The air-to-sea frontier: which perishables cross next

In 1901 a ship called the Port Morant left Jamaica for Britain carrying 23,000 stems of bananas and a carbon-dioxide machine running in the hold. It worked. A fruit that had mostly arrived spoiled, when it arrived at all, became within a generation an everyday thing on a British stall — and the trade learned a lesson it has been relearning ever since. When a perishable can survive the sea, everything about it changes: the price, the volume, the buyers, the countries that can grow it for you.

Your exporter has just offered you asparagus by sea. The saving is enormous. Nobody in the chain will tell you plainly whether it will still be good asparagus when it arrives.

That is the Port Morant question, asked again about a different crop. It has a general form, and it is the same one for every commodity that has ever made the switch: does the fruit’s quality life exceed the door-to-door transit, with enough days left over to survive the voyage going wrong? Where it does, the commodity converts from air to sea, and freight stops being the number that sets the retail price. The trade then reorganises around the lower cost — bigger consignments, longer planning horizons, different varieties, buyers who could never afford the fruit before. Where it does not, it flies, and the market stays small and expensive.

The line between those two states is the air-to-sea frontier, and it moves. Controlled atmosphere, faster services, better packaging and better postharvest chemistry push it outward every few years, so a commodity that could not cross a decade ago crosses routinely now. This page is where the line runs today. We revise it once a year, because it will not stay there.

What is actually at stake #

Airfreighting produce runs in the region of €2–4 a kilo. The same kilo in a deep-sea reefer is nearer €0.20–0.40. Both numbers move, and reefer rates move the most: a box rate can climb by half in a quarter when vessels are going the long way round, and the gap narrows while it does.

The ratio matters more than either number. A conversion does more than cut the freight cost. It pushes freight down the list of costs that decide the price. Fruit that flew was priced as a luxury because it had to be, and the same fruit on water competes on the shelf with things grown a few hundred kilometres away. That is why a crossing is a commercial event rather than a logistics one, and why exporters chase it for years before the fruit is ready to make it.

The carbon figures point the same way, and the gap is wider. A deep-sea container moves a tonne of cargo one kilometre for something like 10 to 20 grams of CO₂e. Long-haul air freight does the same work for around 600 grams, and the range runs past 1,000. That is thirty to sixty times more, for the same box of fruit.

So the two arguments for crossing point the same way, which is not usually how sustainability and cost behave. It is also why the pressure to convert now tends to come from a retailer’s reporting obligations rather than its buying desk. It also makes an air-to-sea conversion the biggest emissions cut an importer can make, well ahead of sourcing closer to home. Food miles are misleading is where we set that argument out in full. One caveat: a reefer burns considerably more than a dry box, so the gap for perishables is narrower than the headline factor suggests. It is smaller than thirty times, and still very large.

How the line has moved #

A frontier is easier to see looking back. Filled dots are commodities that crossed, hollow ones the technologies that let them, and a bar is a conversion that took years rather than a season.

1901Bananas1966CA in transit1980–2008Avocados2012–22Asparagus2025Blueberries

1. 1901 Bananas — the first crossing

   The Port Morant carries 23,000 stems from Jamaica to Britain in a hold cooled by a carbon dioxide machine. Everything since is a variation on what that voyage proved: pick the fruit unripe, hold it still, and ripen it at the far end.

2. 1966 Controlled atmosphere leaves the warehouse

   Whirlpool spins out TransFRESH to put its Tectrol atmosphere system into moving vehicles rather than cold stores. The technology that opens the long lanes arrives decades before most of the fruit that needs it.

3. 1980s → 2008 Avocados take a quarter of a century

   Mexican exports into the Japanese market took off in the 1980s and all of that cargo flew. The same trade was moving in refrigerated containers by 2008. A crossing is rarely an event; this one took twenty-odd years, and it later carried Peruvian avocados into Europe in about twenty days under CA.

4. 2002–03 1-MCP reaches fresh produce

   American regulators clear it for fruit and vegetables in July 2002, and SmartFresh launches commercially across the main pome-fruit exporters the following year. Ethylene-driven fruit gets a pause button that works inside a box.

5. 2012 → 2022 Asparagus crosses while nobody is looking

   Peru ships roughly a tenth of its asparagus by sea at the start of that decade and roughly three quarters by the end of it. The commodity most often named as the one that could never sail is the clearest proof that the frontier moves.

