AHOY! Cruise ship versus airplane emissions: data and commentary
Last year I inquired how one might take sea passage across the Atlantic. The realistic answer is that there are various ocean liners and transatlantic cruises traveling multiple times per year, none faster than 7 days and many taking closer to 10 or 14. Repositioning cruises, when the ship is being moved from one region to another and you just tag along, are infrequent but the cheapest option.
In February, I will unavoidably be in the United Kingdom. I am flying there, but have not purchased a flight back yet. I am thinking about taking a ship in repositioning from England to the US Eastern Seaboard as there is very conveniently one such ship leaving a couple days after my event is over. That would probably be Southampton to Miami (from there, I would take a train home) and would take 11 or so days, zero of which are at intermediary ports.
I was thinking about a ship over a plane because the last time I flew it was like my ears got blown out for an entire day afterward. I don't know what it is about my sinuses but they have never handled flying well. My height makes the experience particularly cramped and unpleasant and I have gotten sick on every plane I have taken in the last six months. It's the most miserable thing I do to myself on a regular basis.
Unfortunately, all the research I can find on passenger ship emissions seem to indicate that it is worse for the environment on a passenger-mile basis than flying that same route, at least as far as cruise ships are concerned (there is zero research on emissions from being a passenger on a container ship). As of 2006, the Queen Mary 2 ocean liner supposedly emits about 0.43kg CO2e per passenger-mile, compared to 0.257kg CO2e for a long-haul airplane. But emissions estimates vary so ridiculously widely that it is a little bit hard for me to take these figures seriously:
Emissions factors for individual journeys by cruise ships to or from New Zealand in 2007 ranged between 250 and 2200 g of CO2 per passenger-kilometre (g CO2 per p-km), with a weighted mean of 390 g CO2 per p-km.
That's literally an order of magnitude. I think there is some guessing going on here. To translate from p-km to p-mi, that's ~402–3545g CO2e/p-mile or a weighted mean of 628g CO2e/p-mi. I would speculate that a repositioning cruise (which spends no time in intermediary ports because it is specifically supposed to get somewhere efficiently) would be on the lower end of the spectrum. So, honestly, while worse than an airplane in terms of gaseous emissions, it's not... that much worse. From the way articles seem to talk about cruise emissions, I would have thought it would be at least an order of magnitude. (For reference, the difference between a train and a plane is about an order of magnitude.) Which I guess it can be based on the higher figures there, but I am pretty sure that that is derived from non-direct routings (port visits apparently contribute massively to emissions) or from luxury behaviors (i.e. having an enormous stateroom and other amenities that decrease space efficiency).
One may notice that those sources are almost 20 years old. There is more recent research on cruise ship emissions, but the non-academic stuff all seems to cite the Queen Mary 2 statistic (not sure where it even originated). I attempted to discern what cruise ship emissions looked like 20 years ago versus today, but was unable to find any specific information about passenger-mile emissions year-by-year. I don't have institutional access to most journals anymore, so feel free to share if you know anything.
There have definitely been new environmental regulations since 2007. In 2020, some new regulations limiting high-sulphur fuels went into effect. But sulfur dioxide is more of a health concern; it isn't a greenhouse gas. It's toxic to marine life too, and all other life, but wouldn't be considered in a passenger-mile CO2e emissions figure. Apparently the regulation has encouraged more ships to switch to less toxic liquefied natural gas (LNG) fuels rather than the literal bottom-of-the-barrel sludges they've traditionally burned (maybe a 20%-ish improvement at face value), but most ships have just installed scrubbers to continue using the same fuel and emit fewer horrible particulates. Apparently a switch to LNG, while favorable for human health, does not really reduce GHG emissions due to increased methane output.
Aakko-Saksa et al. 2023 seems to be the most comprehensive journal article I can access that covers current strategies in reducing greenhouse gas emissions for ship engines now and in the future. It mainly talks about fuels and technical stuff about engines I don't understand. This paper remarks that the switch to LNG could still be positive; it suggests a 30% reduction in GHG emissions compared to diesel fuel but a 6–23% reduction depending on how much "methane slip" happens; the IEA thinks it's 10% or less. That's still a meaningful reduction, though there is quite a lot of variance. The authors' takeaway is that there could be a significant benefit to switching to LNG and then blending that with greener fuels at increasing proportions over time.
