Integrated Policies to Reduce Australia’s Electricity Sector Greenhouse Gas Emissions to Net Zero by 2050

Abstract

Recent events within the Australian National Electricity Market have demonstrated that the system of an energy-only market (a market that only compensates power that has been produced) is no longer fit for purpose. The rate of change in installed capacity and generation requires better planning to ensure reliability is maintained at the lowest total system cost during the transition to net zero. Australian National Electricity Market participants will need sufficient incentives and confidence to invest in new capacity. This paper assesses a “no constraints” scenario and recommends a range of policy and market mechanisms that could be utilized to achieve a net zero National Electricity Market in Australia by 2050. This paper adopts the perspective of total system cost, which allows multiple factors relating to decision-making to be incorporated. In the absence of a carbon price, this paper seeks to put forward technology-based policy and market mechanisms to incentivise the changes required. The “Modelling Energy and Grid Services” model used in this study has shown that this “no constraints” future grid will need to contain approximately 100 GW of variable renewable energy, almost 20 GW of firm, low-emissions generation, such as carbon capture, utilisation and storage, bioenergy with carbon capture and storage, hydroelectric power, or nuclear power. It will also require more than 10 GW of storage, including pumped hydro energy storage and other energy storage technologies, and over 30 GW of firm, dispatchable peaking plants, including thermal power generation.

1. Introduction

To achieve the targets set by Australia’s commitment under the Paris Agreement of net zero by 2050 [1], the Australian electricity sector will need to undergo a profound transition [2]. While the Federal and State levels of Government are aligned on the target, the approach is fragmented, with each jurisdiction defining their own pathway to net zero [3,4,5,6,7]. This fragmentation can lead to sub-optimum results and more costly solutions for decarbonising the grid [8]. The mechanisms utilised to enact change will need to be carefully considered to ensure a smooth transition. Decarbonising the electricity sector at minimum total system cost has been the focus of recent studies [8,9,10,11,12,13] and has been recognised as an improvement over levelised cost of electricity (LCOE)-based methods [14]. A total system cost approach includes the value of grid resilience, an integral feature of a diverse low-emissions grid [15]. However, LCOE continues to underpin policy and decision-making [16,17], requiring a structural shift.

The current Australian National Electricity Market (NEM) is an energy-only market [18]; however, this is evolving. Changes such as the proposed Snowy 2.0 and Battery of the Nation [19] both increase the amount of pumped hydro energy storage (PHES). Along with the development of large, widespread variable renewable energy (VRE), there is a more diverse portfolio of generation and storage assets than what was available when the current market was designed [20,21,22,23]. Together with emissions reduction targets, these changes will alter the way electricity is valued [11].

Governance of the NEM also continues to evolve with a greater focus on system costs. A recent “Independent Review into Future NEM Security of the National Energy Market: Blueprint for the Future” led by the Australian Chief Scientist suggests the Australian Energy Market Operator (AEMO) may need to adopt a role as a central planner with a whole-of-system focus as a way to transition the grid to a low-emissions future [24]. AEMO’s 2020 Integrated System Plan (ISP) has also stated “the ISP must use long-term total system cost as its primary measure of what is in the interest of consumers” [25]. The Electricity Security Board (ESB) is currently responsible for implementing reform [26]. It was tasked to put forward market system changes to better ensure grid reliability, minimum total system cost, and emissions reductions [27]. To date, a Retailer Reliability Obligation has been established, which is designed to incentivise retailers and some large energy users to contract or invest in dispatchable generation [28]. Consultation is underway for a “Post-2025 Market Design for the National Electricity Market” [27,29]. The use of a capacity market has been recommended as part of this process [30]. To date, the ESB’s advice includes recommendations across four reform pathways that aim to:

• strengthen signals for investment in the right mix of capacity to keep the system reliable, affordable, and secure;
• deliver essential system services to maintain grid stability;
• improve transmission and access arrangements to ensure timely transmission investment, incentivise better use of the network to lower costs for consumers, and reduce investment uncertainty;
• better enable participation of flexible demand side resources and the integration of distributed energy resources (DER) [31].

