A methodology for a combined use of normalised difference vegetation index and CORINE land cover data for crop yield monitoring and forecasting. A case study on Spain

The objective of the experience described in the paper consists of extracting crop yield indicators from a combined use of land cover data, CORINE (CO-ordination of INformation on the Environment), and NOAA AVHHR/NDVI data on a regional scale. First, the paper describes the methodology used to integrate the NDVI and the CORINE data. Then, a case study on Spain for a four year span between 1995 and 1998 is presented in the paper. The methodology thus developed allows a better exploitation of the NDVI time series by partly reducing the problem of mixed values. The use of an independent land cover can increase the information content of the extracted time series of NDVI. It is shown in this experience that indicators based on the CNDVI (CORINE-NDVI) time profiles can be more closely related to crop yield performances than indicators based on simple NDVI. The advantages of the approach are confirmed by the case study: good results in terms of crop yield forecast modelling, good discrimination of annual crop cycles, good desaggregation of average regional NDVI profiles. The application and the results fit the objective of the MARS (Monitoring Agriculture with Remote Sensing) Crop Yield Forecasting System of the EC, within which the experience was conducted. The objective of this system is to contribute to an independent, reliable and timely production forecasts EC system.