Life Cycle Environmental Performance of Renewable Building Materials in the Context of Residential Construction

PHASE I INTERIM RESEARCH REPORT ON THE RESEARCH PLAN TO DEVELOP ENVIRONMENTAL-PERFORMANCE MEASURES FOR RENEWABLE BUILDING MATERIALS WITH ALTERNATIVES FOR IMPROVED PERFORMANCE

JULY 15, 2002
Consortium for Research on Renewable Industrial Materials
(CORRIM, Inc.)

Consortium for Research on Renewable Industrial Materials (CORRIM, Inc.)
Prepared with financial support from the USFS (JV11111169-156) and contributions from a number of private companies based on a research plan developed for the Department of Energy (DOE Agreement No. DE-FC07-96ID13437) as a part of the American Forest and Paper Association's Agenda 2020 priorities. Any opinions, findings, conclusions, or recommendation expressed in this publication are those of the authors and do not necessarily reflect the view of the financially contributing entities

LIFE CYCLE ENVIRONMENTAL PERFORMANCE OF RENEWABLE BUILDING MATERIALS IN THE CONTEXT OF RESIDENTIAL CONSTRUCTION

PREFACE

This report provides interim research results from a study to develop a database and modeling capability to adequately describe the environmental performance of building materials and their uses, addressing key wood materials such as lumber, plywood, composite panels and other wood derived products. The report documents progress on the original research plan developed for the Department of Energy and the American Forest and Paper Association under Agenda 2020 priorities for pre-competitive research needs. The Consortium for Research on Renewable Industrial Materials (CORRIM) with the financial support of its 21 institutional and company members and the Department of Energy developed (1) a research plan, (2) a Data, Standards and Procedures: Guideline for LCI and Economic Analysis, and (3) an organizational approach to conducting the research plan in 1998. A Phase I research plan was designed to pilot-test the development of data and analysis procedures for each stage of processing, while providing data for the primary wood producing regions of the US. This Phase I report covers the first 5 modules of the 22 module research plan focusing on forest resources from the US Southeast and Pacific Northwest and residential construction in a warm climate (Atlanta) and a cold climate (Minneapolis). The USFS Forest Products Laboratory, 14 research institutions and 10 companies are providing financial support.
This report provides environmental and economic data on all life-cycle stages from planting and growing the renewable raw material through the manufacturing of product, design and construction of buildings as well as activities associated with occupation, use and final demolition. The collected data and subsequent analysis follows consistent definitions and collection procedures that facilitate integration of results across the full life cycle for all stages of processing in order to address environmental performance questions. The findings are interim, reflecting a mid-point progress report that intentionally identifies data and procedural inadequacies that need to be corrected before completion of the final report. One of the more substantive impacts of this research effort has been the enhancement of institutional capabilities to support the development of environmental performance data and analysis.

The report is organized as follows: Major points are very briefly summarized in an executive summary preceding the table of contents. Section 1 provides the background, mission, organization of effort, and objectives. Section 2 provides the Life Cycle Inventory and Analysis (LCI/LCA) framework. The findings for each stage of processing are reported on in 7 modules (Appendices A-G). The appendices are essentially stand alone LCI reports for intermediate and final products. The last module, Appendix G, covers the construction aspects, integrating the information from the other modules for a residential structural building shell. Information on final building use, maintenance and ultimate disposal will be completed for the final report. The forest resource module demonstrates the impact of management alternatives on environmental performance. Additional scenarios and sensitivity analysis examining the impacts of changes in management and processing technologies will be included in the final report.

EXECUTIVE SUMMARY

• To create a consistent database of environmental performance measures associated with the production, use, maintenance, re-use, and disposal of alternative wood and non-wood materials used in light construction, i.e., from forest resource regeneration or mineral extraction to end use and disposal, thereby covering the full product life-cycle from "cradle to grave."
  
  
• To develop an analytical framework for evaluating life-cycle environmental and economic impacts for alternative building materials in competing or complementary applications so that decision-makers can make consistent and systematic comparisons of options for improving environmental performance.
  
  
• To make source data available for many users, including resource managers and product manufacturers, architects and engineers, environmental protection and energy conservation analysts, and global environmental policy and trade specialists.
  
  
• To manage an organizational framework to obtain the best scientific information available as well as provide for effective and constructive peer review.

The primary difference between the Minneapolis wood and steel house is the use of materials for floors and walls. Both designs share the same basement and roof elements. For the Atlanta structure, the only major difference between the wood and concrete design is in the exterior wall structure as both designs use concrete floors and wood roofs. Making cross design comparisons at the wall and floor section shows much more dramatic percentage differences than for the buildings as a whole (last two columns of the table below) since some parts of the structure share common materials, and the construction process itself is energy intensive and not that much different across the designs. In effect a substantial environmental performance difference for nearly substitutable products may not seem so great for a completed structure with many common components.

An examination of a change in forest management suggests small but significant changes in index measures. A small increase in PNW management intensity resulting in an estimated 5% increase in forest productivity increases the availability of wood such that the number of wood homes built in Minneapolis increases and the number of those built of steel decreases. It is assumed, for simplicity, that the increased forest productivity forces product substitution within the region rather than imports and exports from other regions or international sources. Consequently there is a 6% reduction in embodied energy and air emissions, a 25% reduction in water emissions and a 1% increase in solid waste. The same forest productivity increase in the Southeast results in a 17.5% increase in the construction of wood homes vs. concrete in Atlanta and a reduction in all output measures ranging from 3% for water emissions to 13% for solid wastes.

To support this environmental assessment for a residential structure and be able to analyze the impact for each process, LCIs, were developed for each wood product (logs, lumber, plywood, and oriented strandboard) used in home construction.

The Forest Resource module (Appendix A) provides the environmental, energy and resource impact data on the growth, management, and harvesting of logs and reforestation of harvested timber for the Pacific Northwest (PNW) and Southeast regions of the US for a representative acreage within each region. The study calculates volume harvested for three general combinations of management intensity and site productivity for each region. The combinations were reduced to a single estimate of yield using weighting factors for the acreage under each management regime. The weighting scheme creates a base case for each region. Alternative weights resulting from increasing the acres under a different level of management intensity are used to produce alternative cases, with a different calculated volume. For example, the increase in merchantable volume under a prescribed alternative case was 21% in the Southeast (including the volume from a partial second rotation when rotation ages were reduced) and a nearly 16% increase in volume in the PNW.

