Skip to main content

Measuring the Impact of Renewable Energy Sources on Power Sector Carbon Emissions in Germany—a Methodological Inquiry

Berechnung der Auswirkungen erneuerbarer Energien auf die CO2-Emissionen des Stromsektors in Deutschland – eine methodische Analyse

Abstract

Renewable energy sources for power production (RES) are an essential element of international climate change mitigation measures. Their contribution regarding emission reduction, however, cannot be directly measured. Several methods have been employed in the literature to calculate the emission reduction of RES, however, validating them is not possible. Hence, the question presents itself as to the relative advantages and disadvantages of the respective methods. To address this question, this paper examines the existing methodological approaches, namely (1) the displacement estimations, (2) an econometric approach and (3) optimization model-based dispatch calculations. In a first step, the respective approaches are discussed and quantitatively compared against each other. Subsequently, all methods are implemented for Germany for the years 2016 and 2017 and the specific emissions displaced by RES are calculated. The results indicate that all methods calculate CO2 reductions for wind onshore between 500–900 kgCO2 per MWh and for solar between 400–700 kgCO2 per MWh, indicating that each can provide valuable insights. For Germany, employing a dispatch model entails advantages since most drivers of energy system-related carbon emissions can be incorporated and the method can be applied to all RES technologies. In particular, the inclusion of cross-border electricity flows and the measurement of dynamic effects, two important aspects with possibly substantial effects on carbon emissions, can be incorporated.

Zusammenfassung

Erneuerbare Energien (EE) sind ein wesentliches Instrument der internationalen Maßnahmen zur Eindämmung des Klimawandels. Ihr Beitrag zur Reduktion der CO2-Emissionen ist jedoch nicht direkt messbar. In der Literatur werden verschiedene Methoden zur Berechnung der Emissionsreduktion von EE eingesetzt, deren Validierung jedoch nicht möglich ist. Daher stellt sich die Frage nach den relativen Vor- und Nachteilen der jeweiligen Methoden. Um dieser Frage nachzugehen, werden in dem vorliegenden Beitrag die vorhandenen Ansätze untersucht, nämlich (1) die Verdrängungsschätzungen, (2) ökonometrische Modelle und (3) Dispatch-Modelle (Optimierung des Kraftwerkseinsatzes). In einem ersten Schritt werden die jeweiligen Ansätze diskutiert und quantitativ miteinander verglichen. Anschließend werden alle Methoden für Deutschland für die Jahre 2016 und 2017 implementiert und die durch EE verdrängten spezifischen Emissionen berechnet. Die Ergebnisse zeigen, dass alle Methoden CO2-Reduktionen für Wind (onshore) zwischen 500–900 kgCO2 pro MWh und für Solar zwischen 400–700 kgCO2 pro MWh errechnen, was darauf hindeutet, dass jede Methode wertvolle Erkenntnisse liefern kann. Für Deutschland bringt der Einsatz eines Dispatch-Modells Vorteile, da die meisten Treiber der energiesystembezogenen CO2-Emissionen einbezogen werden können und die Methode auf alle EE-Technologien anwendbar ist. Insbesondere können die Auswirkungen des Handels sowie dynamischer Effekte, zwei wichtige Aspekte mit möglicherweise erheblichen Auswirkungen auf die CO2-Emissionen, berücksichtigt werden.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

Notes

  1. The range is the result of the different assumptions made concerning the displacement of RES of exports and hydro. For the lower value it is assumed that no additional carbon emissions are avoided by exports and hydro while for the upper value the displacement factor of coal is assumed.

  2. Exports and PSPs are allocated the average emission factor of natural gas and coal-fired power plants.

  3. For the econometric approach exports and PSPs are allocated to natural gas and coal-fired power plants equally.

  4. The main issue is the limited data availability from the power generation abroad especially in Eastern Europe.

  5. The significance of the effect of renewable energy generation is obtained by applying the Wald test for joint significance because the effect is composed of several coefficients. Newey-West robust standard errors are estimated to account for heteroskedasticity and autocorrelation.

  6. New CCGTs have an emission factor of 340 kgCO2/MWh while old hard coal power plants have an emission factor of 950 kgCO2/MWh.

References

Download references

Acknowledgements

The authors would like to thank Jascha Fischer for his helpful comments. The Authors would like to express our appreciation and thanks for the valuable comments and inputs. Further, we thank the reviewers for helpful comments and feedback on this work.

