Big Data Will Solve the Agricultural Problem
Agriculture needs to evolve to meet a growing population and calorie intake. It is predicted that production will need to increase by 60% by 2050. The challenge is heightened by the need to also decrease resource use. For example excessive irrigation lowers the water table, effectively making water a non-renewable resource as it is extracted faster than it can recharge. A promising solution to agriculture’s problems is big data.
Big data can increase yield, decrease resource use, reduce food waste and help food safety. It is beneficial for both the environment and the economy. Sensors collect information on soil quality, moisture level and plant health in combination with cloud data, such as weather reports and information from neighbouring farms. This data helps farmers manage their resources to make better crop management decisions and efficiently distribute their key resources. This is known as precision farming.
Further, big data can help instruct farmers on the best crops to plant. By analysing historic weather patterns and crop yields, along with soil type, farmers can make informed choices over what to plant. Big data can be used throughout the supply chain for optimisation. If AI is brought into this as well, then the entire process can be automated. Researchers at Tufts University predict that AI and big data could save $250 billion a year in agriculture alone.
One space that big data and AI are already being used in agriculture is lettuce production in the UK. In the UK alone, 3 million lettuces are wasted a year at the farm level and globally the FAO estimates that ⅓ of crops are wasted. There is a massive value gap here.
Agricultural scientists are using AirSurf-Lettuce, an open-source computer vision and machine learning tool, to extract data from the crops. AirSurf presents farmers with incredibly accurate (>98%) information on the quantity, quality and size of lettuces  allowing them to be harvested at the optimal moment. There is only a small window of opportunity for harvest, before lettuce becomes inedible. Under normal circumstances this would mean that only 75% of planted lettuce is sellable . However, this yield gap can be eradicated by the data that is provided by AirSurf.
The more data collected means more information for farmers allowing better decisions to be made. Combining big data with AI and machine learning presents farmers with recommended actions to take based on accurate and most importantly, relevant data. As sensors continue to evolve the data will become increasingly accurate and the more prevalent the technology becomes, the more data will be shared. Together, this will mean even more accurate information for farmers.
Yields will continue to increase to match demands and provide economic benefits for farmers. Simultaneously, resource use will become more efficient, minimising agriculture’s environmental impact. Big data and AI is not necessarily a silver bullet for agriculture’s problems, but it is certainly a major step in that direction.
 Bauer, A., Bostrom, A.G., Ball, J. et al. Combining computer vision and deep learning to enable ultra-scale aerial phenotyping and precision agriculture: A case study of lettuce production. Hortic Res 6, 70 (2019).