KISS - Knowledge Infrastructure for Scientific Simulation: A Scaffolding for Agentic Earth Science

TL;DR

This paper introduces a knowledge infrastructure scaffold called KI that externalizes scientific expertise into modular components, enabling non-specialists to run and extend complex Earth science simulations effectively.

Contribution

The paper presents KI and its construction toolkit, demonstrating improved simulation success rates and cross-disciplinary generalization, facilitating broader access and collaboration in Earth science modeling.

Findings

Agents with KI achieved up to 84% successful simulations in a coupled-hydrology benchmark.

KI enabled automation across 117 models in 14 Earth-science domains.

Operational expertise is shown to be structured and extractable, not ad hoc.

Abstract

Process-based simulation models encode decades of scientific understanding across the Earth sciences, yet the communities most exposed to climate risk and resource scarcity are the least able to use them. Here, we introduce knowledge infrastructure (KI), an agent-actionable scaffold that externalizes expertise into validated modelling operators, staged domain protocols, and diagnostic recovery mechanisms. Across a 3,000-trial coupled-hydrology benchmark, agents equipped with KI produced physically plausible, verifiable end-to-end simulations in up to 84% of trials, while agents without KI plateaued below 40%. KI generalizes across disciplines. We packaged its construction into a Knowledge Dissection Toolkit (KDT) that autonomously produced KI enabling end-to-end agent execution of 117 additional process-based models across 14 Earth-science domains. Across all 119 KIs, modelling decisions and failure remedies converged despite different underlying physics, showing that operational expertise is structured and extractable rather than ad hoc. Demonstrations show KI-equipped agents lowering both the access barrier between non-specialist users and process-based simulation, and the integration barrier between modelling communities. Through this scaffold, process-based science can then evolve as a living scientific commons, answerable to whoever needs to know and extendable by whoever can contribute.