Abstract by Gabriel Adams

Personal Infomation

Presenter's Name

Gabriel Adams

Degree Level


Abstract Infomation



Faculty Advisor

Matthew Heaton


A Bayesian Spatial Model and Clustering Algorithm for Variable Rate Irrigation Systems


Maximizing the efficiency of irrigation in agriculture is vital to environmental health and economic vibrancy.  Variable rate irrigation (VRI) systems provide an intriguing alternative to traditional pivot systems in that VRI uses sensors and computers to vary the amount of water released by sprinklers as the pipe rotates around the field.  We propose a Bayesian spatial model to use limited data to predict water sensitivity across an entire field, where water sensitivity is a measure of how additional water at a specific location enhances yield.  We also propose an algorithm loosely based on K-means clustering with a contiguity encouragement component to group together similarly sensitive points.  We then use this clustering algorithm to create irrigation zones for use in VRI.