Yuval Sadeh1, Xuan Zhu1, Karine Chenu2 and David Dunkerley1
1 School of Earth, Atmosphere and Environment, Monash University, Clayton, Victoria 3800, Australia. email@example.com,
2 The University of Queensland, Queensland Alliance for Agriculture and Food Innovation (QAAFI), 203 Tor Street, Toowoomba, QLD 4350, Australia.
Providing reliable, consistent and scalable crop yield data is one of the major challenges in monitoring food security. This study aims to improve in-season wheat yield prediction by coupling crop modelling and satellite images. We have developed a nano satellites-based method to detect crop sowing date of grower’s fields, as well as a technique to fuse PlanetScope images (with a spatial resolution of ~3m) and Sentinel-2 images (10m) to create high-resolution datasets of spatio-temporal variation in crop Leaf Area Index (LAI). Finally, we will attempt to use the detected sowing dates and the LAI datasets with the APSIM-Wheat model to predict wheat yield within fields. We shall attempt to predict yield without ground calibration, in a bid to develop a method that is applicable broadly across environments.