San Francisco's reading looks wrong, and this site says so rather than hide it

San Francisco's satellite reading came back at exactly zero, which is not a real measurement for a city of 827,526 people, it means the sample point most likely missed the lit part of town, landing on open water, a shadowed hillside, or just a bad coordinate. Real cities this size do not read true zero.

This is flagged instead of shown as a normal result because reporting it as "San Francisco's dark sky" would be reporting a data error as a fact, see the data quality guide for how this gets checked.

The nearest reliable reading to San Francisco

Mill Valley, CA is about 30 miles from San Francisco and has a trustworthy reading: rural band. It's the closest usable stand-in until San Francisco's own figure is corrected.

Because San Francisco's own coordinate is what produced the bad reading, that distance figure is itself approximate, not a confirmed measurement.

San Francisco against the rest of the dataset

San Francisco, CA has a population of about 827,526, one of 3,206 US cities in this dataset, 3,187 of which have a usable brightness reading today.