The portion of lumber and pulpwood produced in the Southeast changed slightly under the alternative scenario; lumber share declined 2.1 percentage points as pulpwood share increased 2.1 percentage points. The alternative scenario also leads to a 3.2 year reduction in the average rotation age in the Southeast, which is responsible for the percentage shifts in lumber and pulpwood produced. Lumber output increased 15%, only 68% of the merchantable volume increase. Lumber share remained at 100% in the base and alternative PNW scenarios.

System costs increased by around one percentage point above the percent change in the amount of merchantable volume removed in the Southeast, while increasing about the same percent as merchantable volume removed in the PNW. A similar result is observed for fuel and lubricant consumption during harvesting operations: an 11% increase in the Southeast and 16% increase in the PNW roughly corresponding to the changes in merchantable volumes harvested, although there appears to be a slight increase in the consumption factors in the Southeast due to the change in rotation age and products harvested. The increase in harvested volume resulted from using 4 times more nitrogen and phosphate in the Southeast and 2 times mores nitrogen and phosphate in the PNW as fertilizer inputs in conjunction with the other management changes.

• Measures of carbon emissions, and carbon storage on the forest floor and in wood products for each stage of processing and region, and comparable impacts for policy alternatives that result in substitute products.
  
  
• Identification of alternative methods for reducing emissions with quantified impacts across stages of processing and geographic regions.
  
  
• Identification of performance measures and methods to improve environmental performance in areas such as (1) energy and material-use efficiency, (2) biodiversity and habitat protection indices for uplands or riparian zones as well as other measures of the health and sustainability of forest ecosystems, (3) solid waste reduction, and (4) reduction in the production and emission targeted potentially toxic chemicals.
  
  
• Assessments of the impact of policy proposals on the ability of the forest sector to meet expected consumer demand for products.
  
  
• Cost effective approaches to meet changing environmental goals, and develop investment strategies that are more responsive to those needs.
  
  
• Opportunities to adopt strategies that improve environmental performance where costs are not a limiting factor and support for the development of policy alternatives that could offset cost impacts when necessary.
  
  

LIFE CYCLE ENVIRONMENTAL PREFORMANCE OF RENEWABLE MATERIALS IN THE CONTEXT OF RESIDENTIAL BUILDING CONSTRUCTION

Consortium for Research on Renewable Industrial Materials
(CORRIM)
Phase I Interim Research Report

January 27, 2002

Prepared by:
Jim Bowyer, University of Minnesota¹
David Briggs, University of Washington
Bruce Lippke, University of Washington
John Perez-Garcia, University of Washington
Jim Wilson, Oregon State University

With supporting appendices by:
Forest Resources:	Softwood Lumber:
Leonard Johnson, University of Idaho	Michael Milota, Oregon State University
Bruce Lippke, University of Washington	Ian Hartley, Mississippi State University
John Marshall, University of Idaho	Cynthia West, Mississippi State University
Plywood:
Jim Wilson, Oregon State University	Oriented Strandboard:
Maureen Puettman, Oregon State University	Earl Kline, Virginia Tech University
Eric Sakimoto, Oregon State University
Eric Dancer, Oregon State University	Environmental Impacts of a Building:
Residential Buildings:	Jamie Meil, ATHENA
Bo Kasal, North Carolina University	Bruce Lippke, University of Washington
Pat Huelman, University of Minnesota	John Perez-Garcia, University of Washington

¹ Bowyer is director, Forest Products Management Development Institute, Department of Wood and Paper Science, University of Minnesota; Briggs is director, Stand Management Coop, Lippke is director, Rural Technology Initiative and Perez-Garcia is associate professor, College of Forest Resources, University of Washington; Wilson is professor, Department of Forest Products, Oregon State University.

TABLE OF CONTENTS

LIST OF FIGURES

LIST OF TABLES

1.0 INTRODUCTION

1.1 BACKGROUND

• A consistent database to evaluate the environmental performance of wood and alternative materials from resource regeneration or extraction to end use and disposal, i.e. from "cradle to grave" (Figure 1.1)
  
  
• A framework for evaluating life-cycle environmental and economic impacts
  
  
• Source data for many users, including resource managers, manufacturers, architects, engineers, environmental protection and energy analysts, and policy specialists.
  
  
• An organizational framework to obtain the best science and peer review.

Figure 1.1. Integrated Life-cycle of Biological Materials (CORRIM 1998)
CORRIM is a nonprofit corporation of scientists from the 15 member research organizations (Table 1.1).

Soon after the establishment of CORRIM II, a request for proposals was issued by AF&PA in 1994 as part of its' Agenda 2020 program. The Agenda 2020 program focuses on pre-competitive research needs of the US forest products industry, and the '94 RFP targeted two principal research objectives relative to environmental life cycle inventory and analysis: (1) development of an updated analysis of the environmental efficacy of renewable building materials, including consideration of environmental impacts related to energy consumption and (2) identification of alternatives for reducing environmental releases associated with building materials through their life-cycles. This led to funding of CORRIM II in 1996 by the U.S. Department of Energy ($125,000) and the forest products industry ($54,000) to develop a research plan. This plan, outlining research activity within 22 modules over a 5-year span (Table 1.2) was released in January 1998 (CORRIM 1998). In addition, protocols and standards were described in the research plan to ensure that data collection and analysis would be compatible with ISO life-cycle analysis guidelines (ISO 1997, 1998, 2000a, 2000b) and LCI procedures developed for the forest industry (AF&PA 1996).

Table1.2: CORRIM Research Modules - (from CORRIM 1998) Phase I Research consists of Highlighted Areas (CORRIM 1998)

        All estimated funding (monetary figures) are in $5 000's

Preceding the CORRIM II initiative was a project to develop current environmental performance information for building materials used in Canada. This project, titled "Building Materials in the Context of Sustainable Development", was initiated in 1990 as the ATHENA™ project by FORINTEK Canada Corp, and is now continuing at the ATHENA™ Sustainable Materials Institute. In 1997, an alliance was established between CORRIM and ATHENA™ to take advantage of previous ATHENA™ research and to broaden the geographic and product representation of research.

1.2 OBJECTIVES, MODULAR DESIGN, AND SCOPE OF CORRIM PHASE I

• to develop an adequate database and models of environmental performance measures over the entire life-cycles of structural building materials, beginning with extraction of resources, through product manufacture and transportation, construction of a structure, use and maintenance of the structure, and finally dismantling of the structure and either disposal or recycling of the building components.
  
