In this presentation, I start with large-scale observation of seismicity migration and stress/fault distributions. Multi-scale analysis of seismicity migration allows us to derive hydraulic properties for the Arbuckle group and the crystalline basement. These parameters contributed to large-scale modeling that suggested the importance of poroelastic stress from fluid injection. With detailed analysis of the stress state of seismogenic faults, we find that fault stress state could strongly influence temporal evolution of earthquake sequences. Then, I discuss diverse triggering processes observed during several individual sequences. The Woodward sequence in western Oklahoma suggests that earthquake-to-earthquake triggering drives the fault activation. The Guthrie sequence in central Oklahoma occurred within a complex fault mesh network, and showed evidence of aseismic slip along a normal fault. These detailed observations provide a reference framework for future earthquake rupture simulations within different fault systems, and insights into the triggering processes during fluid-driven natural earthquake sequences.

Abstract:  Rockfalls are a defining feature of the cliffs of Yosemite National Park, with a rockfall occurring every five days on average.  Yosemite is an ideal natural laboratory to study rockfall processes, but it comes at a cost: rockfalls pose considerable hazard to the park’s 4-5 million annual visitors.  Combining traditional field mapping with remote sensing tools such as lidar, structure-from-motion photogrammetry and thermal imaging, this talk will summarize the findings from more than a decade of collaborative rockfall research, ranging from how rockfalls control long-term landscape evolution to the challenge of ensuring visitor safety in one of the world’s most popular parks.

Abstract: Humans are vastly modifying the earth system, with identifiable impacts across the land surface, cryosphere, ocean and atmosphere. Here, we will explore two example applications of how explainable AI (XAI) techniques can help us visualize and quantify these changes over time. Both parts of this talk serve as examples of how viewing our climate through an AI lens has the power to uncover new insights into anthropogenic change – allowing scientists to ask “why?” but now with the power of machine learning.

Talk abstract and summary

Abstract: The surface of Venus is covered with some of the largest volcanoes in our Solar System, including some fresh and unweathered lava flows, based on orbital spectroscopic and radar measurements. However, rocks on Venus are in contact with its hot (~470° C) and caustic (CO₂ with S) atmosphere, which should alter these rocks and produce coatings of iron-oxides and sulfates on the rocks surface. These coatings should be observable from orbital measurements. Therefore, unweathered lava flows are thought to be geologically very young. Future missions VERITAS, DAVINCI, and EnVision will fly in the next decade and will image these volcanoes. However, to constrain the age of these unweathered lava flows from current and future data sets, the rate of alteration and how alteration minerals affect orbital measurements are needed. Here I will discuss what is known about igneous petrology of Venusian lava flows and specifically focus on recent alteration experiments that suggest unweathered lava flows are only a few years or up to a decade old.

Abstract: Alluvial fans may represent one of the last widespread signs of significant fluvial activity on Mars’ surface. Understanding the climatic conditions during the formation of these features may provide key insights on habitability and climate change on Mars.

Numerous studies have used data and theory from terrestrial fans from warm, arid, rainfall-dominated climates to estimate flow discharges, runoff rates, and total water volumes that likely built Martian fans. However, it has been suggested that Martian fans sourced water only from snowmelt, and perhaps under periglacial conditions. Thus, there is a knowledge gap about the dominant processes building fans under periglacial conditions, and importantly for Mars, a lack of understanding about the characteristic flow magnitudes during depositional events.

In this talk I will present results from a field-based terrestrial analog study involving characterization of the sedimentology and geomorphology of a periglacial alluvial fan in the Richardson Mountains, Northwest Territories, Canada. I will (1) qualitatively describe the range of sedimentary processes occurring on a periglacial alluvial fan and compare them to prior observations, (2) report measured flow discharges and runoff rates that occurred during a summer storm event, and (3) show that melt rates suggested for Mars are capable of entraining and transporting appreciable amounts of sediment by fluvial processes.

Multiple new policies have been developed to improve water quality in our lakes and streams; however, water quality remains a persistent problem. In this talk, Nandita Basu will address the nature of water quality challenges, especially focusing on long term legacies of nutrients, and discuss opportunities for addressing some of these challenges through a combination of top-down analysis — using large geospatial datasets to identify watershed functional  traits — and mechanistic modeling, from the reach to the watershed scale. Through this discussion, she will highlight novel approaches for regional- and global-scale solutions to water quality challenges.