View all publications

3 Scope of the LCA

The scope of the study has addressed the following items:

  • the functions of the product systems and the functional unit;
  • the product systems to be studied and system boundaries;
  • allocation procedures;
  • types of impact and methodology of impact assessment, and subsequent interpretation to be used;
  • data requirements;
  • assumptions and limitations;
  • initial data quality requirements; and
  • type of critical review. 

3.1 Product Function

The function of the system describes the performance characteristics of the product systems being compared.

The general function of the lamps being studied is to provide lighting for different applications for private, commercial and industrial applications, in the visible light range 400nm to 800nm (ELCF, 2009).  The lamp types and technology are summarised in Table 1.

Table 1 Summary of Lamps Assessed

Lamp family

Lamp type


Main uses

Brief Description

Fluorescent Linear fluorescent lamp (LFL) Linear fluorescent lamp (LFL) Commercial and Industrial Uses electricity to excite mercury vapour in argon or neon gas, resulting in a plasma that produces short-wave ultraviolet light.  This light then causes a phosphor to fluoresce, producing visible light.  The blend of phosphors controls the colour of the light, and along with the lamp's glass, prevents the harmful UV light from escaping.
  Compact fluorescent lamp (CFL): external (CFLe) and integrated (CFLi) Compact fluorescent lamp (CFL): external (CFLe) and integrated (CFLi) Commercial Residential Operate on the same principles as linear fluorescent, above.  A compact fluorescent lamp may have a conventional ballast located in the fitting (CFLe) or they may have a ballast integrated (CFLi) in the lamp, allowing them to be used in fittings normally used for incandescent lamps.
High Intensity Discharge (HID) High pressure sodium (HPS) High pressure sodium (HPS) Public lighting Industrial Uses sodium in an excited state to produce light.  An amalgam of metallic sodium and mercury lies at the coolest part of the lamp and provides the sodium and mercury vapour in which the arc is drawn.  Because of the extremely high chemical activity of the high pressure sodium arc, the arc tube is typically made of translucent aluminium oxide (alumina). 
  Metal Halide (MH) Metal Halide (MH) Industrial An electric arc is passed through a mixture of gases.  The compact arc tube contains a high-pressure mixture of argon, mercury, and a variety of metal halides.  The mixture of halides affects the nature of light produced, influencing the correlated colour temperature and intensity.
  Mercury Vapour (MV) Mercury Vapour (MV) Industrial The arc discharge is generally confined to a small fused quartz arc tube mounted within a larger borosilicate glass lamp.  The outer lamp may be clear or coated with a phosphor; in either case, the outer lamp provides thermal insulation, protection from ultraviolet radiation, and a convenient mounting for the fused quartz arc tube

Source: (Cited directly from Stewardship Solutions, 2008).

3.2 Functional Unit

The functional unit is the reference unit used to report the inventory analysis and impact assessment results for each lamp type.  As the primary focus is to appraise each lamps life cycle with different end-of-life product stewardship options individually, the functional unit is defined as 100,000 hours of operation in the use phase, for the performance characteristics (of lamp life, output and application) as described in Table 2.  Selecting a time period over which the lamps operate provides a fair basis for assessment of the product stewardship options.  A period of 100,000 hours was chosen because this covers, for all lamp types, more than one total lamp lifespan and therefore includes the implications for variations at end-of-life and other phases of the life cycle. 

The interpretation section of the study has considered findings in relation to, total national lamp usage, lamp performance and establishes where product stewardship messages are common across lamp types, or are in fact different for the individual lamps. 

As mentioned, but highlighted here again, the functional unit does not provide for comparative results across the different lamp types.  This is because the different lamp types offer different functions and performance characteristics relating to lamp lifetime, lamp output (in terms of wattage) and in terms of lamp application. 
The typical performance parameters of the lamps assessed in the study are shown in Table 2 for product life span (measured in hours operation), power rating (measured in watts) and application. 