  
• to examine a range of management, product, and process alternatives to identify strategies that can improve environmental performance.

The Phase I research effort was concentrated on development of life cycle inventory data for wood-based building materials produced in the two regions of the US that account for the greatest production of forest products - the Pacific Northwest and the Southeast. Further, because of funding limitations, the scope of Phase I was limited to consideration of the structural shells or envelopes of residential buildings. Although focused on wood-based materials, Phase I research did address environmental performance of wood-framed buildings in comparison to residential buildings constructed of steel and of concrete block. Data for these other materials was obtained from the Athena Sustainable Materials Institute.

1.3 ORGANIZATIONAL FRAMEWORK TO CONDUCT PHASE I RESEARCH

An organizational structure has been established for CORRIM II to provide the best scientific talent available to complete the research while obtaining competent and independent reviews. This structure is organized around working panels of experts (Figure 1.2) including a Data Standards and Procedures Panel, a Stages of Processing Panel (with 4 technical sub-committees), and a Strategy Assessment Panel.

The Data Standards and Procedures Panel has responsibility for providing data processing capability, assuring consistency and compatibility in the design of data collection procedures across research modules, and integrating the results of management and technology alternatives throughout all stages of processing. Development of LCI data, on the basis of individual unit operations, for a range of wood-based building materials from harvesting and forest regeneration through manufacturing, transport, construction of residential structures, building maintenance, and reuse or disposal at the end of the useful life of the building or building component is the responsibility of the Stages of Processing Panel. The Strategy Assessment Panel is charged with development and evaluation of a number of scenarios that represent various strategies for improving environmental performance of wood based building materials throughout their life cycles.

Figure 1.2. CORRIM Organization Chart (CORRIM 1998)
The specific assignment of scientists from CORRIM member institutions to the Phase I modules is presented in Table 1.3.

Table 1.3: Lead Scientists Involved in CORRIM - Phase I Chair of Technical Steering Committee, Dr. Jim Bowyer, University of Minnesota

1.4 REPORT STRUCTURE

This interim report documents progress through the first half of the Phase 1 research plan. The intent is to fully develop procedures and collect data while identifying deficiencies that must be eliminated before completion of the final Phase I Research Report. The final report will also contain more detailed sensitivity and scenario analysis on the impact of management and technology change.

Section 2 of this report provides the framework for developing Life Cycle Inventory (LCI) data and Life Cycle Analysis (LCA) information. Section 3 summarizes accomplishments to date.

The report is designed around the seven individual research modules comprising Phase I of the project. These modules, presented in full as appendices to this report, are in effect stand alone reports covering each stage of processing through construction of a residential building structure and provide LCI data for each wood product used in the structure. The modules are:

• Module A: Forest Resources - identifies environmental performance measures and presents life cycle inventory data for specific unit operations in woodland management activities in forests of the Pacific Northwest and Southeast regions.


The objectives of this module are to:


• Provide environmental, energy, and resource impact data on the growth, management, harvesting and reforestation of timber under five general scenarios of management intensity for the Northwest and Southeast regions of the United States.
  
  
• Develop case studies to represent a typical range of forest management objectives and stand and site conditions.
  
  
• Provide environmental performance measures including indices of stand structure diversity, habitat, and fragmentation for each of the case study scenarios.
  
  
• Provide inputs for the Processing Modules from the case study scenarios.
  
  
• Module B: Pacific Northwest Softwood Lumber - presents life cycle inventory data for specific unit operations associated with the manufacture of softwood lumber manufactured in the Pacific Northwest region of the United States.


The objectives of this module are to:


• Provide environmental, energy, and resource impact data on the manufacture of softwood lumber in the US Pacific Northwest region.
  
  
  
• Provide benchmarks for these products that will enable comparison to process improvements or new processes.
• Provide input for the Structures Module.
  
  
• Compare fossil versus biomass fuel dependency.
  
  
• Provide a measure of resource use efficiency.
  
  
• Provide a comparison of energy and resource impacts relative to the 1976 CORRIM study.
  
  
• Module C: Southeastern US Softwood Lumber - presents life cycle inventory data for specific unit operations associated the manufacture of softwood lumber manufactured in the Southeast region of the United States.
  
  The objectives of this module are to:
  
  
• Provide environmental, energy, and resource impact data on the manufacture of softwood lumber in the US Southeast region.
  
  
• Provide benchmarks for these products that will enable comparison to process improvements or new processes.
  
  
• Provide input for the Structures Module.
  
  
• Compare fossil versus biomass fuel dependency.
  
  
• Provide a measure of resource use efficiency.
  
  
• Provide a comparison of energy and resource impacts relative to the 1976 CORRIM study.
  
  
• Module D: Softwood Plywood - US Pacific Northwest and Southeast - presents life cycle inventory data for specific unit operations associated the manufacture of softwood plywood manufactured in the Pacific Northwest and Southeast regions of the United States.


The objectives of this module are to:


• Provide environmental, energy, and resource impact data on the manufacture of softwood plywood in the US Pacific Northwest and Southeast regions.
  
  
• Provide benchmarks for these products that will enable comparison to process improvements or new processes.
  
  
• Provide input for the Structures Module.
  
  
• Compare fossil versus biomass fuel dependency.
  
  
• Provide a measure of resource use efficiency.
  
  
• Provide a comparison of energy and resource impacts relative to the 1976 CORRIM study.
  
  
• Module E: Oriented Strandboard - presents life cycle inventory data for specific unit operations associated the manufacture of oriented strandboard manufactured in the Southeast region of the United States.
  
  The objectives of this module are to:
  
  
• Provide environmental, energy, and resource impact data on the manufacture of oriented strandboard in the US Southeast region.
  
  
• Provide benchmarks for these products that will enable comparison to process improvements or new processes.
  
  
• Provide input for the Structures Module.
• Compare fossil versus biomass fuel dependency.
  
  
• Provide a measure of resource use efficiency.
  
  
• Provide a comparison of energy and resource impacts relative to the 1976 CORRIM study.
  
  
• Module F: Design of Residential Building Shells, Minneapolis and Atlanta - outlines the design of typical light-frame residential structures for two different climates: a hot and humid climate represented by Atlanta, GA and a cold climate, represented by Minneapolis, MN, for use in subsequent life cycle analyses of representative structures.