There are many proposed ways to decarbonize the industry. It is not clear to me which of these have been adopted recently. The industry seems to have some interest in decarbonizing, or giving the appearance of having interest, as Norwegian claims to want to "reduce GHG intensity by 10% by 2026 and 25% by 2030, compared to a 2019 baseline with intensity measured on a per Capacity Day basis." And they are actually thinking about it: they have ESG staff and a 2022 ESG Report lays out a few relatively specific and achievable metrics (or so it seems to me, a layperson and a landlubber):
We were very excited to announce in early 2023 that two of our Norwegian Cruise Line newbuilds, expected to be delivered in 2027 and 2028, will be re-configured to accommodate the future use of green methanol. Green methanol is a fuel that we see as a promising future solution. Compared to conventional fuels, it can reduce carbon dioxide emissions by up to 95%, nitrogen oxide emissions by up to 80%, and all sulfur oxide and particulate matter emissions.
Environmental goals:
- Reduce GHG intensity by 10% by 2026 and by 25% by 2030, compared to 2019 baseline, and pursue net zero GHG emissions by 2050
- Decrease fleet-wide fuel consumption of boilers per day by 2% annually, compared to 2016
- 100% of fleet equipped with Waste Heat Recovery by 2027
- Increase the percentage of our fleet with shore power capabilities to 50% by 2024, 70% by 2025, and 100% by 2035
- Increase the percentage of treated wastewater compared to untreated sewage discharged by 2024, compared to 2019
- Reduce bunkering by 4%, as compared to 2019, by 2025
- Decrease the total volume of sludge offloaded fleetwide by 5%, compared to 2018, by 2023
Green methanol is apparently a real thing. The figures Norwegian uses are lifted directly from the Methanol Institute. It does seem like methanol production capacity is increasing and is on track to continue increasing, according to this source. If a cruise ship uses about 250 tons of fuel per hour (91250/year), and current green methanol production is just shy of 1 million tons/year, then if all of that production were directed toward cruise ships then it could fuel about... 11 ships. Hmm. If by 2027 production increases (generously) by an order of magnitude, that's still only about 100 ships, or less than 1/3 of the total fleet worldwide across all cruise firms. Better than nothing...?
Norwegian claims to be "on track" for all of these metrics, especially fuel consumption, but of course they will say that. I can't really figure out what their "-50%" and "-80%" figures mean and I suspect they mean nothing. Frankly most of these commitments are ridiculously insufficient, though it isn't reasonable to expect revolutionary changes to happen in just a couple years. I think we should take these commitments with a grain of salt, but it does make sense from an operational perspective why industry would be interested in improving their own efficiency, especially to avoid potentially crippling regulations from governments who they can clearly see are honing in on emissions.
From what I can tell, many or most of the emissions are not just from the transportation itself (i.e. the burning of fuel for the purpose of moving mass from point A to B) but rather from the hoteling aboard the ship (12x more emissions than land-based hoteling) and from other luxury-related activities. Intuitively, if we know how energy-efficient it is to transport goods by sea, this should come as no surprise: if we're comparing fuel costs, it simply takes less fuel to move objects by water than by air. This is why ferries have so few emissions per passenger-mile. Ship fuel is particularly nasty stuff as far as human health is concerned, but many resources appear to primarily emphasize the non-fuel waste produced by these enterprises. So the CO2e emissions of cruise ships would seem to originate not just from fuel but rather from the inefficiency of human habitation at sea. Norwegian is at least vaguely calculating multi-scope emissions (p. 15) with, for example, "purchased goods and services" apparently making up ~21% of total emissions in 2022 (fuel itself is about 55%, and "fuel and energy-related activities," whatever that means, being another 12%). I assume "capital goods" (18% of their emissions) are the emissions from the ships themselves, which is more of a decarbonization question for manufacturers.
There are also significant non-GHG environmental impacts due to operational procedures taken by cruise ships. Wikipedia has a whole page on the environmental effects of shipping (not just cruise ships, but they are included). Waste dumping, noise pollution, etc. Those externalities are different than the externalities produced by airplanes; same idea, but apples to oranges, so I don't know how to compare them.
Anyway, this is all to say: greenhouse gas emissions from cruise ships are pretty rough. Given the relative lack of information on repositioning cruises specifically, the age of much of the data, and newer emissions reductions which are maybe not yet reflected in the literature, I am going to speculate that such a trip has an approximately equal GHG impact as a long-haul flight, assuming a typical stateroom and a direct voyage. On average, it probably works out to somewhat more emissions, though I personally think ships have a clearer (easier/faster) path toward net-zero (ish) emissions than airplanes given the limitations of each mode.
