Notes
How to tell a reefer stack from a dry container stack in satellite imagery
If you're scanning a terminal apron on a fresh pass and trying to figure out which blocks are carrying reefer boxes, color isn't going to save you. Reefers ship in every paint scheme a line uses for dry stock. The tells are in layout and plug infrastructure, not box color.
What separates a reefer block from a dry stack
Reefer yards get built around power, not aesthetics. Look for blocks with wider aisle spacing than the dry stacks around them. That extra gap is there for genset trucks or for a worker to walk the row checking temperature readouts, and it shows up clearly at sub-0.5m resolution as a consistent lighter strip between rows where dry stacks would be packed tight.
Second cue: proximity to a substation or a row of power pedestals. Reefer blocks cluster near the yard's electrical infrastructure because running cable further than necessary isn't something a terminal operator wants to deal with. If you see a dense, regularly-spaced block sitting right up against a utility run along the perimeter or a fence line, that's usually your reefer zone.
Third, orientation. Reefer containers get stacked with their plug ends facing an access row, so the pattern of gaps repeats in a way dry stacks don't bother with. Dry stacks get packed for density first. Reefer stacks get packed for access first, and that trade-off is visible as soon as you compare block geometry side by side on the same apron.
What you can't reliably do from a single RGB pass is count individual plug connections or read condenser housings on container ends. At VHR resolution you can sometimes make out texture differences on the container ends facing the camera, a slight shadow or protrusion pattern, but calling that a reliable per-box count from one image is overselling what the imagery gives you. Block-level layout is the honest read. Box-level plug status isn't, at least not from one frame.
Why this matters for a trade-flow read
Reefer block density is a decent leading indicator for perishable and pharma trade specifically, which moves on its own calendar separate from general dry cargo. A citrus season ramping in the Southern Hemisphere, a produce surge ahead of a holiday, a pharma cold-chain contract starting up: all of these show up first as more boxes sitting in the reefer blocks at origin ports, well before any of it clears customs or gets reported in a trade release. If you're trying to get ahead of a perishables story, the dry-cargo count on the same terminal tells you almost nothing. The useful signal comes from watching the reefer rows on their own, often enough to catch the fill pattern changing week over week.
That's the part a single pass can't give you. One image tells you a reefer block exists and roughly how full it looks that day. It doesn't tell you whether that fullness is building, holding, or already rolling off toward a vessel call. For that you need the same blocks imaged on a daily cadence, stacked into a time series you can trend against prior weeks. That's closer to what Trade Flow Index is built to deliver: a running count of stack activity and vessel calls at named ports, pulled from daily high-res passes, so a change in a specific yard shows up as a line moving before it shows up in a port authority release weeks later.
A practical checklist for the next pass you review
- Wider, regular aisle spacing inside a block, relative to neighboring dry stacks
- Block sitting adjacent to visible power infrastructure or a pedestal row
- Consistent plug-end orientation across the block rather than dense random packing
- Don't trust color or box brand markings as a reefer indicator on their own
- Treat any single-day read as a snapshot, not a trend
None of this replaces a terminal's own EDI data on reefer plug occupancy, which you're not going to get as an outside analyst anyway. What you can get from repeated imagery is a reasonable proxy: is the reefer footprint at this port growing or shrinking week over week, independent of whatever the dry-cargo numbers are doing. For a desk trying to get a few days of lead time on a perishables or cold-chain trade flow, that proxy is often the only outside read available.
If you want that read running on a schedule instead of pulling single frames yourself, that's the gap Trade Flow Index is built to close.