The objectives of this module are to:


• Provide environmental, energy, and resource impact data on the manufacture of floor, wall, and roof components for residential and light commercial structures, both site constructed and factory built.
  
  
• Provide environmental, energy, and resource impact data on the manufacture of representative single residential structures, both site-constructed and factory built, comprised of floor, wall, and roof components (with considerations for windows, doors, roofing, and siding based on ATHENA data).
  
  
• Provide input for the Use and Disposal Module and future Structures Modules that will include internal components and fixtures.
  
  
• Provide data input for the Integration Module.
  
  
• Provide benchmark data for these processes that will enable future comparison of process improvements or to new processes.
  
  
• Provide a comparison of energy and resource impacts to the 1976 CORRIM study for floor, wall, and roof components.
  
  
• Compare fossil versus biomass fuel dependency.
  
  
• Provide a measure of resource use efficiency.
  
  
• Module G: Building Construction, Use, Maintenance, and Disposal - presents the results of a comprehensive analysis of environmental impacts related to the use of wood and non-wood materials in residential structures, and the durability of such structures relative to longevity and eventual replacement, or final use and disposal.


Objectives of this module are to:


• Define the durability of wood materials used in the structural shell of a building and to determine the longevity (useful life) of wood materials used in the construction shell as impacted by a given set of parameters. The parameters that impact the durability of the materials will be defined in the module. Wood materials to be considered in this module are softwood lumber, softwood plywood, oriented strandboard (OSB), laminated veneer lumber (LVL), parallel strand lumber (PSL), glulam beams, composite beams, and trusses. Only exterior doors and windows will be considered in addition to basic construction shell materials.
  
  
• Define those variables that influence durability of wood materials used in the construction shell and develop a methodology for measuring the impact of each material relative to its own durability and subsequently to the durability and life of the structure as a whole.
  
  
• Address durability from both a material performance and aesthetic perspectives (i.e. exterior shell component remodel for aesthetic reasons).
  
  
• Estimate the final disposition of components upon dismantling of a building, along with associated impacts on life cycle inventories.
  
  
• Provide input data for other modules, especially energy use and carbon storage comparisons.

2.0 LIFE CYCLE ANALYSIS FRAMEWORK

2.1 INTRODUCTION

2.2 CORRIM FRAMEWORK AND GUIDELINES

2.3 LIFE CYCLE ANALYSIS (LCA)

Figure 2.1 presents the general flows in a typical "cradle-to-grave" LCA and Figure 2.2 illustrates the major components of an LCA study. An LCA begins with an initiation phase during which a problem is defined along with the scope, system boundaries, data categories, and review process. Initiation is followed by three interrelated phases that may be conducted simultaneously or in a sequence that best suits the problem being studied: a life cycle inventory phase that identifies and quantifies the energy, resource use and environmental effects of a particular product, service, or activity; an impact assessment phase which investigates the potential environmental consequences of energy and natural resource consumption and waste releases associated with the system being studied, and an improvement assessment phase where opportunities to reduce environmental impacts and resource use are investigated. Figure 2.2 highlights some of the factors likely to be considered within each phase and emphasizes the iterative nature of the process.

Figure 2.1. General Flows in a "Cradle-to Grave" LCA System (Franklin Associates, 1990)

Figure 2.2. Main Components of an LCA Study (ATHENA 1997a)

2.4 CASTING THE CORRIM FRAMEWORK IN THE LCA CONTEXT: LIFE-CYCLE STAGES

The similarity between the CORRIM framework in Figure 1.1 and the generic life-cycle approach presented in Figure 2.1 is evident. There are, however, two rather unique features of the CORRIM framework.

First, while some LCAs involving agricultural products or bio-energy have expanded the generic model by including a crop growing stage prior to raw material acquisition (Andersson & Ohlsson 1999, Mann & Spath 1997), forest growth is more complex. The relative complexity of forest growth is due to 1) the much longer time frame involved, 2) the use of intermediate harvests (thinning) to yield products, 3) the broader array of joint products arising from a single tree (saw, veneer, and pulp logs) and stand (due to mixed species having different use preferences), and 4) the unique set of forest CO-products (water, recreation, berries and mushrooms, etc.) and environmental effects (such as water quality, species diversity, wildlife habitat, and carbon sequestration).

Second, buildings, which account for the largest use of wood and main focus of this study, are unique in their size, complexity, and longevity. For example, a residential home is constructed, used for a long period of time, and eventually demolished. The period of use and occupancy involves cycles of maintenance and repair (re-roofing) and may involve a series of owners each of which may remodel the structure to accommodate changes in desired functionality and aesthetics. As a result, the time frame between when tree seed germinates and when a home is demolished could be on the order of one to several centuries or more. Thus, the temporal distribution of events and associated environmental effects during the seed to demolition life-cycle must be considered; merely summing all of the events and effects would produce the naïve and meaningless result that all of the activities and associated impacts occur simultaneously.

Table 2.1 highlights the major differences between the generic LCA model and the CORRIM framework. The major stages in the generic LCA model evolved for investigating consumer products and packaging materials with short lives where a simple temporal summation of effects is reasonable. Table 2.1 also indicates a rough estimate of time associated with each of the components of the CORRIM framework with comments identifying some of the associated activities and environmental effects. The procedure developed by CORRIM to deal with these temporal issues in LCA is discussed more fully in Section 2.6. Identification of these temporal aspects of the forestry and housing components leads to Figure 2.3, which shows the major stages in the cradle-to-grave cycle of a building such as a residential home and also illustrates the two long term stages (forest growing and use of the building) that are important for carbon storage.

Table 2.1: Comparison of the Generic LCA Model and the CORRIM Research Framework

Generic LCA Model	CORRIM	Comment
	Forest Growth Time frame: 25-100+ years	Nursery, planting, thinning, fertilizing, during the growth cycle. Effects on carbon sequestration/global warming, diversity, habitat, streamside conditions, etc.
Raw Material Acquisition	Harvesting Time frame: < 1 year	Logging during commercial thinning or final harvest. Effects on soil compaction and productivity, diversity, habitat, siltation, etc.
Manufacturing	Manufacturing Processes Time frame: < 1 year	Individual products (lumber, plywood, LVL, OSB, etc) Assemblies of products (trusses, glulam beams, I-joists, etc.) Air & water emissions, solid waste.
	Construction of Structures Time frame: < 1 year	On-site or factory built components (floor, wall, roof) and finished structure. Solid waste.
Use/Reuse/Maintenance	Service Life & Use Time frame: 40-100+ years	Maintenance cycles (painting, re-roofing, siding, etc.) and remodeling. Energy use & associated emissions/waste, energy & emissions associated with repair/remodel products.
Recycle/Waste Management	Recycling & Disposal Time frame: < 1 year	Teardown, segregation of materials, recycle, combust for energy, landfill. Energy use & substitution, air & water emissions, solid waste/carbon sequestration.

