Oregon Harvested Wood Products Carbon
Inventory 1906 – 2018
Todd A. Morgan, Thomas S. Donahue, Thale Dillon
University of Montana, Bureau of Business and Economic Research, Forest Industry Research
Program
Andrew Yost
Oregon Department of Forestry
Jeremy Groom
Groom Analytics, LLC
April 2020
Prepared for:
USDA Forest Service, Forest Inventory and Analysis Program
Portland, OR
and
Oregon Department of Forestry
Salem, OR
Report completed through Agreement No. 18-CO-11261979-074 between the U.S. Forest
Service, Pacific Northwest Research Station and the Oregon Department of Forestry; and
Agreement No. 18-CR-11261979-095 between the U.S. Forest Service, Pacific Northwest
Research Station and the University of Montana, Bureau of Business and Economic Research.
2
Acknowledgements
The authors wish to acknowledge funding from the Oregon Department of Forestry and support
from the Pacific Northwest Research Station Forest Inventory and Analysis Program analyst
Glenn Christensen. The authors also thank CalFire research program specialist Mark
Rosenberg; Utah State University’s programmer/analyst Chris Garrard; research forester Nate
Anderson of the Forest Service’s Rocky Mountain Research Station; Dan Loeffler; and the
University of Montana’s Bureau of Business and Economic Research senior research associate
Eric Simmons for their assistance with this project. The authors also acknowledge the critical
reviews of the manuscript from four independent reviewers.
3
Introduction: Forest Carbon Accounting in Oregon
Forest ecosystems are an important and dynamic component of the carbon cycle. While
reducing emissions from combustion of geologic sources of energy (e.g., coal, oil, and natural
gas) remains the most effective and direct way for humans to control rising levels of
atmospheric carbon (Holl and Brancalion 2020), managing forests to increase carbon
sequestration and storage is often cited as a major element in addressing the climate crisis
(McKinley et al. 2011). Understanding the capacity for forests to mitigate fossil fuel emissions
requires a reliable inventory of the stock and flux of carbon in forests and the wood products
generated from timber harvesting. Indeed, there has been a decades-long demand for a reliable
forest carbon accounting framework in Oregon to better inform forest managers about the
capacity for increasing carbon sequestration and storage. For example, legislation passed as
early as 2001 required the State Forester to develop a forest carbon accounting system (ORS
526.783). Through the 2011 Forestry Program for Oregon, the Board of Forestry established
goals for developing a system for monitoring carbon in forests and harvested wood products
(HWP). Further, the Oregon Global Warming Commission (OGWC) was mandated to track and
evaluate the carbon sequestration potential of Oregon’s forests and the carbon stored in tree-
based building materials (ORS 468A.250(1)(i)). More recently, the Forest Carbon Accounting
Report (OGWC 2018) described Oregon’s long-standing need for a reliable forest carbon
accounting system to monitor the status and trends of carbon storage and annual changes (flux)
in carbon storage in forest ecosystems and HWP.
Recognizing the need for proper forest carbon accounting in Oregon, the Forest Ecosystem
Carbon Report (Christensen et al. 2019) presented the forest ecosystem dimension of the
accounting framework. That report provided estimates of the amount of carbon stored in seven
pools of forest carbon based on the Forest Inventory and Analysis (FIA) Program’s annual
inventory of field plots distributed across Oregon’s forests. These were initially measured from
2001 to 2010. The report also provided estimates of the amount of carbon flux to and from those
pools based on measurements of the same field plots revisited from 2011 to 2016. Estimates of
forest carbon flux in the ecosystem report were based on calculations of tree growth, mortality,
and removals between the two periods. The report was a result of a partnership between the
Governor’s Office of Carbon Policy, Oregon Department of Forestry (ODF), and the Pacific
Northwest Research Station (PNW) of the USDA Forest Service.
Reliable estimates of stocks and flux of carbon in forest ecosystems and HWP provide a basis
for analyses of tradeoffs between carbon storage and other forest management objectives,
which can inform forest managers, policy-makers, and the public (Galik and Jackson 2009,
Ryan et al. 2010, McKinley et al. 2011). Although HWP C constitutes a relatively small fraction
of forest carbon relative to ecosystem carbon (Loeffler et al. 2019), it is a fundamental
component of carbon accounting and base of information for evaluating various strategies to
reduce atmospheric carbon dioxide (CO
2
) concentrations.
Forest C removed by timber harvest is not released immediately into the atmosphere because
timber harvests transfer a portion of the C stored in wood to a "product pool." Once in a product
pool, the C is emitted over time mostly in the form of CO
2
as it decomposes. When discarded
wood products are burned, other greenhouse gases (CH
4
, N
2
O, CO, and NOx) are emitted. The
4
rate of emission varies considerably among different product pools. For example, if timber is
harvested for fuelwood to produce energy, combustion releases C immediately but if timber is
harvested and used as lumber in a house, it may be many decades or even centuries before the
lumber decays and C is released to the atmosphere. If wood products are disposed of in solid
waste disposal sites (SWDS), the C contained in the wood may be released many years or
decades later or may be stored almost permanently in SWDS (EPA 2020).
This report provides estimates for the storage and flux of carbon in wood products
manufactured from trees harvested in Oregon since 1906 (Andrews and Kutara 2005, Simmons
et al. 2016). It was made possible through a partnership with ODF, PNW-FIA, and the Bureau of
Business and Economic Research (BBER) at the University of Montana.
Objectives
The objectives of this analysis were to:
1) Use the production accounting approach to generate estimates and confidence intervals
for the stocks and flux of carbon in the HWP pool for timber harvested from 1906 to 2017
within the State of Oregon.
2) Generate estimates and confidence intervals for the cumulative emissions of CO
2
from
burning fuelwood for energy capture and from the decay and burning of discarded wood
products.
3) Compare the amount of carbon in HWP among the major Oregon forest ownerships that
contributed harvested timber to the pool of HWP from 1962-2017 harvests.
4) Combine the estimate of stock and flux for the HWP pool with the average in Oregon’s
forest ecosystems to evaluate the total stocks and flux in forests and HWP from 2001-
2016.
5) Provide a reporting framework for subsequent HWP C analyses in Oregon that can be
applied to other regions with available timber harvest and end-product data.
This report neither advocates any particular course of action for forest carbon management, nor
does it include a life-cycle analysis (LCA) of wood products, estimates for the substitution
benefits of wood products, leakage, or estimates of greenhouse gas emissions associated with
timber harvesting, wood products manufacturing or related transportation.
Methods
Harvested wood products include lumber, panels, paper, paperboard, and wood used for fuel,
and the HWP C pool includes both products currently in use, products-in-use (PIU), and
products that have been discarded to SWDS. Additions to the HWP C pool are made through
5
harvesting timber for wood products, and emissions that occur from the decay and combustion
of wood products subtract from the pool.
The HWP C analysis described in this paper was based on timber harvested and processed in
the State of Oregon, as well as timber that was harvested in Oregon and processed (or burned
as fuelwood) outside the state. It does not include timber from outside Oregon that was
processed by Oregon facilities (mills) or used directly by residents of Oregon. Material left in the
forest after harvest (i.e., logging residue) is not accounted for in this report as HWP C but can
represent approximately 32% of total tree biomass as treetops, branches, and foliage (Ganguly
et al. 2020).
Nearly all (over 99%) of the bark on logs delivered to primary processing facilities is burned for
energy or used for mulch and other landscaping products (Brandt et al. 2006, Simmons et al.
2016, Simmons et al. 2019). Therefore, bark is considered a short-lived by-product with
negligible contribution to the pool of HWP C. Nevertheless, approximations were made for the
amount of biogenic emissions occurring from utilization of bark following Harmon (1996).
Oregon records timber harvest with the Scribner decimal c log rule (32-foot log West OR, 16-
foot log East Oregon), which uses inside-bark measurements to quantify wood volume. Total
harvest volumes were converted from board foot volumes to cubic feet based on the conversion
factors listed in Table 1 (page 8) and the carbon in bark was approximated with the following
equation:
= ×
1
All timber harvested in a specific year is utilized as PIU, SWDS or fuelwood for energy capture
throughout the year of harvest with the final amounts for each pool reported in the subsequent
year. The model will generate estimates for emissions with energy capture from discarded wood
products when reliable information for parameterization becomes available.
Model
As pointed out by Stockmann et al. (2012) and Loeffler et al. (2019), monitoring systems for
HWP C have been implemented at the national level. There are well-established and robust
inventory-based methods for estimating carbon stocks and fluxes in forest ecosystems, with
several tools available to forest managers (Galik et al. 2009, Smith et al. 2004, Smith et al.
2006, Zheng et al. 2010). However, some of these tools, such as the U.S. Forest Carbon
Calculation Tool (Smith et al. 2006), do not provide estimates of HWP C, while others, such as
WOODCARB II (Skog 2008) are restricted to national level HWP C accounting. Neither of these
models are accessible or practical tools for estimating and monitoring stocks and fluxes in HWP
C at the state level (Ingerson 2011, Stockmann et al. 2012).
6
Estimates of HWP C for the state of Oregon were calculated using the Intergovernmental Panel
on Climate Change (IPCC) Tier 3 production accounting approach
1
, which only considers timber
harvested in a particular area of study. The HWP C model has been used by several National
Forest System (NFS) regions to quantify carbon stored in wood products manufactured from
NFS timber (Anderson et al. 2013, USURS 2013, Butler et al. 2014, Loeffler et al. 2014,
Stockmann et al. 2012). It has also been used to produce state-specific estimates for California
(Loeffler et al. 2019). The software to run the HWP C model was programmed in R by Groom
Analytics and available online at www.ODF.gov
.
The HWP C model uses a series of calculations to estimate carbon storage and emissions for
timber that is harvested and used for wood products. Stockmann et al. (2012) provide a flow
chart of the model (Figure 1) demonstrating how HWP C is tracked through a product’s life, from
timber harvest through timber products, primary wood products, end use products, and finally to
disposal. For a more detailed description of the model assumptions and calculations, see
Stockmann et al. (2012) and Loeffler et al. (2019).
Figure 1. Diagram of the harvested wood products model used to quantify carbon
storage pools and emissions.
Source: Stockmann et al. 2012
1
IPCC Tier 3 denotes the availability of highly detailed data (e.g., from field plots) and the use of simulation models,
whereas Tier 1 and 2 use more general data that result in higher uncertainty.
7
Four sets of inputs are required to run the HWP C model: 1) annual timber harvest volume; 2)
annual timber product ratios that allocate harvest to different timber product classes; 3) annual
primary product ratios allocating the timber products to a variety of primary products and residue
uses, and 4) end use ratios that allocate the primary products to a larger set of end use
products. For this Oregon HWP C analysis, the input files were developed from Oregon-specific
data, the sources for which are outlined below.
