LCAs have come a long way in the past 20 years, but the concept still faces many challenges before it will be accepted as transparent and credible

by Leslie Webb

Get a life cycle assessment

Life cycle assessment (LCA) is now formally defined by ISO as "a compilation and evaluation of inputs, outputs and potential environmental impacts of a product or system throughout its life cycle". It is thus no more (and no less) than a sort of environmental accountancy, which provides a structured framework for adding up the totality of a system's burdens and impacts on the environment. This is illustrated at a simple level for pulp and paper products in Figure 1.

Figure 1 - The Pulp and Paper Cycle Chemical

SETAC and ISO

Like a lot of the other trappings surrounding LCA, this definition owes much to the pioneering work of SETAC - the Society of Environmental Toxicology and Chemistry. It was this group that was responsible for broadening the application of LCA into the environmental field in the early 1990s. In 1993, ISO established its environmental management technical committee (TC207), which has since been responsible for developing a range of standards in the field of environmental systems and tools with one of its six sub-committees devoted to LCA.

All ISO standards have to progress through their own seemingly endless life cycle of a working draft, committee draft, draft international standard (DIS) and final draft international standard (FDIS) before becoming a fully-fledged ISO standard. The LCA documents already published and those being developed are summarised in Table 1.

ISO 14040 is the lead document and sets the general principles and framework for conducting LCAs. The draft managed to negotiate its way to publication within four years and was finally published at about the same time as ISO 14001 on environmental management systems. ISO 14040 is probably as far as most people will need to go unless they are keen LCA practitioners. The overall structure of life cycle assessment under ISO 14040 (see Figure 2) follows the original SETAC model, but SETAC's final "improvement" phase has been replaced by the broader "interpretation" phase under the ISO system. ISO 14040 also includes sections on LCA reporting and the need for critical reviews in order to validate the LCA findings.

Figure 2 - ISO Model for Conducting LCA Defining LCA

The ISO standard recognizes several useful applications of LCAs:

• to help industry prioritize key environmental aspects for improvement
  
• to identify aspects with significant impacts for use as overall performance indicators (there is a separate ISO standard in this area - ISO 14031)
  
• to support claims in the marketing of products and in developing eco-label criteria.

   

If we go back to the early 1990s, manufacturers were then beginning to make claims about the environmental advantages of certain products. But producers often concentrated on specific aspects of that product in maybe just one part of its life cycle, conveniently ignoring less desirable attributes in others. One example involved the claims made by some paper manufacturers. The groups noted that their involvement in forestry operations countered the greenhouse effect through the take-up of atmospheric carbon dioxide, but failed to mention the return of the carbon through decomposition or incineration of the used product or the CO2 release from the use of fossil fuels. These cases happened to pick on one of the more complex environmental issues of our times and one which many LCA studies have returned to time and time again.

Inventories and impacts

Because of the many potential pitfalls in the LCA process (at least to non-practitioners), the initial stage of setting goals and defining the absolute protocol to be adopted needs to be undertaken carefully. It should certainly not be rushed. Some of the issues specific to the pulp and paper sector have been investigated within European COST action E1 on Paper Recyclability and the results were presented at a final meeting in Gran Canaria last November.

The work on LCA applied to forestry and forest products is now continuing within COST action E9 through to 2001. Within the E1 action, the LCA working group carried out a case study in which each of their different methodologies were compared for one product system. This highlighted problems such as simple calculation errors by all participants and the need for harmonization of the presentation of inventory data. Most participants used their own software, some of which is commercially available. An important decision for LCAs of paper products based on virgin fiber is whether the forest - or other land in the case of nonwood fibers - is included within the system boundary. There is a general consensus in favor of forestry being included, but this then raises many further questions. For example, the allocation of burdens between the various products of the forest is an issue. Does the sawn timber from a sawmill carry all the burdens and associated impacts from its input roundwood, or should some of this be allocated to the sawmill "waste" - the off-cuts that are usually sent for pulping? Once these types of question have been resolved, other elements arise, such as the appropriate impact categories for forestry and how to handle the carbon flows into and out of the forest.

The ISO standards give no specific assistance in answering such questions and recommend avoiding the problem by redefining the system. If this cannot be done, allocation has to reflect causal relationships where possible, the simplest and most widely used being in relation to mass.

The COST action E1 working group recommended that forest burdens should be allocated on the basis of the economic value of the various outputs. Allocation difficulties also arise with recycling operations, particularly in the so-called "open" recycling systems where the first product or waste is used as a raw material for a different product.

Impact assessment is perhaps the most contentious area within LCA and this probably explains the lengthy gestation period for ISO 14042. There is a general consensus about the impact categories to which the burdens are allocated, ie the classification stage of impact assessment in Figure 2. Among the examples are the depletion of biotic and abiotic resources and the pollution effects from various emissions such as global warming, ozone layer depletion, toxicity to humans and eco-systems, acid deposition, eutrophication of waters, etc.

