Episode 2: Low-Hanging Fruit
“It is now cheaper to save the Earth than to ruin it”
Solar and wind have won the energy race. In most countries, they are now the cheapest form of new electricity generation, surpassing all fossil fuels and nuclear energy. Despite only contributing 10% of global electricity generation, they currently account for three-quarters of all new global electricity generation. As coal and gas plants retire following the end of their useful life, solar and wind will start to dominate the energy sector.
This is great news for the transition to net-zero. Electricity generation contributes 40% of global carbon dioxide emissions, the largest sector by far. With electric vehicles dominating the future transportation sector, we will see a doubling or even tripling of electricity use, multiplying the benefits of solar and wind. And, as we electrify everything from transport to heating and cooking, clean electricity will supercharge the decarbonisation of other sectors as well.
But is there more to the story?
Energy generation: a fragile balancing act
In most countries, virtually every building, house and factory is connected to a web of underground copper wires called the grid. Power supply to the grid must exactly equal the power demanded by the entire population at all times.
But power demand is not constant. It increases in the early morning as people wake up, make breakfast, shower, and leave their homes. It then eases for the next few hours before ramping up to a peak around dinner time when people cook, watch TV and get ready for bed.
Current grids, powered largely by coal, nuclear, gas and hydro, can cope with this consumption profile. Coal-fired power stations burn coal to produce steam, which turns a turbine, generating electricity. Interestingly, coal turbines only work when the turbine is already spinning. For that reason, they take a long time to ramp up and down. Therefore, coal-fired power stations end up running at a constant level for the entire day, providing base-load power to the grid. Similarly, nuclear plants are slow to ramp up and down and are uneconomical to have running for anything less than the entire day. Coal and nuclear produced almost half of global electricity in 2019, almost all of which was base-load power.
Hydropower involves damming a reservoir, letting water only through a small tunnel called a penstock, which turns a turbine and generates electricity. Sluice gates can increase or decrease the amount of water flowing through the penstock to control the power output.
Source: Student Energy
There are two types of natural gas plants: gas peaking and gas combined-cycle. Gas peaking turbines work in more or less the same way as coal, except that they can be quickly ramped up and down according to demand. Gas combined-cycle plants also work this way, except waste heat from the first turbine’s exhaust is then used to turn a second turbine, making these plants far more efficient. Whilst gas combined-cycle plants take longer to ramp up and down, their power output can still change considerably over the course of the day.
Hydro and gas produced 40% of the world’s electricity in 2019. They are load-following energy sources, meaning they can react to changing demand.
The combination of base-load and load-following power has meant supply almost always equaled demand, and grid operators have rarely had to force blackouts to shed excess load.
The big problem with solar and wind
Now, that isn’t the same when you start adding wind and solar to the equation. In a world in which more than a third of our power is from solar farms, there will be massive over-supply during the sunny hours of the middle of the day, and massive under-supply during the night. This gives us the ‘duck curve’ we have below, which plots California’s power generation from non-solar sources across the day, in different years.
Source: California ISO / Bloomberg
As more solar is added, we need to be able to ramp up non-solar power sources quickly in the late afternoon in order to keep up with consumption.
Renewables can also be unreliable and unpredictable. Often the wind just doesn’t blow and the sun just doesn’t shine. A cloudy hour with little wind happens almost daily. In that time, if most of your electricity is coming from wind and solar, the grid would provide well below the required energy, causing electricity price hikes or even outages. Tiny amounts of under-generation can force outages across an entire grid. If you’re like Australia, which has the entire east coast on the same grid system, this could mean a lot of the country is without power.
With climate change causing more frequent and severe weather events, we could easily see natural disasters hindering or even destroying exposed solar and wind farms. Funnily enough, wind turbines are often turned off because wind speeds are too high, not too low, for fear of damage. We will need other energy sources during these times as well.
And you cannot run nuclear power plants only at night, or when solar and wind aren’t generating, as some might like to suggest. As I mentioned earlier, nuclear power plants are only economically viable when running close to 24 hours per day. Plus, they take too long to ramp up and down.
The bottom line is: variable renewables themselves do not provide the control required to meet demand.