This paper contributes to knowledge by assessing a “no constraints” scenario through a policy lens, recommending a multitude of policy and market mechanisms that may be required to achieve a net zero NEM by 2050. Considering the NEM operates as an energy-only market, this paper recommends changes are made at a policy level to ensure adequate grid services available to the grid to ensure a sustainable transition to net zero. This paper adopts the perspective of total system cost as new metric, which allows multiple factors relating to decision making to be incorporated. In the absence of a carbon price, this paper seeks to put forward technology-based policy and market mechanisms to incentivise the changes required. While this paper and the model that underpin it do not incorporate the use of artificial intelligence or machine learning for load forecasting and peak-hour energy cost minimisation, the use of an Artificial Neural Network model could be used for future studies to predict the hour-ahead load using meteorological data to investigate system optimisation [32].

2. Methodology

Until recently, the Australian government’s approach to energy policy has been variable [33]. Australia was an early mover by introducing an emissions trading scheme in 2012; however, this was subsequently abolished with a change in government [33]. “A Case Study of Australia’s Emissions Reduction Policies—An Electricity Planner’s Perspective” is a summary of the market and policy mechanisms implemented in the NEM with the aim to reduce emissions [33]. This summary forms the policy foundation for this study.

Total systems cost has been used as the metric by which to measure success in this study. This new metric allows for multiple characteristics of a power generation technology to be assessed as it is added to an existing grid, including energy and grid services [11]. “Total Systems Cost: A Better Metric for Valuing Electricity in Supply Network Planning and Decision-Making” provides a comparison of electricity cost metrics, favouring total system cost as the optimal metric to use to achieve emissions reduction targets [11]. This metric forms the basis of “Decarbonised Electricity”, a book that compiles the developments of the Modelling Energy and Grid Services (MEGS) model [14]. MEGS was the first to demonstrate the trajectory of total system costs in response to power generation technology additions to the NEM [14]. The MEGS model underpins this study to enable a holistic and systematic view from a policy planning and decision-making perspective.

MEGS is an electricity system scenario tool devised to explore technology options for a decarbonised grid [34]. It is a regional electricity system model designed for the NEM, incorporating decarbonisation targets and inertia as a proxy for grid firming capabilities, while minimising the lowest total system cost [34]. The MEGS methodology was published in the journal Energy Strategy Reviews in 2021.

The objective of MEGS is to model both current electricity systems and future decarbonised electricity systems which include a wide range of storage and generation technologies [34]. The MEGS model does this by ensuring:

○

Balance of electricity demand and supply;

○

Sufficient supply of reserve and response services;

○

Sufficient inertia available;

○

Sufficient reliable capacity to meet peak demand.

For each point in time modelled, the solver determines generation and reserve provision from the plant whilst minimising system short run costs which are given by greenhouse gas emissions, energy storage, start-up, fuel, and non-fuel variable costs [8]. The outputs from this model are designed to be useful to policy makers and systems planners [34].

To model future scenarios, MEGS can be configured to use a Monte Carlo analysis to model up to five key uncertainty parameters—referred to as S-MEGS [8]. For this study, 3300 scenarios are run with the five uncertainties shown in in

Table 1. S-MEGS Uncertainty Parameters.

Weather	The weather can be chosen from ten historic years. This affects wind, solar, and hydro generation, and electricity demand in a consistent manner.
Fuel prices	Fuel types can be allocated a level of volatility. All fuels in that group will experience the same random price change chosen annually from a lognormal distribution.
Capital expenditure	New build plants can be allocated a level of volatility. The capex of all plants in that group will experience the same random change chosen annually from a lognormal distribution.
New build projects	New plants can be subject to a 50:50 chance of being built.
Low emissions projects	Lower emissions generation options, such as renewables, gas, and CCS, can be subject to a higher or lower build rate.

.

The output used in this study using the S-MEGS model can be seen in Figure 1, with 3300 scenarios modelled to achieve an electricity portfolio at various decarbonisation rates. Each of the markers on Figure 1 represent a power generation portfolio, or scenario, each achieving the rate of decarbonisation (plotted on the x axis), at various total system costs (plotted on the y axis). The frontier defines the portfolios with the lowest total system cost across the rate of decarbonisation. The “no constraints” scenario in this study (described in detail in the following sections) refers to the scenario with the lowest total system cost while achieving 100% decarbonisation. This scenario is used in this paper as an example technological solution to a net zero electricity sector. The power generation installed capacity and generation for this scenario is detailed further in Figure 2 and Figure 3 below.

