A 3-D printer is creating the parts for affordable weather stations to be used in developing nations, where weather data is very limited.
Data from the stations will be used to help meteorological agencies and companies create improved forecasts that could help save lives and protect property.
Some areas that do not have weather observations are poorer countries, places where people do not live and areas in war or conflict, AccuWeather.com Meteorologist Adam Douty said.
"Syria and Yemen, for example, had much of their network destroyed," Douty said. "Some of the areas with the worst or non-existing networks are Syria, Afghanistan, across the Sahara Desert and central Africa."
Zambia is expected to be the pilot site this summer for the Micro-Manufacturing and Assembly (MMA) Project, created by the University Corporation for Atmospheric Research (UCAR).
The shell and structure of the weather station is 3-D printed with two kinds of plastics, UCAR Project Manager Kelly Sponberg said.
"For the 'brain,' we use the Raspberry Pi, which is a small single-board computer," he said. "This is used to control sensing, data logging and communications. Finally, we utilize a variety of off-the-shelf sensors to measure temperature, humidity and other factors."
UCAR is working with the U.S. Agency for Industrial Development (USAID) and the National Oceanic and Atmospheric Administration on the project.
"Commercial contracts for stations can cost tens of thousands for equipment alone," Sponberg said. "Sometimes, the project is hundreds of thousands in consultant fees, maintenance, etc. The MMA project aims to reduce these costs, but more importantly ensure that maintenance and construction are something a weather office can perform on its own."
The more data you can get into computer models, the better they will perform, Douty said.
"So more weather observations from across the globe can improve forecasts -- short and long term," Douty said. "If you have weather observations from areas, you can compare the models performance to what is actually going on to find biases. Then, you can either correct the model (or the model output) to better the forecast."
The 3-D stations are currently being field tested in Boulder, Colorado.