There may be a cascade of primary, secondary and tertiary impacts from one particular burden. For example, the release of biodegradable substances to water causes oxygen depletion, which may in turn lead to fish kills and the release of odorous/toxic substances to air causing further possible effects.

The two contentious steps are characterization and valuation. Characterization involves aggregating burdens within impact categories and valuation, then ranking the impact categories. Some burdens are already combined together through their method of measurement, eg BOD measures all substances that utilize dissolved oxygen. But others have to be combined through appropriate weighting factors or models. Accepted schemes are the combination of greenhouse gases into carbon dioxide equivalents using their global warming potential over a particular time period (usually either 20 or 100 years) and the combination of acidic gases into proton equivalents using their acidification potential. There are also techniques such as the Swiss critical volumes approach for aggregating quite disparate effects within the air and water compartments into a single score, but these are not universally accepted.

There is less agreement about methods of aggregating different types of resource depletion. An important burden within nearly all LCAs is fossil fuel consumption, which can be combined in different ways as a parameter corresponding to depletion of known reserves. The most contentious step is valuation as this involves judgements outside the purely scientific realm, including the views of society at large about the acceptable risks of different environmental issues. Within this wider context, there is usually a desire for a simple, single index of harm or potential damage across a system's life cycle, but it looks as though the final version of ISO 14042 will not allow the combination of impact categories into a single environmental score.

Paper LCAs

There are two mainstream paper groups that have been the subject of most LCAs - packaging and newsprint. The focus on packaging is hardly surprising as many businesses in the packaging chain are trying to reduce its environmental impact and thus wish to compare the environmental effects from a diverse range of packaging materials such as paper/board, various plastics, composites, etc.

The EU directive on Packaging and Packaging Waste specifically looks towards LCAs to help establish a hierarchy between reusable, recyclable and recoverable packaging materials. In this way, it aims to set more informed recovery and recycling targets required by 2006. The idea of any universally applicable hierarchy for packaging materials is strongly resisted by the packaging industry. Many companies have been quick to point out the general shortcomings of LCAs and the wide range of local and regional factors that must be taken into account before arriving at the "best" option.

In the context of the packaging directive, a study on the application of LCAs across five distinct packaging scenarios was carried out for the European Commission in 1997. One area that has caused difficulties in the application of the directive concerns the measures adopted by certain countries to favor or mandate the use of reusable drinks containers. The EU study looked into this, but failed to come up with a clear overall conclusion - a result that was much to the liking of the industry. Paper-based cartons came out of the study with a low impact rating for both small and large containers, but only with a very high recycling rate in the latter case. On what has become a favorite analysis scenario for LCA practitioners, the study also found that recycling of segregated waste was "better" than incineration with energy recovery of mixed refuse.

This latter comparison has also been very popular in the various LCAs conducted around newsprint during the 1990s, for example:

• the study by the International Institute for Applied Systems Analysis (IIASA) looking at Western Europe
  
• a Finnish study by KCL looking at some of the environmental aspects of the newsprint flow between Finland and Germany
  
• the study by Jaakko Poyry for the British Newsprint Manufacturers Association (BNMA) looking at the UK situation
  
• the study by the Centre for Environmental Technology at Imperial College in London in the context of sustainable cities
  
• a study by Ecobalance in the UK for Aylesford Newsprint
  
• a collaborative study between Stora (now Stora Enso), Canfor and Springer Verlag on the life cycle impacts associated with newsprint and magazine production
  
• an ongoing study in Germany by various research institutes for the Federal Environment Agency (Umweltbundesamt) in the context of developing new criteria for various grades of graphical paper within the Blue Angel eco-label.

   

The BNMA study looked at various scenarios for used newspapers between landfill, recycling and incineration employing current or best available techniques (BAT). The unusual feature of this study was that, in addition to the normal range of environmental effects modeled, it also looked at economic effects (eg on UK employment) and social effects. The latter necessitated entering the murky waters in which costs had to be attributed to adverse environmental effects. Recycling gave a net environmental benefit when using BAT, but inclusion of private social benefits showed recycling to be superior under all scenarios.

This study aroused some controversy because of its socio-economic emphasis, but the reaction was mild compared to the heat generated by the later Imperial College study. The conclusions of this study were picked up by the magazine New Scientist, the cover of which invited readers to burn the magazine rather than recycle it. As such, it was typical of a spate of recent backlash articles and programs in the popular media against the supposed "goodness" of recycling.

The actual study merited a more considered analysis though. Like the BNMA study, it went beyond assessing environmental burdens as mass impacts by placing a cost on these environmental externalities, but within a systems analysis/optimization framework. Using externality values at either end of a range taken from the literature (eg $4-40/ton carbon dioxide emitted), the model was used to find the optimum (least cost) balance between landfill, recycling, anaerobic digestion and incineration of used papers.