Figure 2.3. A Depiction of CORRIM Phase I Research

2.4.1 LCA Components

Table 2.1 and Figure 2.3 suggest that the LCA of a building system is a composite of many separate, but interrelated, LCA'S. A building system LCA is created through cumulatively embedding the LCAs of many processes and their associated products, sub-assemblies and assemblies, each containing a specific set of life cycle stages from raw materials through recycling and waste management. For some products, certain stages may become deferred and be dependent on the aggregated LCA of the entire building of which they become a part. Examples are wall studs and other main framing lumber products that are covered by other materials and remain unchanged during the building life; these items typically have no maintenance stage of their own. Similarly, the recycling and disposal stage may only apply to only some of the products when the building is finally demolished (i.e. recycling and disposal of these individual products is not separable from the building as a whole).

2.4.2 Initiation and Scope of Phase I

Construction of a complete residential home can be viewed as involving three categories of component LCAs. The first of these includes all of the products and assemblies required to produce the structural skeleton of the building. The second category involves additional products, such as siding, insulation, and roof topping, to complete the building envelope. The third category involves products and activities to complete the interior.
As noted earlier, the scope of Phase I CORRIM research is restricted to a study of a subset of the overall system. This subset includes:

1. Forest growth and harvesting and product manufacturing in the US Pacific Northwest and Southeast. These two regions are the predominant suppliers of domestically manufactured wood building materials.
   
   
2. Wood products and assemblies needed to create the closed-in building envelopes that meet the requirements of prevailing building codes in a warm climate (Atlanta, GA) and a cold climate (Minneapolis, MN).

• Measures of effects on diversity and habitat, stream quality, etc. that are unique to the forest resource stage and provide indications of the impact of the forest activity.



• All of the input and output data compiled by CORRIM Phase I research on forest growing and harvesting activities are based on regional forest growth and yield models and recent studies of harvesting systems reported in the literature. No direct field studies were commissioned by CORRIM.

2.4.3.2 Product Manufacturing & Unit Processes

An objective of Phase I is to treat each major step of each manufacturing process as a separate unit ("unit process") for data collection. For example, a sawmill can be viewed as consisting of the following processing centers (or unit processes): log yard, debarking and bucking, primary breakdown of logs into trimmed, green lumber; kiln drying, planer mill, and grading and packaging. This level of detail allows for recognition of various product grades and joint products from each center that may be sold to other industries without further processing. It also recognizes alternative pathways of primary material through the mill (green lumber to grading & packaging versus green lumber to kiln drying). By performing data collection at the unit process level, many potential issues with respect to co-product allocation can be resolved and differences among various technological configurations for a unit process can be examined. Data for individual unit processes and facilities will vary by plant age and type, and factors such as transportation distances will also vary by region. Sampling individual facilities and unit processes provides a sense of the variation in the industry and permits summarization that provides the range, median, mode and upper/lower quartiles. This permits analyses that can focus on the sensitivity of outcomes to advances in adoption of technology such as indicated by the rhetorical question: "What happens as the industry gradually improves so the average moves up to the current upper quartile?"

CORRIM Phase I involved gathering of current data through questionnaires sent to selected individual manufacturers. Table 2.2 lists wood products and assemblies for which CORRIM data has been collected from manufacturers during Phase I.

Construction of buildings also involves consumption of non-wood materials such as concrete for foundations and steel for nails and fasteners, etc. Life-cycle data for these materials and products were taken from previous life-cycle studies compiled or conducted by the Athena Sustainable Materials Institute. Table 2.3 lists products for which CORRIM is using data from previously published LCAs during Phase I.

Table 2.3: Products Considered in CORRIM Using Data from Studies
Conducted by the Athena Sustainable Materials Institute

2.4.3.3 Building Construction

Construction of a building involves a fundamental shift from individual products to the combination of these into building assemblies (wall, floor, and roof systems) using spans, loads, and other variables that comply with building codes. The activities associated with construction include those involved in producing building materials, as well as new life-cycle steps and impacts tied to construction activity in which materials and energy are consumed, and solid wastes and emissions are produced. Some effects are simply the aggregate of the various materials used in the construction assemblies while others are unique to the construction activities and are not attributable to an individual component product. To integrate various combinations of products into functionally equivalent assemblies and competed structures, CORRIM used the Environmental Design Model of ATHENA™ (1997) which contains more than 50 different assemblies incorporating combinations of concrete, steel, and wood products. Working with ATHENA™ researchers, CORRIM developed additional structural designs based on residential building codes in Minneapolis, MN and Atlanta, GA to represent cold and warm climate conditions, respectively.

2.4.3.4 Building Use and Maintenance

Unlike siding, roofing and other assemblies required to fully complete the exterior of a building, the structural materials and assemblies generally require little direct maintenance during the operational life of a building. Consequently, maintenance requirements are not considered in CORRIM Phase I. Maintenance will be analyzed in detail in later research in the context of a fully completed and occupied building.

Energy consumption during the operational life of the building will be estimated based on the average home in each city hence providing perspective on the relative importance of this stage on overall life-cycle energy consumption and emissions. Further research on energy consumption during the operational life of buildings will be conducted in the future.

2.4.3.5 Building Demolition and Material Recycling or Disposal

Recycling and waste management focus on the individual materials for which recycling and waste management opportunities differ greatly. Energy and emissions are involved in demolition/deconstruction, separation and transport of the materials, and in recycling and waste management processes. Since a building may not be demolished for many years after construction, accurate depiction of this stage would require forecasting technologies and markets for recycling and waste management far into the future. Since these conditions are unknown, CORRIM assumes that current practices are applicable. Furthermore, research suggests that demolition, recycling, and disposal differences between alternative designs are small (FORINTEK & Wayne B. Trusty 1997). Therefore, comparisons between design alternatives will be reasonably accurate in the absence of good data for this stage. However, when absolute effects of individual designs are required, better data for this stage will be necessary, a topic for future research.