Table 2 Typical Lamp Performance Parameters

Product System

Lamp Specification


1. Linear fluorescent lamp (LFL)

  • Life span: 8000 hours
  • Output: 35W, T8
  • Application: Commercial and industrial
  • Mercury content: 4mg

2. Compact fluorescent lamp (CFL): external ballast

  • Life span: 10, 000 hours
  • Output: 11W
  • Application: Commercial residential
  • Mercury content: 5mg

3. Compact fluorescent lamp (CFL): integral ballast

  • Life span: 12, 000 hours
  • Output: 20W*
  • Application: Commercial residential
  • Mercury content: 5mg
High Intensity Discharge (HID):

4. High pressure sodium (HPS)

  • Life span: 20, 000 hours
  • Output: 150W
  • Application: Public lighting and industrial
  • Mercury content: 50mg

5. Metal Halide (MH)

  • Life span: 20, 000 hours
  • Output: 400W
  • Application: Industrial
  • Mercury content: 50mg

6. Mercury Vapour (MV)

  • Life span: 20, 000 hours
  • Output: 250W
  • Application: Industrial
  • Mercury content: 50mg

Source: Adapted from (Stewardship Solutions, 2008). 

* This figure has been updated from that shown in the Stewardship Solutions (2008) report of 11W based on manufacturer feedback of which lamp wattages represent high volume sales in New Zealand. 

Table 3 shows the number of lamps that are manufactured per 100,000 hours of in-use operation for each lamp. 

Table 3 Number of Lamps Manufactured per 100,000 hours operation



1. Linear fluorescent lamp (LFL)
35W, T8

2. Compact fluorescent lamp (CFL)
11W: external ballast

3. Compact fluorescent lamp (CFL)
20W: integral ballast

4. High pressure sodium (HPS)

5. Metal Halide (MH)

6. Mercury Vapour (MV)

Number of lamps manufactured each 12.5 10.0 8.3 5.0 5.0 5.0

Table 4 shows the typical lamp composition of each lamp type in New Zealand (Stewardship Solutions, 2008) that will be assessed in the study. 

Table 4 Typical Lamp Composition (grams)

Product System







Fluorescent: g g g g g g

1. Linear fluorescent lamp (LFL)

120 115 3 -- -- 2

2. Compact fluorescent lamp (CFL): external ballast

55 40 3 -- 10 2

3. Compact fluorescent lamp (CFL): integral ballast

120 65 4 25 25 1
High Intensity Discharge (HID):            

4. High pressure sodium (HPS)

150 105 44.5 -- -- --

5. Metal Halide (MH)

240 195 42 -- -- 3

6. Mercury Vapour (MV)

166 130 9 1 27 --

Note: The table indicates typical values for New Zealand (Stewardship Solutions, 2008, Sylvania, 2002). 

The functional unit accounts for differences in the manufacture of the different lamp compositions, lamp use, lamp life, and disposal, as well as the impacts associated with alternate lamp recycling and recovery options and mercury-content. 

3.3 Product Systems and System Boundaries

3.3.1 System Boundaries

The system boundary determines which life cycle stages and unit processes shall be included within the LCA.  The boundary also determines the environmental releases and inputs to be included.  The product system has been modelled in such a manner that the inputs and outputs at its boundary are elemental and product flows2.

The lamp systems investigated include all significant processes, tracing material and energy flows to the point where material and energy are extracted or emitted to the natural environment.  Waste management processes are also assessed, including the disposal of lamps via landfilling and recycling.  

The following exclusions are made due to these being the same for each typical lamp type as described in further detail below in this Section:   

  • product fittings and control gear; and
  • product packaging. 

The activities that are included in the life cycle for each lamp system are shown in Figure 1. 

Figure 1 Summary Diagram of System Boundaries of the Lamp Systems Assessed


Shows the activities that are included in the life cycle for each lamp system. Life cycle stages covered include the following: raw material production, lamp production, lamp distribution, lamp retail, usage and waste management (incineration, landfill and recycling).

The diagram shows the extraction of resources from the environment (at the top), the consumption of energy and other product systems, and the emissions to the environment (at the bottom.