This paper discusses policy options that may be required to implement the “no constraints” scenario to reach a net zero emissions target for the NEM. This study uses the definition of “policy” as a set of ideas or plans that is used as a basis for making decisions [35], whereas a “market” is defined as any place where two or more parties can engage in an economic transaction [36].

There are several assumptions included in this analysis to support a “no constraints” scenario. To enable the build out rates detailed below, there are no constraints on the ability to construct, commission, and fuel the required capacity of all technologies. Importantly, there is no constraint on the volume and location of CO₂ storage for sequestration. The generation and storage technologies included are current technologies that are costed in the CSIRO GenCost model [37]. No additional technologies from other sources are included. The forecast demand for 2050 is defined by the AEMO [38]. No socio-political constraints have been incorporated into the “no constraints” scenario.

3. The “No Constraints” Scenario

To move from the current average carbon intensity grid (724 tCO₂/GWh) [39] to a net zero emissions system by 2050, the Australian electricity grid will require significant transformation. The “no constraints” scenario is shown below as installed capacity in gigawatts (GW) (Figure 2) and generation in terawatt hours (TWh) (Figure 3). The charts below are generated using the MEGS model. Figure 3 shows how the installed capacity is generating annually in TWh, taking into consideration time of use and technology capacity factor across the year. One important distinction to note in Figure 3 is the change in role for black-coal- and gas-based thermal power generation technologies. Black coal currently provides baseload generation for the NEM [25]; however, within the 2050 net zero grid scenario, these generators are providing the grid firming service as peaking plants with 12 GW of unabated black coal operating with an average capacity factor of approximately 17%. Another distinction between the current grid and “no constraints” scenario, in Figure 2, while the overall installed capacity increases significantly, the amount of dispatchable capacity increases marginally. This key distinction reflects the changing way technologies will be dispatched, as Figure 3 shows that all low-emissions forms of power generation will operate at their capacity.

Brown coal represents 5 GW of installed capacity in the NEM, all of which is in Victoria [40]. AEMO’s 2022 ISP forecasts Victorian brown coal will retire by 2032 [41], in line with the “no constraints” scenario modelled by MEGS that sees no operational brown coal assets.

The “no constraints” scenario is not the only scenario able to meet net zero using the MEGS model, but simply a hypothetical scenario selected to explore the integrated policies required to achieve a net zero outcome, reliably, at the lowest total system cost.

While the NEM operates as a market, changes to the system occur as regulatory change through market rules. Regulation is defined as “any rule endorsed by government where there is an expectation of compliance” [42]. Changes need to be balanced against Australia’s broad strategic objectives, regarding energy security, national competitiveness, and international commitments.

4. The Pathway Forward Using the “No Constraints” Scenario

The discussion below details how market and policy mechanisms can be used to stimulate the development required to meet the capacity and generation output of the net zero “no constraints” scenario by 2050.

4.1. Variable Renewable Energy: Wind and Solar PV

The “no constraints” scenario requires 42 GW of wind power. Between the current grid and 2050, approximately 1.5 GW of new wind power will need to be built each year, year on year (and by around 2045, the earliest builds will require replacement). For comparison, this rate of build is approximately equivalent to three times the largest wind farm project, Coopers Gap Wind Farm, which upon completion will be 453 MW and composed of 123 turbines [43]. This annual target installed capacity also represents almost twice the capacity commissioned in 2019 [44].

Solar PV capacity also increases significantly each year to 2050. The “no constraints” scenario requires 55 GW on the NEM. Both small- and large-scale solar PV will need to achieve a capacity target of 1.9 GW of new generation each year until 2050 to meet the “no constraints” scenario target. For reference, in 2019, the installed capacity was 2.2 GW [45].

