The Virtual IELab – an exercise in replicating part of the EXIOBASE V.2 production pipeline in a virtual laboratory

ABSTRACT

We explore options to replicate the EXIOBASE2 multi-region input–output (MRIO) database in the Virtual IELab cloud-computing laboratory environment. Whereas EXIOBASE2 is constructed using a multi-process reconciliation procedure, we present an alternative compilation technique that uses EXIOBASE2's pre-processed data and final tables in reconciling the IELab MRIO with conflicting raw data information. This approach skips the labour-intensive step of detailing and harmonising country tables. Adherence metrics reveal the EXIOBASE2-based IELab table to be considerably less balanced than the original but with stronger adherence to other constraints data. However, these metrics are not comparable to the original EXIOBASE2 statistics due to the distinctive implementation of constraint sets in the two platforms. IELab's main value-added is its flexibility in tailoring EXIOBASE2-based MRIOs beyond the original recipe. Finally, IELab's global carbon, water and material footprints are shown to be comparable with previously reported resource footprints. In contrast, deviations in land footprints warrant further investigation.

KEYWORDS:

• International trade
• multi-region input–output tables
• EXIOBASE
• virtual laboratory
• global footprints

Acknowledgments

The authors thank Manfred Lenzen for the valuable comments and Sebastian Juraszek for expertly managing our advanced computation requirements. This work was in part done during a visit of R.R. to Leiden University – CML.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 http://www.ide.go.jp

2 http://www.worldmrio.com

3 http://www.exiobase.eu

4 https://www.gtap.agecon.purdue.edu

5 http://www.wiod.org/new_site/home.htm

6 The Project Réunion consortium is the result of a small-scale workshop held in the Island of Réunion of representatives of major institutions involved in the compilation of global MRIO databases, following the 18th Input–Output Conference held in 2010 at the University of Sydney, that originated the idea of global research collaboration for harmonising activities and enhancing synergy and efficiency on MRIO compilation. (http://www.isa.org.usyd.edu.au/mrio/mrio.shtml)

7 CREEA is the acronym for Compiling and Refining of Economic and Environmental Accounts.

8 DESIRE is the acronym for Development of a System of Indicators for a Resource efficient Europe.

9 EU27 includes member states Austria, Belgium, Bulgaria, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, the Netherlands, Poland, Portugal, Romania, Slovak Republic, Slovenia, Spain, Sweden and the United Kingdom.

10 AISHA, an acronym for An Automated Integration System for Harmonised Accounts, is a MATLAB-based tool for constructing large contingency tables developed at the University of Sydney and is the system adopted for the Global MRIO Lab.

11 The root classification refers to the maximum sectoral and spatial disaggregation of the MRIO that provides the highest level of details (Lenzen et al., 2017).

12 Data feeds are purpose-built pieces of code that are used to convert raw data into a format which the MRIOLab suite can process in a fully automated way (Lenzen et al., 2017).

13 This is to avoid discrepancies between disaggregate and aggregate constraints, such as row totals, that might occur in the official SUTs.

14 A measure of goodness-of-fit, R-squared is among the typical metrics for comparing matrices. Although calculated, it is not presented in this work since it captures the same information as correlation, with R-squared equivalent to the square of the correlation coefficient in simple linear regression (Wiebe and Lenzen, 2016).

15 These standard deviation settings reflect EXIOBASE's philosophy of the primacy of adhering to country SUTs data and preserving the balancing relationships.

Additional information

Funding

This work was financially supported by the Australian Research Council through its Discovery Projects DP0985522 and DP130101293, and by the National eResearch Collaboration Tools and Resources project (NeCTAR) through its Industrial Ecology Virtual Laboratory [grant number VL201].