The study has included the following life cycle stages:

  • Raw material production, including:
    • Mining of minerals. The mining of metals, such as phosphor, mercury, sodium and aluminium, is included in the study.  The materials and energy used for mining and transportation, as well as associated emissions, are assessed.  The mining of metals is often associated with the extraction of more than one metal; as such the environmental burden is expected to be allocated between the different outputs associated with mining.
    • Other raw materials.  The LCA study includes impacts of extraction and production of all other raw materials used in the lamp products, such as steel, glass and paper.
  • Transport of raw materials.  The transport of raw materials from the point of extraction to the point of use in lamp production is included. 
  • Conversion processes.  The LCA study also includes the conversion processes associated with all raw materials used in lamp production.  For example, conversion of the metal ores to forms suitable for lamp production.  This includes processes such as metal purification and metal production. 
  • Lamp production.  Processes for the manufacture of each of the lamp types (as listed in Section 2.1), have been assessed.  For each lamp type, the inputs of energy and raw materials consumed, as well as emissions, solid waste and waste water treatment, are included.  This study also accounts for any internal recycling of production waste.
  • Lamp distribution.  An average distribution distance for lamp distribution from production, via importers and distributors, to retail and to use is included in the assessment.  This relates to domestic and commercial use in New Zealand.
  • Lamp consumer use.  The use of lamps is included in the study.  Lamp use is defined by the specification for the functional unit, as described in Section 3.2.    The study has accounted for New Zealand electricity generation mix. 
  • Lamp collection and end-of-life management.  The impact of disposal of the lamps has been investigated in the study.  Including the impacts associated with collection and sorting of the lamps prior to domestic and commercial disposal, and recycling and recovery operations in New Zealand and in Australia.

Those steps excluded from the study boundary were:

  • Fitting and control gear.  The production, packaging, use and disposal of fittings, and control gear that is associated with each lamp type have been excluded from the study.  This is excluded because these components are identical within each lamp type when comparing alternative product stewardship scenarios.  As mentioned in Section 2.1, the goal of the study is to assess the alternative product stewardship options within each individual lamp type.  As such, those activities that are identical for each lamp type may be excluded from the study.  Sensitivity analysis has been used to estimate the significance of this exclusion.  
  • Manufacture, maintenance and decommissioning of capital equipment.  The manufacture, maintenance and decommissioning of capital equipment, such as buildings or machines, are not included in the investigated product systems.  The reason for excluding capital equipment, besides the practical aspects, is that the environmental impact related to the functional unit is considered negligible.  For example, the life time of the capital goods is usually much longer than the life time of the product under study.  Furthermore one production machine or building will be used to mass produce multiple products.  This means that environmental impacts associated with capital goods will need to be allocated across many products and many years of production, making the impacts of capital equipment insignificant.
  • Workforce burdens.  This study has assumed that the impact of the working environment (e.g. the impacts associated with human labour during production) will be insignificant for the studied scenarios and have, therefore, been excluded.  Workforce burdens includes in the energy required for people to do manual work. 

Section 3.5 describes the study limitations, inclusions and exclusions in detail.

3.4 Allocation Procedures

Some processes may yield more than one product and they may also recycle intermediate products or raw materials.  Where this occurs, the LCA study has to allocate material and energy flows, as well as environmental releases, to the different products in a logical and reasonable manner.

According to the ISO standard, allocation should preferably be avoided, which can be achieved through system expansion.  System expansion for recovery and recycling is further described below. 

Where system expansion is not practicable, the inputs and outputs of the inter-related processes are allocated in a manner that reflects the underlying physical relationships between them.  There may be certain circumstances where this is not appropriate or possible when carrying out an LCA study.  In such cases, alternative allocation methods were used and these are documented in the inventory analysis.

Where alternative allocation procedures are applicable then a sensitivity analysis has been conducted to illustrate the consequences of alternate approaches.

3.4.1 System Expansion for Recycling and Recovery

System expansion should be applied in the study when products are recovered, through recycling or energy recovery (ISO, 2006a).  For instance, when recycling of lamps occurs, metals from the lamps are extracted and reused in the recycling process, and an ‘avoided’ production of metals has been included in the system.  ‘Avoided’ production refers to materials not having to be generated from other sources, since the recycled material is used in its place. 

