Monitoring of crop growth for prognostic yield assessment using Sentinel-2 MSI data

The project aim is to develop methods for utilization of optical sensor data from Copernicus Sentinel-2 MultiSpectral Instrument (MSI) for prognostic assessment of crop conditions and agricultural yields. The outcomes of the project will contribute to near-operational capability to track seasonal variation in crop growth and enabling early warnings of drought and other disturbances, thereby forecasting crop yield variations. The project is based on field data collection, time-series processing methodology, and process modeling, to generate: (1) improved knowledge about capabilities and limitations of Sentinel-2 MSI time-series data for agricultural monitoring; (2) methodology for deriving smooth vegetation index data from MSI in near-real time by use of Kalman filtering; (3) methodology for assimilation of seasonal data from MSI into a simple crop model based on biomass allocation; and (4) development of a scheme for integrating ecosystem modeling and remote sensing for crop yield estimation. The methodology is based on long experience within the research group of time-series processing of remotely sensed data, derivation of biophysical parameters, and ecosystem modeling. We will also utilize our established infrastructure for spectral field data collection. The results of the project have potential value for farmers, agricultural authorities, advisory organizations, and insurance companies.

The project aims at improving possibilities to make prognosis of crop yield by combining data from satellites with meteorological data.