I will go to sleep and decide tomorrow whether I will fly or sail home, but right now I am leaning toward the sea for this occasion. I am not sure about future voyages yet. We will see.
It’s really hard to say exactly what the environmental impact of a cruise ship is for exactly the same reasons you mentioned. The good news is that pretty much everyone in the industry who isn’t part of the Carnival group seems pretty concerned about improving it.
I would imagine the biggest source of emissions for a cruise ships is basically from generating electricity. Everything runs on electricity, and that includes the engines these days as well. Add on that they are constantly working on things like water reclamation, waste treatment, laundry, air conditioning, and heating the pools. And the fundamental problem with this is that there isn’t really renewable energy at sea - at least not in a way that is useful for a ship. The good news is that ships are now able to get shore power by hooking up at the port. That’s a pretty good thing because cruise ships are docked a good amount of time, though I would imagine that it’s not available everywhere (especially when it’s not docked at a port).
The biggest problem with your conclusion about cruise ships being roughly equivalent to flights is that flights are direct while cruises rarely are. I don’t think I have ever seen a cruise ship that goes directly to its destination. The closest you could get would be a short round-trip one with a single port, but you’re effectively paying double for that trip since you’re not taking the return trip.
The repositioning cruise that I’m looking at moves the vessel from one region to another at sporadic points, so it isn’t a round-trip purchase. It doesn’t have any intermediary stops. (Some cruises which are marketed as repositioning cruises do have stops in small Atlantic islands, but I’m not considering those for my trip next month.) But in general, I agree that a traditional tourist cruise would be on the higher end of the emissions range I identified.
It seems that shore power, while essential in general, isn’t especially relevant to repositioning cruises. But in general, I’m curious about how even a repositioning cruise could use shore power at its embarking port to charge batteries for the rest of the voyage. From what literature I have read, current battery technology is considered too heavy and spacious for this to be meaningful. But given very recent battery tech breakthroughs, I wonder if the equation has changed—cruise ships have far more space than electric vehicles, which can now manage pretty impressive distances.
I did skim through a few theses and other articles analyzing ways the energy generation needs of the ships could be reduced, and they’re not much different than land-based methods: more efficient HVAC systems, better insulation especially around windows, higher-efficiency machines, etc. It’s unclear to me exactly how much this would reduce emissions, but I imagine considerably.
I think I heard something about next generation sodium batteries using cruise ships as an example application. It’s kind of hard to imagine though; cruise ships are basically cities at sea with the amount of power they use. The scale would truly be crazy. It would also be a huge power draw while at port because they would have to at minimum double the amount of power coming through!
One thing to consider is that in a repositioning cruise, the boat was going to go anyway. And that's more true than it is of the similar "well the plane is going to fly anyway" argument, because the purpose of the cruise is to move the boat, not to sell passenger seats. So from that standpoint, the repositioning cruise seems likely the clearly superior option.
This is even more the case for being a passenger on a cargo ship. Though there's almost definitely more waste in terms of food and other non-moving-the-boat sources of emissions on a repositioning cruise, simply due to there being better amenities than on a cargo ship.
Very interesting that hotelling is 12x the land costs. I was wondering if you'd factored that in, and you had. I tried to skim read the article but couldn't see a reason why? I think I can only see an abridged version.
Just spitballing but I could imagine literally just food could be a huge contributor — if you’re on land in a hotel, it’s probably not too difficult to bring in fresh food every day, and therefore your fridges and freezers only really need to be big enough to hold a days worth (or some factor larger just in case) whereas a cruise ship needs to hold many times that much in order to have enough food to last the entire trip.
