Improving crop yield using artificial intelligence (AI) has been a hot topic as researchers and tech suppliers cast their sights on an industry that isn’t exactly the forerunners in applying technology.
Earlier this year, Australia’s Commonwealth Scientific and Industrial Research Organisation (CSIRO) and rural technology start-up Digital Agriculture Services (DAS) launched a new platform that uses AI and cloud-based geospatial technology to deliver farm data and analytics to farmers.
Called the Rural Intelligence Platform, it is the first of its kind that can assess and monitor rural land anywhere in Australia, drawing on information from trusted data sources on productivity, water access, yield, land use, crop type, rainfall, drought impact and more.
More recently, IBM announced a two-year research collaboration with Thailand’s National Science and Technology Development Agency (NSTDA) to improve the yield of sugarcane in Thailand, the world’s second largest exporter of sugar.
With domain support from Mitr Phol, the largest supplier of sugar in Asia, IBM and NSTDA will pilot an AI-driven intelligent dashboard called Agronomic Insights Assistant to provide insights on crop health, soil moisture, pest and disease infestation risk, expected yield and commercial cane sugar (CCS) index by leveraging weather data.
The dashboard will also make use of local sensing technology customised and enhanced for sugarcane farming in Thailand by NSTDA to deliver those insights.
For a start, it will be piloted in the middle of this year on three sugarcane farms of up to one-million-square-meter. If things go well, farmers may be granted access to information that can help them assess and manage risks early, optimise productivity and ultimately increase their crop yield.
With insights up to two weeks in advance and alerts on pests and diseases, stem borer and white leaf risk, hyper-local, short-term, and seasonal weather forecasts, it is expected that farmers will be able to plan specific actions such as irrigation, fertiliser application, and pesticide spray proactively to fight against threat of yield loss.
While the objectives of a growing number of digital agriculture projects are commendable, given that agricultural development is critical for reducing poverty in developing countries, these technology solutions must be affordable and cost-effective to farmers in the longer term if there’s going to be widespread adoption.
Rather than go it all alone, technology suppliers like IBM could work with the likes of the Asian Development Bank (ADB) to achieve economies of scale and drive down costs for individual farms.
The ADB has been championing the use of technology to improve farm yield, potentially becoming a unifying force not only to fund projects, but also to bring together different players to overcome barriers to technology adoption in agriculture.
These include the lack of coordination across data producers, weak methodological processes, limited human and capital infrastructure, inadequate capacities to collect and analyse data from a policy perspective, and poor-quality metadata and dissemination tools.