Assessing the ability of Sentinel-2 derived vegetation indices to explain inter-field yield variation in the context of index insurance - A case study of paddy rice inHaryana and Odisha, India

Smallholder agriculture in the Global South is characterised by high degree of risk, which disincentivises investment in productivity gains and limits rural development. Index Insurance aims to overcome the limitations of traditional insurance to insurance farmers against exposure to climatic extremes. Based on two study sites in India, Haryana and Odisha, this study contributes to the technical aspect of improving the indices, more specifically on how field level yield can be estimated through Sentinel-2 derived VI variables and which design options are more suitable to create these variables. The study shows that the best variables alone can explain 20% of the inter-field grain yield variation and that the best combination of variables can explain 53%. Furthermore, the main findings of the study suggest that it is beneficial to test different triggering measures and that including variables from phenologically tailored phases and isolating the rice varieties significantly improves the results. Additional research is needed before the approach is suitable for individualised index insurance but compared to alternative data sources the method will likely pose an effective and scalable way to identify yield gaps and to specifically target policy interventions.

Bibliographic Details

Main Authors:	Pedersen, Nicolai Bloch, Østergaard, Kristian Wille
Format:	Thesis biblioteca
Language:	English
Published:	University of Copenhagen 2020-07-02
Subjects:	climate change, agriculture, food security, rice, smallholders,
Online Access:	https://hdl.handle.net/10568/109608
Tags:	Add Tag No Tags, Be the first to tag this record!