VRE represents a considerable increase in installed capacity in the “no constraints” scenario. The design life for both wind turbines and solar PV has been assumed to be 25 years [46]. Given increasing industry footprint, it is expected that local and environmental approvals will become more challenging over time [47,48]. It is likely that large-scale projects will need to access to project fast tracking to ensure prompt commissioning.

The current Renewable Energy Targets (RETs) aim to support an increased deployment of renewable power generation. In Australia, the RETs are split into two forms: the Large-scale RET (LRET) and the Small-scale RET (SRET). Under the LRET, utility-scale facilities can create large-scale generation certificates (LGCs). These can then be sold to entities with liabilities under the LRET, predominantly electricity retailers, to meet their RET obligation set by the renewable power percentage [49]. The renewable power percentage is the mechanism used to meet the RET each year. From 2021 to 2030, the annual target is 33,000 GWh [50]. Entities that do not surrender sufficient LGCs to meet their obligations are required to pay a shortfall charge.

Additional to the federal RET, Queensland, Victoria, New South Wales (NSW), South Australia, and Australian Capital Territory (ACT) have adopted policies to increase sourcing their electricity from variable renewable energy and/or adopt a net zero emissions target [8].

Project development may cause delays to the construction and commissioning of new VRE projects. Project fast tracking can also remove some of the roadblocks in what will already be a lengthy deployment time. Project fast tracking can occur through the Federal Infrastructure Priority List, a list of nationally significant investments [51]. The Infrastructure Priority List is made up of two groups:

• Projects: infrastructure solutions to a defined issue which has a business case.
• Initiatives: potential infrastructure solutions to a defined issue that is at the early stage of study and the business case is not yet completed [51].

To ensure the build-out rate increases to meet the “no constraints” scenario, recommendations include the following:

• Extend the Renewable Energy Target to meet the 2050 generation target of approximately 91 TWh annually of wind and approximately 91 TWh annually of solar PV.
• Invest in transmission infrastructure to enable build out rates of VRE.
• Develop Renewable Energy Zones, proposed by State Governments as well as the ESB, to optimise economies of scale.
• Incentivise smart meters, behind the meter solar PV and batteries, and community storage to enable the aggregation of assets to participate as virtual power plants (VPPs) under the wholesale demand response mechanism.
• Develop a pathway to initiate an e-waste and recycling industry that will match forecast volumes of waste.

To ensure the AEMO has visibility over roof top installations and other behind the meter assets, VPPs [52] are ideal to harness these assets efficiently. A VPP is an aggregation of assets (such as decentralised power generation, storage and demand-side controllable loads) coordinated to deliver services for system management within the electricity market [52]. This may require funding to incentivise the consumer purchase of smart meters and behind the meter storage to increase the capabilities of these assets [53]. These aggregated assets, both large- and small-scale, can be optimised within the NEM through the use of the wholesale demand response mechanism, a new mechanism that commenced in 2021 [54]. The first participant of this mechanism was Enel X who utilised their VPP to create an additional revenue stream while optimising the aggregated assets as a demand response service for the grid [55].

4.2. Energy Storage

The current 1 GW of energy storage on the NEM is comprised of predominantly pumped hydro energy storage (PHES) and the Hornsdale Power Reserve battery (150 MW/193 MWh) [56]. The “no constraints” scenario will require 12 GW to be connected to the grid. The Snowy 2.0 project will add a further 2 GW/350 GWh of PHES [57]. If a ‘Snowy 3.0′ project could be delivered, it could theoretically deliver another 2 GW of PHES [12].

Hydro Tasmania have stated that they could cost-effectively provide an additional 1.5 GW of PHES through the Battery of the Nation project [58]. If this project were to be expanded to deliver the same amount of PHES again, Tasmania could theoretically deliver a total of 3 GW of PHES [12]. The total PHES from the existing Snowy, Snowy 2.0, Battery of the Nation, and the expansions, this only gives the NEM 8 GW of PHES. This means a further 5 GW of other storage would need to be added to meet the 12 GW net zero scenario.