Data Sources
Timber Harvest Data
Data for Oregon’s timber harvest volume is quite extensive but somewhat incomplete for the
earliest years. Andrews and Kutara (2005) compiled state-level harvest volumes going back to
1925 and state-level lumber production as far back as 1849. County-level timber harvest data
with three ownership classes (Tribal, U.S. Forest Service [USFS], private and state combined)
have been compiled since 1953, and with the seven ownership classes currently in use
(industrial private, non-industrial private [NIP], Tribal, State, Bureau of Land Management
[BLM], USFS, and other public) since 1962. Both classifications are from the Oregon
Department of Forestry (Andrews and Kutara 2005, ODF 2019). The earliest vintage allowed by
the HWP C model is 1906 because it was originally developed for use by the NFS, which was
transferred to the Department of Agriculture’s U.S. Forest Service in 1905.
Estimates of Oregon timber harvest volumes without ownership classifications are available
from 1906 to 1961. Harvest volumes by ownership are available for a shorter time—from 1962
to 2017—allowing for an ownership-level analysis for this period. The major limitation
associated with the data for years prior to 1962 is the absence of reliable harvest data for each
ownership and the lack of quantitative information from mill studies necessary for developing
timber product and primary product ratios for each ownership.
The five ownership classes reported in this analysis include USFS, BLM, Tribal, Private
(including industrial and non-industrial lands) and State and other public (including ODF and
other state lands, county and municipal lands). See Appendix 1, table 1A, for Oregon timber
harvest by ownership class. State-level Oregon timber harvest data for 1906 to 2017 are
included in Appendix 1, table 1B.
Because it was originally developed for use by NFS, the HWP C model requires timber harvest
input files with volumes expressed in hundred cubic feet (ccf). Timber harvested in Oregon were
reported in a mix of thousand board feet (mbf) Scribner, log rule and mbf lumber tally (Andrews
and Kutara 2005) and were converted to cubic feet (cf) of logs using conversion factors from
literature and an ordinary least squares regression equation (r-square = 0.9871) formulated
from published bf/cf factors (Table 1).
8
Timber Product and Primary Product Data
Oregon harvest and mill studies (Andrews and Cowlin 1940; Cowlin et al. 1942; Metcalf 1965;
Manock et al. 1970; Schuldt and Howard 1974; Howard and Hiserote 1978; Howard 1984;
Howard and Ward 1988, 1991; Ward 1995, 1997; Ward et al. 2000; Brandt et al. 2006; Gale et
al. 2012; Simmons et al. 2016) were used to develop ratios for timber products (e.g., softwood
sawlogs, softwood pulpwood, etc.) and primary products (e.g., softwood lumber, softwood
plywood, softwood mill residue used for energy, etc.). For years when specific product ratios
could not be determined from the literature, ratios from the previous or following mill study year
were used.
Table 1. Conversion factors used in the Oregon HWP C analysis
Conversion Units Source
8.596 bf per cf, timber harvest 1906 – 1910
Regression Equation
8.141 bf per cf, timber harvest 1911 – 1920
7.6923 bf per cf, timber harvest 1921 – 1930 Andrews and Cowlin (1940)
7.231 bf per cf, timber harvest 1931 – 1940
Regression Equation
6.776 bf per cf, timber harvest 1941 – 1950
6.321 bf per cf, timber harvest 1951 – 1960
5.866 bf per cf, timber harvest 1961 – 1970
5.42 bf per cf, timber harvest 1971 – 1979
Keegan et al. (2010)
5.17 bf per cf, timber harvest 1980 – 1989
4.55 bf per cf, timber harvest 1990 – 1999
4.0674 bf per cf, timber harvest 2000 – 2003 Brandt et al. (2006)
4.1813 bf per cf, timber harvest 2004 – 2008 Gale et al. (2012)
4.0161 bf per cf, timber harvest 2009 – 2017 Simmons et al. (2016)
33 to 42 lbs per cubic foot, primary products
Smith et al. (2006)
2204.6 lbs per metric ton (MT)
0.95 to 1.0 Metric ton wood fiber per metric ton product
0.5 Metric ton carbon (MT C) per dry metric ton wood fiber
0.71 to 0.91 MT C per ccf, primary products
1
1
See Appendix 1, table 1C (embedded .pdf file), for the model’s ccf to MT C conversion for each primary product.
The model utilizes 40 timber product classes (Appendix 1, table 1D embedded .pdf file),
requiring harvest to be allocated to timber product categories. Since Oregon timber harvest
records contain only harvest volume and do not allocate harvest among different timber product
types, harvest and mill studies that quantified the volume of timber used for different products
were used to calculate the proportion of total timber harvest that went into each timber product
category. Quantitative information from two published reports for the period 1906 to 1941 was
9
used to develop timber product ratios for timber from the Douglas-fir and ponderosa pine
regions of Washington and Oregon (Andrews and Cowlin 1940, Cowlin et al. 1942). Timber
product ratios were developed from Metcalf (1965) for the period 1942 to 1961.
Data for developing timber product ratios for harvest after 1961 were more readily available
2
.
The portion of the harvest allocated to each timber product class was calculated for each
harvest year for which data were reported. Although there are 40 timber product classes in the
model, the majority of harvested timber was softwood sawtimber for both periods (Table 2). See
Appendix 2, embedded .pdf file A, for Oregon timber product ratios from 1906 to 2017.
Table 2. Average annual Oregon timber product ratios, 1906-1961 and 1962-2017.
Products class
1906 to 1961 1962 to 2017
Mean Std. deviation Mean Std. deviation
Sawtimber, hardwood
0.003
0.001
0.016
0.010
Sawtimber, softwood
0.818
0.030
0.921
0.061
Pulpwood, hardwood
0.002
0.002
0.011
0.013
Pulpwood, softwood
0.034
0.035
0.042
0.040
Poles, softwood
0.003
0.000
0.004
0.002
Fuelwood, softwood
0.128
0.004
0.002
0.017
Other
0.012
0.010
0.004
0.004
There were 64 primary product classes with ratios, developed from the USFS Pacific Northwest
Region HWP C report (Butler et al. 2014) used for the 1906 to 1941 period, and Oregon-specific
primary product ratios developed from Manock (1970) that were used for 1942 to 1961. For
1962 and later, primary product ratios were developed from the same literature used to
calculate timber product ratios
3
. Mill residues are included as primary wood products, with some
entering solid waste disposal sites (SWDS) immediately, some being burned for energy, and
some being converted into products that rely on mill residues as raw material, such as
particleboard and paper. See Appendix 2, embedded file B, for Oregon’s primary products ratios
1906 to 2017. Estimates of primary products volume (in ccf) were converted to metric tons of
carbon (MT C) using product specific conversion factors (Smith et al. 2006; Appendix 1, table
1C, embedded .pdf file)
2
Metcalf 1965; Manock et al. 1970; Schuldt and Howard 1974; Howard and Hiserote 1978; Howard 1984; Howard
and Ward 1988, 1991; Ward 1995, 1997; Ward et al. 2000; Brandt et al. 2006; Gale et al. 2012; Simmons et al. 2016.
3
Manock et al. 1970; Schuldt and Howard 1974; Howard and Hiserote 1978; Howard 1984; Howard and Ward 1988,
1991; Ward 1995, 1997; Ward et al. 2000; Brandt et al. 2006; Gale et al. 2012; Simmons et al. 2016.
10
End-Use Data
The fate of HWP C is highly dependent on the end use of the primary products. For example,
the release of carbon from lumber used in new home construction has a longer duration than
carbon released from lumber used for shipping containers, which is released into the
atmosphere more quickly through combustion and decay. Fuelwood products are assumed to
have full emissions with energy capture in the year they were produced.
Following the methodology advanced by Stockmann et al. (2012) and Loeffler et al. (2019),
annual primary product output is distributed to specific end-use categories within the HWP C
model according to annual wood product consumption estimates (McKeever 2009, McKeever
and Howard 2011). The model’s primary products and corresponding end use categories are
shown in Appendix 1, table 1E (embedded .pdf file). A national data set was used for a series of
analyses and reports generated for all NFS Regions (USFS 2019) for the distribution of primary
products to end uses (Appendix 2, embedded .pdf file C).
The HWP model has 224 different possible end uses for HWP per harvest year (e.g., softwood
lumber/new housing/single family, softwood lumber/new housing/multifamily, softwood
lumber/new housing/manufactured housing, softwood lumber/manufacturing/furniture, softwood
lumber/packaging and shipping, etc.). The amount of carbon remaining in use during each
inventory year was calculated based on the products’ half-lives (Appendix 2, embedded file D)
and the number of years that have passed between the year of harvest and the inventory year.
An end-use product’s half-life value is the decay rate at which carbon in the PIU category
passes into the discarded-products category, representing the transition between the two pools
(Appendix 1, table 1F, embedded .pdf file). The amount of HWP C remaining in use in any given
year was calculated for each end use from all prior years with the standard decay formula:
=
exp (
()
/
)
where N
t
is the amount of carbon remaining in use in inventory year t, N is the amount of carbon
in the end use category in the vintage year of harvest, t is the number of years since harvest, t
1/2
is the half-life of carbon in that end use, and exp is notation for the exponential function. In our
calculations, the starting amount (N
0
, at n=0) is adjusted downward by an 8% loss factor
(McKeever 2004, Skog 2008) to reflect an immediate transfer to the pool of discarded products
before entering the PIU pool. This “loss in use” accounts for waste when primary products (e.g.,
softwood lumber) are put into specific end uses (e.g., new single-family residential housing).
For a given inventory year, the balance of HWP C that is not in use and not emitted with energy
capture is assumed to be in the discarded products category. Carbon in discarded products falls
into one of five disposition categories: burned, recovered, compost, landfills and dumps
4
. The
proportion of discarded products that ends up in each of these five categories is different for
paper vs. solid wood products, and changes over time. Prior to 1970, wood and paper waste
were generally discarded to dumps, where it was subject to higher rates of decay than in
4
Dumps are open-air disposal sites with high decomposition rates, while landfills are environments with lower decomposition rates.
11
modern landfills. Since then, the proportion of discarded wood and paper going to dumps has
dropped, while the proportion going to landfills has risen, with the remainder going to the other
two disposition categories, composting and recovery. Composting and recovery (i.e., recycling
and reuse) have become a more prominent part of contemporary waste management systems.
In the HWP C model, carbon in compost is assumed to transition directly to emissions while
carbon in recovery has a half-life decay factor that treats the discarded portion as emissions.
