Soil Structure and Land Surface Controls on Soil Hydraulic Properties and Processes: Applications of Machine Learning, Unmanned Aircraft Systems, and Observations from Long-Term Conservation Agriculture Management

Abstract

Soil moisture exerts a strong influence on ecology and energy balance of the environment, yet it is difficult to accurately measure because of its high variability in space and time. Predicting the dynamics of soil moisture is also challenging because of the very complicated relationship between soil hydraulic properties and other physical characteristics. Using machine learning, unmanned aircraft system (drone), and observations from long-term conservation agriculture management, I studied links between soil structure and land surface characteristics with soil hydrology.

I developed machine learning models that predict soil saturated hydraulic conductivity—a property that determines the fate of soil moisture—from soil physical variables. These models significantly improved on the accuracy of currently available models and enabled the investigation of soil structure’s impact on hydraulic conductivity. I also developed machine learning-based methods that interpreted surface soil moisture with reasonable accuracy from multispectral imagery taken by a drone. Important controls on surface soil moisture across a landscape are examined using these models. To describe how changes in soil structure impact soil hydraulic properties, I conducted measurements and numerical simulations on soils that have been under different tillage and cover cropping systems for the past 18-years. While some of the conventional, static, measures of soil hydraulic properties suggested a reduction in quality, results from numerical simulations indicated a contrasting finding that showed that soils under conservation agriculture had a better capacity to capture and store water.

Bio

Samuel is a Ph.D. candidate in the Environmental Systems Program at UC Merced. Samuel has a BSc degree in Land Resources and Environment from the University of Asmara, and an MS degree in Environmental Systems from UC Merced. Samuel is interested in addressing soil and environmental problems across scales using machine learning and spatial analysis methods.

Prior Degrees

• MSc, Environmental Systems
  University of California, Merced, 2014
• BSc, Land Resources and Environment
  University of Asmara, 2007