Sreeram Venkat

Stefan Henneking, Fabian Kutschera, Sreeram Venkat et al.

Near-field tsunami early warning in the Cascadia Subduction Zone is limited by sparse offshore observations. We show that a hypothetical network of 175 seafloor pressure sensors can support real-time Bayesian inference of tsunamigenic seafloor motion and probabilistic tsunami forecasts for two fully-coupled Cascadia earthquake dynamic rupture--tsunami scenarios, a partial rupture and a margin-wide rupture. The complex oceanic acoustic, Rayleigh, and tsunami wavefields in both scenarios are similar during the first two minutes and then diverge. Using an acoustic--gravity inversion with offline precomputation and online assimilation of pressure data, tsunami forecasts are obtained in less than a second. We leverage a Bayesian inversion-based framework that splits the computations into an offline precomputation phase performed with large-scale computing facilities, and an online phase that computes forecasts from real-time data and can be executed on a laptop. Forecast errors remain low at 22.1% for the margin-wide rupture and 19.6% for the partial rupture.

Sreeram Venkat, Stefan Henneking, Omar Ghattas

Real-time tsunami early warning relies on distributed sensor networks to infer seismic sources and seafloor motion. Optimizing these networks via Bayesian optimal experimental design (OED) is exceptionally challenging for systems governed by hyperbolic partial differential equations, which lack the spectral decay required by standard low-rank approximations. We present a scalable Bayesian OED framework for linear time-invariant systems. By reformulating the inverse problem in the data space, we transform OED into dense matrix subset selection. We propose a multi-GPU, Schur-complement-update-based, greedy algorithm that solves the OED problem using a pipelined approach that fully overlaps I/O with GPU computations. Our framework achieves near-perfect weak and strong scaling across hundreds of GPUs on Perlmutter and Frontier. Applied to the 2025 Gordon Bell Prize-winning digital twin for tsunami forecasting in the Cascadia Subduction Zone, we optimize a 175-sensor network, minimizing the uncertainty of a parameter field with over one billion degrees of freedom.

Sreeram Venkat, Ralph C. Smith, Carl T. Kelley

In the construction of reduced-order models for dynamical systems, linear projection methods, such as proper orthogonal decompositions, are commonly employed. However, for many dynamical systems, the lower dimensional representation of the state space can most accurately be described by a \textit{nonlinear} manifold. Previous research has shown that deep learning can provide an efficient method for performing nonlinear dimension reduction, though they are dependent on the availability of training data and are often problem-specific \citep[see][]{carlberg_ca}. Here, we utilize randomized training data to create and train convolutional autoencoders that perform nonlinear dimension reduction for the wave and Kuramoto-Shivasinsky equations. Moreover, we present training methods that are independent of full-order model samples and use the manifold least-squares Petrov-Galerkin projection method to define a reduced-order model for the heat, wave, and Kuramoto-Shivasinsky equations using the same autoencoder.