The model’s disposal of carbon in paper and solid wood products to dumps and landfills
categories is based on discarded products disposition ratios (Appendix 2, embedded file E) from
Skog (2008). Following the passage of the Resource Conservation and Recovery Act of 1976
(RCRA, 42 USC § 6901), a much larger portion of discarded HWP goes into modern landfills
rather than aerobic dumps or disposed through open burning, which were the dominant form of
SWDS prior to RCRA. The model assumes that carbon from discarded products that are burned
or composted is emitted without energy capture due to a lack of reliable data to support the
alternative (Stockmann et al. 2012, Loeffler et al. 2019). Carbon in the recovered category re-
enters the PIU category in the year of recovery. Carbon in products discarded to landfills and
dumps are subject to decay determined by their respective half-lives. Only a fraction of the
discarded products pool in landfills is subject to decay and associated with emission of carbon.
The portion of the discarded products pool not subject to decay is considered “fixed carbon”
(Appendix Table 1F embedded .pdf file). For a given year, the carbon remaining in SWDS is the
sum of fixed carbon and the carbon remaining after decay.
The HWP C model calculates and reports HWP C stock reductions (i.e., emissions) in MT C,
and does not estimate the different forms of C emissions (e.g., as methane, carbon monoxide,
carbon dioxide, etc.). Estimates for carbon emissions from HWP were multiplied by the atomic
weight of carbon dioxide divided by the atomic weight of carbon (i.e., 44/12) to find the carbon
dioxide equivalent (CO
2
e). All landfill and dump emissions are considered emissions without
energy capture. Methane remediation that includes combustion and subsequent emissions with
energy capture at landfills is not modelled. These methods are used to calculate annual HWP C
in the PIU and SWDS pools and emissions for all inventory years from 1906 through 2017.
Uncertainty Analysis
A Monte Carlo (MC) simulation analysis was conducted to estimate the uncertainty associated
with estimates generated with the HWP C model following methods described by Skog (2008)
and used by Anderson et al. (2013) and Stockmann et al. (2012). The goal of the MC simulation
was to produce 90% confidence intervals for the cumulative amount of carbon classified in four
categories: SWDS, PIU, emissions without energy capture (EWOEC), and emissions with
energy capture (EEC) from fuelwood. To achieve this goal, the MC simulation directly altered 16
different variables within the model according to their associated parameters (Appendix 3, table
3a). These 16 variables were allowed to vary by amounts that were based on estimates from
Skog (2008) and professional judgement. Random values were drawn from triangular
distributions that have a peak value of 1.0 (Appendix 3, table 3a) and symmetrically taper to
given 90% confidence interval bounds. The random values from the triangular distribution were
used as proportions for adjusting parameter values for each iteration of the simulation. A full
12
description of the methods for the MC simulation is provided in Appendix 3. Software to operate
the HWP C model and MC simulation was written in the R (2020) programming environment.
Results
Estimates of HWP C stocks and flux for Oregon’s timber harvests from 1906 to 2017 are
reported below, followed by results of the uncertainty analysis. Next, HWP C estimate at the
ownership level are reported for the 1962 to 2017 harvest period. Finally, HWP C stock and flux
estimates are evaluated with results from the Oregon Forest Ecosystems Carbon Report
(Christiansen et al. 2019).
State-wide HWP Carbon Stocks and Flux, 1907–2018
Changes in Oregon’s timber harvest levels by county and ownership have been well
documented (Andrews and Kutara 2005, ODF 2019). Recent trends in harvest ownership,
species, and product mix have been discussed in the context of wood products markets, and
the broader economy, as well as changing forest policy and industry infrastructure (Brandt et al.
2006, Gale et al. 2012, Simmons et al. 2016). The analysis that follows focuses on the
consequences of Oregon’s dynamic timber harvest levels and national wood use and disposal
on the stock and flux of HWP C since 1906. Timber product output (TPO) refers to the quantity
of carbon in harvested timber as estimated by the model.
In 1906, annual TPO was approximately 1.3 MMT C and grew to 5.1 MMT C by 1929 (Figure 2).
Timber harvest steadily increased, after dropping to 1.8 MMT C of TPO in 1932, with greater
demand for wood products during World War II and the housing boom that followed (Andrews
and Kutara 2005, Brandt et al. 2006).
13
Figure 2. Annual and cumulative timber product output (TPO) in MMT C from Oregon's
forests, 1906-2017.
Annual TPO reached a peak in 1972 at 13.3 MMT C and decreased to 8.2 MMT C by 1981 from
three recessions between 1973 and 1982. Annual TPO recovered to 12.5 MMT C in 1986
before its descent during the 1990s when the USFS timber harvest dropped steeply with
implementation of the Northwest Forest Plan (Spies et al. 2019). Recent notable changes
occurred around the Great Recession in 2009, when annual TPO and harvest declined by one-
third, to less than 5.1 MMT C, before rebounding to pre-recession levels of 7.6 MMT C in 2014
(Table 3). In 2017, the last year of timber harvest data at the time of analysis, annual TPO was
7.1 MMT C, and cumulative TPO was 814.4 MMT C (Table 3).
The carbon in TPO is added to the pool of HWP carbon as PIU, which consists of all 224 end-
use products. The carbon in the PIU pool matriculates into the SWDS pool and is emitted to the
atmosphere as wood decays, is burned as fuelwood for energy capture, or burned without
energy capture. For instance, the 1962 total harvest volume of approximately 8.5 bbf Scribner
represented TPO of 10.7 MMT C, with about 7.2 MMT C added to the PIU pool and about 11.4
MMT CO2e emitted from burning fuelwood and wood disposal.
0
150
300
450
600
750
900
1,050
0
2
4
6
8
10
12
14
1906
1910
1914
1918
1922
1926
1930
1934
1938
1942
1946
1950
1954
1958
1962
1966
1970
1974
1978
1982
1986
1990
1994
1998
2002
2006
2010
2014
Cumulative TPO MMT C
Annual TPO MMT C
TPO_MMTC
CumTPO_MMTC
14
Table 3. Oregon timber harvest, timber product output (TPO) and cumulative TPO, 2000-
2017.
Harvest year Timber harvest Timber product output Cumulative TPO
bbf Scribner
MMT C
MMT C
2000
3.9
7.0
696.0
2001
3.4
6.3
702.3
2002
3.9
7.2
709.4
2003
4.0
7.3
716.8
2004
4.5
7.9
724.7
2005
4.4
7.8
732.5
2006
4.3
7.7
740.2
2007
3.8
6.7
746.9
2008
3.4
6.1
753.1
2009
2.7
5.1
758.1
2010
3.2
6.0
764.1
2011
3.6
6.7
770.8
2012
3.7
6.9
777.8
2013
4.2
7.8
785.5
2014
4.1
7.6
793.1
2015
3.8
7.0
800.1
2016
3.9
7.2
807.3
2017
3.9
7.1
814.4
See Appendix 1, table 1B, for all annual harvest and TPO data years.
Carbon in PIU is held in end-use or recovered products, and carbon in SWDS is held in landfills
or in dumps. Over time, stored carbon is emitted with energy capture (e.g., using fuelwood for
energy) or without energy capture (i.e., emitted from landfills, dumps, recovered products,
burning and compost). Emissions from the burning of discarded products with energy capture
are treated as zero in this report because reliable data were not available to parameterize that
portion of the HWP model (Appendix 2, embedded .pdf file C). Approximately 1,700.5 MMT
CO2e (463.8 MMT C) was emitted through combustion and decomposition of wood products
since 1906 (Table 4). Thus, total biogenic emissions from HWP account for 56.9% of cumulative
TPO. The cumulative HWP stock was 350.7 MMT C, about 43.1% of cumulative TPO since
1906. Products-in-use account for 57.3% (201.1 MMT C) of the total HWP C stock, while
products in SWDS account for 42.7% (149.6 MMT C) of the total HWP C stock.
15
Table 4. Cumulative TPO, HWP C stocks, and emissions from Oregon timber harvests,
1906-2017.
Cumulative TPO MMT C
1906-2017
814.4
Cumulative storage as of 2018
Products-in-use
201.1
End-use products
194.9
Recovered products
6.2
Products in SWDS
149.6
Landfills
138.2
Dumps
11.4
Total HWP stock 350.7
Cumulative emissions as of 2018 MMT C MMT CO
2
e
Emissions with energy capture
155.6
570.4
Emissions without energy capture 308.2 1130.2
Landfills
30.6
112.3
Dumps
130.0
476.5
Recovered products
46.4
170.0
Burning
92.6
339.7
Compost
8.6
31.7
Total emissions
463.8
1700.5
The model output reveals a serrated pattern of annual change to the PIU pool (Figure 3) that
generally reflects the history of timber harvest and TPO (Figure 2). For most years between
1906 and 1992, changes to the PIU pool were positive – more C entered the pool through TPO
and HWP manufacturing than was transferred to SWDS or emissions (Appendix Table 1G). The
time series for this pool can be divided into three basic time periods. Up until 1940 the annual
change to the PIU pool averaged about 1 MMT C/year but then climbed to about 3 MMT C/year
up until 1992, after which it declined to 0.4 MMT C/year until the end of the time series. The
annual change in the PIU pool dropped to below zero during the Great Depression in 1933, near
the end of the Federal timber harvest declines in 1995-2000, and in response to the Great
Recession and housing bust in 2009-2011. During periods of negative annual change, more
carbon transitioned into SWDS than was added to the pool of PIU from harvests, usually as a
result of one or more years of steeply declining timber harvest.
The pool of C in SWDS has continued to grow as HWP from earlier harvest years matriculate
out of PIU and into SWDS (Figure 3 and 4). This movement of HWP C into SWDS, particularly
landfills, has contributed to the overall positive net change in HWP C. The annual change in the
SWDS pool (Figure 3) was positive across the time series and increased in a linear fashion up
until its peak in 1990 at approximately 3.3 MMT C/year. With the decline in timber harvest from
federally owned forests in the 1990s, the annual change in the SWDS pool dropped to
approximately 1.3 MMT C/year by 2002, after which it climbed to about 1.9 MMT C/year by the
16
end of the time series. Consequently, the cumulative amount of C in the PIU pool increased at a
smaller rate after 1992 (Figure 4). The PIU pool increased by approximately 10 MMT C in the 25
years from 1993 to 2018, whereas in the 12 years from 1980 to 1992 it increased by
approximately 30.7 MMT C.
Figure 3. Annual additions to the products-in-use (PIU) and solid waste disposal sites
(SWDS) pools from timber harvested in Oregon 1906-2017. Annual emissions with and
without energy capture (in MMT C/yr) are represented as negative values. “Net”
represents the sum of annual additions and annual emissions.
