Microsoft Research is open sourcing its agricultural data and connectivity technologies in the hopes of boosting farmers' yields while reducing costs and helping the agricultural industry take responsibility for its role in climate change.

The tech suite, dubbed Project FarmVibes, builds upon data analysis and integration work with Microsoft's customers like Land O' Lakes and Bayer by incorporating the cloud provider's latest research in precision and sustainable agriculture, Microsoft's Jake Siegel explained in a blog.

By the middle of the century, food production will need to double to feed the planet, Microsoft's Managing Director of Research for Industry Ranveer Chandra said, adding that Microsoft is open sourcing these tools so researchers and data scientists beyond those in Washington, DC can work to tackle the globe's growing food problem.

“We believe one of the most promising approaches to address this problem is data-driven agriculture,” Chandra said.

The first piece of Project FarmVibes to be open sourced is FarmVibes.AI, which is a sample set of algorithms running on Microsoft Azure meant to inspire the research and data science communities to advance data-driven agriculture across the board, Siegel said.

The algorithms are designed to guide decisions at every step of the farming process, starting before seeds are planted and continuing until after harvest. Siegel explained this looks like determining the right amount of fertilizer or herbicide to use on which areas of the fields, forecasting temperature and wind speeds, identifying the ideal depth to plant seeds based on soil conditions, and advising on types of crops and best practices for keeping carbon sequestered in the soil.

Specific capabilities included in the artificial intelligence (AI) arm of Microsoft's FarmVibes project are Async Fusion, which pairs sensors and satellite imagery to create nutrient heat and soil moisture maps; SpaceEye, which removes clouds from satellite imagery using AI; DeepMC, which utilizes sensor data and weather forecasts to predict the temperatures and wind speeds of a farm's microclimate; and a "what if" analysis tool that anticipates how variances in farming methods will affect the soil's carbon sequestration abilities.

"Agriculture is a cause of climate change, [and] it is most impacted by climate change, but with help from technology it can also be a solution to climate change," Chandra said.

The Project FarmVibes suite has been tested on a 7,500-acre farm operated by fifth-generation farmer and software engineer Andrew Nelson, who partnered with Microsoft Research to trial the tech portfolio. He touted the beneficial impacts of the project's AI in agricultural environments, including saving him time and money while improving damage control capabilities.

In addition to AI, Microsoft plans to open source other components of the FarmVibes portfolio focused on edge and connectivity services in the future.

FarmVibes.Connect, which Nelson is currently testing, will bring broadband connectivity to rural and remote areas by leveraging the unused spectrum of TV white spaces. Nelson's farm today is mostly covered, thanks to solar-powered TV white space antenna acting like WiFi routers, Siegel explained.

Nelson is also testing FarmVibes.Edge to intelligently compress vast amounts of data gathered by drones. This solution identifies a certain issue, like weeds in a field, and is able to ignore other details. This results in the efficient construction of small, targeted images that can then be uploaded using FarmVibes.Connect.