AgWise / APPROACHES and data
Approaches
AgWise employs a variety of approaches to generate tailored agronomic recommendations. The specific approach used depends on the type of recommendation required. Explore the different approaches below.
Machine Learning
AgWise uses Machine Learning technology to derive tailored fertilizer advice. Machine Learning can be coupled to the QUEFST crop model and yield ceiling information for refined recommendations.
Read more …
Crop modeling
AgWise determines the optimal planting dates and best-fit varieties for different cropping systems based on soil and climate variability. This is undertaken by coupling diverse crop models with spatial soil and forecast weather data.
Read more …
Remote sensing
Within AgWise, remote sensing (and especially Earth Observation) is integrated with statistical models to complement crop models, particularly in generating crop type maps and estimating planting dates.
Read more …
Data sources
AgWise generates recommendations using biophysical and socio-economic data, primarily sourcing standardized public field trial data through tools like Carob and GARDIAN. It also integrates data from public databases and proprietary sources accessible through partnerships with organizations.
New data may also be collected if the need arise upon aggregating and exploring available data.