• Home
• Doing Business in NZ
• Eco-verification
• Publications
• Life Cycle Thinking: Long Version
• Practitioner Interview Summary

• Eco-verification

• Publications

• Eco-verification
• Eco-verification action
• Eco-verification progress
• Life Cycle Thinking: Short Version
• Life Cycle Thinking: Long Version
• Life Cycle Inventory Review

The structured practitioner interviews are summarised in the following section with visual summaries illustrating the responses to key areas defined in Table 1 Key Interview Topics.

LCA Type

There are three 'types' of LCA study as broadly prescribed (Baumann & Tillman, 2004)

• Stand-alone – Single and exploratory
• Change Oriented - Comparative and prospective
• Accounting - Comparative and retrospective

Stand-alone studies are the most common type of study internationally. They are often used to describe a single product and are used to identify 'hot spots'.

Change-oriented studies are primarily used in a development process (prospectively) to provide decision-making support. This usually applies to the development of products and services.

Accounting types are undertaken after the fact (retrospectively) and are well-suited to comparative decision-making or scenario analysis that could support plans for future improvement.

The distinction between these general category types can be difficult to maintain if the intent of a study is both to identify and then to mitigate (as is often the case). The descriptions of LCA types are included in this document as they provide a useful insight into the 'Intent' of the studies being undertaken. Table 4 NZ Practitioner Profile (By LCA Type) provides a useful glimpse of practitioner depth, as well as illustrating how the more experienced practitioners operate across all three broad LCA type categories.

Table 4 NZ Practitioner Profile (By LCA Type)

System Boundary

A system boundary is defined during the goal and scope step. It affects the inventory analysis phase by determining what data should be considered, and what can be ruled out (scope).

The establishment of a clear and defensible system boundary is crucial to the creation of a robust and effective LCA study. Boundaries are defined in relation to:

• Natural systems (Biosphere)
• Technical Systems (Techno-sphere)
• Geographic Boundaries
• Time Boundaries

System boundaries appear largely driven by the individual circumstance of any project, especially the goal and scope. The boundary can be widely affected by the client organisation's application and use of the study. Practitioners stated this would be a difficult area to harmonise nationally but that a normative 'approach' to determining system boundaries might be a useful mechanism to harmonise LCA practice within New Zealand.

Inventory Class was defined early as a term from the literature search which defined if an LCI was cradle to gate, gate to gate, or cradle to grave. This term did not have currency amongst New Zealand practitioners and became part of the system boundary discussion.

Figure 3 Inventory Class or approach

Even with strong support of a 'Cradle to Grave' approach within the practitioners work, they stated that end of life was a problematic area which led to some concern around results that were generated from a 'cradle to grave' scenario. The 'cradle to gate' and market studies represent predominantly food and primary sector exports, with evidence suggesting these studies were moving into cradle to market type studies (particularly in the fast moving consumer goods sector).

Functional Unit

A functional unit is the principal measure which 'frames' the environmental assessment. An example would be the comparison of a plastic milk bottle versus a glass milk bottle or a tetra pack. The functional unit for the LCA could be 'the delivery of 1,000 litres of milk' to enable an effective consideration of the total life cycle, including bottle washing, transport, and disposal or recycling of any parts.

All respondents cited the identification and handling of functional units as a core part of their work with the exception of one, whose interest remained only in the LCI area and not in a total LCA. A range of functional units were described, providing some insight into the different sectors practitioners are operating in.

The definition of the functional unit appears to be controlled by the audience or client, and in some cases several functional units are needed within a single study to cater for the different stakeholders of a project (for example wool growers and producer boards) (Barber, 2008). Table 5 Example Functional Units provides a selection of functional units utilised by New Zealand practitioners.

Allocation

Allocation is defined as "partitioning the input or output flows of a process to the product system under study" (ISO14040 1997).

There are many processes that often have multiple outputs or uses. Allocating all the environmental burden to a single process would be inaccurate and also lead to duplication. A simple example would be a log which could be split into a percentage allocated to milled graded timber and a percentage is allocated to pulp and paper production. The allocation method selected determines how these different streams are dealt with; this can be achieved through partition, avoided burden or other mechanisms.

Allocation is one of the most difficult areas within the LCA methodology. It complicates the data collection process and can greatly increase the threshold of data required within any given system. The different types of allocation seen across the New Zealand practitioner community are summarised below.

