pyMPSLib: A robust and scalable open-source Python library for mutiple-point statistical simulation

• Published in: Earth science informatics Vol. 16; no. 4; pp. 3179 - 3190
• Main Authors: Chen, Qiyu, Zhou, Ruihong, Liu, Cui, Huang, Qianhong, Cui, Zhesi, Liu, Gang
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.12.2023; Springer Nature B.V
• Subjects: Algorithms; Coding standards; Configuration management; Data analysis; Earth and Environmental Science; Earth Sciences; Earth System Sciences; Information Systems Applications (incl.Internet); Libraries; Maintainability; Ontology; Open source software; Parameter sensitivity; Programming languages; Python; Sensitivity analysis; Simulation and Modeling; Space Exploration and Astronautics; Space Sciences (including Extraterrestrial Physics; Heterogeneous structures; Stochastic simulation; Multiple-point statistics; Open-source Python library; Geostatistics
• ISSN: 1865-0473; 1865-0481
• DOI: 10.1007/s12145-023-01086-5

Python has become an essential programming language for scientific computing and data analysis and processing. Various multiple-point statistics (MPS) algorithms are used to characterize complex heterogeneous structures and phenomena in earth sciences. However, there is currently no Python library that integrates mainstream MPS methods for simulation and computation in geosciences. Aiming to establish a stable MPS tool, we developed an open-source Python library of commonly used MPS methods, named pyMPSLib. pyMPSLib consists of ENESIM, SNESIM, and DS algorithms and provides a flexible and convenient API interface. To ensure the maintainability of pyMPSLib, the Python objects and toolkits of MPS algorithms are defined and implemented. To improve the compatibility and extensibility of the presented library, uniform coding standard is adopted in pyMPSLib. We performed the parameter sensitivity analysis under multiple configurations to validate the performance of the library. This open-source library also provides optional tools to quantitatively evaluate the realizations of the integrated MPS methods.