Events for 03/04/2019 from all calendars

Geometry Seminar

Time: 3:00PM - 4:00PM

Location: BLOC 628

Speaker: Tingran Gao, U. Chicago

Title: Manifold Learning on Fibre Bundles

Abstract: Spectral geometry has played an important role in modern geometric data analysis, where the technique is widely known as Laplacian eigenmaps or diffusion maps. In this talk, we present a geometric framework that studies graph representations of complex datasets, where each edge of the graph is equipped with a non-scalar transformation or correspondence. This new framework models such a dataset as a fibre bundle with a connection, and interprets the collection of pairwise functional relations as defining a horizontal diff‚usion process on the bundle driven by its projection on the base. The eigenstates of this horizontal diffusion process encode the “consistency” among objects in the dataset, and provide a lens through which the geometry of the dataset can be revealed. We demonstrate an application of this geometric framework on evolutionary anthropology.

Industrial and Applied Math

Time: 4:00PM - 5:00PM

Location: BLOC 220

Speaker: Tarun Verma, Los Alamos National Laboratory

Title: Variability and Predictability of the Arctic Freshwater System in Community Earth System Model (CESM) Initialized Decadal Predictions

Abstract: The perennial presence of sea ice and low salinity waters in the Arctic Ocean makes it the largest oceanic reservoir of freshwater. Sea ice (solid freshwater) in the Arctic Ocean regulates the climate by reflecting back most of the incoming solar radiation and insulates the deeper ocean from wind-driven stirring. The low salinity waters (liquid freshwater) beneath the sea ice can strengthen ocean stratification, and thus prevent convection. The changes in the freshwater storage of the Arctic Ocean can imply 1) an increase/decrease in freshwater exchange with the adjacent oceans, or 2) a change in surface freshwater sources like precipitation, river runoff, ice sheet melt etc. These can further be linked to large-scale changes in oceanic and atmospheric circulation. In recent decades, the Arctic Ocean freshwater system has experienced dramatic changes due to anthropogenic climate change. The sea ice volume has shrunk considerably, while the surface ocean has warmed, and freshened at a rate greater than anywhere else over the globe with implications for future climate change, and economic activity in the Arctic, e.g. shipping routes. In this talk, I will present an overview of historical variations in the Arctic liquid freshwater content using an observationally forced ocean-sea ice model simulation, followed by an evaluation of a fully-coupled climate model in predicting these changes. These simulations are part of a large ensemble of initialized decadal hindcasts that were performed at National Center for Atmospheric Research (NCAR), and use fully-coupled Community Earth System Model. Some of the relevant challenges in making climate predictions on decadal timescales will also be discussed.