Funding

The here presented work is result of a research project founded by the German Environment Agency (FKZ 37EV 16 126 0). Preliminary stages of this research have been presented at Student Chapter of GEE and at the 42nd International Conference of the IAEE.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Carl-Philipp Anke.

Supplementary Information

Appendix

Appendix

Appendix I—Overview of ELTRAMOD

The electricity market model ELTRAMOD is an optimization model for the analysis of European electricity markets. Therefore, the model can determine both investment decisions and the power plant dispatch using a myopic approach with a time resolution of 8760 h. In its basic spatial resolution, the model includes the EU27 states as well as Switzerland, Norway and the countries of the Balkans, each with a detailed representation of regional energy supply structures. The respective power plant park, consumption curves and feed-in time series of RES (wind onshore and offshore, photovoltaic and run-of-river power plants) are deposited in an hourly resolution. Trading activities between the individual market areas are limited by net transport capacities (NTC). Similar to trading on the day-ahead market, network congestions within a market region are neglected (“copper plate”). The European Union Emissions Trading System for CO2 is taken into account by CO2 certificate prices. The fundamental model ELTRAMOD is based on a linear optimization with the targeted minimization of the decision-relevant system costs.

$$TC=\sum _{p,t}G_{p,t}\cdot c_{p,t}+LC_{p,t}^{UP}\cdot c_{p,t}^{UP}+LC_{p,t}^{\mathrm{DOWN}}\cdot c_{p,t}^{\mathrm{DOWN}}$$
(6)
P :

Set of power plants

T :

Set of time periods

TC :

Total Costs

\(G_{p,t}\) :

Generation of power plant p

\(c_{p,t}\) / \(c_{p,t}^{UP}\) / \(c_{p,t}^{\mathrm{DOWN}}\) :

Cost of generation/Load change

\(LC_{p,t}\) :

Load change of power plant p

For this purpose, the actual production capabilities of the technologies are represented by technical and economic restrictions. The technology-specific parameters include e.g. efficiency, emission factors and availability. The model also takes into account the sector coupling to the heat area and thus the coverage of the heat generated by the generating units by means of heat demand time series and installed power plants with a minimum generation for heat extraction (cf. Eq. 5).

$$G_{p,t}\geq \mathrm{inst}_{p}\cdot \mathrm{chp}_{p,t}\cdot \text{avail}_{p,t}$$
(7)
\(\mathrm{inst}_{p}\):

Installed generation capacity

\(\mathrm{chp}_{p,t}\):

CHP factor

\(\text{avail}_{p,t}\):

Power plant availability

The high-temporal resolution in ELTRAMOD enables the mapping of intertemporal relationships in the use of generating and storage units, taking into account the feed-in and load conditions that occur. The model contains reservoir and pumped storage plants. In order to be able to reflect adequately the properties of a pumped storage plants, both the pumping and generating as well as maximum storage levels are modelled individually.

Appendix II—Validation of the Dispatch Model

Fig. 10 compares the historical dispatch of the power generation portfolio with the modelled dispatch for Germany. It is shown that in particular the base load technologies and the generation from gas-fired power plants are well matched. The deviations that occur are below 2%. This means that in particular the technologists which effect the calculation of the displacement effects are modelled very well. However, there are minor deviations in PSPs generation in particular. This is partly due to the chosen model approach, which underestimates the need for flexibility, as not all technical restrictions of reality are represented.

Fig. 10
figure 10

Comparison of the German power mix in 2016 and 2017 [TWh]

Appendix III—Overview of Input Data

Table 5 Overview of source for the input data

Appendix IV—Input Data for Regressions Analyses

Table 6 Summary statistics for dependent and independent variables

Appendix V—Results of Econometric Models

Table 7 Total substitution effects retrieved from econometric models
Table 8 Average substitution effects retrieved from econometric models

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Anke, CP., Schönheit, D. & Möst, D. Measuring the Impact of Renewable Energy Sources on Power Sector Carbon Emissions in Germany—a Methodological Inquiry. Z Energiewirtsch 45, 1–23 (2021). https://doi.org/10.1007/s12398-020-00292-8

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12398-020-00292-8

Keywords

  • Carbon emissions
  • Emission reduction
  • Renewable energies
  • Energy system analysis
  • Power system modelling