Salar in Motion

What changed on the surface of the Salar del Hombre Muerto (Argentina) between June and October 2026, seen by the NISAR L-band radar.

Date

–km² brighter than 10 June
–km² darker than 10 June
–km² very dark surface

Bars: area whose HH backscatter rose (blue) or fell (red) by more than 3 dB against 10 June. The swing between dates shows that part of the change is seasonal.

The date selector updates the figures, the chart, the change map and the "very dark surface" layer.

Layers

Overlay opacity

Change zones

#Directionkm²dB

Click a row to go to the zone. Sorted by area.

Method
  • NISAR GCOV L-band scenes, orbit 118 ascending, frame 165, five dates.
  • Area of about 61 by 62 km at 20 m pixels; 5 x 5 speckle filter.
  • Change = HH backscatter of each date minus HH of 10 June 2026, in dB.
  • Zones: change above 3 dB between the first and the last date, and area of at least 0.1 km².
  • Persistent change: same direction in at least 3 of the 4 later dates, above 1.5 dB.
  • The map overlay starts at 2 dB, and the pond sector was followed with a 2 dB threshold.
  • Main zones checked against Sentinel-2 optical images.
Limits
  • Five dates from winter to spring cannot separate a trend from the seasonal cycle.
  • A change in radar signal does not say by itself what changed.
  • No field validation; thresholds are provisional.
  • NISAR has no data between 27 July and 10 August 2026, so the exact day of the change in the pond sector is unknown.
  • Radar sees the surface: no groundwater, no water quality.
  • This tool describes change. It does not assess impacts or assign responsibilities.

Data: NISAR (NASA/JPL and ISRO) via the Alaska Satellite Facility DAAC; Sentinel-2 (ESA; contains modified Copernicus Sentinel data 2026). Software: Leaflet. Built by Team Satellite Miner for the NASA Space Apps Challenge 2026, challenge "Dancing with the SARs". Not endorsed by NASA. Source code: github.com/SatelliteMiner/salar-in-motion.