Most practitioners adhere to the ISO standard prescription for the procedure of allocation which is as follows:

1. Don't allocate if at all possible
2. If required use system expansion
3. If not possible use physical allocation
4. As last resort use economic allocation

Figure 4 Allocation Methods

Allocation is currently an area of significance to New Zealand. This is largely due to the recent development of embodied greenhouse gas emission labelling, or carbon reduction labelling(Carbon Trust UK, 2008). The driving force behind much of this emergent practice is based on the consumer-driven development of carbon labelling through the British-led BSI Publicly Available Specification (PAS) 2050 (British Standards International (BSI), 2008). The current methods of allocation within the PAS2050 are fluid and have been the subject of much discussion. The initial consultations of PAS2050 have specified the default allocation method as Economic.

Table 6 Allocation Methods

Impact Categories

This report does not pursue the aspect of 'Impact Assessment' methodology, but it was raised by respondents as an area to consider for future investigation and discussion.

Impact Categories are selected in the Life Cycle Impact Assessment (LCIA) phase of the LCA process, and define what environmental impacts will be assessed. The LCIA phase is conducted after the LCI phase, and is not considered a part of the formal LCI phase.

The separation of the LCI & LCIA phases is evidenced by leading Swiss practitioners, who divide the science of Inventory (data collection & aggregation), from the science of Impact Assessment (classification & characterisation). This indicates the natural delineation between the LCI phase and the LCIA phase.

Definition

The ISO classification of impact categories provides three broad categories or groupings. In these groupings there are a range of specific impact categories where assessments can be undertaken.

The choice of impact categories is an area of concern in the New Zealand context at the time of writing. There is an overriding focus on greenhouse gas (GHG) emissions and energy use, as evidenced in Table 8 Impact Category Use in NZ.

Table 8 Impact Category Use in NZ

→ Full size version of Table 8 [21 kB JPG]

Practitioners have noted that the restriction of impact categories through the focus on energy and green house gas (GHG) emissions have negatively impacted data collection. This was partly attributed to budget constraints and the current demand for climate change oriented studies and information.

Experienced practitioners (Nebel, 2008) expressed that the selection of limited Impact assessment categories should not affect the integrity of data collection, as all critical inputs and outputs should be collected for an assessment to be undertaken. This appears to warrant further investigation, as the demand for GHG accounting is likely to increase with the implementation of embodied greenhouse gas labelling.

This issue would endorse the importance of a common data collection policy to prevent endemic data loss over time due to studies not collecting all useful data. Consideration would also need to be given to deciding what the most important impact categories are. This could be decided by a group of experts, and should assist practitioners generally to work with a greater focus and more confidence.

If as demonstrated, LCIA is impacting on LCI, then it would be prudent to further investigate the Life Cycle Impact Assessment (LCIA) phase of LCA to better understand whether there are unique aspects in the New Zealand environmental context that need further consideration. There are impact assessment methods which are specific to certain sectors such as the 'BRE methodology (BRE Global, 2007) for environmental profiles of construction materials, components and buildings' (Jaques, 2008) or the pre-calculated Eco-Indicators (PRe Consultants, 1996-2008) method, which is useful in a product development environment where common materials and processes are used and ultimate data quality is not required.

Uptake of Tools

Development of LCA methodologies has resulted in the creation of software tools which enable dynamic modelling of system models. These allow the formation of calculations and visualisations, making the process easier and more streamlined. The adoption of these tools within New Zealand is patchy, with a range of practitioners still using self-generated Excel spreadsheets, mainly due to the intermittent nature of LCA projects and the capital cost of software purchase (rather than willingness to uptake).

Table 9 Tool Use

Both GABI and SimaPro are advanced Life Cycle Assessment software tools with a range of available databases for different industries. A greater number of participants used SimaPro, although GABI is a recent tool in the New Zealand context and is developing a following here.

Excel, though not an LCA tool, was commonly used. Advanced practitioners used Excel to devise their own formulas and dataset creation and organisation, and are using it to control their data and ensure they have maximum transparency during the LCA process. The novice users seemed to use Excel under the guidance of more experienced practitioners.

Everdee is another LCA tool that was used. Everdee was created through the EcoSMEs (Eco SME & Italian National Agency for new Technologies, Energy and the Environment) initiative that also gave rise to 'Tespi' for product innovation. The practitioner's rationale for selecting this tool was because of its industry-based electronics database, which was representative (Vickers, 2008).

The Landcare Research Carbon Zero programme has developed their own tool which integrates GHG protocols for assessments on climate change.

