A multivariate spatial clustering method for partitioning tree-based orchard data into homogenous zones
In: Precision agriculture '15Search for other papers by A. Peeters in
Current site
Google Scholar
PubMed
Search for other papers by M. Zude in
Current site
Google Scholar
PubMed
Search for other papers by J. Käthner in
Current site
Google Scholar
PubMed
Search for other papers by M. Ünlü in
Current site
Google Scholar
PubMed
Search for other papers by R. Kanber in
Current site
Google Scholar
PubMed
Search for other papers by A. Hetzroni in
Current site
Google Scholar
PubMed
Search for other papers by R. Gebbers in
Current site
Google Scholar
PubMed
Search for other papers by A. Ben-Gal in
Current site
Google Scholar
PubMed
Purchase instant access (PDF download and unlimited online access):
A combined spatial-aspatial clustering approach for partitioning tree-based data in orchards was developed. The method employs the Getis-Ord Gi* statistic applied to the analysis of individual tree data in a grapefruit orchard located near the city of Adana, Turkey. Analyzed tree-variables included yield (total fruit weight per tree) and two possible yield-determining variables, tree size measured as tree trunk circumference (cm) and soil properties measured by the soil apparent electrical conductivity (ECa (mS/m)). Data were collected from 179 trees. The developed method was applied to the analysis of ‘hot-spots’ (clusters of high data values) and ‘cold-spots’ (clusters of low data values) in orchards and compared to the k-means clustering algorithm, an aspatial clustering method widely-used in agriculture. The combined method improved results by both discriminating among feature values as well as representing their spatial structure and therefore represents a superior technique for identifying homogenous spatial clusters in orchards. The approach can be used for delineating management zones for optimal precision management of tree crops.