Papers for
environmental modeling teams
Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.
GPU mixed precision speeds up spatial confidence mapping for environmental data
Mixed-Precision GPU Acceleration for Confidence Regions of Spatial Excursion Sets
Abstract: A spatial excursion set identifies where a field exceeds a threshold. With incomplete and noisy observations, inner and outer confidence regions quantify uncertainty in its spatial extent through lower and upper bounds at a specified confidence level. Constructing these regions requires repeated conditional simulations accounting for joint spatial uncertainty, making dense covariance factorization and simulation major computational bottlenecks. We present a GPU-accelerated framework that combines adaptive tile-based mixed-precision arithmetic with distributed task execution to accelerate universal kriging, prediction-uncertainty estimation, conditional Gaussian simulation, and confidence region calibration. In our parallel mixed-precision simulation workflow, triangular matrix-matrix multiplication (TRMM) inherits the adaptive precision layout of the Cholesky factor and uses FP16 simulation operands, while covariance scaling and compensating operand rescaling accommodate the numerical range of reduced precision arithmetic. With appropriate precision tolerances, empirical coverage is close to the double-precision reference and the nominal 90% confidence level across weak, medium, and strong spatial correlation. Ocean surface-salinity experiments show close agreement with the double-precision reference in spatial predictions and confidence regions. On four NVIDIA A100 GPUs (40 GB each), mixed-precision TRMM reaches 6.36x to 6.93x the theoretical FP64 peak throughput. On four NVIDIA GH200 Hopper GPUs, it reaches 10.20x to 11.03x. Each AMD MI250X node has eight GPU dies. Aggregate throughput rises by 9.1x to 11.5x from 1 to 16 nodes under weak scaling and by 1.6x to 1.9x from 4 to 16 nodes under strong scaling. Adaptive mixed-precision accelerates large-scale excursion set analysis while preserving the statistical accuracy required for spatial uncertainty quantification.
Retain and repair neural operators improve real world simulation predictions
Beyond Simulation: Retain-and-Repair Neural Operators for Real-World Adaptation
Abstract: Neural operators increasingly benefit from pretraining on numerical simulations, yet adapting them for real-world prediction remains challenging. We introduce the Retain-and-Repair Neural Operator (R$^2$NO), a framework for adapting simulation-pretrained operators to real-world data while retaining useful pretrained structure. The pretrained operator is first finetuned on real data and then frozen to provide a source prediction, and a shared repair module learns a sequence of refinements from the same observations. Using orthogonal Fourier projections, a spectral ensemble fits a small ridge regression within each cell of the Fourier domain and combines the refinements by weights fitted on a held-out split of the real data. The cells are defined jointly by radial ranges, angular sectors, and measured channels, allowing refinement depth to vary with frequency magnitude, with orientation, and across channels. Including the source prediction as a candidate makes retention available in every cell, and independently trained repair modules enter the same combination as additional candidates. On all RealPDEBench systems and six backbones, R$^2$NO consistently outperforms full finetuning and iterative refinement. The framework treats adaptation depth as a cell-specific choice learned from real data.