To prioritise energy storage investment, a Contract-for-Difference (CfD) can be used. A CfD is a financial mechanism that can be used to promote investment in a nascent technology that would otherwise be at a competitive disadvantage to mature technologies in the market [59]. The CfD mechanism can be used to correct failures in an energy market where issues exist regarding absent policy and absent or incomplete markets [60]. A strike price is defined per project and represents the investment required in a particular low-emissions technology as dollars per megawatt hour [61]. This strike price allows the generator to sell the electricity into the market as per usual, while being protected from the fluctuations of a spot market [62].

PHES represents an important type of energy storage: long-term storage, as opposed to the short-term storage provided by batteries. To ensure the build-out rate increases to meet the “no constraints” scenario, recommendations include the following:

• Introduce a firm capacity market to give large-scale storage facilities certainty in revenue.
• Introduce a CfD scheme for large-scale storage projects to secure revenue certainty.

4.3. Unabted Gas

Combined cycle gas turbines (CCGTs) power plants currently represent 3 GW of capacity within the NEM (This includes a minor contribution from combined heat and power plant capacity). The “no constraints” scenario requires 5 GW of CCGT capacity. This increase of 2 GW is a relatively minor increase and would require few incentives to achieve this build out. A firm capacity market, such as the Western Australian Reserve Capacity Market, may incentivise the investment in gas plants. A Reserve Capacity Market operates in the Wholesale Electricity Market (WEM) to ensure there is dispatchable generation to meet demand at all times, and runs separate to the WEM [63]. The Reserve Capacity Market ensures demand is met by:

• Setting a Reserve Capacity Requirement two years ahead, published in the WEM Electricity Statement of Opportunities [63];
• Allocating Certified Reserve Capacity and Capacity Credits based on a facility’s ability to dispatch [63];
• Administering a Reserve Capacity auction if the allocation of Capacity Credits through the trade declaration process is insufficient [64];
• Testing capacity providers twice per year (once in summer and once in winter) for facilities, and once per year for demand side participants (DSP) [65];
• Assigning an Individual Reserve Capacity Requirement to each Market Customer, based on contributions to the system peak, to allocate the cost of Capacity Credits fairly among Market Customers [63].

Additionally, installing hydrogen-ready gas capacity, such as the proposed Kurri Kurri plant, may further incentivise investment [66]. However, the issue of gas price and supply will continue to be of significance going forwards [67]. To ensure the build-out rate increases to meet the “no constraints” scenario, recommendations include the following:

• A “pipeline” of gas supply to be developed in a futures market to ensure supply. This can be incorporated into the firm capacity market outlook, securing gas supply contracts the year prior, and the National Gas Infrastructure Plan [68];
• Introduce a firm capacity market to incentivise peaking facilities.

Introducing an emissions standard for firm plants, if set at the emissions profile of gas, will incentivise the build out of high-efficiency CCGTs. However, this may have unintended consequences of reducing investment in open cycle gas turbines (OCGTs) and instead using CCGTs as peaking open cycles through bypassing the heat recovery steam generator [69]. Another unintended consequence of removing a technology option, such as OCGTs, is that the NEM will not track along the lowest total system cost pathway. OCGTs provide an important service to the grid as peaking plants. The current fleet of OCGTs in the NEM is 9 GW at an average capacity factor of 10% [46]. A total of 15 GW of capacity is required in the “no constraints” scenario at approximately 12% capacity factor.

The ancillary services market, currently in operation, could allow OCGTs to operate in the NEM as a provider of grid services. The AEMO uses ancillary services to operate the power system reliably. These services support fundamental technical characteristics of the grid, including standards for system restart processes, frequency, network loading, and voltage [70]. There are eight separate markets for the delivery of Frequency Control Ancillary Services (FCAS), Network Support Control Ancillary Services (NSCAS), and System Restart Ancillary Services (SRAS) under agreements with service providers [70]. Ancillary services providers receive payments for availability and delivery of these services [70]. Payments are dependent on the volume and type of service required.

To ensure the build-out rate increases to meet the “no constraints” scenario, recommendations include the following:

• A “pipeline” of gas supply to be developed in a futures market to ensure supply;
• Introduce a firm capacity market to incentivise peaking facilities;
• Strengthening of the ancillary services market to allow this mechanism to incentivise investment in facilities that provide grid service.

