Draught Survey

A ship floats at the depth where the weight of the water she pushes aside equals her own total weight. Her builders computed, for every draught, how much water her hull displaces at that depth — the hydrostatic tables, which are specific to that ship and travel with her. So a draught reading, via the tables, is a weight reading. The ship is a scale that was calibrated at the shipyard.

The survey uses this twice. Before loading, the surveyor determines the ship’s displacement — everything she weighs, empty of cargo. After loading, they determine it again. The difference is what was added, and if everything else aboard is accounted for, what was added is the cargo. The subtraction is doing the real work here, and it is where the survey’s accuracy is won or lost: the cargo figure is a difference between two large numbers, so an error in either propagates directly into the answer.

The visible part is reading the draught marks — painted scales at the bow, the stern, and amidships on both sides. Six readings, not one, because a ship is rarely level and never symmetrical in practice. The readings are combined to find the true mean draught, corrected for the ship’s trim and for the fact that her hull is not straight: a loaded hull sags in the middle or hogs upward, and the amidships reading is what reveals it.

Draught marks

Scales on the hull showing how deep she floats. Read at six points, because a single reading assumes a level, undeformed ship that does not exist.

Trim correction

Adjustment for the ship not floating level fore and aft. Necessary because the hydrostatic tables assume a reference condition she is not in.

Hog and sag correction

Adjustment for the hull bending under its own loading. The reason amidships readings are taken at all.

Water density

Measured on the day, at the ship. Displacement depends on what she is floating in, and dock water differs from the tables’ reference — an uncorrected assumption here is a large error.

Deductibles

Ballast, fuel, fresh water, stores, and everything else aboard that is not cargo. Each must be established, and each is a place the cargo figure can go wrong.

The deductibles are where surveys are actually contested. Ballast water is the largest and the most awkward: tanks must be verified, and a tank thought to be empty that is not empty transfers directly into the cargo figure. Bunkers are measured by sounding. Every one of these is a measurement with its own uncertainty, and the survey’s result inherits all of them. This is why the surveyor’s attention goes to the tanks rather than to the marks — the marks are the easy part.

It is normal for a draught survey figure, a shore scale figure, and an outturn figure to differ, and understanding why is more useful than treating any of them as the truth. They are measuring different things by different means at different times: a conveyor scale measures what crossed it; a draught survey measures what the ship’s displacement changed by; an outturn measures what was recovered at discharge. None is wrong, and none is exact.

• Sea state and swell, which make the marks unreadable except as an average over motion
• Marks obscured, fouled, or repainted inaccurately — the scale itself can be wrong
• Ballast not fully accounted for, the single most common source of a material error
• Water density taken from the wrong place, or in a berth where fresh and salt water are layered
• Hull deformation beyond what the standard corrections handle
• Hydrostatic tables that no longer reflect a hull modified, repaired, or aged since they were computed
• The survey being a difference between two large numbers, so small errors in either end up whole in the answer
• Cargo moisture changing between load and discharge, which changes the mass without anyone losing anything

It is fair to ask why an industry that can track a container in real time still determines its largest consignments by looking at paint on a hull. The answer is that nothing better is available at the scale. Weighing a shipload directly would require weighing equipment of a size nobody has; shore scales exist but measure the shore’s side of the transaction and are the terminal’s instrument rather than a neutral one; and the ship is present at both ends by definition. The draught survey is neutral, needs no infrastructure, works at any berth in the world, and is understood by everyone in the trade.

It is also, importantly, a survey — carried out by a person exercising judgement, usually engaged by a party with an interest, and often attended by surveyors for the other side. Its authority comes from the method being transparent and from both parties being able to watch it being applied. That is a different kind of reliability from instrumental precision, and for a commodity trade that has resolved quantity this way for a very long time, it has proved sufficient.