Airlines
Airlines face a unique challenge when it comes to going green: where electric battery technology is the bedrock of transition in sectors like electric vehicles (EVs), these are simply too heavy to put on a plane. Some smaller aircraft can (and do) use electric batteries. But they won’t work for the vast majority of planes, and therefore won’t provide the means to reduce almost all of the industry’s emissions. That’s a severe handicap.
The solution lies in developing Sustainable Aviation Fuels (SAF) like biofuels, which come from food waste and other biomass. The good news is that the science is basically there, and airlines have been allowed to use a blend of fuels that include SAF on commercial flights for more than a decade. The problem now becomes one of scale, and that’s the second challenge airlines face.
To understand the difficulty, think about how hard it is proving we can create a reliable network of EV charging points. Now imagine doing something similar with airplanes. The adoption of SAF at scale involves adapting all the existing infrastructure behind airports and planes to handle the new fuels. That requires huge amounts of capital investment and, crucially, time.
Unfortunately, I don’t see a way that consumers won’t bear the brunt of the associated costs. As well as the capex demands, SAF themselves are now significantly more expensive to manufacture than regular fossil fuels.
That sounds bad, but I’m actually quite positive on the shape of the market, largely because companies are more aware of the challenges than they were, say, five years ago. Every airline and aerospace company I speak to is desperate for a more environmentally-friendly source of fuel – if they found one, they’d put their competitors out of business overnight. The cost of paying for carbon credits, like those established through the European Emissions Trading scheme, provides more financial motivation to transition at pace, as does an increasingly weighty regulatory burden.
– Marcel Stotzel
Utilities
There has been very little investment in the electricity grid across the developed world over the past two decades. But given the energy transition means working with more complex renewable energy sources and new demand from an increasingly electrified society, this needs to change quickly. We’re already seeing much more pressure on the system from data centres, and from countries looking to onshore manufacturing activities, for example. It’s all the harder given growth in electricity demand was so lacklustre over the last few decades as systems such as lighting and computing became more efficient.
Upgrading to a superior, decarbonised grid requires a huge amount of capital, but it also means tackling issues around permitting and policy. There’s also the headache of finding sufficiently trained labour and increasing the production of equipment such as transformers to cope with higher demand.
There is some good news – for example, the bottlenecks around equipment manufacturing are now easing, and capital is coming in, although countries that have set aggressive targets for decarbonising their grids, such as the UK, are looking overly optimistic.
Shifting to more intermittent renewable energy means a complete change to our system architecture, away from a base loaded system. Again, there exist options such as batteries or pumped storage hydropower, but operators continue to look for a better long- term solution. This is a sector that’s very good at solving its own problems though – even if they’re not perfect.
– Alexander Laing
Shipping
It’s been suggested that for shipping to be compliant with a 1.5-degree pathway the sector will need to cut the carbon emitted in the transportation of every container box by 50 per cent from its 2019 levels by 2030. At any one moment, there are around 20 million container boxes on the move on our oceans, and our existing fleet of container ships is not yet set up to run on clean fuels.1
In 2023 the International Maritime Organization (IMO) produced an industry-wide strategy to reduce the carbon intensity of international shipping by at least 40 per cent by 2030.2 It’s not clear what power the IMO has to enforce the target, however.
Given that most ships typically have a life cycle of 20 to 25 years, many don’t see the financial benefit of paying to upgrade their vessels before they’re due for the scrapyard – especially given there’s no significant regulatory pressure in play yet.
Nevertheless, some shippers are trying. Retrofitting existing vessels to dual-fuel engines is one option being explored, although this can only be done with newer and larger ships. Maersk is buying ships that run on methanol, which emits a fraction of the carbon of conventional fuel. Liquified natural gas (LNG) can also help with particulate emissions, although it still has a relatively high carbon footprint.
In the longer term, hydrogen is being touted by many as a potential solution – although it’s a 10 to 15-year time horizon before it can be fully implemented. Green ammonia is another exciting prospect given it’s a clean, carbon-free fuel. Some companies are already exploring that route.
