---
id: ontologies/world-earth-taxonomies
about:
  label: Taxonomies for structuring World–Earth models
  description: >-
    A 2021 Earth System Dynamics paper by Donges and colleagues proposing three
    taxa of modelled subsystems (biophysical, socio-cultural, socio-metabolic)
    and a scheme for classifying their interactions.
  tags:
    - ontology
    - taxonomy
    - earth system
    - world-earth models
    - social-ecological
    - paper
article:
  tier: 2
  url: 'https://esd.copernicus.org/articles/12/1115/2021/'
  published: '2021-11-12'
  publisher: Earth System Dynamics
  authors:
    - Jonathan F. Donges
    - Wolfgang Lucht
    - Sarah E. Cornell
    - Jobst Heitzig
    - Wolfram Barfuss
    - Steven J. Lade
    - Maja Schlüter
domains:
  - climate
  - ecology
open:
  source: open
  data: unknown
  api: unknown
  participation: unknown
  note: The paper is open access at the journal.
provenance:
  origin: agent
  sources:
    - title: >-
        Taxonomies for structuring models for World–Earth systems analysis of
        the Anthropocene: subsystems, their interactions and social–ecological
        feedback loops
      url: 'https://esd.copernicus.org/articles/12/1115/2021/'
  created: '2026-10-19'
  reviewed: '2026-10-19'
  reviewed_by: >-
    Claude, from the abstract and section headings of the paper; the body was
    not read in full. Suggested by Anselm.
  status: draft
---
Earth system models have left social processes outside the model. The authors argue that in the Anthropocene this blocks understanding of Earth resilience and of how to govern global environmental change. The Anthropocene is the present era, in which people are a main force changing the planet. The modelling landscape is wide and mixed. It runs from purely biophysical Earth system models through macro-economic integrated assessment models, which couple economies to climate, to many models of socio-cultural dynamics. Models that simulate entangled human and Earth processes together do not yet exist. Building them means selecting and combining components from several modelling traditions, and that needs a structured way to identify, classify and critique the components.

The proposal is a set of taxonomies. Modelled subsystems fall into three taxa:

- **Biophysical.** Dynamics are usually written as laws of physics, chemistry or ecology.
- **Socio-cultural.** Dominated by human behaviour, decision-making and collective dynamics such as politics, institutions and social networks.
- **Socio-metabolic.** The material interactions between the two, such as human bodies, natural resources and agriculture.

Higher-order taxonomies then classify the interactions between two or more subsystems. With them the authors point at the social–ecological feedback loops where new modelling work is needed. The sections of the paper walk every pairing: environment to environment, environment to metabolism, culture to environment, and so on.

The worked example is a stylised World–Earth model of a greenhouse-gas emissions game. In it the discount rate, how much players value the future against the present, is transmitted socially. The example shows the effect of feedback loops that current models leave out.

The taxa give a vocabulary for saying which parts of a place a model covers and which it has externalised. The interaction taxonomies are a checklist of the couplings a whole-system twin would have to carry. Volume 12, pages 1115 to 1137; DOI 10.5194/esd-12-1115-2021.