2.5 IMPROVEMENT ANALYSIS AND SCENARIOS

Improvement analysis refers to using LCA results to suggest opportunities for environmental improvement or to evaluate proposed changes to a product, process, or design. Since improvement analysis often involves different technical expertise (process design engineers, product designers), and broader considerations (economic and market factors) than information gained from a life-cycle study, it is often conducted independently of LCA or as an adjunct to an LCA. Through sensitivity and scenario analyses and studying material substitutions, CORRIM research will provide comparisons and insights that will assist analysts, designers, managers, and policy makers in understanding relative benefits and costs associated with alternatives and understanding which stages in the life-cycle provide the greatest leverage for improvement.

2.6 TEMPORAL ISSUES IN THE LCA CONTEXT

Many life cycle inventories are performed for non-durable products for which the time from resource extraction until post-consumer disposal was so short that the temporal distribution of events and associated impacts can be ignored. In contrast, the life cycle for CORRIM research covers a very long time frame, beginning with planting a forest stand, cultivating it to maturity, harvesting and converting logs into products that are assembled into buildings, tracking the activities during occupancy and use of the building until its eventual demolition and disposal. For CORRIM, the time dimension becomes great and presents difficulties that are not well discussed in the LCA literature. Some activities, such as the growth of a forest stand and associated carbon removals from the atmosphere and the consumption of energy for heating and cooling of a building, occur in relatively small annual quantities; the aggregate total when trees are harvested or when a building is demolished would be misleading since the totals and associated emissions did not occur in these specific single event years. Other events such as fertilization of the forest or re-roofing a building only occur at intervals so a total, expressed as if these events occurred simultaneously does not make sense. To properly deal with the timing of events and emissions and avoid misleading aggregations, CORRIM developed a method for portraying the life-cycle as a series of cohorts (forest stand and building age classes) that move through time. Correct accounting at any time point is the sum of the activities and emissions of those cohorts that coexist at that time. To illustrate, consider a simplified CORRIM life-cycle consisting of the following phases:

• Forest Planting and Growth

• Nursery



• Planting on the forest site



• Early age culture: weeding, pre-commercial thinning, fertilizing, pruning



• Middle age culture: commercial thinnings, fertilizing



• Mature stand until final harvest

• Harvesting, Product Manufacture, and Building Construction

The series of conversion and building processes from "stump to structure" completion that commonly takes 2 years or less

• Building Occupancy and Use
• Building occupancy and use consumes annual energy for lighting, heating, and cooling plus periodic maintenance events. The building may be used for 25 to 100 years or more. Consider the following building age cohorts.



° young relatively little maintenance, replacement, or remodeling occurs during the early use of a new home.
° middle-aged: the first cycle of maintenance and replacement activities begins, hence there is demand for materials to accomplish them as well as disposal issues for the replaced materials.
° old-aged: maintenance and replacement cycles may intensify as the house becomes outdated and homeowners undertake various remodeling steps.

Figure 2.4 portrays the "cradle-to-grave" perspective for the cohorts with the time-line shown underneath; "h" is the age when the forest is harvested and "T" is the time when the building is demolished. The problem with the timeline perspective is that it creates some logic issues and violates the tenet that LCI is based on empirical, not forecast data. At any time chosen as the present, data for all activities beyond that time point are forecast data that must rely on assumptions regarding changes in technology and design. If the present is assumed to be the time of planting seed, can we assume that today's building codes, design practices, and materials will be relevant when the home is built from these trees at h+2 years in the future? If not, what will the relevant codes, designs, and materials be? If thinnings are conducted, is it logical to pool them with the final harvest as inputs into the "stump to structure" stage? In the real world, no mill stores thinnings from a stand for 15 or 20 years in order to process them with the logs from the final harvest of that stand.

Figure 2.4. The Timeline of the CORRIM Life-Cycle

To illustrate a different type of problem, consider the final disposal of the building. The disposal, materials recovered, and emissions to the environment of a new building that will not be built until far into the future, and which will not be torn down until much later, are likely to be very different than the problems associated with tearing down and disposing of materials from the today's obsolete buildings. A question to be raised is "are we really interested in the cumulative impacts associated with planting seeds in a nursery and, perhaps 200 years into the future, tearing down and disposing of a building made from them?

The difficulties with the timeline approach in Figure 2.4 can be overcome by recognizing that all cohorts exist simultaneously in a time period (Figure 2.5). Today, nurseries are producing seedlings for planting on lands that were just harvested, there exists a mix of forest age classes receiving silvicultural treatments (pruning, thinning, fertilization), some are producing logs from thinnings and some from final harvesting for manufacturing processes that produce products, subassemblies and housing developed with known, current technologies and design methods. Owners of relatively new buildings are dealing with relatively minimal upkeep and repair; owners of, say, 15-30 year old buildings, are involved in many maintenance activities using current materials and code requirements; owners of much older homes are dealing with the unique problems they present; and some of the existing housing inventory is being demolished and replaced. The environmental impacts of demolition of today's obsolete homes reflect the actual materials of the past that were used to build and maintain them. This includes problems such as lead-based paints and asbestos that will not be used in future designs but are impacting current disposal and recycling options and will continue to have an impact in the near future.

Summation of activities performed on today's stocks of forest lands and housing, coupled with today's processing, construction, and demolition and disposal methods, provides a more realistic "bottom line" inventory report on the current status of resource and energy consumption and releases to the environment. Furthermore, empirical data can be collected on each of these activities because they are occurring now. Thus all data is empirical in the true spirit of a life-cycle inventory. Similarly, it is relatively easy to collect data on current costs of labor, capital, raw materials (both virgin and recycled) and energy associated with each activity and on market prices for products and develop benefit/cost information.

Figure 2.5. Simultaneous Cohorts in the CORRIM Life-Cycle System
Figure 2.5. Simultaneous Cohorts in the CORRIM Life-Cycle System

2.7 SCENARIO ANALYSES: POLICY CHANGES AND ECONOMIC LINKAGES

If one wishes to study the effect of a policy change, such as lengthening forest rotation cycles to store additional carbon, this can be easily accommodated with the cohort approach. Implementing such a policy would alter forest practices whereby each of today's age classes in Figure 2.5 becomes the start of a process where that class evolves over time in response to practices that would implement the policy (Figure 2.6). Thus, the young stands that were just thinned and fertilized in the present will evolve toward maturity over time under the influence of this policy.