4.4. Unabated Black Coal

The current fleet of unabated black coal fired power stations has an installed capacity of 18 GW across the NEM. In the “no constraints” scenario, this will be reduced to 12 GW by 2050. However, the way in which coal fired power stations contribute to the NEM will alter significantly. These plants will go from operating as firm power plants, with 60–90% average capacity factors across the NEM [71], to peaking plants, running at approximately 15% capacity factor on average. The unabated black coal capacity on the system appears counterintuitive; however, as shown in Figure 4, the emissions from unabated black coal are offset by the negative emissions of BECCS. The use of unabated black coal in the NEM is contingent on the availability of both CCUS and BECCS, and if these technologies are not available for deployment, firm, low-emissions power generation technologies would need to be provided by nuclear.

The announced coal retirements as reported to the AEMO in the 2022 ISP, shown in Figure 5 [72], reduce coal capacity to approximately 1.7 GW by 2050. However, the ISP central scenario, Step Change, forecasts the last coal plant to exit the NEM in 2042 [72]. The rate of announced retirements means either a portion of existing plants will need to have the asset life extended, or new plants will be required to meet the 12 GW capacity target in the net zero scenario. Alternatively, a portion of the coal fleet could be replaced with other thermal generation, such as gas-fired power generation; however, this will likely increase the cost of the system.

To ensure the build-out rate increases to meet the “no constraints” scenario, recommendations include the following:

• Plant modernisation to extend the life of existing plants and allow them to reliably operate in a flexible manner. A modernisation fund, such as the European Union’s Emissions Trading Scheme (ETS) Modernisation Fund [73] and/or the CEFC Grid Reliability Fund [74], would enable the transformation of coal units into peaking units. This could also be achieved through innovation grants for plant modernisation to meet flexibility requirements.
• A firm capacity market to allow for plants to operate in both the energy and capacity market, which can provide a financial incentive to remain on the system.

However, current market trends such as increased wholesale price volatility and increased intra-day operational demand place further pressure on coal-fired power generators to operate more flexibly at a high cost [75]. This has resulted in discouraging market conditions for thermal operators, with coal plants such as Eraring and Yallourn bringing forward their scheduled retirements [75]. The MEGS model does not take into account these market trends and associated operational costs.

4.5. Carbon Capture, Utilisation and Storage, and Bioenergy and Carbon Capture and Storage

Carbon capture on fossil fuel and biomass energy sources are two technologies that require the initial focus from policy makers as there is a long lead time for implementation due to the storage component of these technologies [76]. The “no constraints” scenario sees 7 GW of CCUS capacity installed, and 4 GW of BECCS capacity installed.

If CCUS technologies are to be deployed, pre-work needs to commence approximately 10 years prior to the first CCUS deployment at scale. Additionally, to support the deployment of BECCS, feedstock reserves will need to be identified and expanded to ensure security of supply, as there is still a level of uncertainty around the size of the industry due to its dispersed nature [77]. To gain a firm understanding of the Australian CO₂ storage portfolio, each potential storage site will need seismic assessment and drilling to understand the flow rate and volume that can be harnessed over time [78].

The cost of retrofitting existing plants or building new plants is significant [79]. A tax credit, such as the United States’ 45Q, would enable project owners to claim a tax credit to assist in maintaining profitability [80]. The 45Q tax credit encourages investment in CCUS through a credit that increases over ten years from USD 10 to USD 60 per tonne of CO₂ stored through enhanced oil recovery (EOR) and from USD 20 to USD 85 per tonne geologic storage [81]. This tax credit assists in closing the gap between the cost of CCUS and the amount that entities are willing to pay for CO₂ use and/or storage [80].

The CCUS plant operates most efficiently as firm capacity; however, work is underway to increase the flexibility of CCUS plants [82,83,84]. To ensure the new build or retrofit is utilised most efficiently, dispatch needs to be guaranteed. This can be done via merit order reform in the NEM to ensure the CCUS plant is running at a certain capacity. A firm capacity market or CfD can also provide a revenue stream to ensure the CCUS is built.

