Food Early Solutions for Africa - Micro Insurance (FESA)

The FESA project is one of the Milennium projects of the Dutch Minister of Development Cooperation. The objective of the project is to develop a Meteosat based drought micro-insurance system that can reach every farmer in Africa. Partners in the project are MicroEnsure , RABO Development and Ecorys.

The FESA approach is based on 30 years of hourly Meteosat data, which have been processed to climatic data products, in particular temperature, radiation and evapotranspiration. These are then used to generate crop yield estimates or indices and to derive the necessary drought probability statistics for every location on a 3 km grid. The indicators used for drought and crop failure are the growing season relative evapotranspiration (RE) and relative yield (RY).

The first phase of the project has been completed. This phase adressed data base development, data validation and elements of insurance design. A comparative burn-study was carried out for 29 locations in Tanzania, using ground measured precipitation and satellite derived evapotranspiration respectively. The study has revealed great opportunities for reducing basis risk, scaling up and cost reduction.

The results have been consolidated in the printed report "FESA Micro-insurance: methodology, validation, contract design". The report price is 150 euro. It can be ordered by sending us an email through our ordering page. Please write "FESA report" in the first line. Complete ordering form.

user_images/Report.png

Report contents

	FOREWORD by Prof Kees Stigter Founding president of International Society for Agricultural Meteorology	7

1	INTRODUCTION	9
1.1	Traditional crop insurance	10
1.2	Index-based insurance	10
1.3	Satellite indices	11
1.3.1	Reflection indices	12
1.3.2	Precipitation	12
1.3.3	Evapotranspiration	13
1.3.4	Crop yield	13
1.4	Report objective and scope	13

2	DERIVING CLIMATIC DATA FROM METEOSAT	15
2.1	Rainfall monitoring	15
2.2	Evapotranspiration monitoring	16
2.2.1	Calibration	16
2.2.2	Atmospheric correction	16
2.2.3	Air temperature mapping	17
2.2.4	Observation height air temperature	17
2.2.5	Net radiation	18
2.2.6	Sensible heat flux	18
2.2.7	Actual evapotranspiration	18
2.2.8	Relative evapotranspiration	19

3	CROP YIELD ESTIMATION AND FORECASTING	21
3.1	Conversion of solar energy into dry matter	22
3.2	Water limitation to growth	22
3.3	Light use efficiency	22
3.4	Dry matter production	23
3.5	Respiration loss	23
3.6	Net dry matter production	24
3.7	Crop calendar	24
3.8	Relative and difference yield	25
3.9	Combination with geographic information	26

4	VALIDATION OF METEOSAT DERIVED DATA	29
4.1	Validation approach	29
4.2	Validation pitfall	30
4.3	Validation results	31

5	DATABASE GENERATION AND DATA PROPERTIES	33
5.1	Extracting a long term data set from Meteosat	33
5.2	Similarity and difference of evapotranspiration and precipitation data	34
5.3	Comparison of evapotranspiration and precipitation time series	34
5.4	Mass balance of evapotranspiration and precipitation	35
5.5	Phase shift between evapotranspiration and precipitation	37
5.6	Distribution of decadal evapotranspiration and precipitation data	37
5.7	Determination of percentile trigger values	37

6	ELEMENTS OF CONTRACT DESIGN	39
6.1	Current state of the art	39
6.1.1	Starting the growing season	41
6.1.2	Contract parameters vary considerably	41
6.1.3	Critical dependence on growing season start	41
6.2	Using historic data series for timing the growing season	43
6.3	Determination of the sowing window	43
6.4	Comparing evapotranspiration and precipitation based insurance performance	45
6.4.1	Trigger percentiles	46
6.4.2	Growing season structure	47
6.5	Towards scaling up and cost reduction	49
6.5.1	Trigger modelling	49
6.5.2	Zoning approach	51
6.6	Building trust	51
6.7	Summarized findings in this chapter	53

7	SUMMARY AND CONCLUSIONS	55

	ACKNOWLEDGEMENTS	59

	REFERENCES	61

ANNEX A: FESA VALIDATION OF EWBMS CLIMATIC DATA		65
A.1	Validation of air temperature	65
A.2	Validation of radiation	70
A.3	Validation of sensible heat flux	72
A.3.1	Validation with CARBOAFRICA data	72
A.3.2	Validation with Marconi FLUXNET data	73
A.4	Conclusion	77

ANNEX B: FESA VALIDATION OF ECGM CROP YIELDS		79
B.1	Validation data sources	79
B.2	Validation of satellite derived crop yield	79
B.3	Crop yield validation results	81
B.3.1	West Africa	81
B.3.2	Southern Africa	81
B.3.3	Burkina Faso	82
B.3.4	Tanzania	84

ANNEX C: VALIDATION RESULTS FROM RECENT PROJECTS		87
C.1	China: YellowRiver Basin	87
C.2	Mongolia	90
C.3	Netherlands	93