Managing uncertainty in integrated environmental modelling: The UncertWeb framework

  • Authors:

  • Lucy Bastin;Dan Cornford;Richard Jones;Gerard B. M. Heuvelink;Edzer Pebesma;Christoph Stasch;Stefano Nativi;Paolo Mazzetti;Matthew Williams

  • Affiliations:

  • School of Engineering and Applied Science, Aston University, Birmingham, United Kingdom;School of Engineering and Applied Science, Aston University, Birmingham, United Kingdom;School of Engineering and Applied Science, Aston University, Birmingham, United Kingdom;Wageningen University, Wageningen, The Netherlands;Institute for Geoinformatics, University of Muenster, Germany;Institute for Geoinformatics, University of Muenster, Germany;National Research Council of Italy, Institute of Atmospheric Pollution Research, Rome, Italy;National Research Council of Italy, Institute of Atmospheric Pollution Research, Rome, Italy;School of Engineering and Applied Science, Aston University, Birmingham, United Kingdom

  • Venue:
  • Environmental Modelling & Software
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

Web-based distributed modelling architectures are gaining increasing recognition as potentially useful tools to build holistic environmental models, combining individual components in complex workflows. However, existing web-based modelling frameworks currently offer no support for managing uncertainty. On the other hand, the rich array of modelling frameworks and simulation tools which support uncertainty propagation in complex and chained models typically lack the benefits of web based solutions such as ready publication, discoverability and easy access. In this article we describe the developments within the UncertWeb project which are designed to provide uncertainty support in the context of the proposed 'Model Web'. We give an overview of uncertainty in modelling, review uncertainty management in existing modelling frameworks and consider the semantic and interoperability issues raised by integrated modelling. We describe the scope and architecture required to support uncertainty management as developed in UncertWeb. This includes tools which support elicitation, aggregation/disaggregation, visualisation and uncertainty/sensitivity analysis. We conclude by highlighting areas that require further research and development in UncertWeb, such as model calibration and inference within complex environmental models.