Transportation accounted for 28% of global final energy demand in 2017. The sector released 8.2 Gt [1 Gt = 1 billion tonnes = 1 trillion kg] of CO2 in 2018 alone, which represents roughly 23% of global CO2 emissions across all sectors the same year. Transportation’s CO2 emissions have risen by 41% from 2000-2018. Alas, burning fuel doesn’t just emit CO2 – it also releases toxic pollutants.

Its high energy consumption and pollution make the transportation sector an innovator’s dream. Between artificial intelligence and electric vehicles, huge investments are being made to make this industry more sustainable. Both these advances have the potential to lower emissions, if applied globally at the proper scale. Unfortunately, we don’t have the time to wait around until they do. There are many policies and solutions that governments and companies can implement to reduce the sector’s global impacts effectively, instead of placing the blame solely on individuals. That being said, individuals have lots of opportunities to reduce their personal transportation impacts.

Individual car needs are often linked to urban expansion and the proportional weakening of public transport services. In other words, as we move further away from the city, a car becomes increasingly necessary. It’s also true that as a city expands, more amenities get built far from the city center. Thus, urban populations also find themselves wanting cars. Both these factors [and many more] help explain why individuals were responsible for around 63% of total transportation energy use in 2012.


We often overlook the transportation required for everything that surrounds us. Freight transport [vs. passenger transport] accounted for the remaining 37% of the transportation sector’s energy use in 2012. Freight planes, boats, trains, and trucks supply us with food, raw materials, and anything else we buy far from home. For equal transportation methods, greater distances imply more fuel and therefore more emissions. However, we’ll see that some methods of transport are far more efficient than others, which we must take advantage of in the future.

We should note that globalization has some wonderful aspects that allow us to share items worldwide and increase quality of life in many areas of the world. However, the environmental cost is already proving too much to bear. We need to reduce our impacts on the environment by shortening the distances we [i.e. individuals, food, other products] travel while choosing more efficient transportation methods.


Planes are astounding polluters due to the type of fuel used and the long distances travelled. Interestingly, countries don’t take responsibility for the pollutants that international airliners emit. Instead, international flights are bundled together in a separate category that countries don’t really need to worry about – since emissions aren’t counted in their national totals.

In 2018, international and domestic flights combined emitted 0.93 Gt of CO2. That represents roughly 11% of the transportation sector’s total CO2 emissions the same year. To put this in perspective – those emissions have the same weight as 9.3 billion giant pandas [with 1 panda = 100 kg] being released in the atmosphere.

This has not been getting better. Aviation’s global CO2 emissions in 2018 were roughly 38% higher than they were in 2000.

Planes use kerosene and other lightweight jet fuels [that have high energy density] to power the engines without adding too much weight to the aircraft. Unfortunately, these fuels release more CO2 emissions than standard fuel – but only 78% more. This doesn’t mean jet fuel isn’t as bad as everyone makes it sound, it means that we don’t realize how polluting gasoline is. One of the most interesting areas of innovation within the aviation sector is lowering GHG emissions by opting for alternative fuels.

That’s much easier said than done. We use the fuels we use because they have high enough energy densities [i.e. energy per unit of mass] to allow aircrafts to fly. And although hydrogen-fueled planes have received support from governments and aviation companies due to hydrogen’s extremely high energy density and apparent eco-friendliness, these types of planes currently face numerous challenges that make them an unlikely option in the near future.

The graph below shows the amount of energy required by different passenger transportation modes – per passenger for an equal travelling distance.

The only reason that planes aren’t that much worse than cars – as far as energy required per passenger-km, is that they can hold many people at once and are often full [that’s why private jets, for example, are such terrible polluters]. On the other hand, cars are often half-empty. Also, note that a single plane trip emits far more than a single car trip, but that’s really just due to the difference in distances travelled.

Ships and Boats

Boats are pretty big polluters as well. In 2018, shipping emitted 0.86 Gt of CO2. That represents over 10% of the transportation sector’s total CO2 emissions the same year – equivalent to roughly 8.6 billion pandas emitted.

This has not been getting better. Shipping’s global CO2 emissions in 2018 were nearly 38% higher than they were in 2000.

Additionally, oil spills can occur when ships crash or leak. That can be absolutely devastating to local populations and biodiversity. Media coverage has been present for a few of these oil spills in the past [while some went completely unnoticed] – so we know exactly how terrible these spills are for local health, food, and economy.

As far as international freight transport goes, there’s no question that boats emit much less GHGs per unit weight of cargo than aircrafts. While both boats and planes end up emitting roughly the same amounts of CO2 every year, we have to put that in perspective with the massive quantities of goods that ships transport. For example, maritime shipping accounted for roughly 75% of imported/exported goods [by weight] between the EU and the rest of the world in 2019. On the other hand, aircrafts were mostly used to transport higher-value products.

If our goal is to reduce the transportation sector’s emissions as a whole – then we’re going to have to transition away from using airplanes for freight transport.