Large bold cells represent pathways (left to right) from present age classes of forest lands and housing. Bold italic cells represent pathways based on future plantings after stocks of existing forests are harvested.

As the forest evolves, it yields logs from thinning and final harvest at any point in time that are processed by industry into products, subassemblies, and houses. The changes in forestry to accommodate the policy may alter the quantity, size, and quality of the logs available to industry thereby forcing changes in processes, products, and designs which create new homes for occupancy. Concurrently, maintenance, replacement, repair/remodel, and demolition/disposal is continuing on older homes produced in previous periods. Thus the effect of a policy produces changes in the forest which in turn have lagged effects on the characteristics of materials which later may affect technologies used, products, and design. By forecasting these changes to a time of interest in the future and summing the columns for the time periods, one can obtain inventory summaries and summarize the costs, revenues, and other benefits that occur. These "bottom line" summaries can then be compared in terms of change in physical parameters of interest and in terms of economics.

This approach, in which cradle-to-grave aspects of forest products are regarded as simultaneously occurring cohorts, allows analysts to develop a valid life-cycle-inventory status report on the current situation, develop economic measures of the current situation, and then use forecasting techniques and sensitivity analysis to estimate how changes in any system element lead to changes in environmental and economic measures over time.

This perspective provides a means for reconciling difficulties with temporal aspects of life-cycle inventories associated with forestry and housing and provides a means by which the tool of life-cycle inventory can be used in conjunction with the tools of economic analysis. It is also consistent with the approach commonly used by individual firms in applying and combining life-cycle inventory and improvement analysis methods in identifying cost-effective ways to provide products with improved environmental performance.

3.0 PHASE I ACCOMPLISHMENTS

The section describes the accomplishments achieved in Phase I in summary form. A major accomplishment is the preliminary environmental assessment for the construction of a single family building shell. Other accomplishments include life cycle inventories for wood and other processes that produce the materials and energy used to construct the single-family shell. Detailed information on these processes has not been available until now. Phase I also initiated research into other processes that have yet to be addressed elsewhere including cost analysis, carbon inventory measurements and dynamic issues associated with forest resources.

One of Phase I's major realizations is the preliminary environmental assessment for a building shell using two sets of different materials. Five indices are constructed and used to summarize the many output measures from the LCI on the building shell. The indices measure embodied energy, greenhouse warming potential, air pollution, water pollution and solid wastes. All measures, except one, indicate lower values for the wood design in Atlanta and Minneapolis. The one exception is in Minneapolis for solid waste, where the steel design produces about one ton less solid waste than the wood design. One can infer from the environmental assessment that the wood design results in lower energy and pollution measures with the one exception noted above. An examination of a change in forest management suggests small but significant changes in the index measures. A near 10% increase in the construction of wood design shells in Minneapolis associated with a 5% increase in land productivity results in index measure declines of 4% for the global warming index, 6% for the embodied energy and air pollution indices, 25% reduction in the water pollution index and a 1% increase in the solid waste index. A 5% increase in land productivity leads to a 17.5% increase in the construction of wood design shells in Atlanta and a reduction in all output measures ranging from 3% for water pollution to 13% for solid wastes.

To complete the environmental assessment Phase I produced several LCIs. Life cycle inventories were produced for forest resources, lumber manufacturing, plywood production, and oriented strand board manufacturing.

The Forest Resource module provides the environmental, energy and resource impact data on the growth, management, harvesting and reforestation of timber for the Pacific Northwest (PNW) and Southeast regions of the US for a representative acre within each region. The study calculates volume harvested for three general combinations of management intensity and site productivity for each region. The combinations were reduced to a single estimate of yield using weighting factors developed for this research. The weighting scheme creates a base case for each region. Alternative weights, such as more acres under a higher level of management intensity, produce an alternative case, with a different calculated volume. For example, the increase in merchantable volume under the alternative case was 21% in the Southeast (including volume from the second rotation when rotation ages are reduced) and 16% in the Pacific Northwest.

The portion of lumber and pulpwood produced in the Southeast changed slightly under the alternative scenario; lumber output share declined 2.1 percentage points and pulpwood increased 2.1 percentage points. The alternative scenario also leads to a 3.6 % decrease in the average rotation age in the Southeast, which is responsible for the percentage shifts in lumber and pulpwood produced. Lumber output increased 15% in the Southeast, only 68% of the total volume increase. Lumber output share remained at 100% in the base and alternative PNW scenarios.

System costs increased by around one percentage point above the percent change in the amount of merchantable volume removed in the Southeast; however, the shorter rotation may more than offset that increase in a discounted cash flow analysis. System costs increased about the same percent as merchantable volume removed in the PNW. A similar result is observed for fuel and lubricant consumption during harvesting operations: an 11% increase in the Southeast and 16% increase in the PNW roughly corresponding to the changes in merchantable volumes harvested, although there appears to be a slight increase in the consumption factors in the Southeast due to the change in rotation age and products harvested. The increase in harvested volume resulted from using 4 times more nitrogen and phosphate in the Southeast and 2 times mores nitrogen and phosphate in the PNW as fertilizer inputs in conjunction with the other management changes.

Along with harvested volumes, the study also produces harvesting cost data and air emission estimates related to stand growth and harvesting. Resources used to produce the harvested volume, such as fuel, fertilizer and herbicides, are quantified and their impacts reported in the Environmental Impact Report during the extraction phase of the single family shell construction.

The Forest Resource Module also produces estimates of tree biomass by component. These estimates are used to approximate the standing and removed carbon pool over time. Other environmental performance measures including indices of stand structure, diversity, habitat and fragmentation using a landscape approach to forest management will be developed in subsequent phases of the project.