The developed world has become accustomed to the idea of constant power. Rethinking that assumption would have drastic economic consequences. In truth, it really isn’t an option. Either we come up with a reliable renewable energy grid, or nothing will happen.
Fortunately, we have the technologies right now to deal with this problem.
Diversification is key!
Balancing supply and demand in the grid is like choosing a portfolio of stocks. The more renewable sources you add to the grid that aren’t completely correlated with each other, the less variable the supply of power will be. Take the extreme case: if the entire grid was powered by one wind farm, and the wind stopped blowing in that area even for a few minutes, you would lose your entire power supply. It’s far less likely that everywhere experiences no wind at the same time, so if you had many wind farms across the country, you would get a far smoother supply of power over time.
The goal is to have many different renewable sources in as many different places as possible, all powering the same grid. That way, the grid is predictable, consistent and reliable. It is for this reason solar and wind are so complementary to each other. Solar produces electricity almost entirely during the day; wind speeds pick up at night.
It is also a big reason for the rapid growth of offshore wind, which adds a number of benefits to a grid system:
Offshore wind represents a strong source of diversification; farms are placed well away from onshore wind instalments.
Wind speeds over the sea are stronger and often more consistent than those over land.
Although offshore wind is currently more expensive than onshore wind, it is getting cheaper. Because no one can see these installations, and because you can transport larger structures by boat than by road, offshore wind turbines are a lot, lot bigger than onshore turbines — and they’re getting bigger. With larger size comes higher efficiency and lower unit costs, making offshore wind even more attractive.
4. Many countries without significant land space for solar and onshore wind farms, such as the UK and Germany, can supply a significant amount of their energy from offshore wind. In fact, the International Energy Agency (IEA) has estimated there are enough suitable areas for offshore wind to power the world 18 times over. Offshore wind could easily dominate the energy sector in a few decades time.
But, with a diverse set of renewable resources (solar, onshore wind, offshore wind, and hydropower) comes a huge challenge for grid operators: they all need to be connected up to the same grid.
Thankfully, High Voltage Direct Current (HVDC) transmission lines are an effective solution we have right now. They can transport 12GW of power over 3000kms at a power loss of just 10%, half that of normal AC transmission lines at the same voltage.
As well as protecting against unpredictable weather patterns, east-west HVDC lines smooth solar and wind production across the day, whilst north-south HVDC lines smooth across the seasons. International, and sometimes undersea, transmission lines will also be essential in balancing renewable grids.
But upgrading existing grid networks will take a lot of investment. So to will building networks in areas of the developed world without access to power. From 2016 to 2020, the world spent US$300bn per year on energy infrastructure. The IEA says that needs to triple by 2030 if we are to get to net-zero.
However, it is important to realise diversification and more consistent production don’t provide the whole answer. All they do is make variable renewable production more like base-load power. Energy demand is still variable across both the day and the seasons. For that, we need an enormous source of load-following power.
We know hydropower can be ramped up and down according to demand. And in fact, hydro accounts for 16% of the world’s electricity generation, the largest renewable source by far. But there is a natural limit to the amount of hydro we have: the number of rivers not already being used for hydro is diminishing rapidly. Plus, there are many concerns about the environmental impact of hydro damming. Because of this, only countries like China and Ethiopia have been adding significant hydro capacity. In fact, the IEA thinks hydropower will decrease as a proportion of the world’s energy mix over the coming decades.
We need another source of scalable load-following power: energy storage. By storing electricity during the day when power supply exceeds demand, and releasing it back into the grid when supply falls short, batteries, pumped hydro and hydrogen can make sure there is always enough energy to meet demand, preventing extremely volatile energy prices and disruptive outages.
Batteries: the holy grail or way over-hyped?
There are four ways in which batteries are being used to balance the grid:
Renewable power installations, such as solar and wind farms, can have battery farms installed onsite. The big advantage of batteries is that they can switch from consuming power to fully releasing it in a matter of milliseconds. This means they can counter very short-term fluctuations in the power output of the renewable energy source they service. When a cloud flies over a solar farm, an onsite battery installation could fill the shortfall in generation for that short, 10 minute period. Otherwise, the solar company could be penalised by the grid operator for having overly unreliable production levels.
