Date of Award
Civil and Environmental Engineering
Uncertainty in water availability is a significant challenge to the agriculture industry. Farmers and irrigators depend on novel uses of sensors and data to maximize water efficiency. Documented studies have demonstrated scheduling irrigation is a straightforward, deterministic means of achieving water efficiency. Irrigation scheduling uses several parameters to determine the moment of crop water stress due to available water in the soil. However, sensors and data for soil moisture and matric potential, a parameter describing water available to plants, have the potential to train machine learning algorithms to forecast water irrigation needs based on previous measurements. Satellite remote-sensing is another developing technology that describes the environmental conditions that enable irrigation scheduling and provides data on crop health by allowing for calculations on collected field images.
This project trains a learning machine with soil moisture and home-brew tensiometer information. To create a water management system that avoids exposing crops to stress, the learning machine uses previous soil water conditions to forecast crop water demand. This machine learning model informs the farmer of the moment maximum water depletion will occur, providing the farmer opportunity to irrigate in advance of crop water stress conditions. Additionally, this research evaluates the value of soil moisture, matric potential, and trained machine learning against characteristics of the specified agricultural undertaking. Because larger agricultural undertakings can be managed with remote-sensing of crop health, this research investigates the viability of ground-sensing against satellite remote-sensing. Sensor-improvements would be more viable for an urban agriculture system. Understanding scenarios in agriculture to tailor technological development will allow farmers to further maximize crop yield and quality with their increasingly limited water availability.
Consalvo, Paolo Alexander, "Designing Technology for Different Scales of Irrigation Scheduling" (2020). Undergraduate Honors Capstone Projects. 475.
Copyright for this work is retained by the student. If you have any questions regarding the inclusion of this work in the Digital Commons, please email us at .