
The Smart Indigenous Weather App (SIWA) is a climate resilience initiative developed by the DIPPER Lab at Kwame Nkrumah University of Science and Technology (KNUST) in collaboration with the Ghana Meteorological Agency. This mobile platform bridges the gap between traditional ecological knowledge and modern science by integrating ancestral weather indicators, such as the movement of birds, insect behavior, and tree phenology, with scientific satellite data. By using machine learning to digitize and validate these local observations, SIWA provides smallholder farmers with highly localized and reliable climate information, including precise planting alerts.
This project specifically addresses the growing unreliability of traditional forecasting signs caused by climate change and environmental degradation from illegal mining, or galamsey, in Ghana. Unlike previous systems that required physical travel for data collection, SIWA allows for remote data access, which minimizes data loss and enables researchers to continuously refine the machine learning models based on real-world farmer inputs. With pilot sessions recording forecast accuracies as high as 82%, the application aims to reduce annual crop losses by up to 30%, ultimately enhancing food security and economic stability for thousands of people in rural communities.
Scientific Director
climate & ecosystems service monitoring
researcher, mphil
researcher, mphil
Join us in our mission to develop innovative solutions through research and technology.
Get in Touch