Авторизоваться
Для зарегистрированных пользователей

Based on the experience of the Monitoring Agriculture with Remote Sensing (MARS) project in Europe, in 2001 the European Comission decided to enlarge the monitoring activities to other regions of the world. The MARS-FOOD group was established to support the Food Aid and Food Security policies of the European Commission.
The methodologies developed for crop monitoring combine remote sensing and meteorological data, which can be used directly in a GIS environment or as input for crop growth simulation models (CGMS and FAO-CSWB). Several techniques are used for extracting crop specific temporal profiles of the Normalized Difference Vegetation Index (NDVI) and aditionally dry matter production maps are calculated according to the Monteith approach. Both remote sensing and agrometeorological indicators are then used together with agricaltural statistics for crop yield forecasting.
A new method has been recently developed by MARS-FOOD to produce quantitative yield forecasts for regions with generally low data availability by using a multistep procedure. The method starts with a time series analysis of the historical yields to have a first broad prediction range. Similarity analysis is then used to identify the year with the most similar crop growing conditions over a time series. The next step is a regression analysis involving the different meteorological and remote sensing indicators available. Finally, based on the results of the previous 3 steps, the most likely yield value is estimated. The method is still in a test phase, but is already used in a preoperational way in the MARS-FOOD crop monitoring bulletins for the Mediterranean Basin, Russia and Central Asia and Somalia.