ROLE
React Frontend
Dash of Backend
Design
Research
BACKGROUND
Around 5000 stars illuminate the night sky, but where are they visible? Meet STELLARGAZE: it identifies low-light pollution areas, ranked by moon phase and nightly weather forecasts. Suitable for space enthusiasts and casual sky gazers alike, it marked my entry into React after completing my Computer Science degree.
Made via collaborative sessions in Hamilton, Ontario cafes with my friends, it spurred my journey into the professional web development industry.
HOW IT WORKS
STELLARGAZE assists users in North America, Australia, and New Zealand by locating nearby parks with optimal conditions for stargazing. These parks are evaluated based on factors such as light pollution, cloud coverage, humidity, and moon illumination. Our proprietary algorithm, endorsed by an astronomer from McMaster University, assigns a score indicating naked-eye star visibility. A score of 80% or higher denotes excellent visibility.
DATASET
Satellite Data - Utilizes satellite images from the Earth Observations Group (EOG) at the National Centers for Environmental Information (NCEI). These images, captured by the Day/Night Band, provide average radiance values for North America, Australia, and New Zealand. Extracted pixel values correspond to Bortle scale ratings, offering an approximate measure of local sky brightness.
GIS/Park Data - Ranks parks within up to 140km of the user based on factors such as light pollution, cloud coverage, humidity percentage, and moon phase. Prioritizes parks located in lower light-pollution zones.
Weather Forecast Data - Obtains weather forecasts from OpenWeather. Due to data limitations and cost considerations, utilizes a k-means clustering algorithm to group nearby parks and share forecasts. The centroid of each cluster serves as the forecast request point, ensuring all parks within a cluster receive roughly the same forecast.
For additional information, visit the
FAQs.