Nonetheless, boats still pollute since they burn fuel [often burning highly polluting residual fuel left after oil refining processes]. Large and heavy ones especially, so it’s not very surprising that cruise ships are being scrutinized extensively for the damages they incur to the environment. After all, who would have thought that running a huge hotel solely on oil as an energy source could pollute so much?

Trucks and Cars

Trucks and cars emit way too many GHGs and toxic pollutants. In 2018, passenger and freight road vehicles emitted 6.0 Gt of CO2. That represents over 73% of the transportation sector’s total CO2 emissions the same year – equivalent to roughly 60 billion pandas emitted.

This has not been getting better. Road transport’s global CO2 emissions in 2018 were over 41% higher than they were in 2000.

Heavier vehicles like trucks consume more fuel and emit more GHGs per km than lightweight cars. However, there are other factors that can increase our individual emissions as well, like speed. Speeding on a highway burns more fuel than driving at the speed limit, even when considering the amount of fuel saved from the trip being completed early.

Additionally, while gasoline may not be scrutinized as much as jet fuel, it remains a terrible polluter that nearly releases the same amount of GHGs per kg of fuel consumed.

Road vehicles are astounding polluters because they are individualized. As such, the easiest thing to do to cut our transportation emissions in half is to carpool with one other person. Divide by three? Two people. This is why buses, although heavier, can emit much less CO2 per passenger-km. Improving shared services like public transit is by far one of the best ways to ensure cleaner mobility in cities and neighboring areas.


Trains have been one of the cleanest transportation modes for a while now, ever since most of them ditched coal many moons ago. In 2018, rail emitted 0.08 Gt of CO2. That represents just 1% of the transportation sector’s total CO2 emissions the same year – equivalent to roughly 0.8 billion pandas emitted.

This is getting better, despite passenger rail usage almost doubling from 1996-2016. Rail’s global CO2 emissions in 2018 were 2.4% lower than they were in 2000. That’s partly due to around 75% of passenger trains worldwide relying on electricity as a power source in 2016– a considerable increase from 60% in 2000.

Trains are without a doubt the most eco-friendly way to travel long distances. Despite accounting for 7% of freight and 8% of passenger transportation worldwide, rail only accounted for 2% of the transportation sector’s energy demand in 2016.

Even for short distances, urban rail is proving very energy efficient. However, for trains to really show how eco-friendly they are, we need people to use them. And that means reducing the cost of a ticket considerably. What message are we sending if the cost of low emission travel is terribly expensive? And for that matter, why are we even charging for public transit? Reducing individual transportation is one of the eco-friendliest solutions out there, but it simply can’t happen until public transit offers cheap fares for short and long distances alike. Government policies and subsidies for cleaner travel would be far more effective if concentrated on improving shared services, instead of other individualized solutions.

Electric Vehicles

Electric vehicles [EVs] run on electricity stored chemically in batteries in the car ‘floor’ [other types of EVs exist, but battery EVs are the most common]. These batteries are extremely heavy for the amount of energy stored [low energy density], whereas gasoline has decent energy density. As a result, an EV weighs much more than a similarly sized combustion vehicle [CV].

EV manufacturers are always looking to improve the range of their vehicles. We could just load the car with more batteries, but this would increase the car’s weight. And, if the mass of the vehicle increases, then so does power consumption. That’s part of the challenge of increasing EV range. There doesn’t seem to be an easy solution, since loading up more batteries would increase both the energy supply and consumption of an EV.

The solution for increasing the range isn’t in the number of batteries, it’s inside the batteries. Researchers worldwide are working on increasing the energy density and lifetime of batteries. In 2018-2019, the battery energy density in average electric vehicles was 20-100% higher than it was in 2012, while battery costs have decreased by over 85% since 2010. Governments and companies are also working on increasing the number of charging stations along popular travel routes to fight against range anxiety.

2 Important EV Problems

While EVs are becoming increasingly common, they have their fair share of challenges.

First, batteries require loads of components and metals. For example, an estimated 19 kilotonnes of cobalt, 17 kt of lithium, 22 kt of manganese, and 65 kt of nickel were required to make batteries for EVs sold in 2019. If over a million pandas-worth of metals doesn’t seem like a lot, stay with us.

As of 2019, the total worldwide supply of lithium was 77 kt per year across all industries, with 21% of that being used for EV production. By 2030, lithium demand for EVs alone is required to increase to at least 365.3 kt per year to align with the Sustainable Development Scenario – much higher than the 17 kt required in 2019.

That’s just for lithium and EVs. We already know that renewables depend on this metal for energy storage, as do high-tech devices. And unfortunately, similar projections exist for other metals that are critical to battery production – like cobalt, manganese, and nickel.

Now this doesn’t mean that we won’t be able to meet this projected demand by 2030. However, we’re going to need to increase mining activity considerably – which will increase energy consumption and other mining impacts as well.

Alas, it won’t stop there. There were ‘only’ 7.2 million EVs on the road in 2019, compared to over 1 billion CVs. The massive increase in mining activity described above is only projected to help EVs reach 13.4% of the total car share by 2030 – up from 0.8% in 2019. If EVs are supposed to replace all CVs in the future, we can expect this increasing mining activity to continue for decades.

