Closing the gap? Top-down versus bottom-up projections of China’s regional energy use and CO2 emissions☆
Highlights
- •First global model comparison with harmonized socio-economic assumptions for China.
- •First soft-linking study that down-scales global energy scenarios for three regions of China.
- •Key results for China are benchmarked with 23 global models from the Asia Modelling Exercise.
- •A highly transparent, interdisciplinary and open-data scenario analysis approach.
Abstract
As the world’s largest CO2 emitter, China is a prominent case study for scenario analysis. This study uses two newly developed global top-down and bottom-up models with a regional China focus to compare China’s future energy and CO2 emission pathways toward 2050. By harmonizing the economic and demographic trends as well as a carbon tax pathway, we explore how both models respond to these identical exogenous inputs. Then a soft-linking methodology is applied to "narrow the gap" between the results computed by these models. We find for example that without soft-linking, China’s baseline CO2 emissions might range from 15–20 Gt in 2050, while soft-linking models results in 17 Gt. Reasons for the results gap between the models are discussed subsequently, such as model structure and statistical inputs. At a sectoral level, the gap can be mainly traced to China’s future coal use in electricity production. The study finds that it is beneficial to soft-link complex global models under harmonized assumptions. Although this study fails to "close the gap" between the two models completely, the experiences and insights shared here will be beneficial for researchers and policy makers that are drawing conclusions from the results of China and global scenario studies.
Introduction
Two classic modelling approaches are widely used to study the impacts of energy and climate policies on a future energy and economic system. Both approaches, either a macro-economic top-down (hereafter TD) or a technological bottom-up (hereafter BU) approach, have their own strengths and weaknesses [1], [2]. Strengths of BU models include their ability to represent a large number of discrete energy technologies in a partial equilibrium framework and to assess the trends and financial costs of different technological options. However, due to their structure, BU models are unable to capture the full macro-economic impacts of energy policies. By contrast, TD models, including computable general equilibrium (CGE) models, are widely used to represent interactions between national economic sectors and agents. TD models are able to calculate macro-economic costs and impacts of energy policies, however at the expense of specific sectoral or technological details. Not surprisingly, due to disciplinary and structural differences these two types of models demonstrate wide ranges of future projection of energy and emission trajectories at both global and national levels [3], [4], [5], [6].
To "narrow the gap" between computed results from TD and BU approaches, soft-linking existing TD and BU models emerged as a pragmatic methodology to assess a wide range of energy and climate policies [1], [7]. Soft-linking is typically one starting point in coupling existing TD and BU approaches due to its transparency, flexibility, and learning [8]. Various country-specific energy model soft-linking studies were recently carried out. Ref. [9] used the MARKAL-MACRO model to generate energy and emission scenarios for China until 2050 and concluded that a carbon emission ceiling is unacceptable for China due to its high economic costs. Ref. [10] illustrated low-carbon scenarios for the UK toward 2050 and found that the carbon abatement cost is lower in the soft-linked MARKAL-MACRO model than in the standalone BU MARKAL model. Ref. [11] soft-linked a CGE model with a TIMES model to analyze several low-carbon scenarios for Portugal and concluded that different models suggest different carbon reduction strategies. Ref. [12] demonstrated a highly detailed and transparent method for soft-linking a CGE model with a TIMES model for Sweden. Many studies integrated TD and BU models by representing detailed energy technologies within the TD framework in a specific sector, such as the electricity sector [13], [14], housing sector [15], or transport sector [16]. Other studies tried to complement each type model’s advantages and disadvantages by exchanging information between a detailed BU model with simplified macroeconomic module [17], [18]. Moreover, [19] a hard-linked AIM/Enduse model with a global AIM/CGE model resulted in hundreds of detailed energy end-use technologies in all sectors explicitly represented in a CGE model.