6. 2018 Cherries get their own service

   Chile opens the Cherry Express, a direct run to China that takes roughly thirty days down to about twenty-three. The textbook unforgiving fruit gets a lane — one that only works when nothing slips.

7. 2025 Blueberries, and the line running backwards

   About 97% of Peruvian blueberries now move by sea. At the same time air volumes are climbing again for the premium early window — proof that a crossing is a commercial choice and can go back the other way.

8. Now On the line

   Mangoes and papaya on the long lanes, lychees in perpetual trial, passion fruit where the cultivar allows, fresh figs, herbs, and the far edge of the berry lanes. Each is waiting on days that do not exist yet.

9. Probably never Cut flowers on most lanes, delicate leafy produce

   The clock is not close, nothing in prospect closes it, and for some of it nobody in the chain wants the cheaper freight badly enough to try.

The four questions that decide it #

Score a commodity against a specific lane, never in the abstract — Peru to Rotterdam and Peru to Shanghai are different answers about the same box of fruit. Four things decide it, and they are not equally weighted. The first is a hard constraint. The other three are economics, and economics can wait for next season.

The clock #

Quality life is the number to use, and it is not shelf life. Shelf life is how long the fruit stays sellable. Quality life is how long it stays good: firm, coloured, tasting of itself. Quality life ends first. If you plan on shelf life, the cargo can arrive technically fine and still be rejected against a retail specification.

Compare it against door-to-door transit, not port to port. Conversions are lost at the ends: two days on a quay waiting for a plug, four days of inspection and cold treatment on arrival, a road leg north from Rotterdam that nobody counted. The sea leg is the part everyone quotes and the part that varies least.

The buffer #

A commodity does not cross when quality life equals transit. It crosses when quality life exceeds transit by enough days that a normal bad week does not write off the consignment.

That slack is the whole difference between a lane that works and a lane that works until it doesn’t. A missed connection at a transshipment hub, a berth queue, a canal running dry: on a mature short lane these are annoyances. On a lane opened by spending the fruit’s last spare days on the distance, any one of them can cost the whole consignment. Frontier lanes are, by construction, the lanes with no margin for error — see transit time is shelf life for what those days cost.

Value density #

Price per kilo decides whether anyone bothers. Very high-value fruit absorbs air freight comfortably and feels no pressure to move. Cargo worth less than the airway bill never flew at all, so there is nothing to convert. The frontier matters in the middle: fruit expensive enough to have justified flying, cheap enough that the freight bill is what caps the market.

Value density also sets what a failure costs. A high-value, low-volume container on a narrow temperature margin (dragon fruit out of Ecuador is the clean example) turns a drifted setpoint from a quality complaint into a serious loss.

The technology trajectory #

A crossing is usually the last few days arriving, not a breakthrough. Controlled atmosphere matters most, because it can open a lane outright. Modified atmosphere liners do a cheaper version for fruit that only needs a little. 1-MCP blocks the ripening signal chemically, which is what makes several managed-ripening programmes survivable at distance. Faster services and fewer transshipments work on the transit instead, and take days off it.

So: a commodity with live trials and a technology still improving is a candidate. One where the postharvest science has not moved in twenty years is where it will stay.

Where each commodity sits today #

Already crossed #

Bananas set the template and avocados proved it generalised. Both are now so completely sea trades that nobody thinks of them as conversions at all, which is what a finished crossing looks like.

Blueberries went further than most people realise: nearly all of the Chile and Peru crop now sails. The interesting part is the fruit going back the other way. Air volumes are rising again for the premium early window, because early-season prices are high enough to pay for the flight. A crossing is a commercial choice, and it can be reversed.

Asparagus is the one worth arguing about. It is the commodity most often held up as proof that some things cannot sail, and it sailed. No single breakthrough did it. It was the slow build-up of small gains that this framework predicts — better packaging, faster and more reliable services, lanes short enough that the fruit’s few days were enough. But the lane matters, as it always does in this framework. That success is on the short run into North America, where about 86% of the fruit now sails. On the long haul the answer is reversed: roughly 71% of the European volume and nearly 87% of the Asian volume still flies. Same fruit, three different answers, because the lane is part of the question.

Two smaller crossings are recent enough that you can see how a crossing happens. Dragon fruit moved to sea as growers learned what it tolerates, out of Ecuador into Europe and out of Vietnam into Hong Kong. Goldenberries crossed because the husk does the preservation work for free. The fruit’s own structure can do the work of technology, so not every frontier is pushed by equipment.

