Speaker: Yakun Zhang, assistant professor, Department of Soil and Environmental Sciences, UW–Madison
Soil is a nonrenewable natural resource, essential for maintaining food security, sequestering carbon, cycling water and nutrients, and providing physical support for living organisms. For the sustainable use of soil resources, it is important to quantify soil variability in space and time and understand how soil interacts with environmental and human factors. Recent advances in large geospatial soil datasets, in situ soil monitoring networks, proximal and remote sensing technologies, and machine learning (ML) algorithms offer new opportunities to map soils at high resolution and monitor their changes under land use and climate change. In this talk, Zhang will present three case studies that illustrate the spatial and temporal changes in soil organic carbon, soil thickness, and the growing role of AI in soil science. First, she will show how legacy soil observations, remote sensing, climate data, and ML can be combined to reconstruct 150 years of soil organic carbon change across Wisconsin, revealing widespread carbon losses following land conversion and partial recovery under improved management. Second, Zhang will examine how topsoil and whole-soil thickness vary across the United States in response to climate, land use, and erosion. She will also show that soil thickness is not merely a dynamic attribute but also controls ecosystem memory that helps buffer water stress, sustain productivity, and enhance resilience to climate extremes, especially in arid systems. Finally, Zhang will highlight our recent projects that integrate proximal and remote sensing and spectroscopy with ML and AI to advance soil monitoring, prediction, and mechanistic understanding of ecosystem processes in the era of big data