Neptune Atlas
Draft survey: weighing cargo by displacement
A draft survey does not weigh the cargo. It weighs the ship, the ship is loaded, the ship is weighed again, and the difference is taken to be the cargo. Everything that makes the method useful, and everything that makes it a source of argument, follows from that: the quantity is never measured directly. It is worked out from how much deeper the hull sits in the water.
The method
The surveyor reads the draught (the depth of the hull below the waterline) at six marks: forward, midships and aft, on both sides. The surveyor then takes the mean. These readings must be corrected before they are used, because a ship is not a rigid box sitting level:
- Trim is the difference between the forward and aft draught. It needs a correction because the draught marks are not exactly at the ends of the ship's measured length, and the centre of flotation is not exactly in the middle.
- Hog and sag describe the hull bending upwards or downwards in the middle. This is why the midships readings are taken at all. A "mean of means" gives a better answer than a simple average of the forward and aft readings.
- Density. The ship's hydrostatic tables are worked out for a standard water density. So the surveyor takes a sample of the dock water and corrects the displacement to the density the ship is actually floating in.
The corrected mean draught is looked up in the ship's hydrostatic tables, which give the displacement, meaning the total weight of the ship and everything on it. From that figure the surveyor subtracts everything on board that is not cargo: ballast water, fuel, fresh water, lubricating oil, stores, and the constant. The constant is one figure for all the weight the ship carries that nobody lists separately. What is left is the empty ship plus the cargo, and the difference between the first and the final survey is the cargo figure.
Where the error comes from
The West of England P&I Club's loss prevention guidance is direct about how precise the method is: draught surveys are not an exact science, and the calculated cargo figure may typically be off by between 0.5 and 1.0 per cent. On a Panamax cargo, one per cent is worth enough money to start a dispute.
The Club's worked example shows how much a small error matters. An error of two centimetres in the mean draught changes the displacement by about 68 tonnes on a Handysize and by about 300 tonnes on a Capesize. Two centimetres is not a careless reading. It can come from a small swell, a rusty mark, or a reading taken from the quay instead of from a boat.
| Source of error | Why it causes error |
|---|---|
| Reading the marks | Swell, waves from passing boats, marine growth and paint over the numbers all get in the way. Reading from a small boat is better than reading from the deck |
| Water density | In a river berth, fresh water often sits on top of salt water. One sample there measures only one layer, not all the water the ship floats in |
| Ballast soundings | Ballast is the largest single deduction, and a tank measured while the ship is tilted is the easiest figure to get wrong |
| The constant | Stores nobody declared, mud in the ballast tanks and built-up sludge. If the constant differs between the two surveys, the whole difference appears in the cargo figure |
| Squat and moorings | A ship that is moving at the berth, or pulling hard on its mooring lines against a current, is not sitting at its true resting draught |
Two honest figures that disagree
The shore has its own figure, from a scale on the conveyor belt, a weighbridge for the trucks, or a count of units. The ship's figure and the shore's figure will not match, and neither one is automatically wrong. They measure different things by different methods, each with its own margin of error, and a difference of a fraction of a per cent is normal in the trade. It is not evidence that cargo was lost.
What matters commercially is which figure the contract treats as binding. Freight is usually earned on the bill of lading quantity, and the bill of lading is normally signed on the shore figure. If the master believes that figure is higher than what really came on board, the master does not simply refuse to sign. Instead, the master issues a letter of protest that records the ship's own figure and protects the ship's position. A protest costs nothing at the time, and it is often the only document written at the time that still exists when a shortage claim arrives eighteen months later.
Why the discharge survey is the one that hurts
A shortage claim compares the figure at the discharge port with the bill of lading figure. Suppose the bill at the loading port was signed on a generous shore figure, and the discharge port then carries out a careful draft survey. The ship will appear to have lost cargo it never carried. Nothing was stolen and nothing was spilled; two different measuring systems were compared as if they were the same instrument.
All the defences are documents, and all are cheap if they are prepared at the time: a draft survey at loading, a protest against any figure the master disputes, the density record, the ballast soundings and the surveyor's working sheets. The parties who lose these arguments are usually the ones who had the right answer but cannot prove they had it at the time.
The clauses that decide it
Because the method has a known margin of error, the contract is where the allowed tolerance is set. A sale contract often sets a franchise, which is a percentage within which no claim can be made. A charterparty may say whose survey counts, whether a joint survey is required, and who pays the surveyor. These terms are negotiated, and they are worth reading before the cargo moves, not after.
The standard procedure most often referred to is the United Nations Economic Commission for Europe code for draught surveys of coal cargoes. It was written for coal, but it is used far more widely because no better public standard exists. It was created because standard forms and a standard order of steps remove the differences that would otherwise turn into business disputes.
The one thing to carry away
A draft survey figure is worked out from a displacement, and it is only as good as the deductions made from it. When a quantity is disputed, the question is almost never whether someone misread a mark. The question is whether the ballast, the constant and the density were treated the same way at both ends of the subtraction.
References
- West of England P&I Club, Loss Prevention Bulletin: Inaccuracies in Draught Surveys. westpandi.com
- Britannia P&I, Loss prevention guidance: preventing bulk shortage claims, 2025. britanniapandi.com
- United Nations Economic Commission for Europe, Code of Uniform Standards and Procedures for the Performance of Draught Surveys of Coal Cargoes, 1992.
- IMO, International Convention on Load Lines, 1966. imo.org
- IMO, International Maritime Solid Bulk Cargoes (IMSBC) Code. imo.org
Related
Neptune Atlas
Chartering software for a broking desk, covering the whole cycle: the world merchant fleet, position lists, cargo enquiries, fixtures, laytime and demurrage, and who to write to. Every paid plan starts with 7 free days. A card is needed to start them, and cancelling before they end costs nothing.
Try 7 days free Ship chartering software for brokers What each plan includes