After this earthquake, where did the ground shift and the lights go out?

A big earthquake leaves two fingerprints satellites can read from orbit: the **ground physically moves**, and **infrastructure goes dark**. Together they map where the shaking did the most.

A big earthquake leaves two fingerprints satellites can read from orbit: the ground physically moves, and infrastructure goes dark. Together they map where the shaking did the most.

What you can answer

Ground deformation, to the centimetre. InSAR compares the phase of two Sentinel-1 radar passes before and after the quake; the difference (an interferogram) reveals how the surface shifted — the signature of the rupture.
Where the lights went out — Black Marble nightlights before vs after flag neighbourhoods that lost power, a fast proxy for damage and disruption.
Visible destruction — clear-sky optical pre/post imagery confirms collapsed structures.

What you can NOT answer (easily)

Building-by-building damage — EO gives area patterns; precise structural assessment needs higher-res tasking or ground survey.
InSAR through dense vegetation or with too much change — decorrelation degrades the phase; works best on bare/urban terrain.
Casualties — these are physical-impact proxies, not human-impact counts.

How you’d approach it

Form an InSAR pair across the event for the deformation field, difference Black Marble nightlights for outages, and overlay both on population/buildings to triage. Advanced (InSAR processing is involved). Supports the Respond phase of the NASA Disasters program.

⚲ How a scientist answers this

Parameters

Ground deformation from a Sentinel-1 InSAR pair: interferometric phase difference between one pre-event and one post-event SAR acquisition (C-band, ~5.6 cm wavelength), unwrapped to line-of-sight displacement in centimetres (one fringe ≈ 2.8 cm half-wavelength). Power-loss proxy from Black Marble VNP46A2 moonlight-corrected, gap-filled nightlight radiance (nW·cm⁻²·sr⁻¹) pre vs post; optical Sentinel-2/Planet pre/post for visible collapse.

Method

Form the co-event interferogram, remove flat-Earth and topographic phase (DEM-corrected), unwrap to recover the line-of-sight deformation field around the rupture; separately difference post-minus-pre Black Marble radiance to flag neighbourhoods that went dark, then overlay deformation and outage on population/building layers to triage shaking impact.

Validation

Check interferometric coherence and mask decorrelated (vegetated/heavily-changed) pixels; sanity-check the deformation field against the published fault model/aftershock distribution, and confirm nightlight drops are not artefacts of cloud, moon phase, or snow before attributing to outage.

In plain EnglishRadar measures how far the ground physically shifted (down to centimetres), night-light images show which neighbourhoods lost power, and clear-sky photos confirm collapsed buildings — together they map where the shaking hit hardest.

Make it yours → Set your event date and AOI, pick the bracketing Sentinel-1 acquisitions and the pre/post Black Marble nights in the notebook.

⌨ Run the core method · no login

The Before-After-Control-Impact vs a noise floor at the heart of this question — runnable on synthetic data, right here. The full earthaccess code template further down does it on real NASA data (needs an Earthdata login).

editable · runs in your browser

import numpy as np
rng = np.random.default_rng(1)
control_change = rng.normal(0, 0.5, 200)     # natural change at UNAFFECTED sites = the noise
effect = 0.9                                  # change at the impacted site (edit me; try 0.2)
sd = float(np.std(control_change)); floor = 0.5*sd
print(f"effect {effect:+.2f}   noise floor {floor:.2f}   SNR {abs(effect)/sd:.2f}")
print(">>>", "real signal — clears the floor" if abs(effect) > floor else "no clear change — below the floor")