition has benefited by a decoupling of gas and oil prices. Worryingly, the most recent reports out of the EIA indicate that with US gas prices now rising again, there is already a switch back to coal, with 2013 electricity sector emissions already rising).

Beyond the decadal-timescale, driving emission reduction in the electricity sector will need an aggressive build out of zero-emission generation capacity using nuclear and/or renewables. Creating new demand through electrification of transport and a move away from domestic gas use should help provide renewed incentive for new build (as well as help drive emission reduction in the transport sector).

While it has been just five years, in just one sector, in just one country, the changes in the Australian electricity sector emissions since 2008 are quite phenomenal. They show how collective direct action can lead to the emissions freefall warranted by the best wisdom of science.