A framework for characterising and evaluating the effectiveness of environmental modelling

Research output: Contribution to journalArticleScientificpeer-review

Researchers

  • Serena H. Hamilton
  • Baihua Fu
  • Joseph Guillaume

  • Jennifer Badham
  • Sondoss ElSawah
  • Patricia Gober
  • Randall J. Hunt
  • Takuya Iwanaga
  • Anthony J. Jakeman
  • Daniel P. Ames
  • Allan Curtis
  • Mary C. Hill
  • Suzanne A. Pierce
  • Fateme Zare

Research units

  • Brigham Young University
  • Charles Sturt University
  • Australian National University
  • Edith Cowan University
  • Queen's University Belfast
  • University of New South Wales
  • Arizona State University
  • United States Geological Survey
  • University of Kansas
  • University of Texas at Austin

Abstract

Environmental modelling is transitioning from the traditional paradigm that focuses on the model and its quantitative performance to a more holistic paradigm that recognises successful model-based outcomes are closely tied to undertaking modelling as a social process, not just as a technical procedure. This paper redefines evaluation as a multi-dimensional and multi-perspective concept, and proposes a more complete framework for identifying and measuring the effectiveness of modelling that serves the new paradigm. Under this framework, evaluation considers a broader set of success criteria, and emphasises the importance of contextual factors in determining the relevance and outcome of the criteria. These evaluation criteria are grouped into eight categories: project efficiency, model accessibility, credibility, saliency, legitimacy, satisfaction, application, and impact. Evaluation should be part of an iterative and adaptive process that attempts to improve model-based outcomes and foster pathways to better futures.

Details

Original languageEnglish
Pages (from-to)83-98
Number of pages16
JournalEnvironmental Modelling and Software
Volume118
Publication statusPublished - 1 Aug 2019
MoE publication typeA1 Journal article-refereed

    Research areas

  • model evaluation, model assessment, model performance

ID: 33282350