Time-Series Of Remotely Sensed Data To Estimate Wheat Grain Yield At Field Level

Trying to foretell crop yield has been a key issue since the first farming systems were set, it's even more compelling now because of climate change (increasing uncertainty in the system) and a growing population (forcing more accurate and timely estimates). Data acquired by satellite earth observation systems give us a unique opportunity to observe crop growth at adequate spatial resolution with a quasi-weekly temporal frequency. Indeed, remote sensing imagery has been extensively used in data-driven approaches to estimate yield employing a variety of approaches: parametric regressions, machine learning algorithms, and statistical models. Past scientific literature highlights some important advice for this study such as the significant exploitation of time-series of bio-physical parameters rather than vegetation indices. In this context, the main goal is to estimate the yield of winter cereals at field level exploiting a multi-year dataset of ground data, meteorological variables, and satellite derived time series of Leaf Area Index (LAI) with a non-parametric regressive approach.