Utility-scale battery farms can load-follow over time periods of under 4-8 hours. They can store excess solar power during the day for release during the afternoon and night, and can store excess wind energy during the middle of the night for release in the morning. Utility-scale battery farms are starting to be built on the site of retired coal and gas power plants to take advantage of existing transmission lines. However, it is important to know that these batteries are not yet capable of load-following at a low enough cost for more than 4-8 hours hours. That is where pumped hydro takes over.
Remote batteries (the household Tesla Powerwall being the obvious example) can smooth out the production of rooftop solar panels on residential, commercial or industrial real estate. In a renewable world, household batteries can store energy from the grid during the day and release it back at dinner time, earning a sweet credit on the homeowner’s power bill whilst helping the grid to balance.
Electric vehicle batteries act in the same way as household batteries, providing an extra source of storage for grid balancing when connected to power. The Australian Energy Market Operator (AEMO) says Vehicle-to-Grid energy storage could account for as much as two-thirds of Australia’s total energy storage by 2050, measured in terms of power output.
The democratisation of the grid, in the sense that consumers are active players in the grid’s optimisation, will be essential to the reliability of a 100% renewable future. For instance, financial incentives to charge electric vehicles during the day are types of demand management, a practice crucial to smoothing out variable renewable power supply.
Promisingly, the cost and capacity of batteries have been improving consistently and significantly over the past few decades. It’s for this reason that 100% renewable systems have become so economically attractive in recent years. There are still a few problems with batteries, notably the availability of critical metals and the treatment of harmful chemical waste. We will need to address both of these problems if we are going to sustainably achieve net-zero.
Pumped hydro: the neglected powerhouse
Despite all of the hype around batteries, pumped hydro represents a staggering 99% of global energy storage right now. Often the neglected older sibling of battery storage, pumped hydro will play an enormous role in balancing renewable electricity grids over time periods ranging from a few hours to a number of weeks.
Pumped hydro converts electrical energy into gravitational potential energy, which can then be converted back into electrical energy whenever required. These installations are enormous batteries.
‘On-river’ pumped hydro is more or less the same as a normal hydro plant built on a river. The only difference is that the tunnel through which the water travels is reversible. When power is cheap, they pump water from the lower reservoir (downstream, below the dam) to the upper reservoir (upstream, above the dam). When power is expensive, they let gravity push the water back through the same tunnel, which spins a turbine to produce electricity.
‘Off-river’ pumped hydro is very similar, except that it isn’t built on an existing river system. The upper and lower reservoirs can be lakes, separate dams, or even abandoned open-cut mines. One big benefit of these installations is that they don’t damage river ecosystems.
Source: Australian Renewable Energy Agency (ARENA)
I spoke to Paul Graham, Chief Energy Economist at the CSIRO, whose research finds that pumped hydro is cheaper than batteries for more than 8 hours of energy storage. The cost of battery farms is derived from both the inverter (which converts DC power into AC) and the batteries themselves. In order to double the energy storage of a battery farm, you do not need to have another inverter, but you do need double the number of batteries. And hence, costs increase considerably with increased energy capacity.
On the other hand, by far the most significant cost of a pumped hydro project is the turbine. In order to increase the energy storage of pumped hydro, all you need to do is increase the water capacity of the reservoirs, a relatively low-cost exercise. The number of turbines only changes the amount of power generated, not the energy capacity (which is power multiplied by time). Therefore, pumped hydro is currently the cheapest method of storing power over the medium to long term. But with decreasing costs, batteries will start to cut into pumped hydro’s market share quite considerably.
Even more important than pumped hydro’s cost competitiveness is its enormous capacity potential. I had the privilege of meeting with Professor Andrew Blakers, an ANU engineer specialising in 100% renewable systems. His research has identified 616,000 potential pumped hydro sites around the world. These sites consist of two reservoirs differing in height by 200-800m, sufficiently close together for a tunnel to be built, and not in national parks or residential areas.
Source: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8836526&tag=1. Dots represent the 616,000 potential pumped hydro locations.