Figure 2 shows the system boundaries of the expanded system.  The processes shown in shaded grey indicate the additional activities that are included due to expansion of the system to include the benefit associated with avoided materials.

Figure 2  System Diagram Showing the System Expansion

Shows the system boundaries of the expanded system. The following processes are the additional activities that are included due to expansion of the system to include the benefit associated with avoided materials: extraction/processing of substituted energy source from incineration; extraction/processing of substituted energy source from landfill; and extraction/processing of substituted materials from recycling.

Where closed loop recycling occurs (as is presented in sensitivity analysis section) the inventory data shall reflect those inputs and emissions arising from the product specific recycled material content or recycling rate, as shown in the calculation given below (PAS2050, 2008):

Emissions / unit = (1 - R1) x EV + (R1 x ER) + (1 - R2) x ED


R1 = proportion of recycled material input;
R2 = proportion of material in the product that is recycled at end-of-life;
ER = emissions arising from recycled material input per unit of material;
EV = emissions arising from virgin material input per unit of material; and
ED = emissions arising from disposal of waste material per unit of material.

3.5 Inclusions, Exclusions and Assumptions

It is important to identify the main assumptions and limitations of the LCA so that the study results can be interpreted appropriately and in context. 

The results contained in this report are based on the following main assumptions:

System boundary:

  • Life cycle stages.  The study includes all main phases of the life cycle from cradle-to-grave, as described previously. 
  • Lamp composition.  The material composition of an average lamp varies due to the many different manufacturers and designs being available on the market.  This study has defined a typical material composition of a mercury-containing lamp based on the report titled New Zealand Lighting Industry Product Stewardship Scheme PHASE 1: Assessment and Review (Stewardship Solutions, 2008) which provides estimated material compositions for New Zealand.  That report provides a generic level material breakdown for lamps (as shown in Table3).  That report also identifies the difficulty in determining an average lamp composition; however, through use of similar European studies and in agreement with the Lighting Council New Zealand (which represents the major New Zealand lighting importers) a typical material composition was determined which has been used for this study.  Additionally, this composition has been supplemented by data from manufacturers material safety data sheets (MSDS) which identify levels of toxic ingredients in the lamp composition, such as heavy metals (Philips, 2001, 2002 and 2005 and Sylvania, 2002).  Gases used in the lamps have been omitted from the study. 
  • Lamp manufacture.  All the lamps are manufactured overseas.  ERM has contacted all New Zealand based importers of mercury-containing lamps.  However, there was a lack of specific data to describe the production of each lamp type.  The results have therefore estimated electricity input for production based on a report by European Lamp Companies Federation (2008b) which provides a figure for electricity usage for CFL lamp manufacture.  This is likely to be an overestimate.  For other lamp types, this has been estimated based on mass of each lamp and electricity usage for a CFL.  The electricity mix is based on the countries of origin of manufacture for each lamp type.  Raw material inputs are included based on composition and 99% conversion efficiency.    
  • End-of-life disposal and recycling.  This study has focused on lamp collection and disposal in New Zealand and Australia.  For recycling, primary data has been gathered to describe the collection and transfer of waste lamps in New Zealand and the processes for recycling and recovery in Australia.   All lamps that are not collected for recycling are sent for landfill disposal in New Zealand.  The emissions from landfill have been generated by using ecoinvent (2007) landfill models which have been manipulated to reflect specific conditions for New Zealand and to account for lamp composition.  No lamps are disposed of to incineration. 
  • Avoided materials.  ‘Avoided’ materials refer to those raw materials that do not have to be produced from other sources, because the recycled material is used in its place.  For lamps, avoided materials include the following that are produced from recycling processes in Australia: aluminium; glass granulate (for insulation); mercury (for dental amalgam); plastics; and phosphorus.  The avoided impacts for these materials have been estimated using adapted ecoinvent (2007) datasets to represent location of material production and use.   
  • Electricity generation.  Electricity generation in New Zealand is based on a generation mix from the Ministry of Economic Development (MED, 2007) and inventory data to describe these processes uses adapted datasets from ecoinvent (2007).  The datasets account for the composition of coal in New Zealand, for efficiency of NZ coal fired power stations.  Surrogate data were used for geothermal electricity generation in the absence of a specific inventory (refer to Appendix B for further details).  Generation from gas accounts for generation efficiency and onshore/offshore gas supply.  Coal generation is of particular interest due to the release of mercury from combustion of coal for electricity generation.  For electricity generation in other countries, such as China and Thailand for lamp production and Australia for lamp recycling, the generation mix is based on International Energy Agency (IEA, 2009) data and ecoinvent datasets. 