Figure 3 - Optimal Management Scenario for Used Papers

The results (Figure 3) are supposed to substantiate the New Scientist's "burn me" headline, but they depend critically on the cost level for the externalities, whether or not wood-derived carbon dioxide is counted and on many other assumptions built into the model. Even if you buy this study's conclusions, the paper industry should not be criticized for using an economically available resource that would otherwise (at least in the UK at present) not be incinerated, but landfilled, as there are very few waste incinerators. It is up to others outside the paper industry to build waste incinerators, although it would clearly be possible for a mill to use combustible wastes for on-site steam/power production as well as recovered papers as a fiber source. Indeed, many recycled mills (as in the Aylesford Newsprint (ANL) example below) already do this in part by recovering energy from incineration of the deinking/cleaning residues.

LCA in the news

In contrast to these two very general studies covering the whole of a country and using average data within each scenario, the study for Aylesford was very specific. The work covered actual production conditions and the mill's real customer base. As the goal of the study was to compare recycling at ANL with incineration and energy recovery of the same materials, the study was described as a "gate to grave" system. The functional unit was one ton of used newspapers and magazines. The methodology is described as additive in that electricity generation elsewhere has to be added to ANL's newsprint production and newsprint production elsewhere added to the incinerator electricity generation so that the outputs from both systems are identical.

The resultant inventory occupies several pages and ranges from materials with very large flows (tons of water) to those with minute mass flows (small fractions of a gram of some metals). The main impacts modeled are global warming through the greenhouse effect; acidification from atmospheric gases; water use and eutrophication of surface waters; and depletion of non-renewable resources (see Figure 4).

Figure 4 - Selected Results of ANL Study

The main contributor to global warming is carbon dioxide from fossil fuel combustion. But methane and its much greater global warming potential is significant in both systems. The national electricity grids make significant contributions to both product systems, so assumptions about alternative sources of newsprint in the incineration scenario could affect these results significantly.

Acidic gases are mainly a mixture of sulfur and nitrogen oxides and are low numerically as they are expressed collectively as equivalent (eq) protons rather than in terms of the mass of the actual gas. Water is always considered to be an important (potential) vector of mill burdens on the environment, as water use (m3/ton product) is an approximate indicator of some of these impacts. Water use in these two systems is largely at the respective mills, with ANL again coming out well due to its above average efficiency for a newsprint mill. In terms of wastewater components (not shown), suspended solids, COD and nitrogen are lower for the ANL system, but phosphorous is somewhat higher. Despite this, the modeled eutrophication ability is greater for the incineration system.

The environmental effects of resource use are some of the more difficult to model within LCA - both for renewable and non-renewable resources. Use of renewable resources (trees) during newsprint production at competitor mills was not quantified during this study. But use of non-renewables (mainly fossil fuels) was modeled using reserve depletion (kg/yr) as the criterion. Using this methodology, the ANL system has a slightly greater effect than the incineration system. The biggest contribution in both systems was due to natural gas consumption.

Using a different model based on fractions of reserve (not shown), the relatively high use of uranium in the incinerator system (for nuclear power production) turns the result round. Irrespective of its form (feedstock, fuel or total primary), energy use is lower in the ANL system. A specific study on chemical use showed that its overall contribution to the impacts studied was negligible.

LCA of a life cycle

The overall conclusion of the above study (with suitable provisos and qualifications) was that recycling of used newspapers and magazines at ANL is environmentally preferable to incineration elsewhere. One of the characteristics of all LCAs is that, despite the focus on the life cycle of a product, the actual assessment does not have a time element, but coalesces impacts that occur over quite a large timescale into an instantaneous snapshot. The time displacement of carbon released from harvested trees through to its re-uptake by the replacement trees as they grow leads to a temporary accumulation or depletion of atmospheric carbon, depending on the relative rates of carbon release and sequestration.

An interesting variation in the application of LCA was investigated recently by Jaakko Pöyry Consulting. The group looked at the trend in the environmental burdens from a hypothetical paper machine over its 20-year lifetime in what they described as an LC2A. The newsprint machine was converted to an SC machine in mid-life and there were other changes in fiber and energy sourcing, as well as waste treatment. The study introduced some interesting concepts such as the impact clock to indicate when half the lifetime emissions had been released and the medical monitor showing releases over time.

It is clear that LCA has slowly come of age over the last 10 years and the imminent completion of the ISO standards should ensure that future LCAs are carried out credibly and transparently. With the development of environmental management to embrace health/safety, sustainability and social aspects of industrial development, LCA has a challenging future to incorporate measures of these new parameters within its framework, as well as developing ways of expressing its outcome in economic terms.

Leslie Webb would welcome information on how mills have utilized LCA. He can be contacted in the UK by telephone/fax at +44 1372 276599, or via email on leswebb@compuserve.com.