Introducing emissions standards for new plants might incentivise the construction of CCUS over unabated thermal options. In Canada, performance standards have been applied to all new and existing coal-fired power stations, which are stringent enough to ensure no new coal-fired power stations can be built without CCUS [85]. The Federal regulation requires power stations to meet 420 tCO₂/GWh by 2029 [86]. Existing facilities will either need to shut down, convert to natural gas or retrofit with CCUS to continue operation past this date. This can be implemented in Australia to ensure that only a low-emission plant is added to the system. However, this may have unintended consequences as gas plants would then become the cheapest thermal capacity and could gain the most from this incentive without allowing the grid to decarbonise beyond the emissions intensity of gas.

By classifying a CO₂ hub as a Priority Development Area, hubs can access State Infrastructure Funds [87] to build out what would be a government-owned asset. CO₂ hubs are central collection and storage distribution systems for CO₂ [88]. Collection and storage hubs provide point-to-point transportation for compressed/supercritical CO₂, thereby reducing the risk and cost of transport infrastructure between the individual point-source emitters and individual points of injection into geological storage.

To ensure the build-out rate increases to meet the “no constraints” scenario, recommendations include the following:

• Ensure bioenergy feedstock reserves and expand security of supply;
• Update the Geoscience Australia CO₂ storage atlas to include storage based on sustainable flow rate and duration;
• Introduce a tax credit mechanism akin to the 45Q;
• Introduce emissions standards for new thermal plant;
• Establish a firm capacity market;
• Introduce CfDs for new build CCUS plants;
• Include CCUS plants in low-emissions merit order dispatch;
• Classify CO₂ hub infrastructure as a Priority Development Area.

4.6. Unabated Brown Coal

In a lowest total system cost, zero emissions environment, unabated brown coal is to be phased out before 2050. In the current grid, there is 5 GW of installed capacity, all of which is in Victoria [40]. Brown coal generation is retired ahead of the “no constraints” scenario in 2050.

4.7. Hydro Power

The levels of hydro power generation stay constant throughout the low-emissions modelling period, which includes the new build out of Snowy 2.0 and Battery of the Nation [34]. These capacity additions are captured in the energy storage section, as the increases to capacity are in the pumped hydro storage component, included in the demand forecast.

4.8. Nuclear Power—Additional Considerations

Nuclear power is currently not an available option to be included in the Australian NEM power generation fleet due to Federal Government legislation [89]. While this technology does not appear in the scenarios, if CCUS or BECCS were not available or were constrained, nuclear power would be required to provide firm, low-emission electricity to the NEM if a net zero is to be met. To hedge or reduce future risk of achieving net zero, changes to the Australian Radiation Protection and Nuclear Safety Act 1998 (the ARPANS Act) and the Environment Protection and Biodiversity Conservation Act 1999 (the EPBC Act) [90] would be required to allow nuclear power to be considered as an option in a technology neutral future grid. If nuclear is required in a low-emissions grid, the planning and eventual development of this technology in Australia would likely need to begin immediately due to a long lead time [91].

Despite the lack of inclusion in the “no constraints” scenario, there are both positive and negative considerations of nuclear power in Australia. The positives include the following:

• High energy density: nuclear power plants can produce a large amount of electricity from a small amount of fuel, making them relatively efficient [92].
• Low emissions: nuclear power does not produce CO₂ emissions and is considered a low-emissions power generation technology [93].
• Reliability: nuclear power plants are designed to run continuously, which makes them a reliable source of electricity [92].
• Waste disposal: Australia already has an operational radioactive waste disposal system, managed by the Australian Radioactive Waste Agency. A single waste disposal facility is currently the subject of a development study [94].
• Technological advancements: nuclear power technologies are continuously improving internationally in both power generation and monitoring [95]. Reactors currently under construction have simpler designs as a means to reduce capital cost, as well as increasing fuel efficiency and safety [95]. Additionally, advancements in modelling power systems including nuclear reactors has advanced, with a new methodology for the estimation of the long-term impacts of spent nuclear fuel that can be included in whole of system modelling [96].

