Space meets the agriculture sector – the SEOM Project

We are pleased to announce the successful completion of the Project SEOM COMMONS for Land Cover Change Detection and Monitoring Methodologies Based on the Combined Use of Sentinel-1 and Sentinel-2 for Natural Resources and Hazard Management. This is an ESA project where CLS and TRE ALTAMIRA have participated as Prime and developers of crop monitoring techniques using SAR and optical images. Partners of the project are the Italian Research Institute IRPI and the EURAC Research Centre.

The agriculture sector is facing increasing challenges in our fast-changing world. In recent years considerable efforts are being made to ensure that satellite data are exploited to their full potential to address agricultural needs.

Yet the question remains how we can bridge the gap between technology research and the actual conversion of data into valuable information for farmers.

SEOM is a further step in this regard. It was conceived with the objective of producing a prototype implementation of a new change detection methodology for land cover. The key applications covered by the project are landslides, floods, forest change, snow and agriculture.

In agriculture the frequent coverage of Sentinel-1 (and Sentinel-2) are especially useful for monitoring crops constantly changing and allows different analysis to be conducted. For instance, by spatially analysing the parcel anomalies in the expected growth stage of the crops can be detected.

spatial analysis with Sentinel data

Spatial analysis of a selected parcel (a), dual polar ratio distribution (b) and (c), (d) NDVI distribution of a selected parcel.

An additional temporal analysis of the radar and optical parameters over the crop cycle can help detect variations in the parcel from the expected phenological cycle.

SEOM methodology for data validation

SEOM methodology applied to AgriSAR 2009 data for validation purposes: mean and standard deviation (measured and modelled) of dual pol ratio for a particular AgriSAR 2009 parcel along the crop cycle vs Day of year (top) and phenological stage with their associated model (bottom).

Sentinel-1 co-polar, cross polar etc.

Sentinel-1 co-polar, cross polar and its ratio backscatter coefficient evolution of one AgriSAR 2009 parcel (top) and Sentinel-2 NDVI evolution (bottom).

The agriculture sector can benefit from the accurate and continuous information provided by satellites. These data support farmers to better assess the stages of their crops and the impact of their decisions and can also be used for a better protection of the environment.

We at TRE ALTAMIRA are committed to provide value-adding services that transform satellite data in solutions to be used by the community.

More information and a summary of the main results achieved can be found at the SEOM COMMONS webpage.