The amount of annual biogenic emissions from burning fuelwood with energy capture (EEC)
peaked in 1942 at 3.6 MMT C/year (13.2 MMT CO
2
e/year) but then declined in a linear fashion
after 1963 to the end of the time series (emissions are represented as negative values in Figure
3 and 4). Annual biogenic emissions from the decay and burning of discarded HWP without
energy capture (EWOEC) peaked in 1980 at 4.8 MMT C/year (17.7 MMT CO
2
e/year) and have
averaged about 4.3 MMT C/year (15.8 MMT CO
2
e/year) since then. From 1906 to 1975, EEC
was greater than EWOEC but since 1975 EWOEC exceeded EEC (Figure 3). These facts are
likely due to several related factors: 1) the SWDS pool continues to grow as a function of
cumulative TPO and PIU reaching the end of their useful lives (Figure 4); 2) emissions from the
SWDS pool are all EWOEC because the model lacks data on EEC from SWDS; 3) industrial
and residential fuelwood use and the burning of wood waste, as reflected by Oregon’s timber
product ratios, primary product ratios and annual EEC, have generally declined; and 4) markets
17
for mill residue (e.g., sawdust and chips) in Oregon, as reflected in the primary product ratios,
are mainly for pulp and particleboard, rather than biomass energy.
Figure 4. Cumulative stocks of HWP C from timber harvested in Oregon 1906-2017 for
products-in-use (PIU) and solid waste disposal sites (SWDS). Cumulative emissions with
energy capture (EEC) and emissions without energy capture (EWOEC) are represented
as negative values.
The net difference between the annual change in the pool of HWP C and biogenic emissions—
from burning fuelwood for energy capture, and decay and burning of discarded products—was
positive at the start of the time series up until 1914, during three instances between 1944 and
1953, from 1972 to 1975, and again from 1986 to 1990 (Figure 3).
Based on the methods from Harmon (1996) and the BF/CF conversion factors in Table 1, the
cumulative amount of carbon in bark associated with logs harvested from 1906 to 2017 was
approximately 92.21 MMT C (338.15 MMT CO
2
e). To better understand how much bark was
generated at Oregon facilities in a contemporary year we can compare the estimates for bark
from Brandt et al. (2006) with the bark estimate using Harmon’s (1996) method. According to
the former, Oregon’s facilities generated about 1.5 MMT (1.422 million bone dry units (BDU)) of
bark during 2003 while processing timber, 80 percent of which was used as fuel, with nearly all
18
the remaining 20 percent used for decorative bark or soil additives (Brandt et al. 2006). This
amount of bark constitutes a possible lower limit of approximately 2.84 MMT CO
2
e/year in
biogenic emissions, whereas the estimate for 2003 using Harmon’s (1996) method was slightly
higher at 3.24 MMT CO
2
e/year, constituting a possible upper limit. Emissions from bark
utilization for energy production are provided in the Oregon Sawmill Energy Consumption report
(Donahue et al. 2021). Finally, the HWP C model does not treat bark as a wood product and
carbon in bark is accounted for and integrated within the “cut” estimates in the Oregon Forest
Ecosystem Carbon Inventory (Christensen et al. 2019). Therefore, the above annual
approximations for bark associated with harvested wood products would be included in the “cut”
estimates of the Forest Ecosystem Carbon Inventory. To avoid double counting biogenic
emissions from the combustion and decomposition of bark, these bark quantities should not be
added to HWP C unless they are subtracted from the Forest Ecosystem Carbon Inventory.
The computational methods associated with the IPCC production accounting approach requires
carbon to be followed through the duration of a product’s life from harvest through disposal,
applying appropriate ratios and half-lives at each stage (Stockmann et al. 2012). Hence, the
HWP C model accounts for all of the carbon that enters the system as TPO each year and
follows that carbon through its duration as new and recovered PIU, then into the landfills at the
end of product usefulness, and finally as emissions with and without energy recovery. Indeed,
the cumulative amount of all carbon that entered the HWP C model as TPO through 2017
(814.4 MMT C) was accounted for in each of the disposition categories through 2018 (Figure 5;
Appendix 1, table 1H).
19
Figure 5. Carbon Mass Balance showing that the cumulative amount of carbon entering
the HWP pool, as Timber Product Output (TPO) for 1906-2017, is equal to the sum of the
cumulative amounts of carbon in each disposition category through 2018.
Monte Carlo (MC) Simulation Results
After 2,000 iterations, the MC simulation stabilized on values for the mean and the 90%
confidence intervals for the cumulative amount of carbon in the combined HWP pools of PIU
and SWDS (Figure 6). For 2018, the MC estimate for the mean of these two pools combined
was 349.7 MMT C with a lower bound of 291.0 and an upper bound of 410.0 MMT C. The width
of the confidence intervals through the time series reflected the effect of altering parameter
values for amounts of carbon entering the model and being emitted or distributed to different
pools. The precision of parameter estimates for harvest and the ratios for timber and primary
products was modelled as improving over time (Appendix 3, table 3a). Between 1906 and 1920,
the difference between the lower bound relative to the simulated mean averaged about 37%
and the upper bound about 42%. From 2010 to 2017, both the lower and upper bounds
averaged about 17% relative to the mean. The difference between the MC simulation means
and the HWP C model values for the combined pool of PIU and SWDS were less than 1% for
most of the time series, except for the years prior to 1944, when it was slightly larger.
20
Figure 6. Monte Carlo simulated mean and 90% confidence interval for the cumulative
amount of HWP C in PIU and SWDS pools combined.
Confidence intervals were generated separately for the cumulative amount of carbon in the PIU,
SWDS, EEC, and EWOEC pools (Figure 7). The small differences (< 2 MMT C) between the
MC simulated means for these pools in 2018 and the respective estimates from the HWP C
model (Table 5) indicate that the simulation operated as expected given the parameters in
Appendix 3, table 3A. The 2018 MC confidence intervals for EEC, relative to the mean, were the
widest of these four disposition categories at about 24% and 26% for the lower and upper limits,
respectively. Conversely, the width of the 2018 confidence intervals for PIU were the narrowest
at about 18% and 19% of the mean for the lower and upper limits, respectively.
21
Figure 7. Monte Carlo simulation mean and 90% confidence intervals for the cumulative
amount of carbon in products in use, solid waste disposal sites (SWDS), emissions with
energy capture and emissions without energy capture for Oregon’s1906-2017 timber
harvests.
Table 5. Estimates of HWP C (in MMT) for PIU, SWDS, EEC, and EWOEC for the 2018
HWP C model output compared with means and 90% confidence intervals from the MC
simulation.
Disposition Categories
HWP C Model
Output
MC Simulation Mean and
90% Confidence Interval
Products in Use (PIU) 201.1 163.3 < 199.3 < 236.8
Solid Waste Disposal Sites(SWDS) 149.6 122.5 < 150.4 < 179.3
Emissions With Energy Capture (EEC)
155.6
118.7 < 156.0 < 197.3
Emissions Without Energy Capture (EWOEC)
308.2
249.9 < 309.4 < 373.2
22
Stocks and Flux of HWP Carbon by Ownership, 1962-2017
Consistent timber harvest data by ownership class are available for Oregon since1962, which
provides the opportunity to understand how much each ownership has contributed to the pool of
HWP C (Figure 8). From 1962 to 2017, the cumulative TPO for all ownership classes was 514.5
MMT C, from approximately 346.9 bbf Scribner of harvested timber. The volumes and relative
proportions by ownership class vary from year to year and are documented in historic and
recent reports (Andrews and Kutara 2005, Simmons et al. 2019). For example, in 2016,
industrial timberlands accounted for 63.3% of harvest, followed by NIP timber (13.1%), USFS
(9.0%), State (7.3%), BLM (4.7%), Tribal (1.5%) and other public (1.1%). However, the average
for each ownership across the time series shows that timber from industrial ownerships
accounted for 47.2%, USFS 29.5%, BLM 10.2%, NIP 7.7%, State and other public 4.1%, and
Tribal 1.3%.
Figure 8. Cumulative HWP C stock as products-in-use (PIU) and at solid waste disposal
sites (SWDS) by ownership from timber harvested in Oregon, 1962-2017.
Note: The ownership group “Private” includes industrial and non-industrial private forest owners, and “BLM, State, Tribal” includes
BLM, State and other public, and Tribal forest owners (see Appendix 1, Table 1J).
23
Net additions to the HWP stock have generally diminished with declines in total timber harvest
in Oregon from the peaks of the 1970s and ‘80s. This is especially the case for the USFS
ownership that significantly curtailed timber harvest in the 90’s and the consequent decline in
the accumulation in HWP C. Indeed, the annual net increase in HWP C from USFS forests up to
1990 averaged about 2.8 MMT C/year whereas the annual average dropped to -0.2 MMT
C/year after 1990 (Appendix 1, table 1J). Nevertheless, the all-owner total HWP C pool
continued to increase, albeit at a lower rate, with timber harvest from other ownerships. Since
1962, PIU account for about 57% (164.4 MMT C) and SWDS account for about 43% (123.6
MMT C) of the total HWP C pool. About 288.0 MMT C (56%) of the TPO since 1962 remains in
the HWP pools, while 44% (226.5 MMT C or 830.6 MMT CO
2
e) was emitted through burning or
decay.
Net change in HWP C originating from BLM forests increased at an annual average of 0.9 MMT
C/year from 1963 to 1990 but the annual average was slightly negative (-0.04 MMT C)/year)
after 1990 (Figure 9). Net change in HWP C from State and other public forests averaged 0.2
MMT C/year through 1990 and increased to 0.3 MMT C/year after 1990. Net change of HWP C
from Tribal forests increased steadily at an annual average of about 0.1 MMT C/year from 1963
to 2018 (Appendix 1, table 1J). However, the annual change was slightly negative for 2018.
Figure 9. Cumulative HWP C stock as products-in-use (PIU) and at solid waste disposal
sites (SWDS) for the BLM, State and other, and Tribal owners from timber harvested in
Oregon, 1962-2017.
24
Evaluating estimates of HWP C relative to Oregon’s forest ecosystem carbon
Measurements from field plots monitored by the FIA Program indicate that for the 2016
reporting period Oregon’s forests contained approximately 3,239.7 ± 32.8 MMT C in living and
dead trees, understory vegetation, downed wood, roots, forest floor and forest soils across all
ownerships (Christensen et al. 2019; table 4.13a). About 1,118.3 MMT C were in the
aboveground parts of live and dead trees. For the 2017 HWP inventory year (which includes
carbon from timber harvests from 1906 through 2016), the model estimated the amount of
carbon accumulated in the HWP pool was 348.1 MMT C (Table 6), which is equivalent to
approximately 10.7% of the total C in Oregon’s forests and about 31.1% of the aboveground
portion of live and dead trees (1,118.3 MMT C). Finally, the total combined amount of carbon in
Oregon’s forests and HWP in 2016 was approximately 3,587.8 MMT C.
