LUNDIsim: model meshes for flow simulation and scientific data compression benchmarks

2025-08-19

AI Summary

To address the growing challenges of computability, interpretability, and sustainability arising from data explosion in Earth science numerical simulations, this paper introduces LUNDIsim—the first open-source geological mesh benchmark dataset tailored for multiphase flow simulation. Built upon an enhanced SPE10 model, it features a faulted, multiscale mesh integrating porosity and permeability fields, enabling cross-resolution flow simulation and compression algorithm evaluation. A novel HexaShrink multiscale representation ensures geometric and topological consistency across resolutions, and the dataset encompasses four representative subsurface scenarios. Fully compliant with FAIR principles, LUNDIsim is publicly hosted on Zenodo and accompanied by reproducible use cases. It fills critical gaps in benchmarking data compression, upscaling, and machine learning for complex geological models, establishing a standardized evaluation platform for high-performance data processing in Earth sciences.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Data Mining & Knowledge Management: Data Compression

Application Category

• Web Mining and Content Analysis: Web data generation and simulation
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

The volume of scientific data produced for and by numerical simulation workflows is increasing at an incredible rate. This raises concerns either in computability, interpretability, and sustainability. This is especially noticeable in earth science (geology, meteorology, oceanography, and astronomy), notably with climate studies. We highlight five main evaluation issues: efficiency, discrepancy, diversity, interpretability, availability. Among remedies, lossless and lossy compression techniques are becoming popular to better manage dataset volumes. Performance assessment -- with comparative benchmarks -- require open datasets shared under FAIR principles (Findable, Accessible, Interoperable, Reusable), with MRE (Minimal Reproducible Example) ancillary data for reuse. We share LUNDIsim, an exemplary faulted geological mesh. It is inspired by SPE10 comparative Challenge. Enhanced by porosity/permeability datasets, this dataset proposes four distinct subsurface environments. They were primarily designed for flow simulation in porous media. Several consistent resolutions (with HexaShrink multiscale representations) are proposed for each model. We also provide a set of reservoir features for reproducing typical two-phase flow simulations on all LUNDIsim models in a reservoir engineering context. This dataset is chiefly meant for benchmarking and evaluating data size reduction (upscaling) or genuine composite mesh compression algorithms. It is also suitable for other advanced mesh processing workflows in geology and reservoir engineering, from visualization to machine learning. LUNDIsim meshes are available at https://doi.org/10.5281/zenodo.14641958

Problem

• Addressing scientific data volume explosion in simulation workflows
• Evaluating compression techniques for large-scale scientific datasets
• Providing benchmark datasets for flow simulation and compression algorithms

Innovation

• Lossless and lossy compression techniques
• HexaShrink multiscale representations
• FAIR principles open datasets