<i>d2ix</i>: A Model Input-Data Management and Analysis Tool for MESSAGE_<i>ix</i>

Abstract

Bottom-up integrated assessment models, like MESSAGEix, depend on the description of the capabilities and limitations of technological, economical and ecological parameters, and their development over long-time horizons. Even small models of a few nodes, technologies and model years require input-data sets involving several hundred thousand data points. Such data sets quickly become incomprehensible, which makes error detection, collaborative working and the interpretation of results challenging, especially for non-self-created models. In response to the resulting need for manageable, comprehensible, and traceable representation of input-data, we developed a Python-based spreadsheet interface (d2ix) that enables presentation and editing of model input-data in a concise form. By increasing accessibility and transparency of the model input-data, d2ix reduces barriers to entry for new modellers and simplifies collaborative working. This paper describes the methodology and introduces the open-source Python-package d2ix. The package is available under the Apache License, Version 2.0 on GitHub.

Keywords

• MESSAGE_<i>ix</i>
• reproducibility
• collaborative work
• open modelling and data
• data-handling
• integrated assessment modelling
• data pre- and post-processing