Tonnage

What a ship can actually lift

Published 23 August 2026 · updated 22 September 2026

A charterer wants 55,000 tonnes of iron ore. An owner offers a 58,000 deadweight Supramax. Those two numbers look compatible and they are not, because deadweight is not cargo and the port is usually shallower than the ship. Working out what she can really lift takes four subtractions and a comparison, and the answer is often several thousand tonnes below the figure on the circular.

The arithmetic, written out

Supramax, iron ore. Illustrative figures, consistent with each other
and not a description of any particular ship.

  Deadweight at her summer marks                     58,000
  Less bunkers for the voyage                         1,400
  Less fresh water                                      200
  Less constants                                        290
                                                   --------
  Deadweight cargo capacity                          56,110

  Her summer marks                                    12.80 m
  The load berth can work                             12.00 m
  Short by                                             0.80 m = 80 cm
  TPC 52.8, so 80 cm of draft is                      4,224 mt

  Cargo she can lift THERE                           51,886

  Wanted                                             55,000
  Short by                                            3,114

Nothing in that is difficult. It is simply four numbers that are not on the circular and one comparison the circular cannot make, and it decides whether there is a fixture.

Deadweight is a total, not an allowance

Deadweight is everything she can carry when she is down to her marks: cargo, and also fuel, water, stores, spares and crew. Three things come off it before a tonne of cargo goes on.

• Bunkers. These scale with the voyage rather than with the ship: an ocean passage carries well over a thousand tonnes of fuel and a coastal run a fraction of it. It is the one figure the voyage estimate already knows, which is why the estimate and the intake calculation belong in the same conversation.
• Fresh water. A couple of hundred tonnes, unremarkable, and worth a couple of hundred tonnes of freight.
• Constants. Everything aboard that is not fuel, water or cargo: stores, spares, paint, sludge, mud in the ballast tanks, unpumpable water, the crew and what they own. A few hundred tonnes on a bulk carrier.

Deadweight less those three is the figure a broker wants and it has its own name: deadweight cargo capacity. Constants are the line people argue about, and the reason is that nobody measures them directly. The constant is what is left over when the lightship and the cargo are subtracted from the displacement, which means it silently absorbs the entire error of the draft survey that produced it. A careful survey is accurate to something like half to one per cent, and on a Panamax cargo that is more tonnes of pure measurement noise than the whole constant. Which is also the useful diagnostic: if the constant comes out materially different at load and discharge, the answer is almost never that the ship gained weight. It is that one of the two surveys is wrong.

Three ceilings, and the answer is the lowest

There are exactly three things that stop cargo going aboard, and every intake question is working out which of them binds first.

• Weight. Deadweight cargo capacity, as above. The ceiling everybody quotes, and frequently not the one that decides.
• Draft. What the shallowest water on the voyage will let her draw. This binds most part cargoes and is invisible on a deadweight figure.
• Volume. Whether the holds fill first. This binds on everything light.

Draft, and the number that converts it into tonnes

A port that can work 12.0 metres does not care that the ship is marked for 12.8. Every centimetre she cannot use costs her cargo, and the conversion rate is TPC, tonnes per centimetre immersion: the weight it takes to push her one centimetre deeper into the water.

TPC is a property of the hull at that draft and it comes off her capacity plan. It is the area of her waterplane, so where the plan is not to hand it can be approximated from length and breadth to within a few per cent, which is fine for deciding whether to offer and not fine for declaring a quantity. It also changes with draft, so read it at the draft in question.

The consequence is worth stating baldly. On a Supramax, TPC is in the low fifties, so one metre of draft restriction is over five thousand tonnes of cargo. There is no rate negotiation on a Supramax cargo that moves five thousand tonnes, which is why a broker who checks the deadweight and not the berth has checked the smaller number.

The season moves the mark itself

A ship’s load line is not one line. She has summer, winter and tropical marks, and the convention that sets them is arithmetic: the winter mark sits below the summer mark by one forty-eighth of her summer draft, and the tropical mark sits the same distance above it. On the Supramax above, one forty-eighth of 12.8 metres is close to 27 centimetres, which at her TPC is about fourteen hundred tonnes of cargo that exists in July and does not exist in January.

Which mark applies is decided by where she is and what month it is, under the load line zones and seasonal periods, and the North Atlantic in winter is the case everybody meets. It is a fact about the calendar and the chart, not about the ship, it moves the intake by more than most rate arguments do, and it appears nowhere on a deadweight figure. Brokers who trade the North Atlantic in winter know it cold. Everybody else forgets.

Fresh water gives some of it back

Load lines are cut for salt water. In fresher water a ship floats deeper for the same weight, so she is entitled to load past her marks by the difference: the fresh water allowance, scaled by how far short of salt the dock water actually is. At a river berth this is not a rounding. On the ship above it is around a third of a metre in fully fresh water, which is over a thousand tonnes of cargo.