Lamp performance:

  • Lamp life.  This study has assumed for each typical lamp type an average lifetime (hours of operation) based on published sources, as shown in Section 3.2.  However, differences in design, such as high-end or low-end products, will result in potential differences in lamp lifetime.  Lamp lifetime has been tested in the sensitivity analysis, to indicate the change in impacts of a reduced or extended lamp lifetime of ±50%.
  • Lamp mercury level. The study has assumed typical lamp mercury content based on published sources, as shown in Section 3.2.  These levels may potentially vary depending on regulated levels in country of manufacture and on differences in lamp design.  This assumption has been tested in the sensitivity analysis in order to indicate the change in impacts of increased or reduced mercury content in the lamp.  It has been assumed that future technologies which deliver lower mercury level in lamps will provide the same functional performance (of lifetime, wattage and light quality).  Refer to Appendix B, Section B11.1 for further details.
  • Lamp energy consumption.  Some research indicates (Parsons, 2006) that there is a warm-up effect for fluorescent lamps which increases power required in the initial period of operation compared to rated power.   The study does not account for these effects, but uses the rated lamp wattage to determine energy consumption.  Lamp warm-up effects have been considered in a sensitivity analysis in order to establish their potential significance for the environmental impacts appraised.

Cut-off criteria:

  • Process inputs.  In order to maintain a feasible scope for the study, and in conformance with the ISO standard, ERM has applied ‘cut-off criteria’ to ensure focus is directed at the major environmental contributors for each lamp type.  In the case of individual life cycle stages and processes material inputs known to be less than 1% by mass of the product output of a particular process, and for which no appropriate inventory data are available, have been excluded.  However, the total cut-off was not more than 5% of input of all materials in reference to the functional unit.  Ideally, cut-off criteria should be based on environmental relevance.  However, cut-off by mass has been applied as it is impractical to define cut-off criteria based on environmental impact, since data for a process need to be collected in order to understand the environmental impact of that process or the entire life cycle.  Nonetheless, where considered relevant, such as in the case for potential ecotoxicity impacts, a test of environmental significance has been used based on professional judgement.  Material inputs excluded through cut-off are documented and reviewed for significance.

Impact categories:

  • Land occupation.  The study excludes the environmental implications of land occupation and use, for example, we excluded the implications of alternative land use and the effects of land use changes between the options. 
  • Litter.  Littering is a potential environmental impact.  However, this is difficult to quantify both as an environmental impact but also as an aesthetic impact. Therefore, this impact has been excluded from the LCA study.

3.6 Data Collection

The data required to perform the LCA are listed in the following Sections, which describes:

  • data categories to be collected;
  • primary data requirements;
  • secondary data requirements; and
  • data quality requirements.

The lamp systems assessed in the study are representative of those available on the New Zealand markets.  To ensure representative product systems, detailed questionnaires have been sent to lamp manufacturers and lamp collection and recycling companies.  Where specific production processing data for a lamp were not available, secondary data has been used, together with estimates based on the data gathered for the other lamps.

Appendix B provides a summary of the data sources, assumptions and limitations used for the LCA study. 

3.6.1 Data Categories

The following data categories are included in the study:

  • raw materials;
  • chemicals;
  • transport;
  • energy;
  • other physical inputs (e.g. water);
  • products and co-products;
  • emissions to air, water and soil;
  • solid waste; and
  • waste water.

3.6.2 Primary Data Requirements

Specific data were collected for the production of the lamps and for the generation of wastes and emissions in each of the main life cycle stages.

