← U.S. mission mapNew York City · 8 h response target

Incoming mission / New York City

Click the envelope to open your letter.

Open the full flight brief and science sources
Archived Earth observation of New York City

Mission control / priority uplink

● ACTIVE TRAINING SCENARIO

Flash flood alert: New York City

This training scenario replays the remnants of Hurricane Ida on September 1, 2021, when record rain fell on New York in one hour.

MISSION OBJECTIVEFind the flooded neighborhoods through the clouds and deliver the map within 8 hours.

Why space? Clouds cover parts of the flooded area. Your spacecraft must find useful observations and transmit them to the ground team.

Archived NASA imagery · not a live emergency · source & date ↗

Mission science, assumptions & data sources

What the mission models

Find the flooded neighborhoods through the clouds and deliver the map within 8 hours.

Storm clouds cover the city during the downpour.

Results come from the deterministic simulation. Surface temperature is different from air temperature, and radar does not directly measure surface heat.

Source provenance

  • NASA GPM IMERG daily precipitation (via NASA GIBS) · NASA-observed reference · Observed rainfall, 31 Aug-3 Sep 2021 (remnants of Hurricane Ida). Flood cells are a heavy-rain PROXY, not mapped flood water: the wettest 20% of populated cells by IMERG event total (at least 125 mm). eventDay is the day a cell's cumulative IMERG rain first reached half its event total (mission day 0 = 2021-09-01; earlier lead-in rain counts as day 0).
  • NASA SEDAC Gridded Population of the World v4, 2020 density (via NASA GIBS) · NASA-observed reference · Cells with at least 1000 people per square kilometre count as populated neighbourhoods.
  • NASA MODIS NRT Global Flood Product, 2-day composite (via NASA GIBS) · NASA-observed reference · Cells with 20+ MODIS flood pixels (about 5% of the cell) in a composite after 2021-08-31, and almost none on 2021-08-31 (masks shoreline false alarms), are added as observed flood cells (20 cells). Checks: 2021-08-31 69% no-data, 0.1% flood; 2021-09-01 89% no-data, 0.0% flood; 2021-09-02 3% no-data, 0.2% flood; 2021-09-03 1% no-data, 0.9% flood.
  • Cloud cover and sensor detectability (educational model) · Educational model · Not observed: per-cell cloud probability is a model (storm-cloud base, higher where IMERG rain was heavier) and detectability is a seeded model. The heavy-rain-to-flood rule is also a simplification.