Early cotton yield assessment by the use of the NOAA/AVHRR derivedVegetation Condition Index (VCI) in Greece

Abstract

Satellite data can significantly contribute to agricultural monitoring.The reflected radiation, as recorded by satellite sensors, provides an indicationof the type, density and condition of canopy. A widely used index for vegetationmonitoring is the Normalized Difference Vegetation Index (NDVI)derived from the National Oceanic and Atmospheric Administration/AdvancedVery High Resolution Radiometer (NOAA/AVHRR) data provided inhigh temporal resolution. An extension of the NDVI is the Vegetation ConditionIndex (VCI). VCI is a tool for monitoring agrometeorological conditions,providing a quantitative estimation of weather impact to vegetation. The primaryobjective of this paper is the quantitative assessment of the cotton yieldbefore the end of the growing season by examining the weather effects as theyare depicted by the VCI. The study area comprises several cotton producingareas in Greece. Ten-day NDVI maximum value composites (MVC)are initially utilized for the period 1982–1999. The correlation betweenVCI images as extracted from NDVI and the 10-day intervals during thegrowing season is examined to identify the critical periods associated mostlywith the yield. Empirical relationships between VCI and yield are developed.The models are tested on an independent dataset. The results show that anearly estimation of the cotton yield trend is feasible by the use of the VCI.