I think we will end up seeing a combination of these fuels forming a solution, while there will also be innovations that allow ships to reduce their energy needs. A prototype is in development for a container ship with sails, for example. While wind can’t move the current generation of container ships on its own, there could be some marginal benefits of hybrid systems that save one or two per cent on fuel consumption.
– Jonathan Neve
Artificial intelligence
It won’t surprise anyone to hear that demand for energy from the artificial intelligence (AI) industry is growing. But this is steady growth rather than a gigantic spike. As a rule of thumb, one graphical processing unit (GPU) – the chip at the centre of AI operations – uses as much power in a year as one US consumer. The industry is currently adding 4 to 5 million new GPUs per year, meaning global GPU demand is equivalent to an additional 1.5 per cent annual growth in the US population. This figure could increase as GPUs become more powerful and their energy demand increases, although all the AI companies in our coverage are very much focused on reducing the power requirements of their operations. Often, the cost of maintaining, powering, and operating the chips used in data centres is more than the upfront cost of the chip itself, so there’s an economic incentive to try to increase power efficiency.
Obviously where marginal energy demand is tight, a couple of percentage points growth in the requirements on a system can have an impact on the supply-demand balance. It’s therefore important at the moment to place data centres in locations where there’s more power capacity available. Further down the line – and so long as the demand for AI continues – we’re going to have to increase supply capacity.
However, the challenge is not only in the amount of energy needed, but also the nature of the supply. Data centres run 24 hours a day, seven days a week, and so need a steady base of power. That’s difficult when you’re considering the intermittent nature of renewable sources of energy. Nuclear power, which has zero carbon emissions and provides steady base level power, is likely to be used in combination with battery storage or pumped hydro to smooth over that generation versus demand gap.
There are also solutions being developed to cool data centres more efficiently, and to improve chip architecture. We’re already seeing some suppliers creating more specialised chips. Google has custom-developed its own tensor processing units (TPUs), which run better and at lower cost than the generic GPU. As the sector matures and the technology is standardised, more individual firms could be encouraged to create their own specialised solutions.
– Jonathan Tseng
Mining
The easiest way to reduce emissions from the mining industry is for the sector to make greater use of renewable energy. This could include more electric trucks, which have progressed significantly in the last couple of years. Indeed, in August Fortescue announced it is teaming up with Liebherr to develop the world’s first self-driving electric mining truck. It could also involve trucks powered by hydrogen such as those trialled by Anglo American.
But the biggest challenge facing the industry is removing carbon from the processing. Intensity varies dramatically from one metal to another – producing nickel and aluminium, for example, is highly carbon intensive. Alternative low-carbon processes can often cost more and, in many cases, will only be economical if those products attract a premium. In practice this means consumers accepting to pay that premium. Low-carbon aluminium has had some early success here, but progress is uneven and the same cannot really be said of, say, nickel.
At times, there may be a need to support the transition more directly – the large amounts of capex now being announced to decarbonise Europe’s steel plants demonstrate how financially painful this can be. Policymakers have a role in providing a supportive regulatory framework. It’s difficult to see how the sector shifts to an acceptable carbon intensity without widespread carbon taxes, for example. These have been effective in changing behaviours in Europe but, without wider adoption, risk leaving affected European industries exposed and uncompetitive.
The bottom-line is that abating the emissions of the metals and mining sector will need to be paid for. The energy transition won’t happen without the materials the sector produces. Supplying them in the necessary quantities using low-carbon, sustainable processes will cost more than using traditional methods. The obvious example is green steel. Producing low-carbon steel is technically feasible, using green hydrogen or electricity from renewable sources instead of metallurgical coal.
But it comes at a premium which varies depending on where you produce it, and even a relatively modest premium can make more sustainable alternatives economically unviable in the absence of external support.
– James Richards and Oliver Hextall
1 Science Based Target Setting for the Maritime Transport Sector, Science Based Targets, May 2023
2 2023 IMO Strategy on Reduction of GHG Emissions from Ships
