|
|
Prediction of crop yields using satellite remote sensing
Lukas, V. ; Trnka, Miroslav ; Semerádová, Daniela ; Rajdl, Kamil ; Balek, Jan ; Štěpánek, Petr ; Zahradníček, Pavel ; Hlavinka, Petr ; Žalud, Zdeněk
Knowledge of the crop yield with sufficient lead time prior to harvest is crucial for both the farm management and the agro-food sector policy. The aim of this study was to test feasibility of developing crop yield forecasting model in Czech Republic for winter wheat, spring barley and oilseed rape based on 2000-2014 database of vegetation indices Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI2) from Moderate Resolution Imaging Spectroradiometer (MODIS) on Terra satellite in form of 16-days composites. An average yield data were collected for 14 selected districts in the Czech Republic that represent the regions with more intensive agricultural production among varying climate and topographic conditions. The viability of the concept was proven in years with significant yield decline i.e. 2000, 2003, 2006 and 2012, when yields of cereals were significantly affected by occurred drought periods. More stable regression results were achieved in the most productive areas such as Olomouc and Prerov, whilst models in highland regions were influenced by lower acreage of three modelled crops and higher prevalence of fodder crops. In most cases, EVI2 showed higher correlations to the crop yield together with using an average value of all composites during vegetation period.
|
guest ::
