Introduction

Multiphysics simulation has been growing in the past decades due to computational capabilities. It involves translating the physical partial differential equations to an efficient numerical code. Implementing good Scientific software development practices help in reducing computational power. Testing applicability of multiphysical computational models can prove to be challenging due to thermodynamic and numerical constraints. Therefore, improving the code reproducibility through automation and object oriented concepts is beneficial.

Caption

This project aims to improve an existing 1D advection-diffusion sea ice simulation model using python to improve the benchmark testing for various model parameters. The automation of the model is performed using hydra python package. Hatch python project manager is used for testing, static analysis checks, and for creating reproducible build ecosystem. OOP concepts are leveraged and the mediator design pattern is implemented to improve sustainability of code.

Diagram of mediator pattern implemented in this package