Geography is the constraint

Longitude coverage

A direct-to-Earth station only sees a satellite when the pass crosses its sky, so a constellation's contact rate depends on how evenly stations are spread in longitude — not on how many exist. The longitude lens stripes the map into 24 one-hour bands and counts operational stations in each.

Most empty bands are open ocean, where the absence is physics rather than a market gap.

A 15° bin is narrower than the longitude a single station actually works: at mid-latitudes a low-Earth pass stays usable across roughly ±20–25°, so an empty band flanked by occupied ones is generally covered by its neighbours and only a run of consecutive empty bands is a real hole. The bins also say nothing about latitude — two stations one band apart but 60° apart north to south see quite different passes. Read the bands as a distribution, not as a coverage model.

The GEO arc is already covered

GEO feeder links are the opposite story. A GEO satellite sits at a fixed longitude, so a station's usefulness is fixed too: it can reach the slots where the satellite stays above a workable elevation. The GEO lens draws each operational GEO-capable station's reach as a fan down to the arc, taking 20° elevation as the floor — where fans overlap, more than one station can work the same slot.

Fewer independent sites than stations

Cloud cover is correlated over a few hundred kilometres, so two stations at the same site are one weather bet, not two. The sites lens collapses operational stations within 100 km of each other into a single marker, numbered by how many stations it holds.

Backup exists on the map, not on paper

A station has a weather backup only if a second site can take the same pass under different sky. That means a partner roughly 500 to 2,000 km away: far enough that one frontal system rarely covers both, close enough that a single LEO pass crosses both horizons. The weather lens draws every operational pair that qualifies, and marks the few an operator has actually documented.

Both distances are stated assumptions, not measurements: 500 km is a mid-range figure for how far cloud cover stays correlated, 2,000 km approximates where a low-Earth pass stops being mutually visible. Neither is isotropic, and the lens treats them as if they were. Cloud decorrelates over tens of kilometres in convective tropical air and over well beyond a thousand in a North Atlantic frontal system, so a single 500 km threshold is generous at low latitudes and tight in the storm tracks. Separation is also measured as plain great-circle distance, which ignores how a pair is oriented: a near-polar ground track runs north–south, so two sites on that axis share a pass quite differently from two the same distance apart east–west. See how weather is handled. A documented pair is drawn whatever its separation, because an operator's stated arrangement outranks our geometry. And an undocumented pair is possibility only: nothing here says those two operators have agreed, scheduled or ever tested a handover.

Hemispheres