Specific data were required for the following:

  • lamp production operations (based on published data and questionnaires);
  • transport distances of raw materials to the lamp production facilities, and types of transport;
  • electricity generation mix for the relevant countries; and
  • lamp collection and waste management.

As mentioned in Section 3.5, all lamps are manufactured outside of New Zealand and are imported into the country (primarily from China and Thailand).  Primary data were not available relating to emissions, raw materials used; solid waste generated and waste water data from the production of lamps is anticipated to be limited for New Zealand conditions.  Manufacturers have been contacted as part of the study, however, due to a lack of data this has been estimated based other published sources for Europe.  Raw materials have been included using the composition of the lamps.  Previous studies of fluorescent lamps, such as those published by European Lamp Companies Federation (2008) and Ramroth (2008) indicate that the manufacturing phase of the life cycle contributes approximately 5% of total life cycle impacts across the measured environmental indicators. 

Transport distances have been determined based on specific country locations and estimated distances for retail and use in New Zealand. 

Specific data on electricity mix in the geographical area where the lamp is produced, used and disposed of has been sourced from International Energy Agency (IEA, 2009) data and from the Ministry of Economic Development (MED) (2008) for New Zealand.

As mentioned in Section 3.5, waste management data for New Zealand and the recycling operations that occur in Australia have been sourced from questionnaires to waste operators.  The availability of data to describe landfilling and incineration of lamps and processing wastes for New Zealand and Australia has been based on ecoinvent (2007) datasets that have been amended to reflect New Zealand and Australian conditions.  The final results will describe specific conditions for New Zealand and account for lamp composition. 

3.6.3 Secondary Data Requirements

For the production of commodity materials and energy, and the operation of transport processes, generic data were used where no specific information was made available by the data suppliers.   

Secondary data has been used for the following:

  • production of raw materials and chemicals;
  • general waste management operations;
  • electricity generation methods; and
  • emission data from transport.

Secondary data have generally been sourced from relevant databases or published sources, such as the Ecoinvent (2007) database, International Journal for LCA, US life cycle inventory (LCI) databases (NREL, 2008, EPA, 2008) and the Australian LCI database (AusLCI, 2008).  For example, the ecoinvent database contains international industrial life cycle inventory data on energy supply, resource extraction, material supply, chemicals, metals, agriculture, waste management services, and transport services, including any offset benefits of avoided production through recycling operations. 

The geographical base for most of the ecoinvent database is European, but for a limited number of inventory data, world data are available.  The data represents the most up-to-date and available public inventory data.  Where relevant, processes used from the generic databases have been adapted to represent the appropriate geography, by adjusting electricity generation mix and transportation steps to the appropriate country of origin. 

For electricity generation relating to New Zealand conditions, published data which describes trace metal composition of coal in order to estimate the specific mercury-emissions from electricity generation from known fossil fuel combustion sources. 

Data Quality Requirements

Data quality requirements specify in general terms the characteristics of the data needed for the study, based on the ISO 14044 (ISO, 2006a).  Descriptions of data quality are important to understand the reliability of the study results and to interpret the outcomes of the study.

In general, the data collated for the study has been assessed on a qualitative robustness test as shown in Table 4.

Listed below are specific details of data quality requirements for the lamp types being assessed in the study:

  • Geographic boundaries.  All lamps are imported into New Zealand and must represent appropriate geographic locations of manufacture (e.g. China, Thailand, Germany, Hungary, Taiwan and others).  Lamp distribution, use and waste management data are based on New Zealand conditions.  All lamp recycling occurs in Australia and should reflect those conditions.
  • Technological boundaries.  The study has assumed that the data collated via questionnaires for specific manufacturing plants and data used from the secondary LCA databases are representative of New Zealand conditions and representative production technologies.
  • Time boundaries.  The systems investigated should represent the situation in 2007, as this represents the last full calendar year for which data were available for the study.  With regard to landfilling, the decomposition of any biomass is assumed to take place within the time boundaries of the study. 