T
able 6. Annual accumulation of HWP C stock, since 2001, in solid waste disposal sites
(SWDS) and products-in-use (PIU) for Oregon timber harvested from 1906-2017.
Inventory
Year
PIU
SWDS
Total HWP C
MMT C
2001
189.5
122.3
311.8
2002
190.0
123.6
313.6
2003
191.2
125.1
316.3
2004
192.6
126.5
319.1
2005
193.9
127.9
321.8
2006
195.1
129.4
324.6
2007
196.2
131.0
327.2
2008
196.4
132.5
328.9
2009
196.1
134.1
330.2
2010
195.1
135.6
330.7
2011
195.0
137.1
332.1
2012
195.7
138.7
334.3
2013
196.4
140.3
336.7
2014
197.9
142.1
339.9
2015
199.1
143.9
343.0
2016
199.7
145.7
345.5
2017
200.5
147.6
348.1
2018
201.1
149.6
350.7
The net v
egetation flux of carbon in Oregon’s forested ecosystems was approximately 30.5
MMT CO
2
e/year (8.3 MMT C/year), after accounting for the transfer of 25.3 MMT CO
2
e/year
(6.9 MMT C/year) to dead wood pools from tree mortality, and 34.8 MMT CO
2
e/year (9.5 MMT
C/year) from trees being cut (Christensen et al. 2019, tables 4.6 and 4.7a). Comparing the FIA
estimates for the amount of carbon cut from forests with estimates for the average TPO is
challenging due to live trees being cut between field plot measurements, with growth equations
being used to estimate tree diameter and height at the midpoint of the measurement interval in
25
order to calculate C at the time of cutting. Moreover, harvesting of timber represented in the FIA
cut estimates could have occurred in any year from 2001 to 2016. The estimate for average
annual TPO output for 2001 to 2016 was approximately 7.0 MMT C/year (25.5. MMT CO
2
e),
approximately 73% of the average cut volume reported in the FIA Oregon Forest Ecosystem
Carbon Report (Christensen et al. 2019).
Considering the model estimates for average HWP emissions with energy capture from burning
fuelwood (EEC = 1.7 MMT CO
2
e/year) and emissions without energy capture (EWOEC = 15.5
MMT CO
2
e/year) for the 2001-2016 time period, the balance of carbon in the pool of HWP (8.4
MMT CO
2
/year) was negative at approximately -8.8 MMT CO
2
e/year. Therefore, the combined
average net change for the pool of carbon in Oregon’s forests and HWP for this time period
(including PIU and SWDS) was approximately 21.7 MMT CO
2
e/year (5.9 MMT C/year).
Discussion
Oregon has a long history of timber harvest and manufacturing of wood products that continue
to retain carbon throughout their use and disposal. Just like the various pools of carbon in forest
ecosystems, the pools of HWP C have dynamic rates of input and output. The analysis
described in this report used the IPCC production approach to generate estimates for the stocks
and flux of carbon in HWP for timber harvested within the State of Oregon from 1906 to 2017. It
also provides estimates for the cumulative emissions of CO
2
e from burning fuelwood for energy
capture and from the decay and burning of discarded wood products. Reliable estimates for
timber harvest among the different ownerships in Oregon are not available prior to 1962,
therefore, this report provides estimates of carbon in HWP for the major forest ownerships that
contributed to HWP C pools from 1962-2017 harvests. The results of the HWP study were
combined with those from the Oregon Forest Ecosystem Carbon Report (Christiansen et al.
2019) to calculate the total stocks and flux of carbon in both forests and HWP. This report
provides the HWP dimension of an overall forest carbon accounting framework for Oregon that
can be used for subsequent HWP C analyses and applied to other regions (e.g., California,
Washington; Christensen et al. 2018) with timber harvest, timber product, primary product and
end-use product data.
Carbon emissions from fossil fuels used in harvest, transportation and manufacturing of HWP
were not estimated and therefore were not deducted from the HWP pools. Although the HWP C
model generates estimates of emissions from burning HWP with energy capture, this study did
not make estimates for substitution of fossil fuel carbon, potentially reducing fossil fuel
emissions in some scenarios (Jones et al. 2010). Furthermore, the HWP C model does not
address carbon or energy associated with substituting HWP for construction materials such as
metal or concrete. While emissions trade-offs from product substitution are outside the scope
and purpose of this report, there are well-developed methods of life cycle analysis (LCA) that
account for all carbon emissions associated with manufactured products and that facilitate the
comparison between wood products and alternative products (Rebitzer et al. 2004, Lippke et al.
2011, CORRIM 2018). The production approach used in this analysis provides information that
could be used to inform an LCA of wood products generated in Oregon.
26
The dynamics in TPO for Oregon generally reflected changes in the economy and changes in
NFS land management through the 1906 to 2017 period. TPO increased from less than 2 MMT
C/year in 1906 to upwards of 13 MMT C/year in 1972, followed by a steep decline upon
implementation of the Northwest Forest Plan. Part of the annual additions to the SWDS C pool
are proportional to additions to the pool of PIU because the model transfers 8% of the carbon in
primary products directly to the discarded pool to account for waste generated during
manufacturing and construction of end-products each year. Wood products with a short half-life
such as crates, pallets, concrete forms, etc., enter the waste stream relatively quickly after
production, which also maintains the waste stream along with the regular demolition of products
with longer durations of use. Consequently, the waste stream of wood products was relatively
smooth during the time series, although it reflects declines following major decreases in harvest.
Many discarded products were disposed of in SWDS rather than burning, which has resulted in
significant quantities of HWP C transferred to these long-term storage pools rather than being
rapidly released into the atmosphere (Skog 2008). The net change in the HWP pool follow the
pattern of annual additions to PIU, slightly lagged, with periods of negative changes following
downturns in the economy and the large decline in federal harvest starting in 1987. The current
negative balances from 2001-2016, reflect the federal harvest decline, followed by the Great
Recession in 2009-2011.
Approximately 25% of the 814 MMT C in the cumulative TPO output since 1906 is currently in
PIU, 18% in SWDS, 19% has been emitted back to the atmosphere from burning fuelwood for
energy capture, and about 38% has been emitted through decomposition or burning without
energy capture. The contribution of carbon to the atmosphere from discarded HWP could be
significantly reduced with further advances in wood waste management. Indeed, end-of-life
management of wood products is perhaps the single most significant variable for the full life
cycle carbon profile of wood products (Sathre and O’Connor 2010).
Estimates of HWP C pools by ownership would be different than reported here if consistent,
reliable harvest information by ownership were available prior to 1962. However, HWP C
storage estimates by ownership for harvests since 1962 show that 59% of the HWP C pool in
2017 originated from private forestlands, 35% from federal forests, and 6% from State, Tribal
and local governments.
Comparing the stocks and flux of carbon in Oregon’s forests and HWP with estimates at the
national level is problematic because the flux of carbon across U.S. forests is based on stock
changes and reported annually, whereas C flux values reported in the Oregon Forest
Ecosystem Carbon Report were based on changes in measurements of individual trees and
averaged over the six years that plot re-measurements were available (EPA 2018; Christiansen
et al., 2019). Nonetheless, Oregon’s forests have been accumulating carbon with an annual
average total forest carbon flux of 30.5 MMT CO
2
e/year (8.3 MMT C/year). This represents
about 5.5% of the average national total forest carbon flux for 2016 of 565.5 MMT CO
2
e/year
(154.2 MMT C/year).
The EPA (2018) estimated that the total stock of HWP C for 2016 (2,591 MMT C) was about
4.7% of the total forest carbon stock for the continental U.S. plus coastal Alaska (55,592 MMT
C). The total stock of HWP C for 2016 in Oregon was 345.5 MMT C (Table 6), which is
equivalent to approximately 10.7% of the estimate for the total C in Oregon’s forests and about
27
13.3% of the national stock of HWP C. For 2017, the total annual input to HWP C from Oregon
was 9.7 MMT CO
2
e/year (Appendix 1, table 1G), which represents about 10.1% of the 2017
national HWP flux of 95.7 MMT CO
2
e/year. These results demonstrate that Oregon contributes
a significant amount to the national pool of HWP C.
Differences between the FIA estimates of cut from Oregon forests and TPO can be attributed to
a combination of differences between forests and HWP accounting methodologies, and
“additional” items in the FIA removals estimate that are not part of TPO. For example, logging
residue generated during timber harvesting, such as needles, branches, tops, and other
removals of non-merchantable material during operations such as pre-commercial thinning, are
included in FIA’s estimate of cut. Of the harvested tree, the stem represents about 67.54%,
while residues (tops, branches, and foliage) represent about 32.46% of total biomass (Ganguly
et al. 2020). In contrast, the HWP C estimates are calculated exclusively from log (wood fiber)
volumes delivered to timber-processing facilities. (i.e., mills). It is also important to note that the
HWP and FIA estimates were developed from very different data sets and methodologies that
prevent direct comparisons. For example, FIA’s estimates of average annual cut from forests
are based on re-measurements of a sample of field inventory plots over a 10-year cycle, while
the HWP estimates are based on annual timber harvest records. Completion of the full re-
measurement cycle for all FIA field plots by 2021 will increase the confidence in the estimates of
the stocks and flux of forest carbon, which should also transfer to comparing and combining
estimates of HWP carbon. There has not been a comprehensive analysis comparing FIA
estimates of removals to timber harvest records in Oregon since FIA began using the current
plot measurement system in 2001.
Conclusions
The pool of carbon in harvested wood products (HWP) is significant in size and should be
considered in decision making associated with carbon monitoring and climate change
adaptation and mitigation (Stockmann et al. 2012). This report provides an empirical accounting
of carbon in timber harvested in Oregon since 1906, with annual estimates of HWP C stocks
and changes in PIU and SWDS. It also includes estimates of C emissions from HWP with and
without energy capture. The information presented here is fundamental to the wood products
dimension of Oregon’s forest carbon accounting framework. It is consistent with the IPCC Tier-3
production accounting approach that was used for the HWP C section in the California Forest
Carbon Inventory (Loeffler et al. 2019, Christensen et al. 2018) and multiple USFS regions and
forests (Stockmann et al. 2012, Anderson et al. 2013, Butler et al. 2014, Loeffler et al. 2014).
The data, modeling framework and results are fundamental to evaluating how the dynamics of
Oregon’s past and current timber harvest levels have influenced the annual growth of the pool
of HWP C and total forest carbon stock levels. The modeling framework does not include
estimates for the emissions associated with timber harvesting, log transportation, HWP
manufacturing, substitution effects of wood products, bioenergy, or leakage.
While this analysis was generated at the state level with details for ownership class back to
1962, it provides an analytical framework that can be used for smaller regions in the state as
28
well.
