No abstract

No abstract

Varying the rate of application of agronomic inputs generates many positive economic and environmental impacts. Increasingly, technologies that enable variable rate application are becoming a distinctive feature of precision agriculture. Nonetheless, a prerequisite, and crucial challenge, remains the optimal and operational designation of distinct application zones for differing agronomic operations. Core to this challenge is the conflation and fusion of diverse data sources ranging from satellite imagery to realtime in-situ data from farms. At present, zones for variable rate application are often defined manually by agronomists and farmers. This paper proposes a novel methodology for the automatic definition of zones for variable rate application. This approach comprises multi-dimensional spatio-temporal data integration methods, clustering-based data classification and a zone creation and representation procedure. In this way, the harmonization of heterogeneous data sources, augmented with different clustering algorithms, enable the delineation of management zones and subsequent construction of maps for potential variable rate applications. Experimental results confirm the effectiveness and efficiency of the proposed approach.

No abstract

No abstract