OpenWeatherMap weather service is based on the VANE Geospatial Data Science platform for collecting, processing, and distributing information about our planet through easy to use tools and APIs.
VANE platform permits OpenWeatherMap to process and distribute billions of weather data points every second. Number of servers processes more than 10,000 calls a second that are received from OpenWeatherMap service. This is possible due to cutting edge Big Data technologies that is being applied to the VANE platform.
It's not a secret that cost of transactions and cost of storages are dropping rapidly. This opens up extraordinary opportunities for small and medium IT-companies. A few years ago the cost of developing of large scaled system that support multiple terabytes and load thousands of transactions per second was absolutely unaffordable for them, to say nothing of enthusiasts. Now everything is changing. Big Data is becoming available. And this is why OpenWeatherMap weather data is continuing to be free and open.
Open API for developers makes it easy to embed weather information into variety of applications including web and mobile weather applications, solutions for insurance, advertising, agriculture, sport and many others. For those who just want to know weather conditions for their location we provide current weather, forecast, maps and a lot of charts on our web-site publicly.
VANE platform collects processed weather forecast data. Calculation of the global forecasts require incredible computing power, and unfortunately we don't have an IBM Watson (yet). Fortunately many meteorological services share our love of free data and as a result their data is freely available. We mostly use data from NOAA GFS model and Environment Canada. These models are global and they see their mission in not a prediction of specific weather situation at a certain point of land but in determining of the total atmospheric dynamics on the whole planet. They have a large grid size - about 50 km and longer ranged forecast - 5-7 days. Then, they calculate more detailed models for specific regions. We combine data of multi-scaled forecasts, from the average and global to local and more accurate. That is why OWM interactive maps are very convenient. You can use global forecasts for large scale, and you can get more and more detailed data just by zooming.
We are collecting data from weather stations and forecasts of meteorological services and research laboratories, combining long-term and short-term forecasts with real-time data from weather stations, processing them and immediately updating current weather and forecasts in our API. Data is stored in the VANE database and processing with our unique algorithms to create interpolated data of current weather conditions anywhere in the world, as well as into a variety of weather maps. Finally, OWM provides our API which allows access to all the weather data including a variety of maps.
OpenWeatherMap service collects data from professional and private weather stations. Today we have more than 40,000 weather stations. Most of them are professional stations which are installed in airports, large cities etc. But it is also important for OpenWeatherMap to collect weather data from non-professional stations that are assembled and installed by fans and weather enthusiasts. OpenWeatherMap APIs wouldn't be available without meteorologists sharing share our love of free software. Many models that we use are OpenSource - WRF model, for example.
Environment Canada http://www.weatheroffice.gc.ca
European Centre for Medium-Range Weather Forecasts (ECMWF) http://www.ecmwf.int/
気象庁 Japan Meteorological Agency http://www.jma.go.jp
METAR data from airports http://en.wikipedia.org/wiki/METAR
APRS network http://www.aprs-is.net/