To the left, to the left; is Liddell irreplaceable through the day?

In Issue 52, considering the recent announcement from the federal government to build a 250MW (previously 1GW) gas-fired plant in New South Wales (NSW), we investigated price volatility trends in the state and the opportunities for storage to fill the impending energy gap when Liddell retires in 2023.

In this chart of the week, we take a further look at the situation in the state and how the market has responded since AGL reaffirmed it will be shutting down the station back in 2017. How has the market evolved since this announcement, and given the state of the market, is gas an ideal replacement for Liddell?

As shown in Figure 1, the NSW market has undergone significant change since 2017 both on the demand and supply side. The growth of rooftop solar in NSW is having a substantial impact on time-of-day demand in the state. In just two years, demand between 10AM – 3PM has dropped by an average of ~547MW; 46% of Liddell’s current average generation through the day. In that time variable renewable energy (VRE) – solar and wind – output during the day has also grown by an average of ~490MW; about 41% of Liddell’s current average daily output. This two-sided market transition is having a significant negative impact on total coal generation in NSW. NSW coal generation (excluding Liddell) has dropped by an average of 1GW between 10AM-3PM since Liddell announced its exit. The decrease in NSW coal is almost exactly equal to the increase in rooftop solar and grid-scale renewables. This evidence suggests that the current energy-only market might actually be working as intended.

While the market seems to have replaced Liddell’s 1.2GW generation profile during the middle of the day, this still leaves potential shortfalls in capacity/reserve during the morning and evening peaks as demand in the state after sunset (and before sunrise) has held steady; with coal only being displaced by new wind generation including during peak periods. Interestingly, on average, gas generated 13MW less over the day (including 32MW less during the evening peak) in FY20 than it did in FY18; suggesting an increase in existing  peak gas reserves since Liddell announced its exit.

This confirms our previous CoTW findings that in FY20, NSW had reserves below 1GW only 2% of the time, but when removing the availability of Liddell, NSW would have had reserves below 1 GW 24% of the time. However looking forward, with ~1.7GW of wind and solar projects currently committed in NSW, and another 6.2GW proposed, these projects alone may be able to provide a significant amount of the energy required when Liddell retires in 2023 (allowing displaced coal, gas and hyro to act as strategic reserves).

Exactly how much NSW can rely on other states for increased support during its peak periods post-Liddell is uncertain with NSW increasing its daytime imports (mainly from Victoria) between FY18 and FY20 by twice its increase in evening peak imports. With Victoria and Queensland facing retirements of their own, it is left to be seen how much support can come from South Australia through EnergyConnect in the future.

NSW is likely to experience a stronger need to shift energy through the day, increased price volatility, and higher network support requirements as more VRE connects; suggesting that dispatchable assets (such as storage) satisfying the trilemma of affordability, reliability and security may be better suited for the state as it continues its journey through the transition.