He estimates these off-river pumped hydro sites can provide enough storage (23 million GWh) to entirely support a 100% renewable world more than 100 times over. In other words, less than 1% of these sites would actually need to be built in order to support a renewable world. And this analysis doesn’t even look at ‘brownfield sites’ — old mining pits or existing reservoirs able to repurposed for pumped hydro. Contrary to common criticism, the water used by these off-river pumped hydro projects is a fraction of the water we currently use in coal-fired power stations.
But regardless of how promising batteries and pumped hydro are, they don’t entirely solve the problems posed by 100% renewables. We need super long term storage, backup power during peak demand, and a source of clean electricity in countries without enough renewable energy. Thankfully, hydrogen can solve these problems — and many more. But because of just how important hydrogen is for the transition to net-zero, I will leave a more in depth discussion of it for a later article.
Unfortunately, for the time being, we may also have to keep some natural gas plants in order to provide backup load-following power as we transition to variable renewables. Often described as a ‘bridge fuel’, natural gas is cheaper, more efficient, and emits less carbon dioxide than coal. And on top that, gas plants can gradually reduce their emissions by co-firing with increasing proportions of clean hydrogen or ammonia, or can be retrofitted to use entirely biomethane, a renewable power source. The IEA has said the global average blend of hydrogen and ammonia in gas-fired power stations needs to be 85% in order to achieve net-zero.
So, what are the big takeaways?
Solar and wind are dominating the energy scene. Soon they’ll be supplying the majority of electricity globally. But to support massive amounts of variable power generation, we need three things.
We need diversified renewable energy sources, connected by upgraded transmission infrastructure.
We need vast amounts of energy storage in the form of batteries, pumped hydro and hydrogen, and potentially backup from increasingly clean gas-fired power stations as well.
Or, we could just build more and more renewables such that we always produce more than we demand. However, that would result in enormous amounts of wasted energy, which we call spillage.
Professor Blakers says a country can interchange between these three support systems and achieve more or less the same result at the same cost. For instance, he believes the US, which currently runs on a regional grid system, would require only one-fifth of the energy storage if it were to connect to a national grid. But because of individual states’ concerns about energy security, we’re unlikely to see a national U.S. grid anytime soon.
Most amazingly, the CSIRO has found that even including the cost of transmission lines, storage and spillage, solar and wind are still the cheapest form of electricity generation in Australia. Gas combined-cycle plants are competitive, but they note this analysis doesn’t take into account both the difficulty of funding a gas project these days and the potential for carbon pricing in the future.
On top of that, solar and wind have almost zero marginal cost, meaning electricity prices could become incredibly cheap once the storage and transmission infrastructure is up and running. Imagine the productivity boost of cutting electricity prices in half.
“It is now cheaper to save the Earth than to ruin it.”
But, are we really getting there?
Despite everything we’ve done to accelerate the move towards renewables, we still aren’t doing nearly enough. The IEA estimates the world will add 305GW of renewable power per year from 2021 to 2026. This is their ‘main case’ (their best guess).
Under their ‘accelerated case’, which assumes a massive policy shift, huge investor backing, and decreasing commodity prices, we add 380GW per year in the same time period. This is their best-case scenario.
To achieve net-zero, annual growth needs to be 550GW…
Source: International Energy Agency (IEA)
And in 2021, we added 295GW. Not a great start.
The West has spent the last few years sitting around telling itself “hey, look at how much renewable power we’ve added, we’re going to have entirely clean energy in a couple of decades, aren’t we good!”.
News flash, we aren’t good.
Electricity generation is actually the easiest sector to decarbonise. The IEA says advanced economies need to be net-zero in electricity generation as early as 2035. Countries like Australia, with abundant solar and wind resources, need to be there even earlier.
We need to buy time whilst we decarbonise harder to abate sectors, make countries like Russia wake up to this crisis, and assist developing nations with this mammoth challenge. Plus, we need clean electricity to make the electrification of transport and real estate at all impactful.
Renewables are the low-hanging fruit of the net-zero transition. But we’ve spent too long wandering the orchard picking fruit from the ground.
We need to do so much more. Otherwise we’ll have lost this race before it even started.
More from The Climate Project:
The party is over