STS crane vs mobile harbor crane: telling them apart in satellite imagery

If you're building a crane inventory off imagery instead of site visits, the first sorting problem you hit isn't manufacturer, it's type. An STS crane and a mobile harbor crane can sit thirty meters apart on the same apron and look like the same gray blob at 0.5 m GSD until you know what to check. Here's what actually separates them from directly overhead.

Rail footprint and the portal structure

Ship-to-shore cranes run on rails set into the quay, and the rail gauge shows up as two long parallel lines running parallel to the berth, usually 15-18 m apart. The crane itself sits on a rigid portal frame straddling those rails, four corner legs, sometimes more on twin-lift machines. You'll see that footprint repeat identically down the quay if there's more than one STS on site, same spacing, same orientation, because they're all built to the same rail gauge.

A mobile harbor crane has no rails. It sits on a slewing ring over a crawler undercarriage or a set of rubber tires, and in plan view that reads as a compact, roughly circular or square base with none of the long parallel rail lines. It can be parked anywhere along the apron, nose into the quay or set back from it, and its position often isn't aligned with anything else nearby. A crane sitting slightly off the waterline edge, not locked to a rail line, is your first tell that it's mobile.

Boom shape and backreach

This is the part that trips people up on compressed or hazy imagery: the boom geometry.

STS cranes have a long horizontal boom extending out over the water, with a matching backreach boom over the landside rail. From directly above, this reads as a long straight line (sometimes with a slight upward kink where the boom articulates) running perpendicular to the quay, often 40-65 m out over the vessel berth. The A-frame or portal structure holding it up shows as a wide rectangular shadow block at the base.

A mobile harbor crane's boom is a luffing boom, a single angled arm that pivots up and down rather than running flat. In plan view it reads shorter, doesn't extend nearly as far waterside, and has no separate backreach boom on the landside, since the whole upperworks rotates on the slewing ring instead of splitting into waterside and landside sections. A stubby boom line relative to the base footprint, or one that points in a different direction between two image dates, marks a mobile unit rather than a rail-mounted STS.

Shadow length as a quick height check

Height differences compound the other signals. STS cranes on a modern container terminal commonly run well over 40 m to the boom tip in the stowed position, and their shadows at a low sun angle stretch out dramatically, often several times the length of the crane's own footprint. Mobile harbor cranes are shorter, and their shadows are noticeably more compact. Two cranes sitting side by side with one shadow twice the length of the other, at the same time of day, points to a real height difference, and it usually lines up with the rail-footprint and boom-shape checks you've already made.

None of this tells you the manufacturer by itself. Rail gauge, boom shape, and shadow length get you to the type bucket, STS versus mobile harbor versus yard gantry, before you start looking at counterweight housing shape, cab position, or trolley style for the maker-level call. Getting the type right first matters more than people give it credit for. Mis-bucket a mobile harbor crane as an STS and every downstream count, reach estimate, and competitive read for that terminal is wrong.

Doing this terminal by terminal across a whole market, by hand, in a browser tab with imagery pulled up next to a spreadsheet, adds up fast, and it's easy to miscount a busy apron. Crane Fleet Mapping runs this classification across every terminal in a market once a year and hands back a terminal-by-terminal inventory with type, maker attribution where it's visible, and a height read already done. If you're scoping a market crane inventory and want to skip the manual imagery triage, that's worth a look.