Published literature with a relatively long time-series of county-level harvest volume data is
available for Oregon (Andrews and Kutara 2005, ODF 2019), including quantitative information
needed for allocating harvest estimates to timber and primary product distributions to reflect the
changes in wood product manufacturing. Clearly, the pool of carbon in HWP generated from
Oregon’s past timber harvests has grown to a large, active pool and, as such, is a significant
consideration when monitoring or developing (forest) carbon management policy. Total forest
carbon integrates the carbon dynamics of forest ecosystems and wood products manufactured
from timber, as well as changing forest management policies and wood product markets.
Periodically updated HWP end-use estimates would allow better HWP C storage and flux
estimates; however, no regional or state-level variations of product end use have been identified
in the literature (Loeffler et al. 2019).
The analysis presented above provides the historical perspective of carbon dynamics in the pool
of HWP C harvested from Oregon’s forests. Given the limitations already stated, the model was
not applied to potential future timber harvest levels. There are several different forest modeling
systems that can be used to predict the outcomes of alternative forest management regimes for
carbon mitigation. The ODF is currently working with partners from the USFS, California,
Washington and British Columbia in evaluating these different forest-modeling systems to
compare their capabilities for generating the carbon outcomes of alternative scenarios for
managing forests and HWPs across this large geographic region. The Carbon Budget Model
(CBM) of the Canadian Forest Sector is currently being tested for forests in Oregon and
California in collaboration with American Forests, ODF, and CalFire, with results expected in
2022. The CBM model will be used to quantify and understand the carbon outcomes of
alternative strategies for managing forests and wood products that integrate new technologies
and policies for recycling and waste disposal.
29
Literature Cited
Anderson, N., Young, J., Stockmann, K., Skog, K., Healey, S., Loeffler, D., Jones, J.G.,
Morrison, J. 2013. Regional and forest-level estimates of carbon stored in harvested
wood products from the United States Forest Service Northern Region, 1906-2010. Gen.
Tech. Rep. RMRS-GTR-311. Fort Collins, CO: USDA, Forest Service, Rocky Mountain
Research Station. 114 p.
Andrews, H.J., Cowlin, R.W. 1940. Forest Resources of the Douglas-fir Region. Miscellaneous
Publication No. 389. Portland, OR: USDA, Forest Service, Pacific Northwest Forest and
Range Experiment Station. 179 p.
Andrews, A., Kutara, K. 2005. Oregon’s timber harvests: 1849–2004. Salem, OR: Oregon
Department of Forestry. 154 p.
Brandt, J.P., Morgan, T.A., Dillon, T., Lettman, G.J., Keegan, C.E., Azuma, D.L. 2006. Oregon’s
forest products industry and timber harvest, 2003. Gen. Tech. Rep. PNW-GTR-681.
Portland, OR: USDA, Forest Service, Pacific Northwest Research Station. 53 p.
Butler, E., Stockmann, K., Anderson, N., Skog, K., Healy, S., Loeffler, D., Jones, J. G.,
Morrison, J., Young, J. 2014. Estimates of carbon stored in harvested wood products
from United States Forest Service Pacific Northwest region, 1909-2012. Unpublished
report. Missoula, MT: USDA, Forest Service, Rocky Mountain Research Station,
Forestry Sciences Laboratory. 28 p.
Christensen, G.A., Gray, A.N., Kuegler, O., Yost, A.C. 2019. Oregon Forest Ecosystem Carbon
Inventory: 2001-2016. Report submitted to Oregon Department of Forestry. Portland,
OR. 347 pp.
Christensen, G.A.; Gray, A.N.; Kuegler, O.; Tase, N.A.; Rosenberg. M; Loeffler, D.;
Anderson, N.; Stockmann, K.; Morgan, T.A. 2018. AB 1504 California Forest Ecosystem
and Harvested Wood Produc
t Carbon Inventory: 2017 Reporting Period. Final Report.
U.S. Forest Service agreement no. 18-CO-11052021-214, 17-CO-11261979-086,
California Department of Forestry and Fire Protection agreement no. 8CA04056 and
8CA03714 and the University of Montana. Sacramento, CA: California Department of
Forestry and Fire Protection and California Board of Forestry and Fire Protection. 539 p.
CORRIM (Consortium for Research on Renewable Industrial Materials) 2018.
[https://corrim.org/].
Cowlin, R.W., Briegleb, P.A., Moravets, F.L. 1942. Forest Resources of the Ponderosa Pine
Region of Washington and Oregon. Miscellaneous Publication No. 490. Portland, OR:
USDA, Forest Service, Pacific Northwest Forest and Range Experiment Station. 113 p.
Donahue, T., Morgan, T.A., Dillon, T. 2021. Oregon Sawmill Energy Consumption and
Associated Emissions, 2017. Report to the Oregon Department of Forestry, Bureau of
Business and Economic Research, Missoula, MT. 26 p.
30
EPA (US Environmental Protection Agency). 2020. Inventory of U.S. greenhouse gas emissions
and sinks: EPA 430-R-20-002. US EPA, Office of Atmospheric Programs, Washington,
DC. https://www.epa.gov/sites/production/files/2020-04/documents/us-ghg-inventory-
2020-main-text.pdf
EPA (US Environmental Protection Agency). 2018. Inventory of U.S. greenhouse gas emissions
and sinks: 1990–2016. EPA 430-R-18-003. US EPA, Office of Atmospheric Programs,
Washington, DC. https://www.epa.gov/ghgemissions/inventory-us-greenhouse-gas-
emissions-and-sinks-1990-2016.
Gale, C.B., Keegan, C.E., III, Berg, E.C., Daniels, J., Christensen, G.A., Sorenson, C.B.,
Morgan, T.A., Polzin, P. 2012. Oregon’s forest products industry and timber harvest,
2008: industry trends and impacts of the Great Recession through 2010. Gen. Tech.
Rep. PNW-GTR-868. Portland, OR: USDA, Forest Service, Pacific Northwest Research
Station. 55 p.
Galik, C.S., Jackson, R.B. 2009. “Risks to forest carbon offset projects in a changing climate.”
Forest Ecology and Management 2009, 257: 2209-2216.
Galik, C.S., Mobley, M.L., Richter, D.B. 2009. “A virtual “field test” of forest management carbon
offset protocols: the influence of accounting.” Mitigation and Adaptation Strategies for
Global Change 14: 677-690.
Ganguly, I, Pierobon, F., Hall, E.S. 2020 Global warming mitigating role of wood products from
Washington state’s private forests. Forests 11, 194; doi:10.3390/f11020194
Harmon, M.E., Harmon, J.M., Ferrell, W.K., Brooks, D. 1996. Modeling carbon stores in Oregon
and Washington forest products: 1900-1992. Climatic Change 33: 521-550.
Holl, K.D., Brancalion, P.H.S. 2020. Tree planting is not a simple solution. Science Vol. 368,
Issue 6491, pp. 580-581.
Howard, J.O., Hiserote, B.A. 1978. Oregon’s forest products industry: 1976. Resour. Bull. PNW-
RB-79. Portland, OR: USDA, Forest Service, Pacific Northwest Forest and Range
Experiment Station. 102 p.
Howard, J.O. 1984. Oregon’s forest products industry. 1982; Resour. Bull. PNW-RB-118.
Portland, OR: USDA, Forest Service, Pacific Northwest Forest and Range Experiment
Station. 79 p.
Howard, J.O., Ward, F.R. 1988. Oregon’s forest products industry: 1985. Resour. Bull. PNW-
RB-149. Portland, OR: USDA, Forest Service, Pacific Northwest Research Station. 90 p.
Howard, J.O., Ward, F.R. 1991. Oregon’s forest products industry: 1988. Resour. Bull. PNW-
RB-183. Portland, OR: USDA, Forest Service, Pacific Northwest Research Station. 91 p.
Ingerson, A. 2011. “Carbon storage potential of harvested wood: summary and policy
implications.” Mitigation and Adaptation Strategies for Global Change 16:307-323.
31
Jones, J.G., Loeffler, D, Calkin, D., Chung, W. 2010. “Forest residues for thermal energy
compared with disposal by onsite burning: Emissions and energy return.” Biomass and
Bioenergy 34(5):737-746.
Keegan, C.E., III, Morgan, T.A., Blatner, K.A., Daniels, J.M. 2010. “Trends in lumber processing
in the Western United States. Part I: Board foot Scribner volume per cubic foot of
timber.” Forest Products Journal. 60(2): 133–139.
Lippke, B., Oneil, E., Harrison, R., Skog, K., Gustavsson, L., Sathre, R. 2011. “Life cycle
impacts of forest management and wood utilization on carbon mitigation: knowns and
unknowns.” Carbon Management 2(3): 303-333.
Loeffler, D., Anderson, N., Stockmann, K., Morgan, T.A., Tase, N.A. 2019. AB 1504 California
Forest Carbon Stored in Harvested Wood Products: 1952 – 2017. Final Report.
Prepared for: USDA Forest Service and California Department of Forestry and Fire
Protection. University of Montana Bureau of Business and Economic Research, Forest
Industry Research Program, 54 p.
Loeffler, D., Anderson, N., Stockmann, K., Skog, K., Healey, S., Jones, G.J., Morrison, J.,
Young, J. 2014. Estimates of carbon stored in harvested wood products from United
States Forest Service Alaska Region, 1910-2012
[https://www.fs.fed.us/rm/pubs_other/rmrs_2014_loeffler_d002.pdf].
Manock, E.R., Choate, G.A., Gedney, D.R. 1970. Oregon timber industries: wood consumption
and mill characteristics, 1968. Salem, OR: Oregon Department of Forestry. 122 p.
McKeever D.B. 2004. Inventories of woody residues and solid wood waste in the United States,
2002. The Ninth International Conference on Inorganic-Bonded Composite Materials
Conference, October 10-13, 2004. Vancouver, British Columbia, Canada. 12 p.
McKeever, D.B. 2009. Estimated annual timber products consumption in major end uses in the
United States, 1950-2006. GTR-FPL-181, USDA, Forest Service, Forest Products Lab,
49 p.
McKeever, D.B., Howard, J.L. 2011. Solid wood timber products consumption in major end uses
in the United States, 1950-2009: a technical document supporting the Forest Service
2010 RPA assessment. GTR-FPL-GTR-199, USDA, Forest Service, Forest Products
Lab, 39 p.
McKinley, D.C., Ryan, M.G., Birdsey, R.A., Giardina, C.P., Harmon, M.E., Heath, L.S.,
Houghton, R.A., Jackson, R.B., Morrison, J.F., Murray, B.C., Pataki, D.E., Skog, K.E.
2011. “A synthesis of current knowledge on forests and carbon storage in the United
States.” Ecological Applications 21:1902-1924.