The intermittent nature of LCA work until recently appears to have been an impediment to the uptake of more advanced life cycle engineering tools (Ogilvie, 2008) due to up front capital outlay and training. In addition, a range of practitioners stated they did not have a requirement to conduct a full LCA so were content with using spreadsheet tools and formula's for completion of data collection and inventory analysis.

The Importance of Data

The inventory analysis process hinges on the availability of high quality, effective data to form an accurate picture of a system being assessed. The credibility of any LCA or GHG assessment is substantially dependant on data quality.

Most of the work in generating data within New Zealand has been biased toward the primary sector that has produced a growing body of information up to farm gate. As markets have become more interested in transport distances and the impact on climate change of entire product systems, these studies have extended to consider the whole supply chain, all the way through to market and in some cases end-user disposal. This is seen with the emergence of cradle to market studies seen in the food exporting sector outlined in Figure 3 Inventory Class or approach.

The evidence gathered in this study shows that manufacturing and value-added industries have little New Zealand-specific data of a general nature to draw on as yet and have tended to source data from international datasets. As a result, there is a nationwide need to create better country-specific and industry vertical data.

Data Quality

When discussing data quality the study looked at the following areas:

Data quality appears to be variable, with some intensive and highly detailed studies in certain areas (e.g. Pastoral) where real-time data was collected on-farm to give a high degree of data quality (Barber, 2008). Whilst in other studies data was used from existing literature research (which may be of uncertain age and provenance).

Generally most practitioners cite data quality as a serious issue, both in terms of availability and accuracy within the New Zealand context. In addition they had real concerns about the cost of data collection and maintenance, which were seen as a recurrent hurdle cost to SMEs especially (Smith, 2008), and even to larger enterprises. This hurdle is due in part to the requirement for continuous data collection that requires technically-sophisticated methods or devices and specialised staff.

There are a range of approaches taken by most participants which could be broadly categorised as:

• Created Datasets: researched and created for the environmental analysis.
• Modified Datasets: existing datasets modified for a particular application.
• Existing Datasets: using existing datasets for a particular application.

Most of the advanced practitioners have been, or are involved in the creation of homogenous datasets, as well as using modified international databases of materials and processes where required.

Created datasets

The datasets created range from highly detailed studies (direct measurement), to studies which are based on historic research papers. One such study went to a huge amount of detail to include the whole capital cost of building a plant and amortised the environmental impacts across the lifespan of the structure, and allocated this to the end product on a per unit basis (Crawford, 2008).

Modified datasets

Modified datasets are primarily international (originating from the EU or the USA), that are generally linked to, or embedded in, LCA software tools. These databases, such as Ecoinvent (Ecoinvent Centre, 1998-2006), contain a wide variety of materials and processes which enable modelling at a schematic level and in some cases offer detailed levels at which there is parity in application and context.

Existing datasets

The commonly referred and used international data sets are as follows:

• Ecoinvent (Swiss) – Uses literature so it can be open and transparent, but may not have the industry accuracy.
• GABI (German) – Uses industry data and (therefore) is aggregated and cannot be fully transparent, but is considered accurate.

There are other databases used which are more industry specific, a useful exercise may be defining all the major sector databases which would be considered applicable to the New Zealand context. This information could be made available to all stakeholders so they may find relative datasets with greater ease.

Compatibility

There are a number of data formats available which tend to be application specific. Exchange formats are used to port the data between applications; the most widely used exchange file format is 'XML' (extended mark-up language) (Nebel, 2008). This supports a fast and easy exchange of inventory data into different software packages, such as GABI or SimaPro, as well as handling in open source and Internet environments.

Availability

Many of the datasets are private (or not accessible), which prevents them from being openly exchanged or used for the benefit of other companies. Research organisations often retain the right to use aggregated data for other applications (within their own projects and for their own benefit). This type of practice would appear to work against wider transparency, discourage openness, and encourage competition.

Some data is published, and is therefore partly available in aggregated form (although not accessible at a detailed application level) and some data is made available on specific request to students and other practitioners.

Currently there is no central repository for inventory data and therefore no easy way to determine if a study has been conducted in the area or if there is other useful data in New Zealand that could be applied in other studies.

Research also plays a key role in the creation and support of assumptions and the development and formation of new datasets. It would also be useful to have a reference of critical industry or sector level research that can be applied to inventory analysis at a national level. This research could then be monitored for efficacy and relevance as industries evolve.