报告人：Prof. Scott B. Jones，美国犹他州立大学

报告时间：2023年12月22日（周五）上午10: 00‒11: 00

报告地点：信电学院443交流厅

联系人：盛文溢  邮箱：wenyi.sheng@cau.edu.cn

报告人简介：

Scott B. Jones博士任美国犹他州立大学植物、土壤和气候系的环境土壤物理学教授。Jones教授长期致力于土壤物理领域先进测量与建模方法，以提高农业和环境研究中对土壤性质和过程的理解。Jones教授为美国土壤学会会士（Fellow），曾任美国土壤学会土壤物理与水文分会主席，在土壤物理领域共发表期刊论文110余篇，总被引9000余次，h指数45。

报告内容简介：

Efficient water management plays a vital role in sustainableagriculture, especially in regions facing water scarcity. Conventional irrigationscheduling approaches typically rely on fixed scheduling, which can result ininefficient water usage and suboptimal crop yields. We are developing anapproach for optimizing root zone water management through the integrationof targeted sensor placement, historical sensor data, and machine learningalgorithms. We employ crop root density profiles to inform strategic single soimoisture sensor depth placement within the root zone. Additionally, historicawater content data provides a pathway to estimation of the soil profilehydrodynamics. Machine learning algorithms are used to analyze these complexdata patterns and to learn relationships between our historical sensor data andlocal soil hydraulic properties. By leveraging historical sensor data, we derivesoil-specific hydraulic properties leading to estimation of root zone soil profilewater content. The integration of targeted sensor depth, historical sensor data,and machine learning algorithms holds great potential for precision irrigationand sustainable water management. By accurately estimating soi hydraulicproperties and predicting soil profile water content, this approach canrevolutionize water management practices, conserving water resources andincreasing food production in water-limited regions.