Preprints
https://doi.org/10.5194/egusphere-2026-4681
https://doi.org/10.5194/egusphere-2026-4681
07 Oct 2026
 | 07 Oct 2026
Status: this preprint is open for discussion and under review for Geoscientific Model Development (GMD).

AccreDiff v1.0: a modular framework for coupling planetary accretion and core–mantle differentiation

Zhihui Kong, Beibei Liu, Ke Chen, Qiyu Wang, and Kang Shuai

Abstract. Planetary accretion and core–mantle differentiation are commonly modelled as separate processes, although the chemical evolution of a growing terrestrial planet depends directly on its impact history. We present AccreDiff v1.0, a modular Python framework designed to couple N-body accretion histories with multi-stage metal–silicate differentiation calculations. AccreDiff uses time-ordered collision records and particle histories as input, reconstructs parent–product relationships and source-reservoir contributions, assigns impact-dependent equilibration conditions, and updates mantle and core compositions through event-by-event core–mantle differentiation.

The model workflow consists of four main components. First, initial compositions are assigned using either CI-based refractory-enriched compositions with prescribed oxygen fugacity or reservoir-based meteoritic endmembers. Second, accretion histories are reconstructed from collision records to track planetary growth, parent bodies, and source fractions. Third, impact events are classified into small-impact and large-impact regimes, with corresponding pressure–temperature conditions and magma-ocean effects. Fourth, equilibrating and non-equilibrating material budgets are passed to the differentiation solver, which computes metal–silicate partitioning and records diagnostic quantities such as bulk composition, mantle redox state, and core mass fraction.

The paper documents the governing assumptions, data structures, and reproducible example workflows distributed with the code. The bundled notebooks demonstrate how AccreDiff reads cosmochemical and dynamical inputs, generates intermediate reconstruction products, performs differentiation calculations, and exports outputs for downstream scientific analysis. AccreDiff is intended as a post-processing and coupled geochemical evolution framework rather than as an N-body integrator or hydrodynamic impact model. Its modular structure also supports extensions to alternative gas–disk prescriptions, collision-outcome models, meteoritic reservoir libraries, and sulfur-bearing differentiation chemistry. By providing an explicit and reproducible interface between accretion dynamics and geochemical evolution, AccreDiff offers a flexible platform for investigating how planetary growth histories shape the compositions and internal structures of terrestrial planets.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.
Share
Zhihui Kong, Beibei Liu, Ke Chen, Qiyu Wang, and Kang Shuai

Status: open (until 02 Dec 2026)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
Zhihui Kong, Beibei Liu, Ke Chen, Qiyu Wang, and Kang Shuai
Zhihui Kong, Beibei Liu, Ke Chen, Qiyu Wang, and Kang Shuai
Metrics will be available soon.
Latest update: 07 Oct 2026
Download
Short summary
Rocky planets grow through repeated collisions, while their metal-rich centres and rocky outer layers develop at the same time. AccreDiff is an open computer framework that links these processes by following each collision and the material it adds. The supplied examples show that each stage can be reproduced and checked, and that the framework tracks how a planet's growth history shapes its final make-up. This allows different formation scenarios to be tested and compared transparently.
Share