Cloudy with a Chance of Space Stations

Makers have created countless devices to keep an eye on the weather, but few are as interesting as having a tiny live view of the entire planet sitting on your desk. Cloud Globe V2 from Okubo Heavy Industries does exactly that, combining satellite weather imagery, a day and night terminator, and the current position of the International Space Station on a 240×240-pixel display.

An Adafruit QT Py ESP32-S2 runs the project, driving an ST7789 color TFT that displays a rotating 3D globe. NASA Blue Marble imagery provides the surface texture, while current infrared cloud observations from the National Satellite Meteorological Center and China Meteorological Administration are layered over it. The firmware also calculates the day/night terminator, so the illuminated portion of Earth changes with the actual position of the Sun.

The cloud data comes from NSMC's global GEO infrared product. Cloud Globe checks the service's availability API every 15 minutes to determine the newest frame. Successfully downloaded images are saved to a microSD card, and the firmware can fill gaps in its archive from the previous seven days.

That archive is more than a cache. A pushbutton lets you replay the previous 24 hours, 7 days, or 30 days of weather. During playback, the globe uses each archived frame's timestamp to recreate the corresponding day/night terminator. The normal globe rotation then picks up where it left off when playback ends.

There is also an ISS tracker built into the project. The ESP32 periodically downloads the ISS's TLE orbital data from CelesTrak and stores it on the SD card. AioP13 then handles the actual orbital propagation locally instead of continually relying on an online tracking service. The globe displays the station's current position along with an orbit track covering 60 minutes before and after the present time.

A separate 128×64 OLED provides some additional information. After the user's location is entered in the SD card configuration file, it can display ISS pass predictions and live azimuth and elevation data. The software also estimates whether a pass should be optically visible based on the observer's location, ISS position, and local lighting conditions.

The Cloud Globe hardware is quite simple, and all of the data comes from publicly available resources, making it easy to reproduce. If you’d like to build your own version of this project, check out the GitHub repository as a first step.