«Project no: 502687 NEEDS New Energy Externalities Developments for Sustainability INTEGRATED PROJECT Priority 6.1: Sustainable Energy Systems and, ...»
SIXTH FRAMEWORK PROGRAMME
Project no: 502687
New Energy Externalities Developments for Sustainability
Priority 6.1: Sustainable Energy Systems and, more specifically,
Sub-priority 22.214.171.124.5: Socio-economic tools and concepts for energy strategy.
Technical paper n° 4.1 - RS Ia
“Development of parameterisation methods to
derive transferable life cycle inventories” Technical guideline on parameterisation of life cycle inventory data
Due date of paper:
Actual submission date: 28.02.2009 Start date of project: 1 September 2004 Duration: 48 months Organisation name for this paper: PSI Authors: Thomas Heck, Christian Bauer, Roberto Dones.
(With contributions from Sven Gärtner, Peter Viebahn, Oliver Mayer-Spohn, Markus Blesl).
Project co-funded by the European Commission within the Sixth Framework Programme Dissemination Level Public X PU Restricted to other programme participants (including the Commission Services) PP Restricted to a group specified by the consortium (including the Commission RE Services) Confidential, only for members of the consortium (including the Commission CO Services) 1 Contents Summary
2 Why parameterisation of LCA ?
3 Parameterisation of LCA data - General framework
4 Examples of time-, technology-, and space-dependent parameters
4.4 External costs results example
5 Selected parameters of electricity generation systems
5.1 Advanced fossil
5.4 Solar thermal power
5.7 Fuel cells
5.9 Wave and tidal power
BOP Balance of plants CC Combined Cycle CCS Carbon Capture and Storage CHP Combined Heat and Power DNI Direct Normal Irradiation ECLIPSE Environmental and eCological Life cycle Inventories for present and future Power Systems in Europe EIA Environmental Impact Assessment GCC Gas Combined Cycle GHG Greenhouse Gas GIS Geographical Information System IGCC Integrated Gasification Combined Cycle IPCC Intergovernmental Panel on Climate Change kWh kilo Watt hour kWhe kilo Watt hour electricity LCA Life Cycle Assessment LCI Life Cycle Inventory LCIA Life Cycle Impact Assessment LEC Levelised Electricity Costs LHV Lower Heating Value LRV Luftreinhalteverordnung NEEDS New Energy Externalities Developments for Sustainability NG Natural Gas NMVOC Non-methane Volatile Organic Compounds PAH Polycyclic aromatic hydrocarbons PM Particulate Matter PM10 Particulate Matter with diameter up to 10 micro-metre ppm parts per million PV Photovoltaics 3 Summary Traditionally, life cycle analyses and life cycle impact analyses do not consider space- and time-dependencies. In reality, the environmental performance of energy technologies may vary in space and time, while their main characteristics remain the same. The objective of this work package was the outline of a parameterisation method that facilitates the description of space- and time-dependent life cycle data for energy systems.
In the LCA part of the NEEDS project, the future development of electricity generation technologies and LCA background processes up to the year 2050 has been assessed. The coverage of a broad spectrum of future technologies and a long time scale based on the framework of the large ecoinvent database is a substantial achievement for life cycle assessment of energy systems.
The present technical report provides some new ideas and proposes new methodologies on parameterisation of LCA modelling that are intended to support further extensions of space and time coverage beyond what has been achieved already within the NEEDS LCA modelling. The intention is also to contribute to the improvement of assessment of space- and time-dependent impact and external cost effects in connection with LCI data. Environmental impacts and external costs depend a lot on site-specific conditions. Therefore a higher spatial differentiation of LCI modelling compared to current models is desirable.
Firstly, general aspects of an advanced parameterised LCA system are discussed. It is proposed that the connection of the LCA model to a Geographical Information System (GIS) should be considered because several spatial parameters can be treated systematically in a GIS software. A couple of explicit examples of parameters relevant for energy systems under the perspective of space-dependency, time-dependency and technology-dependency that could be implemented into an advanced LCA system are provided. For a number of advanced electricity generation technologies, overviews on important space- and time-dependent parameters are given. The coverage and depth of the discussion varies for the different energy systems according to the estimated practicability and relevance for LCA. Parameters provided within the NEEDS project for the different technologies are also considered where appropriate.
Generally, it can be concluded that the possibility and appropriateness of parameterisation depends much on the specific energy system, in particular for the space-dependency. The focus of the present work was more on the variety of parameters that have to be considered rather than completeness. The implementation of the proposed methods up to a running advanced LCA model would require deeper investigations of the variety of energy systems and background processes and would need substantial resources. The implementation can proceed in an iterative way because already a partial parameterisation can be advantageous for further extensions of LCA models and databases in view of the large number of parameters in present LCA modelling.
4 1 Introduction The environmental performance of energy technologies may vary in space and time, while their main characteristics remain the same. The objective of this work package was the outline of a parameterisation method that facilitates the description of space- and time-dependent life cycle data for energy systems. Results are intended to support the efficient specification of region- and time-specific LCI datasets in reasonable resolution. This will improve the possibilities to assess space- and time-dependent impact and external cost effects.
2 Why parameterisation of LCA ?
The Life Cycle Assessment (LCA) part of the NEEDS project (stream RS1a) has been based on the ecoinvent database (www.ecoinvent.com). Ecoinvent is probably the most comprehensive life cycle inventory (LCI) worldwide. It includes inventory data on energy supply, material supply, transport services, chemicals, metals, agriculture, waste management services, and resource extraction. The environmental part of the database includes emissions to air, to water, and to soil as well as land use and resources taken from nature.
