Understanding subsurface processes through the Oman ophiolite and AI
Earth science data are typically highly heterogeneous which leads to mixed determined inverse problems and poses challenges to extract process-level information. In this talk I will discuss various adventures I have been on in my attempt to tame the wild data that is collected in a multiborehole observatory and across the globe through IDOP missions. Through taming this wild data, I have found evidence of hydrogen that is likely being actively produced in the Oman ophiolite, I will describe a climate subsurface coupling mechanism that I believe is one of the driving factors of fracture development, and the observation that lithological labeling may not be informative for understanding some of the continous physical processes that drive geology. Along the way I will talk about the challenges we face as geoscientists whose data becomes even more complex, and I will discuss what I think the future of science looks like given the deluge of data and AI tools.