I was also thinking about this. I found a paper (not peer-reviewed?) by Vučinić et al. 2023, which explains the process by which food is stored and preserved for use for an ocean voyage. It doesn't have a lot of useful data about per-capita food waste emissions, but there is this:
This is food waste, not total food produced. It is unclear to me whether the food processed by the human machine is considered in these figures as "waste," but I suspect not. In any case, Michael Lück writes in Nautical Tourism: Concepts and Issues (2007) that a typical cruise ship in that year would go through 18 tons of food per day. For an 11-day voyage, that's 198 tons in total. There is even a breakdown by type of food for a week:
That's a lot of Scotch. And a lot of meat. Someone who is a little more familiar with land-based food production figures (or has more patience than I do at this particular moment) could figure out the land equivalent of CO2e emissions for producing each quantity of food. If we had a little more information on the storage and waste mechanisms (which I'm sure is out there), we could calculate the worst foods to bring on a ship specifically (as opposed to on land; I imagine they are different). Theoretically, I guess an individual could choose not to eat some such food in order to reduce their individual impact as, possibly, uneaten non-perishables would remain frozen and uneaten perishables would be distributed to landlubbers at port, but it's unclear to me what the exact food waste practices are.
It seems to me that it would make a huge impact if cruise ships would go all-vegan. Plants tend to spoil much more slowly than animal products and have a much smaller environmental impact on their own.
Personally speaking it was a trip on a cruise ship that had started me on the path towards a plant based diet when I was not particularly moved by the dishes on the dinner menu that day and so ordered from the vegetarian section instead and somehow lucked out into the best meal I had on that trip. It would be nice if others would have this same experience, but having an all-vegan sailing is sure to turn off most people.
On a cruise ship uneaten food generally doesn’t get re-served. Maybe some of the stuff in the buffet will come down to the crew canteen. Cruise ships are hotels, and as such they usually have a pretty high bar for quality and will not serve food if it doesn’t meet their standards.
I can also only see an abridged version. I agree with @ThrowdoBaggins that food is probably a large portion of it. But it also seems that some of the electrical systems (lighting, heating, laundry) on cruise ships are inefficient. Brækken et al. write in "Energy use and energy efficiency in cruise ship hotel systems in a Nordic climate" (2023) that:
This doesn't completely answer the question about why sea-based hoteling is more expensive than land-based, but I suspect that building codes on land are simply held to a higher standard. Additionally, cruise ships often visit very warm or very cold places, which probably increases average air conditioning use a lot; so a 38% energy efficiency improvement from a heat pump makes sense!
If so had to venture a guess, I would probably say it’s simple extravagance. There is nothing stopping you from running your hot water all day except for the stateroom attendant coming in to clean your room. There is a sign that will ask you to re-use your towels, but they will happily comply if you want fresh ones every day. There is unlimited food that is complimentary with the purchase of your ticket, so you can eat as much as you want nearly 24 hours a day. A bunch of small extravagances really add up. Compare this to a hotel where people are often there for business instead of pleasure and are not going to indulge in these kinds of practices.
A better comparison would be very high end all-inclusive resorts like Aquarius and the like. But even then a lot of the expenses of running those operations are offloaded on land based operations. They do not need to house or provide for their crew. They do not need to do water reclamation, food and human waste management, and the like. Any studies comparing those should have accounted for that with any luck, but keep in mind the scale of government operations might make them more efficient, especially because a lot of those services can be powered by cleaner or renewable energy sources.
But do keep in mind that it’s somewhat unfair to criticize cruise ships for the extravagance. They are vacation destinations. If the people going on these trips were not going there, they would have been extravagant in other ways instead. It’s a lot harder to figure out what that difference in impact might be because everyone is different.
I guess the difference would be due to the fact that an airline packs you in like sardines with minimal amenities, vs a cruise ship where passengers have separate rooms, bathrooms, GOOD dining facilities, and weird things like pools.
Interesting data and thank you for sharing.
I would like to interject a bit of opinion on counting a personal GHG number for the trip.
There is little/no data on passenger GHG for a container ship because being a passenger isn't the purpose of them. The ability to be a passenger on one is simply a small revenue stream for the owner on a trip that was going to happen anyway. The same goes for being a passenger on a repositioning cruise. The ship is going to go from Point A to Point B regardless if you are on it or not. You're effectively hitchhiking. Yes, there are GHG to account for regarding your energy usage, food consumption, etc on the ship, but it is likely negligible if even measurable in the first place.
Taking a cruise/flight is creating a demand for that cruise/flight, thus GHG counts towards you. Jumping on a cruise/flight that has a spare seat and happens to be going in the direction you need to go, isn't creating a demand as it's the people that booked the actual cruise/flight that are causing the repositioning. I would count my GHG if I booked a taxi, I wouldn't count it if I stuck out my thumb and jumped into the passenger seat of an 18 wheeler going the same direction as I didn't order the products being shipped, I just went along for the ride with a near-zero difference in GHG whether I was there or not.