A Community Weather Balloon IoT Network

Leading weather agencies like NOAA, NCAR, etc. have published scientific articles discussing the need for more weather balloon stations (also called radiosonde stations) across Africa to assist with climate and weather forecasting. Efforts to improve the existing networks have proven difficult for several reasons, including 1) apathy of the teams launching the balloons, 2) cost of hydrogen and sensors in remote locations, and 3) reliance on older data dissemination technologies like emails, Excel, and FTP.

We have built open-source IoT weather balloons that cost 80% less than traditional radiosondes using LoRa, 3D-printing, and blockchain. All weather data is stored openly on a public blockchain, and university students receive digital currency rewards in real time as the balloon ascends. We have been demonstrating this in 2 different universities in West Africa, and are now creating a simple machine learning model that makes a 12-hour rain forecast for local small-holder farmers.

We are solving the problem of inadequate data collection across Africa (particularly in the Sahel and West Africa). Additionally, small-holder farmers only need less than 48-hour flood and rain forecasts (due to the small size of their lands), yet they cannot have access to this information. It is silohed away in difficult to interpret satellite data. Additionally, satellite data typically does not have the resolution to capture showers and thunderstorms. And radar is only helpful for predicting movement of a storm AFTER it has formed. By analyzing the weather in the morning with a weather balloon radiosonde, we can get an accurate picture of the probabilities of rainfall for the day in that one town or city.

The problems of cost of maintaining a network come down to expense of the hydrogen, materials, and the cost of working with government officials in corrupt environments. Instead, our balloons are very small (only 30 grams) and the balloon payload boxes can be printed at any university with access to a 3D-printer. Even the IoT technology itself is relatively simple, as we've seen undergraduate electrical engineering students being able to activate the sensors and demonstrate launches.

Our balloons are lightweight (30 grams) and have a 40 gram payload attached. This payload is a simple electronics board that needs a very small 350mah LiPo battery. It has a temperature, humidity, and pressure sensor, as well as a GPS module. The GPS is used to calculate wind speed and possibly recover the payload once the balloon pops in the upper atmosphere. There is also a small LED on the board to guide students to setting up their "balloon miner" account and receive digital currency. It's also used to troubleshoot any network issues.

Once the balloon is filled with hydrogen or helium, the electronics are placed in a small 3D-printed box and attached to the balloon. As the payload ascends it sends the environmental data over a low-power IoT network called LoRaWAN. This data is fed in real time to the blockchain and stored in table format that can be viewed during and after the launch. As the balloon ascends, atmospheric pressure decreases, and the blockchain smart contract continuously sends small amounts of digital currency to the miner's account. 

Internet of Things (IoT) technologies are extremely low-cost and accessible to young people in Africa. But there are limited ways in which they can make use of these skills. By working with university students, we can collect enough data to make a 12-hour rain forecast that is specific to their city.

A short-term localized forecast for a small-holder farmer means being able to save a crop before the area is flooded. It also means being able to bring family members together to harvest the crop before a rain shower. We are trying to connect with farmer groups to understand their farming needs. One avenue we see is to also connect with agriculture drone operators, as they often have close ties with farmer groups in the field.

We believe the tools we are using, such as 3D printing, IoT, and blockchain technologies align well with the challenge. The educational nature of our work improves digital literacy and demonstrates an area of where blockchain can be used for good.

We have signed contracts with Academic City University in Accra, Ghana to routinely do launches on campus. To date, we have done about 15 launches from this location. At least 30 students have participated in launches there, with more aware of the project. We have long-standing equipment there including a hydrogen generator and LoRaWAN gateways. All have been managed by faculty and students.

Additionally, we are working 3 students at the University of Uyo in Nigeria with about 20 launches done here. In Uyo we have demonstrated that we can actually fill our balloons with hydrogen from water using solar panels.

We are working with drone groups in Cameroon, Senegal, Sierra Leone, Nigeria, and Ghana to film the balloons and help market the product. Finally, we are doing a global hackathon currently to develop a weather dashboard for our forecasts.

Our solution is primarily possible because of the following:

1) Smaller and more energy-efficient computing ( less hardware and battery weight means we can utilize very small balloons that require far less hydrogen)

2) Access to global sensors and electronics via AliExpress and similar websites

The number of young people in developing countries with working knowledge of electrical engineering is enormous advantage of this solution. Additionally, young people's motivation to participate in both research and local weather forecasting is an enormous game changer for weather data collection in Africa. Especially in the radiosonde market, where traditional actors are slow-moving government agencies.

We are serving primarily university students for educational purposes now. When our 1-year pilot completes, we will have enough training data to produce our machine-learning rain forecast. From there, we expect to be serving 500 or more farmers in each city we are deployed in. We expect to be in 10 different cities by the end of next year.

In five years, we expect to be serving 1000 per location, with over 200 launch locations across West Africa.

We can measure our project's impact by the accuracy of our forecasts against other players in the space. If we are able to produce better 12-hour rain forecasts to the communities we serve at lower cost, then we know we are headed in the right direction.

Part time:  Three engineers

I have an extensive background in weather and software, including a software engineering position at Boeing and an algorithm development position for the GOES-16 weather satellite at NASA Goddard Space Flight Center. He has interned and worked with NOAA in the past and has pursued graduate-level study of a global climate model at Sandia National Laboratories in New Mexico. He has researched much about the importance of West African weather for global climate.

Blaise Bay is our business contact with Academic City. Since 2010, Blaise has worked as a technology entrepreneur in Ghana, building software products for different market verticals.  He has worked extensively with different APIs across different verticals, especially in the eCommerce and payment industry.  For the past 4 years, Blaise has carved a niche in building blockchain applications on different blockchain protocols such as BitShares, Telos, Steemit and Whaleshares. 

Emmanuel Patrick is located in Uyo, Nigeria. He has a background in Electrical/Electronic Engineering from the University of Uyo and worked as an intern at a local TV broadcasting company called AKBC and Pamosen Engineering, where he got experience in communication Engineering and Hardware engineering. Emmanuel Patrick is also very active in the blockchain sphere since 2017, he is the co-founder and ideator of a blockchain network for Africa and the world called “Telokanda”

I'm a hispanic male located in the United States and see these students as part of the journey to learn about IoT technology myself. My background is software and weather, and they are very talented with IoT. 

We hope to have more gender diversity to our group. We think there's a good avenue for this in the artistic and engineering space, since our solution happens to potentially be very influential in terms of image.

We’re applying to Solve because we are interested in growing our technology into larger pilot projects. Our strengths are our technology background, but we do not have many contacts in the areas of grant making and doing pilots on a larger scale. We are also looking for ideas and connections for revenue generation.

An ideal partner would have a good track record of scaling a project or a team that appreciates low-cost solutions with big impacts that can use our small revenue sources to expand our network.

We don’t have any specific organizations in mind, but any that work with drone operators would be a good fit. Many drone pilots are familiar with grant application processes and are trying to solve related issues with small-holder farmers including soil monitoring and plant observing.

The weather balloons offer a very-hands on way to get students interested in meteorology. I personally recall a few demonstrations with National Weather Service staff doing weather balloon launches when I was in in primary and secondary school.

Our method of storing climate data in a decentralized way reflects on the equal opportunity that it can bring to young people around the world.

We are utilizing simple machine learning to make our forecasts, but hope to incorporate more advanced methods of AI prediction in the future.

This prize would allow us to expand to other areas with ecological obstacles like island nations or coastal regions. These are coincidentally the locations that would could offer the most benefit to upper air observation and forecasting, due to their unique locations.