Open engagements, and standing ground.

Colonias are unincorporated, which is the whole reason the data problem exists. A household outside any service boundary is also outside the record that would say it has no water line. The gap is not that nobody has looked. It is that looking has never been anyone's job.

Vida Water works in these communities directly and asked a question that sounds simple and is not: how many households in El Paso and Hudspeth Counties still have no water infrastructure, and where exactly are they? We are answering it from parcel records, structure detection in overhead imagery, and the service boundaries of every utility that publishes one, then differencing the three.

The second ask is separate and narrower. In Precinct 2, older buildings with aging plumbing may be leaching into the water they carry. No satellite can see leaching. What overhead imagery and parcel records can establish is building age and structure condition, which is a proxy, and saying so plainly is part of the deliverable rather than a caveat buried at the end of it.

On the morning of 26 August 2026 a flood came down the Lhende, crossed into Nepal at Rasuwagadhi, and struck settlements along roughly 72 kilometres of the Trishuli. The Trishuli rose about nine metres in thirty minutes. Seismometers on the other side of the world recorded the initiating collapse before most people in the valley knew anything had happened.

That sequence is the argument for this work in one paragraph. The event was observable. It was observed. It was observed by instruments whose output reached seismologists, and not the valley.

So the question we are working is not whether these floods can be seen. It is what the historical record already tells us about where they recur, and whether that record can be turned into something a district chapter can act on before the next one. We are cross-referencing ICIMOD's outburst database against flood events recorded within ten kilometres of a glacier since 2000, then reading the post-flood corridor for building destruction, population clustering, and damage to the roads that decide who can be reached.

PISFCC collects what people remember. A shoreline that used to be further out. A stand of trees that is gone. A storm that changed the shape of a coast in a single night. Those accounts carry the part of the record no instrument holds, which is what the change did to the people living through it.

What they do not carry is a measurement, and in the rooms where climate obligation is argued, an account without a measurement beside it is easier to set aside. PISFCC took a climate advisory opinion to the International Court of Justice and won it in July 2025. They know exactly what kind of evidence those rooms ask for.

The first three sites are in Vanuatu: Worasiviu on Pele Island, where the community has already built a sea wall against the erosion; Eton on the east coast of Efate; and Blacksand on the south coast, which is the most tractable of the three because slow shoreline retreat is precisely what a long optical time series resolves well. The fourth is Niue, where Cyclone Heta came through as a Category 5 in January 2004 and rewrote the north and west coasts. That one is a different problem: rapid onset, twenty-two years back, and the useful output is less a map of what happened than a trend and a projection that a ministry can put against a resource allocation.

The Thar is the counterexample that makes the case for reading more than one instrument. Look at vegetation indices alone and the desert appears to be greening, which it is. Look at gravimetry and the groundwater that greening is drawn from is being pumped out faster than the monsoon replaces it.

Either signal on its own supports a confident and wrong conclusion. Together they describe something specific: irrigated expansion into arid land, running on a store that is not being refilled. That is a timing problem, and timing problems are the kind a satellite record is actually good at.

Flaring is one of the few environmental signals that is genuinely well measured. VIIRS Nightfire has logged flare locations, temperatures, and radiant heat every twenty-four hours since 2012, and Nigeria publishes the coordinates through NOSDRA's tracker. The measurement is not the gap here.

The gap is that a flare coordinate is not an exposure estimate. Nobody routinely joins the nightly flare record to who lives within a kilometre of it, or to the spill record in the same creek, or to the cloud cover that decides which flares the instrument can see at all. Reading those together is unglamorous and it is the whole contribution.