In cases, where there are missing data, this has been identified and documented in the inventory analysis.  The quality of surrogate data will be characterised and methods used to integrate the data explained.
Table 5 Data Quality Requirements for Inventory Data




Time-related coverage Desired age of data and the minimum length of time over with data should be collected. Data should represent the situation in 2007 and cover a period representing that calendar year.  
Geographical coverage Area from which data for unit processes should be collected. Data should be representative of the situation in the New Zealand.
Technology coverage Technology mix. Technology (for manufacture, product usage and end-of-life management) should be representative of New Zealand conditions and technology.
Precision Measure of the variability of the data values for each data category expressed. Not applicable.
Completeness Assessment of whether all relevant input and output data are included for a certain data set.  Specific datasets will be compared with literature data and databases.
Representativeness Degree to which the data represents the identified time-related, geographical and technological scope. The data should fulfil the defined time-related, geographical and technological scope.
Consistency How consistent the study methodology has been applied to different components of the analysis. The study methodology will be applied to all the components of the analysis.
Reproducibility Assessment of the methodology and data, and whether an independent practitioner will be able to reproduce the results. The information about the methodology and the data values should allow an independent practitioner to reproduce the results reported in the study.
Sources of the data Assessment of data sources used. Data will be derived from credible sources and databases.

Source: (ISO 14044, 2006)

3.7 Inventory Analysis

As discussed in Section3.3, all life cycle stages are included in the inventory analysis. 

Inventory analysis involves data collection and calculation procedures to quantify all relevant inputs and outputs of a product system. 

Section 3.6.2 and Section 3.6.3 detail that inventory data requirements.  Where specific data for lamp production and disposal were not available, secondary data has been used, together with estimates based on the data gathered for the other lamps.

Data validation has been conducted using mass balances, energy balances or comparative analyses of release factors to verify the unit process data.  Any anomalies will be checked to ensure that they comply with the data quality and selection requirements.  No significant anomalies exist.

3.8 Impact Assessment

The impact assessment phase of a LCA provides a system-wide perspective of the environmental issues for the systems being assessed.  The impact assessment quantifies the results of the inventory analysis in terms of several different impact categories. 

The contributions of each system are assessed for the impact categories listed below.  The listed impact categories address a range of environmental issues, and thorough methodologies have been developed for these categories.

The study employs the problem oriented approach for the impact assessment, which focuses on:

  • depletion of abiotic resources;
  • global warming;
  • stratospheric ozone depletion;
  • human toxicity;
  • freshwater and marine aquatic ecotoxicity;
  • terrestrial ecotoxicity;
  • photo-oxidant formation;
  • acidification; and
  • eutrophication.

The contribution that lamp manufacture, use and waste management make on these categories is calculated for each system.

The impact categories listed above are further described in Annex A.

Generally within the science of LCA, for some impact categories, particularly human toxicity, a number of simplifying assumptions are made in the modelling used to derive characterisation factors, relating to fate and effects modelling.  As a result, the adequacy of human toxicity in representing impact is still the subject of international scientific discussion. 

However, this category is still widely used and has been accompanied by caveats describing the deficiencies.  The impact assessment reflects potential, not actual, impacts and does not take into account the local receiving environment.

The modelling method used is that developed and advocated by CML (Centre for Environmental Science, Leiden University) and which is incorporated into the SimaPro LCA software tool.  The version contained in the software is based on the CML (2007) spreadsheet version 3.2 as published on the CML web site.

The final results will evaluate impacts using the following alternative impact methods:

  • TRACI, version 3.0, which is the Tool for the Reduction and Assessment of Chemical and Other Environmental Impacts, developed by the US EPA (Bare, 2003); and
  • IMPACT 2002+ (2003), version 2.04, which is mainly a combination between IMPACT 2002 (Pennington, 2005), Eco-indicator 99 (Goedkoop, 2000), CML (2000) and data from the Intergovernmental Panel on Climate Change (IPCC, 2000).

3.9 Interpretation

The Interpretation Section contains sensitivity analysis which considers lamp performance and addresses key assumptions and limitations within the study.  Sensitivity analysis is essential to ensure the conclusions and recommendations made are supported by the results of the study.

