Satellite soil moisture and rainfall for yield prediction in water limited regions
Mariette Vreugdenhil demonstrates the use of the EUMETSAT H SAF soil moisture (H116, SM) and SM2RAIN (H64) products to predict yields for Morocco and Senegal.
We demonstrate the use of the EUMETSAT HSAF soil moisture (H116, SM) and SM2RAIN (H64) products to predict yields for Morocco and Senegal. Root-zone SM was calculated from SM, and NDVI was used as a vegetation indicator. Data on yields was obtained from the Food and Agriculture Organization of the United Nations.
Yield prediction was done for main crops using multiple linear regression and a time for space approach. SM improved yield prediction, especially early in the growing season, improving early warning capabilities. NDVI showed better predictions later in the growing season. SM2RAIN outperformed other benchmark rainfall datasets.