Metcalf, M.E. 1965. Timber Resource Statistics for the Pacific Northwest as of January 1, 1963.
Resource Bulletin PNW-RB-9. Portland, Oregon: USDA Forest Service, Pacific
Northwest Forest and Range Experiment Station, 40 p.
32
ODF (Oregon Department of Forestry). 2019. Natural Resources: Timber Harvest Data 1942 –
2018. Data.Oregon.Gov. Available online at:
https://data.oregon.gov/Natural-
Resources/Timber-Harvest-Data-1962-2018/kz4p-3qrv .
Oregon Global Warming Commission (OGWC). 2018. Forest Carbon Accounting Project Report
2018. Salem, OR. 52 p. Available online at:
https://www.oregon.gov/ODF/ForestBenefits/Documents/Forest%20Carbon%20Study/O
GWC-Forest-Carbon-Project-Report-2018.pdf . Last accessed 11/30/2019.
ORS 526.783. 2001. Development of forestry carbon offset accounting system. Available online
at: https://www.oregonlegislature.gov/bills_laws/ors/ors526.html
. Last accessed
11/29/2019.
ORS 468A.250(1)(i). 2007. Mandate of Oregon Global Warming Commission. ORS
468A.250(1)(i). Available online at:
https://www.oregonlegislature.gov/bills_laws/ors/ors468A.html
. Last accessed
11/29/2019.
R Core Team (2020). R: A language and environment for statistical computing. R Foundation for
Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.
RCRA (Resource Conservation and Recovery Act) of 1976 § 42 U.S.C. § 6901 (2011).
Rebitzer, G., Ekvall, T., Frischknecht, R., Hunkeler, D., Norris, G., Rydberg, T., Schmidt, W.P.,
Suh, S., Weidema, B.P., Pennington, D.W. 2004. “Life cycle assessment part 1:
framework, goal and scope definition, inventory analysis, and applications.”
Environmental International 30:701-20.
Ryan, M.G., Harmon, M.E., Birdsey, R.A., Giardina, C.P., Heath, L.S., Houghton, R.A., Jackson,
R.B., McKinley, D.C., Morrison, J.F., Murray, B.C., Pataki, D.E., Skog, K.E. 2010. “
A
s
ynthesis of the science on forests and carbon for U.S. forests.” Issues in Ecology 13: 1-
6
.
Sathre, R., O’Connor J. 2010. A synthesis of research on wood products and greenhouse gas
impacts, 2
nd
Edition. Vancouver, B.C. FPInnovations. 117 p. (Technical Report TR-19R)
Schuldt, J.P., Howard, J.O. 1974. Oregon Forest Industries: Wood Consumption and Mill
Characteristics, 1972. Special Report 427. Corvallis, Oregon: Oregon State University
Extension Service, 113 pp.
Simmons, E.A., Scudder, M.G., Morgan, T.A., Berg, E.C., Christensen, G. A. 2016. Oregon’s
forest products industry and timber harvest 2013 with trends through 2014. Gen. Tech.
Rep. PNW-GTR-942. Portland, OR: USDA, Forest Service, Pacific Northwest Research
Station. 58 p.
Simmons, E.A., Rymniak, L., Morgan, T.A., Marcille, K., Lettman, G., Christensen, G. 2019.
Oregon forest products Industry and timber harvest, 2017: Highlights and summary data
tables. Bureau of Business and Economic Research, Forest Industry Research Program.
http://www.bber.umt.edu/FIR/S_OR.asp
33
Skog, K.E. 2008. Sequestration of carbon in harvested wood products for the United States.
Forest Products Journal 58(6): 56-72.
Smith, J., Heath, L., Skog, K., Birdsey, R. 2006. Methods for calculating forest ecosystem and
harvested carbon with standard estimates for forest types of the United States. Gen.
Tech. Rep. NE-343. Newtown Square, PA: USDA, Forest Service, Northeastern
Research Station. 216 p.
Smith, J.E., Heath, L.S., Woodbury, P.B. 2004. “How to estimate forest carbon for large areas
from inventory data.” Journal of Forestry 102: 25-31.
Spies, T.A., Long, J. W., Charnley, S., Hessburg, P.F., Marcot, B.G., Reeves, G.H., Lesmeister,
D.B., Reilly, M.J., Cerveny, L.K., Stine, P.A., Raphael, M.G. 2019. Twenty‐five years of
the Northwest Forest Plan: what have we learned? Frontiers in Ecology and the
Environment. 35: 1319-. https://doi.org/10.1002/fee.2101
Stockmann, K.D., Anderson, N.M., Skog, K.E., Healey, S.P., Loeffler, D.R., Jones, G., Morrison,
J.F. 2012. Estimates of carbon stored in harvested wood products from the United
States Forest Service Northern region, 1906-2010. Carbon Balance and Management
7(1): 1-16.
USFS (USDA Forest Service). 2019. NFS Carbon Assessments. U.S. Department of
Agriculture, Forest Service. [https://www.fs.fed.us/climatechange/advisor/products.html
]
USURS (Utah State University Remote Sensing)/GIS Laboratory. 2013.
https://www.gis.usu.edu/projects/harvested-wood-products/
Ward, F.R. 1995. Oregon’s forest products industry: 1992. Resour. Bull. PNW-RB-207.
Portland, OR: USDA, Forest Service, Pacific Northwest Research Station. 89 p.
Ward, F.R. 1997. Oregon’s forest products industry: 1994. Resour. Bull. PNW-RB-216.
Portland, OR: USDA, Forest Service, Pacific Northwest Research Station. 70 p.
Ward, F.R., Lettman, G.J., Hiserote, B.A. 2000. Oregon’s forest products industry: 1998.
Portland, OR: USDA, Forest Service, Pacific Northwest Research Station. 82 p.
Zheng, D., Heath, L.S., Ducey, M.J., Butler, B. 2010. Relationships between major ownerships,
forest aboveground biomass distributions, and landscape dynamics in the New England
Region of USA. Environmental Management 45:377-386.
34
Appendix 1
Table 1A. Oregon timber harvest by ownership in billion board feet (bbf) Scribner and
cumulative TPO in MMT C, 1962-2017.
Harvest year
Private USFS BLM
State and
other
Tribal
bbf
MMT C
bbf
MMT C
bbf
MMT C
bbf
MMT C
bbf
MMT C
1962
4.1 5.2 3.1 3.9 1.1 1.4 0.2 0.2 0.1 0.1
1963
3.8 9.9 3.2 8.0 1.4 3.1 0.2 0.5 0.1 0.2
1964
4.0 15.0 3.4 12.3 1.6 5.2 0.2 0.8 0.1 0.3
1965
4.0 20.1 3.8 17.1 1.2 6.8 0.3 1.1 0.1 0.4
1966
4.1 25.3 3.3 21.2 1.2 8.3 0.2 1.4 0.1 0.4
1967
3.8 30.2 3.2 25.2 1.1 9.7 0.1 1.6 0.1 0.6
1968
4.4 35.7 3.6 29.8 1.5 11.6 0.2 1.8 0.1 0.7
1969
4.2 40.9 3.5 34.2 1.2 13.1 0.2 2.1 0.1 0.8
1970
3.9 45.8 2.8 37.8 1.0 14.4 0.2 2.3 0.1 0.9
1971
4.2 51.6 3.2 42.2 1.3 16.2 0.2 2.5 0.1 1.0
1972
3.9 56.9 3.9 47.5 1.5 18.3 0.3 2.9 0.1 1.2
1973
3.6 61.9 3.8 52.8 1.5 20.4 0.3 3.3 0.1 1.3
1974
3.8 67.1 3.2 57.1 1.0 21.8 0.2 3.7 0.1 1.4
1975
3.8 72.2 2.7 60.7 0.6 22.6 0.2 3.9 0.1 1.6
1976
3.6 77.1 3.2 65.1 1.1 24.1 0.2 4.2 0.1 1.7
1977
3.6 82.0 2.9 69.0 1.0 25.5 0.2 4.5 0.1 1.9
1978
3.5 86.9 3.2 73.4 0.8 26.6 0.3 4.9 0.1 2.1
1979
3.2 91.2 3.2 77.8 1.0 27.9 0.3 5.2 0.1 2.2
1980
3.1 95.7 2.4 81.2 0.8 29.1 0.2 5.5 0.1 2.4
1981
2.7 99.6 2.0 84.0 0.7 30.0 0.2 5.9 0.1 2.5
1982
3.4 104.5 1.7 86.4 0.3 30.5 0.2 6.1 0.1 2.7
1983
3.4 109.3 2.9 90.6 0.8 31.6 0.3 6.6 0.1 2.8
1984
3.1 113.8 3.2 95.1 0.9 32.9 0.3 7.0 0.1 3.0
1985
3.3 118.5 3.5 100.1 0.9 34.2 0.3 7.4 0.1 3.2
1986
3.5 123.5 3.9 105.6 1.0 35.7 0.3 7.8 0.1 3.3
1987
3.3 128.2 3.5 110.6 1.1 37.3 0.3 8.1 0.1 3.5
1988
3.3 132.9 3.5 115.6 1.4 39.4 0.3 8.6 0.1 3.7
35
Table 1A. Oregon timber harvest by ownership in billion board feet (bbf) Scribner and
cumulative TPO in MMT C, 1962-2017.