The following graph shows how important and metal-intensive batteries will be in the upcoming green revolution. How we decide to adopt EVs [i.e. at an individual scale or for public transit] will have massive implications on mining activity and the speed at which we complete our energy transition.

The second problem is really an electricity production problem. If our global electricity mix is still 37% coal and 63% Big 3, then recharging our EVs still has a considerable footprint [although that will depend on local electricity mixes].

Combining the metal supply and electricity mix problems, let’s see what we’ve got. Unsurprisingly, comparing a mid-sized EV [40 kWh battery pack] with a similarly sized CV [6.8 L/100 km] has shown that the EV emits 33-57% more GHGs during manufacturing, which makes sense since the average EV is made with 6 times the mineral inputs required to make a CV.

Helpful note for the following paragraph: the ‘CO2e’ or ‘CO2 equivalent’ unit accounts for all GHGs and their greenhousiness, not just CO2.

On the other hand, fuel impacts depend heavily on the electricity mix and the fuel consumption of the CV. For an electricity emission factor of 50 g of CO2e/kWh, it’s estimated that EVs would emit 94.7% less than CVs during fuel consumption [with an assumed lifetime of 200,000 km for both vehicles]. However, for an electricity emission factor of 800 g of CO2e/100 km, that lowers to just 15.3% less than CVs during fuel consumption.

Combining both production and fuel consumption, we can conclude that in the low-carbon electricity mix, EVs end up emitting 73.0-76.4% less than CVs over a 200,000 km lifetime – or just 5.0-8.4% less in a high-carbon electricity mix. Consequently, it’s incredibly important that individuals look up their local electricity emission factors to ensure that they’re not doing more harm than good – on top of the whole ‘metal-intensiveness’ issue. For reference, the US national average grid electricity emission factor is around 400 g of CO2e/kWh in 2021 – while the EU had an emission factor of 275 g of CO2e/kWh in 2019 [keep in mind that regional emission factors can be very different from local ones].

It’s important to note that lower fuel-consumption for CVs can also help slash the transportation sector’s emissions significantly. With more research and stronger regulations, we could design and offer affordable fuel-efficient vehicles [e.g. under 2 L/100 km] – which could help us reduce our impacts even further.

EV Conclusion

Although we’ve made great strides and are starting to see electric vehicles on the road more frequently, we should be careful of their collateral damage. It’s great that the public opinion is shifting in favor of decarbonizing the atmosphere. That should be our priority. However, like renewable energy technology – if we are too quick to adopt a ‘half-good’ solution, we’ll end up creating more problems than we started with.

The best solution is using our technology in moderation, where it makes sense to do so. EVs are helping and will continue to help lower the transportation sector’s GHG emissions. This will be even more evident as our global electricity mix further decarbonizes. However, we simply don’t have the resources for everybody to make the switch to EVs. Even if we did, digging up all the metals required would prove incredibly damaging to the environment and could slow down our energy transition.

Our first priority shouldn’t be jumping the gun and banning new CV purchases by 2050, as many countries have ambitiously declared. Neither should it be to reject all EVs because they aren’t as clean as advertised. The most important thing we can do now is reduce our dependence on individual transportation while demanding improved-cheap public transit. At the same time, we can focus on electrifying public transit [e.g. with trams/streetcars, rail, EV buses, etc…], since public transportation is the future of eco-friendly travel.

Artificial Intelligence

Artificial intelligence [AI] has the potential to reduce the transportation sector’s emissions, especially for public transport. It could allow our cities to become well-oiled machines, where efficiency rules. This would reduce fuel waste, encourage the use of public transport, and redefine vehicle sharing apps.

Unfortunately, we’re still years [or decades] of investments, research, and trials away from driverless vehicles becoming available to the public. Additionally, even if the technology were proven and integrated in our societies, a driverless vehicle wouldn’t magically stop consuming fuel. Although fuel consumption could be optimized with efficient AI software and car-to-car communication [e.g. to reduce traffic], that would barely make a dent in the sector’s total emissions. Without mentioning the hidden impacts that this type of AI technology can have [e.g. data storage]. As we’ve stated previously, a shift to public-transit-dominated societies is required.

Nonetheless, driverless vehicles could one day revolutionize public transit by driving people that don’t live near public transit routes to the nearest station. This could help us replace buses that run with little to no passengers with mini shuttles. These driverless shuttles would then operate at lower costs and lower fuel consumption – since they wouldn’t have drivers and would be much lighter than large buses.

Autonomous vehicles would also be pivotal for car-sharing applications when public transit isn’t an option, to reduce the number of cars in the world [by reducing the number of cars not in use]. While this wouldn’t necessarily reduce the number of cars in use at any given moment, AI algorithms could help with that by making carpooling and car-sharing much more convenient.

Alas, driverless vehicles won’t be available for a while, so we’ll have to stick to more concrete solutions for now.