The negatives of nuclear power in Australia include the following:

• High capital costs: the costs associated with constructing nuclear power plants are high, and pose a potential barrier to the adoption of nuclear power [97].
• Public perception: in a poll by the Institute of Public Affairs, 53% of respondents agree that nuclear power should be considered to reduce emissions, while 24% were unsure, and 23% disagree [98]. In 2011, the Lowy Institute polled the same query with only 35% of respondents in favour of nuclear power, 3% unsure, and 62% opposed [99]. While there is an increase in a positive public response, there is still considerable opposition to nuclear power in Australia that will need to be addressed if this technology is to be considered.

5. Discussion

This approach to utilise specific market and non-market mechanisms in parallel is designed to ensure the “no constraints” scenario capacity build out and operating practice. Several distinct conclusions can be drawn out of the above examination. Almost all technologies will have a role to play in an integrated, reliable, lowest cost, net zero grid; however, the role of thermal generation will need to change as we decarbonise. This is evident when comparing the capacity and generation profiles for thermal generation, as these plants transition to providing peaking and grid stabilising services. If a plant cannot maintain the change in load, it is possible they will not remain financially viable.

CCUS will be critical to achieve a net zero grid, if no other low- or zero-emission firm generation is available, such as nuclear. The grid services required by a highly variable grid exceed that of the available PHES. The lead time for new generation commissioning varies between technologies and can be significant. CCUS and PHES will have the longest lead times of the technologies considered. As a result, a whole system pathway is required to maintain optionality when planning, investing in, and commissioning new plants. Long lead times for firming technologies will need to be planned at the whole system level.

The increased installed capacity required to meet the “no constraints” scenario will increase the footprint of the overall grid. Retiring assets, such as VRE, battery storage, and decommissioned thermal plants will all require waste solutions. Recycling opportunities will need to be incentivised to accommodate the amount of waste and e-waste materials that will be generated from the expanding grid.

Further work is required to build on this study. This includes the following steps:

• Further analysis is required to understand the implications of the “no constraints” scenario via a socio-political assessment, which forms the basis of future works;
• Issues of social licence, geopolitics, and volatile markets will likely constrain the outcome of the “no constraints” scenario. Further analysis is required to assess the viability of this scenario from a socio-political perspective, incorporating the ongoing sustainability of coal-fired power generation in Australia;
• Further modelling and analysis would be required to understand the implications of change to the electricity system at a regional level.

6. Conclusions

To achieve a net zero outcome for the NEM by 2050, pathway planning will be critical for the efficiency and effectiveness of the future grid. Investment decisions for the new capacity will need to be made now that are consistent with the lowest total system cost 2050 grid. Furthermore, delaying investment will likely increase the required rate of change which needs to occur later, and this is likely to be constrained via technical, commercial, and/or regulatory conditions. This study demonstrates that to achieve these technology outcomes, a wide range of policies will need to be implemented and coordinated at a whole system level.

A key advantage of this study is the use of the MEGS model as a basis. While MEGS does not model a full techno-economic system, nor detailed electrical network modelling, the granularity of the model is fit for purpose as a decision-making tool [14], as the scenarios modelled are intended to be indicative of a net zero grid. Another key advantage is the use of incorporating inertia as a proxy for grid services [34]. This allows a value to be placed on grid services that are not currently valued in an energy-only market. The “Net Zero Australia” study currently underway The University of Queensland, the University of Melbourne, Princeton University, and management consultancy Nous Group are exploring similar concepts to this study at a more granular level across six industries, including electricity, under several scenarios [100]. Like this study, “Net Zero Australia” will explore scenarios to guide the boundaries of public and policy debate [101]. This study is currently underway. While the “Net Zero Australia” study is comprehensive, it does not recommend specific policy and market mechanisms, like this study, to enable the net zero outcome.

One limitation of this study is that the outputs are not practical in nature. The outcomes of this study, however, are informative to policy and decision-makers to exemplify the size of the policy and market intervention required if the NEM is to achieve emissions reduction targets. These changes must adequately address security of supply, at a lowest total system cost, while achieving a net zero emissions grid. This study is also limited to a theoretical “no constraints” scenario, which does not take into consideration construction time, additional transmission requirements, land access, time for environmental approvals, and other factors that will likely impact the installation of new power generation capacity. Further analysis is required to assess the viability of the “no constraints” scenario from a socio-political perspective, incorporating the ongoing sustainability of coal-fired power generation in Australia.