Research
Urmi’s current research is at the intersection of computational and data science applied to soil mineralogy. Her work revolves around deciphering the intricate relationships between soil mineralogy and various soil properties, including nutrient capacity, soil fertility, erosion, leaching, carbon degradation, water retention, and adsorption isotherms.
Currently, her primary objective is harnessing digital tools to analyze extensive datasets related to soil mineralogy and establishing correlations with property data, notably with the National Soil Inventory of Scotland (NSIS). This work holds immense promise for assessing soil health, a critical factor for food security and carbon sequestration, in an era where these concerns are of paramount importance.
2023-2024 Hutton seedcorn: How should we measure clay? Integrating XRPD with MIR spectroscopy for quantifying clay minerals in soils using Machine Learning. PI
Past research
Urmi is a Geologist with background spanning across Geochemistry, Mineralogy, Ore Geology, Igneous Petrology, and Machine Learning. Urmi’s Ph.D. research primarily focused on elucidating critical aspects of tungsten mineralization, specifically within granite-hosted vein and greisen type deposits. Her research focussed on trace element partitioning between minerals and fluids, shedding light on the processes responsible for trace element mobilization and enrichment by hydrothermal fluids. In addition, her studies involved in-situ major and trace element, B and Li isotope study of minerals such as tourmaline, mica, and skarn garnet as reliable proxies for mineralization, contributing significantly to the understanding of these complex geological systems.
Beyond her doctoral research, Urmi has worked on several Machine Learning projects in the field of hard rock geochemistry.