The framing that keeps it straight: the allowance is not extra cargo. Loading to the fresh water mark in a river gets her to the same deadweight she would have reached on her summer mark in the sea, and a broker told the load berth is fresh who does nothing with it has simply left that cargo on the quay.

Volume, and cargoes that cube out

The third ceiling has nothing to do with weight. Every cargo occupies a volume per tonne, its stowage factor. Divide her hold capacity by it and that is how many tonnes will physically go in.

A cargo that fills the holds before the ship reaches her marks has cubed out. One that puts her on her marks with the holds part empty has deadweighted out. Ore does the second and grain usually does the first.

Take the Supramax from the worked example, at an unrestricted berth this time, with the grain capacity conventional for the class, about 71,000 cubic metres. On soyabeans at 1.35 the holds take about 52,600 tonnes, so she cubes out with thousands of tonnes of deadweight she is not allowed to use. On iron ore she is on her marks with the holds barely a third full. Same ship, same berth, and the only thing that changed was the commodity.

Three refinements matter and each of them has cost somebody a stem.

• Grain capacity is not bale capacity. Grain is measured to the full internal limits of the space, out to the frames and up into the hatch coamings, and is what a cargo that flows into every corner can use. Bale is measured to the inside of the frames or cargo battens and is what a bagged, baled or packaged cargo can use. The conventional gap is seven to ten per cent, though that figure comes from framed multi-deck tonnage and a modern smooth-sided bulk carrier is much closer than that. Use the wrong one and the answer is out by the difference.
• Cubic feet are not cubic feet. Half the market still quotes stowage in cubic feet, and the traditional quote is per long ton rather than per tonne. One conversion is about 35.3 and the other about 35.9, and picking the wrong one moves a Supramax stem by hundreds of tonnes in the direction nobody notices until she is alongside.
• The published figure is a starting point, not a fact about this parcel. The shipper’s declared figure governs, moisture changes it, and the same commodity out of two ports is not the same number.

The one number worth carrying per ship

Divide her grain capacity by her deadweight cargo capacity and the answer is the stowage factor at which she cubes out and deadweights out in the same moment. On the Supramax above it is about 1.27. Anything lighter than that fills her holds first; anything heavier puts her on her marks first. It is a fixed property of the hull and it settles the volume question for every cargo she is offered.

It also falls as ships get bigger, because a geared Supramax is a cubier ship per tonne of deadweight than a Capesize. That is most of the reason grain moves on Supramaxes and Panamaxes and iron ore moves on Capes. And it is why coal is the cargo that decides arguments: coal is quoted anywhere from around 1.1 out of Australia to over 1.5 out of West Africa, which straddles that number, so the same ship deadweights out on one origin and cubes out on another. Never plan a coal stem on a generic figure.

Which port governs

The restriction that binds is the shallowest water she has to work in, and it is not always the load port. A broker who checked only the load end can fix a cargo she cannot deliver. Tidal windows complicate it further: a berth’s figure may be reachable only for part of the day, and a ship waiting for water is a ship on laytime.

The Parana is the clearest case in the trade. On a given day the up-river berths at Rosario and San Lorenzo will take a ship deeper than the falling river will let her leave, and that gap is not an inconvenience, it is a business: she part-loads up river and completes at a deeper port downstream, picking up both the extra draft and the rise from fresh water into salt on the way. The same operation runs in reverse at the discharge end, where cargo is lightered out of a deep-laden ship so that she can enter. The Mississippi below New Orleans and the tidal iron ore berths of north-west Australia are the other two everybody meets, and in Australia what she may sail with depends on the height of tide on the day rather than on any published figure. All of it is real cost and real time, and it belongs in the estimate rather than in a footnote.

The number in the charterparty is not a single number

Because the intake cannot be known exactly in advance, quantity is fixed with a tolerance and an option: a percentage either side of the stated figure, declared by one named party before loading. On a dry cargo enquiry that is MOLOO, more or less in owner’s option, or MOLCHOPT where the charterer declares. Which it is gets negotiated, because it is the right to move the freight the tolerance is worth. Behind it sits the obligation to load a full and complete cargo, with freight earned on the quantity actually loaded rather than on the figure in the box, and deadfreight owed on any shortfall the charterer failed to supply. All of that machinery exists because a single number would be a promise nobody could keep.

None of this is advanced. It is four subtractions, a multiplication and picking the smallest of three. What makes it worth doing properly is that it happens at the front of a negotiation, and the party who has done it is the one who knows whether the enquiry on the screen is a cargo for this ship or an hour about to be wasted.

References

• Britannia P&I, Draft survey calculation on a ship: practical guidance, 2023. britanniapandi.com
• IMO, International Convention on Load Lines, 1966. imo.org
• IMO, International Convention on Tonnage Measurement of Ships, 1969. imo.org
• IMO, International Maritime Solid Bulk Cargoes (IMSBC) Code. imo.org

Related