As the world’s largest emitter of CO2, China is a prominent and important global and national case study for scenario analysis. Numerous previous studies contributed to explore China’s future energy and emission pathways [20], [21], [22], [23]. These studies focused on national scenarios for China in a global context without exploring regional disparities in China’s economic development, industry structure, and population. While research on regional disparities in China is available, such as regional disparities in energy supply and consumption [24], these are not yet discussed in global context [25]. Furthermore the existing studies represent either the TD or the BU perspective. To the best of our knowledge, no global soft-linking exercise with focus on an improved regional analysis of China was carried out yet. Possible explanations for this research gap are the methodological challenges in combining China’s energy and economic data with commonly used international data [26]. In the future, BU and TD models will continue to be used as tools to explore plausible future pathways of China’s energy demand and CO2 emissions, especially in the light of China’s recent commitment to peak its emissions by 2030 [27]. Therefore, a better understanding of China’s future regional economy and energy system pathways in a global context, as it can be represented by different TD and BU models, is needed. In this context this study aims to fill the following research gaps:
- (a)to explore the future energy and emission pathways of China at a regional and a sectoral level with two newly developed global TD and BU models with regional China features;
- (b)to investigate how the two TD and BU models respond to identical economic and population projections;
- (c)to investigate how the two TD and BU models respond to identical climate policies, such as carbon price signals;
- (d)to analyze where and why main results differences between the models appear and if soft-linking both models allows "narrowing and closing this gap"; and
- (e)to transparently share experiences and insights for researchers and policy makers that are drawing conclusions from the results of China and global scenario studies.
For those purposes this study compares the results from a newly developed multi-region dynamic CGE model and a bottom up global energy system model, both of which include three economically different regions of China in their global framework. The socio-economic assumptions are harmonized and the results in the baseline scenario and carbon tax scenario are compared. The results from this study will feed into the ongoing dialogue on formulating future energy policies and setting carbon emission reduction targets, both within China and the United Nations Framework Convention on Climate Change.
The remaining sections of this paper are structured as follows: Section 2 describes the models and underlying databases used in this study. Section 3 summarizes how the socio-economic assumptions are harmonized between both models in their baseline and carbon mitigation scenarios. Section 4 presents the results at a global, a China national and a China regional level. Section 5 discusses where and why differences between the two TD and BU models exist and the challenges encountered in this study "to narrow the gap" between them. Conclusions and an outlook for further research are given in Section 6. More detailed model descriptions are provided in two appendices, the TD model is described in Appendix A and the BU model is introduced in Appendix B. All calculation results are provided as supplementary material.
Section snippets
Two global models with a regional China focus
Two global models that are able to represent similar regional features of China’s energy and economy are selected to provide new insights for China’s future energy system. A joint research effort between Japan’s National Institute of Environmental Studies, the Technical University of Denmark and China’s Energy Research Institute allowed to soft-link the AIM/CGE-China (hereafter CGE) model and the TIMES Integrated Assessment Model (hereafter TIAM). This CGE model is a global TD general
Soft-linking two global models
The soft-linking methodology applied here will serve as an analytical framework for our China-specific global model comparison and scenario analysis (see Fig. 1). This methodology builds on our previous work, which was limited in scope and tested only for the China regions in both models [25]. In this study we expand and document the iterative soft-linking methodology for these two global models and discuss our scenario assumptions for a subsequent model comparison with a focus on China.
Modelling results
One primary research question of this study is to analyze how different the results between TD and BU models under harmonized baseline assumptions are, and to what extent this gap could be narrowed or closed after soft-linking the models. This chapter shows the main results of our model comparison study, highlighting results for three regions of China in a global context, across three different models (the TD CGE model, the BU TIAM model and the soft-linked TD BU model).
We first show the future
How big is the gap – comparison of results with leading global studies
The main purpose of this study is to identify the gap between BU and TD models that are frequently used as tools to explore long-term scenarios of energy use and CO2 emissions. The focus of this section is to identify the gap between TD and BU results as a general trend rather than in highly detailed quantifiable results of CO2 energy and emissions projections.
Previous studies have found that BU and TD models demonstrate wide ranges of future energy and emission trajectories at both global and
Conclusions and outlook
This study compared the gap of China’s future energy and emissions from two newly developed global BU and TD models. A novel improvement from previous studies is that we harmonized the socio-economic assumptions that may govern the overall scale and trend of economic and energy indicators of the baseline scenario in both models. By doing this study compared the differences of both models at regional, national, global and sectoral levels in a highly standardized, transparent, and open-data
Acknowledgements
Financial support for this research from the following sources is gratefully acknowledged: (i) the Sino-Danish Centre for Education and Research (www.sinodanishcenter.dk); (ii) the Danish Idella Fonden; and (iii) the Environment Research and Technology Development Fund (A-1103) of the Ministry of the Environment, Government of Japan.
