1st BEACON Business Case
BEACON toolbox is the result of "insurance market pull" influencing the new product/services evolution and development. As such BEACON products are led by the ‘market’ requirements, involving the potential customers (re-insurers, insurers, underwriters etc.) as well as relevant stakeholders in all phases from the design, to development and validation. In this way, the full range of BEACON services will be tested, evaluated and tailored up to the specific needs of each entity through introduction of 3 BEACON Business Cases.
In Spain, the private sector is integrated into the system and contributes to the coverage of a part of the risk. Crop insurance provides a comprehensive risk coverage and approximately one half of AgI premiums are subsidized by the state. The agri-insurance is provided by insurance companies under the Agroseguro pool management. This system is considered one of the most successful and wide-reaching in the world. The company Agroseguro from Spain represents an early adopter of the BEACON toolbox.
The BEACON approaches for Drought Damage assessment were tested against ground-truth drought damage assessment on rainfed wheat and barley, originating from the rural area of Ávila and Segovia, in Spain; |
The MODIS Terra 8-day composited NDVI at 250 m resolution, was fully exploited for agricultural drought detection, monitoring and crop loss assessment; |
NDVI-Anomaly was calculated, by dividing the average NDVI for a particular 8-day period of a given year, by the multi-year mean NDVI of the specific 8-day interval; |
Machine Learning classifiers and regression models were evaluated for damage classification and yield prediction. The NDVI and NDVI-Anomaly produced throughout the growing season, together with damage percentage and final received wheat and barley yields were used as training data input in the Machine Learning scheme. |
The approach implemented in the operational workflow of BEACON estimates with a great accuracy the final wheat and barley yield, regardless of whether there is a drought incident occurring or not. |
Drought Damage Assessment Relative NDVI – Anomaly
Machine Learning scheme