The results have appraised the following:

  • impact of assumed lifetime of the lamps – ± 50% lifetime;
  • impacts of assumed power rating of the lamps, accounting for potential increases in energy usage from lamp start-up – +0.5%;
  • impacts of improved energy efficiency in the use phase – +10% efficiency;
  • impacts of closed loop recycling;
  • impacts of not attributing offset benefits to avoided materials from lamp recycling operations; and
  • potential impacts of packaging and fittings.

Further details of these sensitivity analyses can be found in Appendix B, Section B.11.  The results of the LCA study are interpreted and conclusions are drawn on the difference in environmental impacts for the studied lamps in the context of the goals of the study. 

General observations and lessons learned from all lamps studied are presented and discussed. These include elements such as comparisons between alternative options for end-of-life product stewardship and mercury levels in the lamps.

Additionally, the interpretation identifies improvement potentials in the individual life cycle stages of the lamp products.

3.10 Modifications to Initial Scope

Conducting an LCA is an iterative process and modifications to the initial scope may be necessary.  Where this is the case, modifications were discussed, agreed upon with the Ministry for the Environment, and documented in the report.  The only modification to the original goal and scope has included the following:

  1. The power rating of 11W for a typical CFL with integral ballast has been updated from that reported in the Stewardship Solutions (2008) to 20W, based on manufacturer feedback that identified which lamp wattages represent mainstream products in New Zealand. 

3.11 Optional Elements Under ISO 14044

ISO 14044 (2006a) provides optional elements for impact assessment relating to the following four areas:

  • normalisation: calculating the magnitude of category indicator results relative to reference information;
  • grouping: sorting and possibly ranking of the impact categories;
  • weighting: converting and possibly aggregating indicator results across impact categories using numerical factors based on value choices (for an ISO compliant LCA, data prior to weighting must remain available); and
  • data quality analysis: better understanding the reliability of the collection of indicator results.

In this study we have not carried out normalisation, grouping or weighting of impact indicator results.  Although these methods can provide additional context to the results, especially in the case of normalisation, they also introduce subjectivity and value-based judgements, especially relating to grouping and weighting.  Handling these value judgements can be an area of controversy and individual opinion.  Weighting is prohibited by the International Standard for use in comparative assertions to be disclosed to the public.  We have excluded these optional elements from the study.

In terms of data quality, the final results for the study will use sensitivity analysis to evaluate uncertainties relating to data, assumptions and methodological choices.

Critical Review

In accordance with the ISO standard for LCA, the study has been reviewed by an external review panel (as shown in Appendix E) consisting of three experts, as follows:

  • Dr. Barbara Nebel, Scion, Wellington;
  • Dr. Sarah McLaren, Landcare Research, Wellington; and
  • Donald Hannah, ERMA, New Zealand.

The review panel’s report, and ERM’s responses, will be included in the final report.

The reviewers will address the issues below.

  • For the goal and scope:
    • ensure that the scope of the study is consistent with the goal of the study, and that both are consistent with the ISO standard; and
    • include this in a review statement.
  • For the inventory:
    • review the inventory for transparency and consistency with the goal and scope and with the ISO standard;
    • check data validation and that the data used are consistent with the system boundaries.  It is unreasonable to expect the review panel to check data and calculations beyond a small sample; and
    • include this in a review statement.
  • For the impact assessment:
    • review the impact assessment for appropriateness and conformity to the ISO standard; and
    • include this in a review statement.
    • For the interpretation:
    • review the conclusions of the study for appropriateness and conformity with the goal and scope of the study; and
    • include this in a review statement.
  • For the draft final report:
    • review the draft final report for consistency with reporting guidelines in the ISO standard and check that recommendations made in previous review statements have been addressed adequately; and
    • prepare a review statement including consistency of the study and international standards, scientific and technical validity, transparency and relation between interpretation, limitations and goal.

2. An elemental flow is material/energy entering the system being studied, which has been drawn from the environment without previous human transformation, or it is a material/ energy leaving the system being studied, which is discarded into the environment.