Crossing now #

Cherries from Chile to mainland China is the live one, and the most instructive. This is the textbook unforgiving commodity, landing on the tightest deadline in the calendar (Chinese New Year), and a share of it sails anyway. It works because the fruit got a service of its own: the Cherry Express, running since 2018, took roughly thirty days down to about twenty-three. It needs a very fast lane, a very short window and everything going right. That is what a crossing looks like from the inside.

Mangoes are the conversion most people are watching next. Much of the trade already sails on shorter lanes; what is still undecided is premium fruit on the long ones, where variety and cultivar matter more than the shipping does. Papaya is on the same list for the same reason, and lychees belong here too, because the trials keep happening without giving a clear answer.

The frontier #

Fresh herbs are the clearest case. They have the clock asparagus used to have, and a fraction of the money behind solving it. That combination keeps a commodity where it is.

Berries on ultra-long lanes belong here too: not strawberries or raspberries, which are not close, but the marginal blueberry lanes at the far edge of what the converted trade already does. Passion fruit is close enough that the answer differs by origin and cultivar.

Figs get named as a crossing candidate more often than the fruit deserves. Fresh figs bruise where they are touched, split when the humidity moves, and have no storage tail at all, which is why almost all of the long-haul trade still flies and the European volume comes by road from Turkey. Sea trials exist. Treat them as trials.

Probably never #

Cut flowers on most lanes, and the most delicate leafy produce. The clock is not close and nothing in prospect closes it. There is a second reason that optimistic versions of this page leave out: for some of it the value density is high enough that nobody in the chain wants the cheaper freight.

The other three tiers are only worth reading because this one exists. A page that says everything crosses eventually has predicted nothing.

What crossing costs you afterwards #

The conversion story usually leaves this part out. Every one of these lanes works because the fruit arrives with just enough life left, which means the buffer was spent on the distance. The technology that opened the lane took away the slack that made delays survivable. You do not get both.

So a programme on a converted lane behaves differently from the air programme it replaced. Air was expensive and predictable. Sea is cheap and variable, and the variability now hits fruit with no spare days. On citrus you would absorb a four-day slip without noticing. On a frontier lane it is the difference between fruit that meets the spec and fruit that gets marked down on arrival.

You cannot prevent the slip. What you can change is when you learn about it — day two of the voyage, when ripening slots can still be moved and customers re-sequenced, or on the quay, when the only decision left is who takes the discount. On a lane the fruit only just survives, that timing decides most of the commercial outcome, and it is the job Trackberry does on the containers you already have on the water.

FAQ #

Which fruits have moved from air freight to sea freight? #

Bananas were the first, in 1901. Mexican avocados to Japan flew entirely in the 1980s and were going by reefer container by 2008, asparagus went from roughly a tenth of Peru’s crop by sea to roughly three quarters between 2012 and 2022, and blueberries are now almost entirely a sea trade. Chilean cherries have had a dedicated express service to China since 2018. Premium mangoes and papaya on the long lanes are the ones being trialled hardest now.

Can asparagus be shipped by sea? #

Yes, and most of it already is: roughly three quarters of Peruvian asparagus travels by sea, against about a tenth a decade earlier. The lane matters more here than the headline figure does. Into the United States about 86% sails, on short, fast runs with controlled atmosphere and no slack for delay. Into Europe the answer is reversed: about 71% still flies, and into Asia nearly 87% does, because there the fruit’s few days run out before the voyage does.

What is the difference between quality life and shelf life? #

Shelf life is how long the fruit remains sellable; quality life is how long it remains good: firm, coloured and tasting right. Quality life ends first, sometimes by a wide margin, and it is the number a sea programme has to be built on. Planning against shelf life produces cargo that arrives technically saleable and is rejected against a retail specification.

Does controlled atmosphere let any commodity travel by sea? #

No. CA slows respiration and buys real days on fruit that responds to it, and several trades exist only because of it. But the gas recipe is commodity-specific, some produce is damaged rather than helped by a low-oxygen atmosphere, and no atmosphere stops the clock — it only slows it.

How much cheaper is sea freight than air freight for produce? #

Roughly an order of magnitude per kilo, though both ends move with fuel, season and lane. The important consequence is not the saving on a given consignment but that freight stops being the cost that sets the shelf price, which changes who can buy the fruit at all.