Harvested volumes from the two forest resource regions are passed on to lumber manufacturers in the Pacific Northwest and Southeast regions. Here the research developed independent LCIs for the two regions. The two regional lumber manufacturing modules were developed to provide the environmental, energy and resource impact data on the manufacturing of softwood lumber. In the Northwest a survey produced the data on sawing, drying and planing processes. The unit process approach produces detailed descriptions of activities and the resource used to produce specific outputs. For example, the sawing process involves log movement within the mill, their sorting and storage, their delivery to debarkers, and bucked to length. The logs are then debarked and sawed into rough lumber producing rough lumber, pulp chips, bark and sawdust. The rough lumber is transported within the mill to stacks for kiln dryers or planer facilities, two additional processes with their corresponding activities. To accomplish the sawing activity maintenance work on equipment and vehicles are also recorded, as well as treatment processes of air, liquids and solid materials. The result of the survey work produced detail information that was then analyzed with the SimaPro life cycle program. The SimaPro analysis consisted of four processes, adding energy generation to the three processes mentioned above. The result of the analysis produces unit factor estimates for one Mbf of planed dry lumber. These unit factor estimates include raw material usage, airborne waterborne and solid emissions, and energy usage.

The Southeastern lumber module produces similar type of data. The survey responses were not detailed enough to break out the sawing processes into three components, hence sawing was considered as a single process. Therefore the SimaPro analysis consisted of only two processes: lumber and energy generation.

Harvested volumes from the two forest resource regions are also passed onto softwood plywood manufacturers in the Pacific Northwest and Southeast regions. The research module for softwood plywood production produces the life cycle inventory data used in the LCI for the building shell. A survey was implemented in a manner similar to the lumber manufacturing modules. The plywood process was defined in terms of six processes: bucking and debarking, block conditioning, peeling and clipping, drying, lay-up and pressing, and trimming and sawing. The result of the survey work produced detail information that was then analyzed with the SimaPro life cycle program. The result of the analysis produces unit factor estimates for one Msf of 3/8-in basis of plywood.

Phase I also produces an LCI for Oriented Strand Board (OSB) production. Using a survey, Phase I collected information on log transport, production of phenol-formaldehyde and MDI resins and wax. The environmental and energy impact data were recorded and passed along to the construction phase of the research.

Phase I provides data that will allow a comparison of energy and resource impacts relative to the 1976 CORRIM study. Data is also available to provide a measure of resource use efficiency. With the data produced by the lumber, plywood and OSB modules, process improvements can also be identified. An analysis of costs and carbon accounting has been initiated in Phase I as well as a preliminary integration of the various processes.

4.0 REFERENCES

AF&PA. 1994. Agenda 2020: A Vision and Research Agenda for America's Forest, Wood and
     Paper Industry. American Forest and Paper Association. Washington, D.C.

AF&PA. 1996. Life Cycle Inventory Analysis User's Guide: Enhanced Methods and
     Applications fort the Products of the Forest Industry. American Forest and Paper
     Association. Washington, DC

Andersson, K., and T. Ohlsson. 1999. Life Cycle Assessment of Bread Produced      on Different
     Scales. Intl J. LCA. 4(1): 25-40.

ATHENA™ 1993a. Raw Material Balances, Energy Profiles and Environmental Unit Factor
     Estimates: Cement and Structural Concrete Products. Prepared by Canada Center for Mineral
     & Energy Technology and Radian Canada, Inc. Athena Sustainable Materials Institute.
     Ottawa, Canada.

ATHENA™ 1993b. Raw Material Balances, Energy Profiles and Environmental Unit Factor
     Estimates: Structural Steel Products. Prepared by Stelco Technical Services, Ltd. Athena
     Sustainable Materials Institute. Ottawa, Canada.

ATHENA™ 1993c. Raw Material Balances, Energy Profiles and Environmental Unit Factor
     Estimates: Structural Wood Products. Prepared by FORINTEK Canada Corp. Athena
     Sustainable Materials Institute. Ottawa, Canada.

ATHENA™ 1997a. The Summary Reports: Phase II and Phase III and Research. Prepared by
     FORINTEK Canada and Wayne B Trusty & Associates. Ottawa, Canada.

ATHENA™ 1997b. Working with ATHENA™: Comparative Manual and Model Case Study
     Assessments. Prepared by The Environmental Research Group, School of Architecture,
     University of British Columbia. Athena Sustainable Materials Institute. Ottawa, Canada.

Curran, Mary Ann (Ed). 1996. Environmental Life Cycle Assessment. McGraw-Hill, New York.

Canadian Standards Association, Z760-94 Life Cycle Assessment: Environmental Technology
     (1994), ISSN 0317-5669, Toronto.

CORRIM 1998. Environmental Performance Research Priorities: Wood Products: Final Report
     on the research plan to develop environmental performance measures for renewable building
     materials with alternatives for improved performance. Consortium for Research on
     Renewable Industrial Materials (CORRIM, Inc.) DOE Agreement No. DE-FC07-96ID13427,
     Subcontract 417346.

CORRIM 2001. Research Guidelines for Life Cycle Inventories (draft). Consortium for Research
     on Renewable Industrial Materials (CORRIM, Inc.) Seattle, WA.

FORINTEK and Wayne B. Trusty, Ltd. 1997. The Summary Reports: Phase II and Phase III and
     Research Guidelines. FORINTEK and Wayne B. Trusty & Associated, Ltd. Ottawa, Canada.

Franklin Associates, LTD 1990. Resource and Environmental Profile Analysis of Polyethylene
     and Unbleached Paper Grocery Sacks: Final Report. Franklin Associates LTD. Prairie
     Village, KS.

IFIAS, 1974, Energy Analysis Workshop on Methodology and Convention, International
     Federation of Institutes for Advanced Study, Workshop Report No. 6, Stockholm.

ISO 1997. Environmental Management-Life Cycle Assessment-Principles and Framework, ISO
     14040. First Edition 1997-06-15, International Standards Organization, Geneva.

ISO 1998. Environmental Management-Life Cycle Assessment-Goal and Scope Definition and
     inventory analysis, ISO 14041. First Edition 1998-10-01, International Standards
     Organization, Geneva.

ISO 2000a. Environmental Management-Life Cycle Assessment-Life Cycle Impact Assessment,
     ISO 14042. First Edition 2000-03-01, International Standards Organization, Geneva.

ISO 2000b. Environmental Management-Life Cycle Assessment-Life Cycle Interpretation, ISO
     14043. First Edition 2000-03-01, International Standards Organization, Geneva.

Mann, Margaret K., and P.L. Spath. 1997. Life Cycle Assessment of a Biomass Gasification
     Combined-Cycle Power System. NREL/TP-430-23076. National Renewable Energy
     Laboratory, Golden, CO.

National Research Council 1976. Renewable Resources for Industrial Materials. National
     Academy of Sciences, Washington, DC. 267pp.