Characterizing the canopy variability of the American Farm School's educational vineyard using precision agriculture methodologies and Normalized Difference Vegetation Index (NDVI)

Abstract

Varianty existing in field crown crops can be caused by a number of factors. Although human eye Visibility 400-700nm visible ration range may not be able to spot any evident differences, sensors used for recording crop characteristics, such as plant canopy reflectance in the infrared spectrum (700-1300nm) may reveal significant areas of difference. Recently due to the development of this agricultural technology and the further development of remote sensing technology a vegetation index named NDVI (Normalized Difference Vegetation Index) is commonly used to identify such differences in the context of Precision Agriculture methodologies. This study was conducted in order to identify and address NDVI variability in 20.7 ha area educational vineyard located in the American Farm School. For research purposes only four(4) out of (10) grape varieties (Vitis vinifera L.), commonly used in viticulture industry of Greece were examined. The study is separated into two major parts, in the first one characteristics of vines and history of viticulture are reviewed as well as the importance of viticulture for Greece. The second part of this research assesses variability at the American Farm School educational vineyard based upon NDVI measurements taken and also the potential variability of two soil chemical properties.soil chemical reaction (pH) and soil's electrical conductivity(EC), which are of high importance and affect vine growth, development and final yield. The entire research design and data collection procedures for measuring vineyard's soil and crop properties took place in the new Perrotis College's Precision Agriculture laboratory and the American Farm School educational vineyard, with the use of Precision Agriculture equipment and other available resources. The results of this study revealed that a distinct level of variability existed throughout the examined vineyard. Regarding NDVI values a distribution curve of variability was found to among three different data sets taken during the experiment period. Additionally it was shown that only a narrow difference in NDVI values existed among the different vine varieties, a fact that characterizes the entire vineyard's relatively uniform. Although certain areas tend to differentiate regarding NDVI values, a fact that requires deeper research regarding the reasons behind that fact Finally, due to the fact that the vineyard is irrigated by three different dosages, soil chemical properties analysis highlighted some facts of major importance regarding vineyard's soil management and existing differentiations between different irrigation treatments. The last part of the present BSc Thesis includes a statistical analysis of results and it is accompanied by the development of NDVI maps with the use of a Geographic Information Systems (GIS) for a visualization of the existed variability throughout the examined vineyard.