Technically, the ecoinvent database has essentially the structure shown in Fig. 2.1. The different research groups contribute input to three types of data which are then organised in three matrices. Firstly, the input process matrix (sometimes called “technosphere matrix”) links a process to other processes of the technosphere. The technosphere processes refer to energy systems, chemicals, transport, agriculture, etc. Secondly, the elementary flow matrix (sometimes called “biosphere matrix”) describes the use of resources from nature and the emissions to nature for each process. Finally, a set of impact assessment and valuation methods is included in the LCIA (life cycle impact assessment) matrix. For the LCA stream in the NEEDS project, mainly external costs have been discussed. The major outputs for the users are the cumulative LCI results which include the direct and indirect flows from and to nature for each process and the cumulative LCIA results for each process.
The input and output of the database uses the so-called “EcoSpold” format (see www.ecoinvent.com). The definition of the EcoSpold format during the development of ecoinvent was a big breakthrough for the communication of LCI data between different software products. As a consequence, the ecoinvent data has been included in many LCA software packages (Umberto, GaBi, Regis, EMIS, Green-e, Bilan Produit, WRATE, SimaPro) in order to serve as a basis for subsequent LCA studies.
Fig. 2.1 Basic structure of the current ecoinvent LCI and LCIA database (m = number of processes, n = number of flows to and from nature, r = number of LCIA methods).
The advantage of the relatively simple data scheme is that different research groups can quickly set up new LCA data in the EcoSpold format and combine knowledge from different research fields.
Nevertheless, as it is, the organisation of large datasets has also disadvantages as one can see from the mere number of entries included in the database. Tab. 2.1 shows the number of input entries for the LCI part of ecoinvent version 1.1 which was the starting point of the NEEDS project. The database comprised already more than 65’000 entries which are formally treated as independent parameters for the calculation of cumulative results. Additionally, the LCI database is supplemented by about 200 LCIA methods (on the subcategory level). NEEDS and other projects further extend the number of processes and entries. Furthermore, different time series for future scenarios have been investigated in the NEEDS project.
Tab. 2.1 Size of the LCI part of the ecoinvent database (v 1.1)
The increasing complexity of the databases is difficult to handle. Moreover, extending the datasets in space and time, while keeping control over consistency and correctness, becomes more and more difficult.
Among the huge number of input parameters, not all are really independent, and not all are really relevant with respect to final results.
6 Thus an incentive for the parameterisation concept is the attempt to reduce the number of parameters in particular for further significant extensions of the databases.
An appropriate parameterisation can also facilitate the establishment of self-consistent LCA databases for different future scenarios. Ideally, a self-consistent database should guarantee that key assumptions about the future development of technologies are not contradictory. For example, the effort for improvements in a technology may depend on the production output in the sense of economies of scale or experience curves. A scenario which is optimistic for a certain technology may be pessimistic for a competing technology. For example if the scenario assumes that renewable energies will strongly expand and the relative share of fossil energy carriers will be reduced, this may imply the fast improvement of renewable energy technologies but at the same time diminish the pressure for improvements of fossil technologies. Via the LCA chain the influence on input materials and other processes can be complex (for example the scenario assumptions about the future capacities of photovoltaics do not only influence the electricity mix but also the production of solar grade silicon in comparison with electronic grade silicon). Currently it is pretty difficult to set up detailed LCA scenario databases while keeping up the consistency over the full chain. A wellstructured parameterisation could improve the situation e.g. if parameters like assumed installed capacities could be used directly in the LCA modelling.
To summarise, the major goals are the following:
• Improvement of transparency,
• Reduction of redundancy,
• Improvement of error checking / consistency checking,
• Facilitation of transferability and generalisation of the LCI data across regions and for different time horizons and different scenarios.
7 3 Parameterisation of LCA data - General framework Traditionally, life cycle analyses and life cycle impact analyses do not usually consider spaceand time-dependencies. Thus the envisaged parameterisation is a pioneering task, in particular in view of the very detailed technological description of the involved systems in present LCA modelling. Consequently, basic methods for parameterisation have to be built up at first. The first task had been the development of a preliminary methodological concept for the parameterisation.
An aspect was the identification of key parameters which are most important for LCA results with respect to space and time. The goal is to find the parameters that are driving future changes and spatial variations of life cycle data.
Because of the very large number of parameters in the full LCA database, a systematic analysis of all parameters of energy systems was far beyond the scope of this work package.
Rather a couple of systems and some key parameters have been selected in order to discuss illustratively the possibility of a parameterisation of LCA data.
A questionnaire on space- and time-dependent parameters in life cycle inventories was developed and distributed within the LCA stream. The goal was to collect specific information on space- and time-dependency related to the single energy systems from the corresponding technology expert groups. The answers to the questionnaire have been analysed. A clear result is that the possible parameterisation will depend very much on the specific energy system, in particular for the spatial parameters.
Steps towards a parameterised representation of LCA data have been made already in the ECLIPSE project for emerging energy technologies (solar photovoltaics, wind, fuel cells, biomass and small combined systems). The idea was to enable end users to take into account the influence of geographic conditions and to model technological improvements (ECLIPSE 2004).
Parameters can be simply representatives for single numeric values. For example, the LCA software SimaPro (www.pre.nl) allows the definition of parameters by name and associated value. The parameter name can be used in formulas which are evaluated within SimaPro.
Thus dependent parameters can be reduced to functions of the defined parameter set.