We would like to thank all Chinese colleagues from the Chinese Academy of Sciences, the Energy Research Institute of the National Development and Reform Commission,
References (53)
- C. Böhringer et al.
Combining bottom-up and top-down
Energy Econ
(2008) - C. Böhringer
The synthesis of bottom-up and top-down in energy policy modeling
Energy Econ
(1998) - D.P. van Vuuren
Comparison of top-down and bottom-up estimates of sectoral and regional greenhouse gas emission reduction potentials
Energy Policy
(2009) - N. Rivers et al.
Useful models for simulating policies to induce technological change
Energy Policy
(2006) - D. Wilson et al.
Exploring the gap: top-down versus bottom-up analyses of the cost of mitigating global warming
Energy Policy
(1993) - T. Martinsen
Introducing technology learning for energy technologies in a national CGE model through soft links to global and national energy models
Energy Policy
(2011) - W. Chen
The costs of mitigating carbon emissions in China: findings from China MARKAL-MACRO modeling
Energy Policy
(2005) - N. Strachan et al.
Hybrid modelling of long-term carbon reduction scenarios for the UK
Energy Econ
(2008) - R.D. Sands
Dynamics of carbon abatement in the second generation model
Energy Econ
(2004) - S. Proença et al.
Hybrid modeling to support energy–climate policy: effects of feed-in tariffs to promote renewable energy in Portugal
Energy Econ
(2013)
MESSAGE–MACRO: linking an energy supply model with a macroeconomic module and solving it iteratively
Energy
(2000)
Energy efficiency and the cost of GHG abatement: a comparison of bottom-up and hybrid models for the US
Energy Policy
(2011)
Development of a global computable general equilibrium model coupled with detailed energy end-use technology
Appl Energy
(2014)
Energy and emission scenarios for china in the 21st century-exploration of baseline development and mitigation options
Energy Policy
(2003)
The role of Asia in mitigating climate change: results from the Asia modeling exercise
Energy Econ
(2012)
The impacts of China’s household consumption expenditure patterns on energy demand and carbon emissions towards 2050
Energy Policy
(2012)
Mapping and benchmarking regional disparities in China’s energy supply, transformation, and end-use in 2010
Appl Energy
(2015)
Modelling tools to evaluate China’s future energy system – a review of the Chinese perspective
Energy
(2014)
Introduction and overview
Energy Econ
(2004)
The synthesis of bottom-up and top-down approaches to climate policy modeling: electric power technologies and the cost of limiting US CO2 emissions
Energy Policy
(2006)
The costs of limiting fossil-fuel CO2 emissions: a survey and analysis
Annu Rev Energy Env
(1993)
Hybrid modeling: new answers to old challenges introduction to the special issue of the energy journal
The Energy J
(2006)
Top-down and bottom-up modelling to support low-carbon scenarios: climate policy implications
Clim Policy
(2013)
A coupled bottom-up/top-down model for GHG abatement scenarios in the Swiss housing sector
Experiments with a hybrid CGE-MARKAL model
Energy J
(2006)
Cited by (90)
Comparison of health and economic impacts of PM2.5 and ozone pollution in China
2019, Environment InternationalOptimal green energy planning for sustainable development: A review
2017, Renewable and Sustainable Energy ReviewsCitation Excerpt :The developed model helps to analyze aggregate energy efficiency and sustainable development indexes. Dai et al. [21] investigated that the future energy consumption and CO2 emission pathways could be measured by top-down and bottom-up models for harmonizing the economic development. This model predicted the CO2 emission rate by using a soft-link methodology to narrow the gap between the researchers as well as policymakers.
Economic Impacts from PM2.5 Pollution-Related Health Effects in China: A Provincial-Level Analysis
2016, Environmental Science and Technology
- ☆
- This article is based on a short proceedings paper in Energy Procedia Volume 161 (2014). It has been substantially modified and extended, and has been subject to the normal peer review and revision process of the journal. This paper is included in the Special Issue of ICAE2014 edited by Prof. J Yan, Prof. DJ Lee, Prof. SK Chou, and Prof. U Desideri.
Copyright © 2015 Elsevier Ltd. All rights reserved.