Mini-grid or national grid: reading light signatures on a night-light map

A disbursement note says "energised." A field report two years later says something different, or nothing at all. In between, you're left guessing whether a settlement lit up because a developer switched on a diesel-hybrid mini-grid, or because the utility's feeder finally reached it. Both show up the same way in a results framework: a dot that was dark is now lit. They do not show up the same way on a monthly light read, if you know what to look for.
Two different events, one visual cue
A mini-grid and a national grid connection are different pieces of infrastructure, built by different actors, on different timelines. The indicator most portfolio teams rely on between surveys, "is there light where there wasn't before," collapses that distinction. Pulling them apart means tracking how the light spreads across several monthly reads, not reading a single lit-or-dark snapshot.
An isolated mini-grid is, by design, bounded. Someone installed a generation asset (solar-diesel, usually) sized to one settlement or a small cluster, and wired it to the households within reach of that plant. There is no feeder running past it to anywhere else. On a monthly panchromatic read, that shows up as a compact patch of new light that appears over one or two readings, climbs in intensity as households connect over the following months, and then holds roughly steady. The area immediately around it stays dark, because there was never a line running through it in the first place.
National grid arrival behaves differently because the infrastructure behaves differently. A feeder extension follows a physical alignment, typically a road or a surveyed right-of-way, from a substation toward the project settlement. On a sequence of monthly reads, you tend to see light appear not just at the destination but along the corridor the line takes to get there: a string of smaller settlements picking up light in the months before or around when the target community does, roughly along a line rather than scattered at random. The destination settlement's own light often arrives in steps too, since transformer energisation and household hookups don't happen on the same day.
What to look for across a few months of reads
Three things tend to separate the two patterns, once you're looking at several months of imagery side by side rather than a single snapshot:
Geometry of spread. A mini-grid stays a dot. A grid extension traces a line, because it follows the feeder route and lights up neighboring points along it, not just the one you're tracking.
Timing relative to neighbors. If the settlement next to your project site, one with no recorded mini-grid of its own, lights up in the same window, that's a strong hint a line passed through rather than a standalone plant switching on.
Pace of intensity change. Mini-grid intensity tends to ramp as a known, fixed set of households connects to a fixed-capacity plant, then flattens. Grid-connected intensity can keep climbing for longer as more households tap in and as commercial load (shops, cold storage, milling) follows the connection, since capacity isn't capped by what one plant can produce.
None of this confirms the technology on its own. At ≥250 m resolution with no thermal band, a monthly panchromatic read tells you where light is spreading and how fast, not what's generating it. What it does well is flag the pattern worth a field visit: a dot that stayed a dot is one story, a line that grew along a corridor is another, and you want to know which one you're looking at before your next disbursement decision rather than at the next full survey.
A monthly night-light read over your programme area sits in that exact gap, flagging which settlements are worth a closer look before the formal survey numbers come in.
If your team is tracking grid extension against financed milestones, it's worth seeing what a few months of reads over your own programme area would actually show.