Where the numbers come from
We use public datasets and name them on the page. Satellite and launch counts come from GCAT, Jonathan McDowell's catalogue. Park locations come from the National Park Service. Cloud records come from NASA's POWER project, and planet figures come from NASA's planetary fact sheets. The full list is on Sources and credits.
We don't type numbers into pages by hand. A script reads the data and writes the page, so a figure on the page can be traced back to a row in the source.
How we calculate
The Milky Way pages work out where the Sun, Moon and the center of the galaxy are in the sky, hour by hour, for each park. The method is on the Milky Way season page, and the code is checked against known moon phases and equinoxes before every run. The scale maths is on the scale page.
What we don't claim
A score isn't a forecast. Our pages rank months and places using averages. They don't know tonight's weather, smoke or moonlight conditions, and they say so. Modelled values, such as sky glow, are labeled as models.
Known, inferred, unknown
On the science pages we sort every claim into one of four boxes: measured, inferred, unknown, and what would settle it. If scientists disagree, we say that.
Photos and illustrations
Photos come from NASA and the National Park Service, from openly licensed sources, or from licensed stock libraries. Each caption says what the picture shows. Where a picture is an artist's concept or a computer illustration, the caption says so.
Corrections
We make mistakes. If you spot one, email [email protected] with the page address and what you think is wrong. We check it against the source, fix the page if you're right, and write back. When a correction changes a number or a conclusion, we say so on the page.
Keeping data current
Each data page shows the date its data runs to. Satellite and launch counts are rebuilt from the source catalogue on a regular schedule, and the Milky Way calendars are rebuilt each year. We only change a page's date when its data really changes.