Harvest year
Private USFS BLM
State and
other
Tribal
bbf
MMT C
bbf
MMT C
bbf
MMT C
bbf
MMT C
bbf
MMT C
1989
3.7 138.2 3.3 120.3 1.0 40.8 0.2 8.9 0.1 3.8
1990
3.2 143.5 2.0 123.6 0.7 42.0 0.2 9.2 0.1 4.0
1991
3.3 148.9 2.1 127.0 0.5 42.8 0.1 9.4 0.1 4.1
1992
3.6 154.7 1.4 129.2 0.5 43.6 0.2 9.7 0.1 4.3
1993
3.6 160.6 1.1 131.0 0.4 44.2 0.1 9.9 0.1 4.4
1994
3.2 165.9 0.6 132.0 0.1 44.3 0.2 10.2 0.1 4.6
1995
3.4 171.5 0.5 132.9 0.1 44.5 0.1 10.4 0.1 4.7
1996
3.0 176.4 0.4 133.5 0.3 45.0 0.1 10.6 0.1 4.8
1997
3.1 181.6 0.5 134.4 0.1 45.2 0.2 11.0 0.1 4.9
1998
2.8 186.2 0.3 134.9 0.1 45.4 0.2 11.2 0.1 5.1
1999
3.0 191.1 0.2 135.3 0.1 45.7 0.3 11.7 0.1 5.2
2000
3.2 196.9 0.2 135.7 0.1 45.8 0.3 12.3 0.1 5.3
2001
2.9 202.2 0.1 136.0 0.0 45.9 0.3 12.8 0.1 5.4
2002
3.3 208.3 0.2 136.3 0.1 46.0 0.3 13.4 0.1 5.5
2003
3.3 214.4 0.2 136.7 0.1 46.1 0.3 14.0 0.1 5.6
2004
3.6 220.8 0.3 137.3 0.1 46.3 0.3 14.6 0.1 5.8
2005
3.5 227.0 0.3 137.9 0.1 46.5 0.4 15.3 0.1 5.9
2006
3.6 233.4 0.2 138.2 0.1 46.8 0.3 15.9 0.1 6.0
2007
3.1 238.8 0.2 138.6 0.1 47.0 0.3 16.4 0.1 6.1
2008
2.7 243.7 0.2 139.0 0.1 47.2 0.3 17.0 0.1 6.2
2009
2.1 247.5 0.2 139.4 0.1 47.5 0.3 17.5 0.1 6.3
2010
2.4 252.0 0.3 139.8 0.1 47.7 0.3 18.1 0.1 6.5
2011
2.7 257.1 0.4 140.5 0.2 48.0 0.3 18.7 0.1 6.6
2012
2.9 262.4 0.4 141.2 0.1 48.3 0.3 19.2 0.1 6.7
2013
3.3 268.4 0.4 141.9 0.2 48.6 0.3 19.8 0.1 6.8
2014
3.2 274.3 0.4 142.6 0.2 49.0 0.3 20.3 0.1 6.9
2015
2.8 279.6 0.4 143.3 0.2 49.4 0.3 21.0 0.1 7.0
2016
3.0 285.0 0.4 144.0 0.2 49.7 0.3 21.6 0.1 7.1
2017
3.0 290.6 0.3 144.6 0.2 50.0 0.3 22.2 0.0 7.1
36
Table 1B. Oregon timber harvest in billion board feet (bbf) with annual and cumulative
TPO in MMT C, 1906-2017.
Harvest Year Harvest (bbf)
Timber Product
Output (MMT C)
Cumulative TPO
(MMT C)
1906
1.5
1.3
1.3
1907
1.6
1.4
2.7
1908
1.7
1.4
4.1
1909
1.8
1.6
5.7
1910
1.9
1.7
7.3
1911
1.9
1.8
9.1
1912
1.9
1.8
10.9
1913
1.9
1.8
12.6
1914
1.8
1.6
14.3
1915
1.8
1.7
16.0
1916
2.1
1.9
17.9
1917
2.5
2.3
20.1
1918
2.6
2.4
22.5
1919
2.9
2.6
25.1
1920
2.6
2.4
27.5
1921
2.8
2.7
30.2
1922
3.0
2.9
33.1
1923
3.6
3.4
36.5
1924
3.8
3.7
40.2
1925
4.5
4.3
44.6
1926
4.5
4.4
48.9
1927
4.4
4.3
53.2
1928
4.7
4.5
57.8
1929
5.3
5.1
62.9
1930
3.6
3.5
66.4
1931
2.6
2.7
69.1
1932
1.7
1.8
70.8
1933
2.6
2.6
73.5
1934
3.1
3.2
76.7
1935
3.8
3.9
80.6
1936
4.8
4.9
85.4
1937
5.5
5.6
91.1
1938
4.4
4.5
95.6
1939
5.6
5.8
101.4
1940
6.3
6.4
107.8
1941
8.1
8.8
116.7
1942
6.7
7.3
124.0
1943
7.2
7.8
131.9
1944
7.3
7.9
139.8
1945
6.0
6.6
146.4
37
Table 1B. Oregon timber harvest in billion board feet (bbf) with annual and cumulative
TPO in MMT C, 1906-2017.
Harvest Year Harvest (bbf)
Timber Product
Output (MMT C)
Cumulative TPO
(MMT C)
1946
6.6
7.2
153.6
1947
8.2
9.0
162.6
1948
8.4
9.2
171.8
1949
6.9
7.5
179.3
1950
7.9
8.6
187.9
1951
8.7
10.2
198.1
1952
9.8
11.5
209.6
1953
8.6
10.1
219.7
1954
8.9
10.4
230.0
1955
9.7
11.4
241.4
1956
9.3
10.9
252.4
1957
7.6
8.9
261.3
1958
7.7
9.0
270.3
1959
8.9
10.5
280.8
1960
8.4
9.8
290.6
1961
7.4
9.4
299.9
1962
8.5
10.7
310.7
1963
8.7
11.0
321.6
1964
9.4
11.9
333.5
1965
9.4
11.9
345.4
1966
8.9
11.3
356.7
1967
8.4
10.6
367.2
1968
9.7
12.3
379.5
1969
9.2
11.5
391.1
1970
8.0
10.1
401.1
1971
9.0
12.3
413.4
1972
9.7
13.3
426.7
1973
9.4
12.8
439.5
1974
8.4
11.4
451.0
1975
7.4
10.1
461.0
1976
8.1
11.1
472.2
1977
7.9
10.8
482.9
1978
8.0
10.9
493.8
1979
7.7
10.5
504.3
1980
6.6
9.5
513.8
1981
5.7
8.2
522.0
1982
5.8
8.2
530.2
1983
7.5
10.7
540.9
1984
7.5
10.8
551.7
1985
8.1
11.6
563.4
1986
8.7
12.5
575.9
38
Table 1B. Oregon timber harvest in billion board feet (bbf) with annual and cumulative
TPO in MMT C, 1906-2017.
Harvest Year Harvest (bbf)
Timber Product
Output (MMT C)
Cumulative TPO
(MMT C)
1987
8.2
11.8
587.7
1988
8.6
12.4
600.0
1989
8.4
12.0
612.1
1990
6.2
10.1
622.2
1991
6.1
9.9
632.1
1992
5.7
9.3
641.4
1993
5.3
8.6
650.1
1994
4.2
6.8
656.9
1995
4.3
7.0
663.9
1996
3.9
6.4
670.3
1997
4.1
6.7
677.0
1998
3.5
5.8
682.8
1999
3.8
6.1
688.9
2000
3.9
7.0
696.0
2001
3.4
6.3
702.3
2002
3.9
7.2
709.4
2003
4.0
7.3
716.8
2004
4.5
7.9
724.7
2005
4.4
7.8
732.5
2006
4.3
7.7
740.2
2007
3.8
6.7
746.9
2008
3.4
6.1
753.1
2009
2.7
5.1
758.1
2010
3.2
6.0
764.1
2011
3.6
6.7
770.8
2012
3.7
6.9
777.8
2013
4.2
7.8
785.5
2014
4.1
7.6
793.1
2015
3.8
7.0
800.1
2016
3.9
7.2
807.3
2017
3.9
7.1
814.4
39
Appendix Table 1C. Hundred cubic feet (ccf) to metric tons (MT) carbon conversion
factors (embedded .pdf file).
Table 1C.pdf
Table 1D. Timber products and corresponding primary products (embedded .pdf file).
Table1D_Timber
productsToPrimaryPro
Table 1E. Primary products and corresponding end use products (embedded .pdf file).
Table1E_PrimaryProd
ToEndUseProds.pdf
Table 1F. Discarded harvested wood and paper half-lives and landfill fixed ratios
(embedded .pdf file).
DiscardedWood&Pa
perHalfLives&FixedRa
40
Table 1G. Annual changes and cumulative stocks in products-in-use and solid waste disposal sites (SWDS) and
annual and cumulative emissions with and without energy capture, from Oregon timber harvested 1906-2017.
Inventory year
Products-in use
(end-use and recovered)
Products in SWDS
(landfills and dumps)
Emissions w/energy
capture
(fuelwood and discard
burning)
Emissions w/o energy
capture
(landfills, dumps, recovered
products, burning and
compost)
Cumulative
stock
(MMT C)
Annual stock
change
(MMT C/year)
Cumulative
stock
(MMT C)
Annual stock
change
(MMT C/year)
Cumulative
emissions
(MMT CO
2
e)
Annual
emissions
change
(MMT
CO
2
e/year)
Cumulative
emissions
(MMT CO
2
e)
Annual
emissions
change
(MMT
CO
2
e/year)
1907
0.72
0.72
0.04
0.04
1.94
1.94
0.05
0.05
1908
1.42
0.70
0.11
0.07
3.97
2.03
0.17
0.12
1909
2.11
0.70
0.21
0.10
6.13
2.16
0.35
0.18
1910
2.84
0.73
0.34
0.13
8.48
2.35
0.61
0.25
1911
3.59
0.74
0.50
0.16
10.97
2.49
0.94
0.33
1912
4.34
0.76
0.67
0.18
13.61
2.64
1.35
0.41
1913
5.07
0.72
0.87
0.19
16.26
2.65
1.83
0.48
1914
5.76
0.69
1.07
0.20
18.91
2.65
2.39
0.56
1915
6.35
0.60
1.28
0.21
21.38
2.47
3.02
0.63
1916
6.95
0.60
1.49
0.21
23.90
2.52
3.71
0.69
1917
7.64
0.69
1.71
0.22
26.73
2.83
4.48
0.77
1918
8.51
0.87
1.95
0.24
30.11
3.38
5.33
0.85
1919
9.39
0.89
2.20
0.25
33.64
3.54
6.27
0.94
1920
10.38
0.99
2.48
0.27
37.56
3.91
7.32
1.04
1921
11.22
0.83
2.76
0.28
41.16
3.60
8.45
1.13
1922
12.18
0.96
3.06
0.30
45.19
4.03
9.68
1.23
1923
13.22
1.04
3.37
0.31
49.53
4.34
11.02
1.34
1924
14.51
1.29
3.71
0.34
54.66
5.13
12.49
1.46
1925
15.88
1.38
4.08
0.37
60.18
5.52
14.09
1.60
1926
17.56
1.68
4.49
0.41
66.67
6.49
15.85
1.76
1927
19.19
1.63
4.94
0.45
73.24
6.57
17.78
1.93
1928
20.68
1.49
5.41
0.47
79.64
6.40
19.87
2.09
1929
22.26
1.59
5.90
0.49
86.45
6.80
22.12
2.25
1930
24.12
1.85
6.42
0.52
94.13
7.69
24.55
2.43
1931
24.97
0.85
6.93
0.51
99.33
5.19
27.10
2.55
1932
25.40
0.43
7.39
0.46
103.37
4.05
29.71
2.62
1933
25.33
-0.07
7.78
0.39
106.00
2.63
32.35
2.64
1934
25.81
0.48
8.13
0.35
109.96
3.96
35.04
2.69
1935
26.59
0.78
8.47
0.34
114.70
4.73
37.79
2.76
1936
27.77
1.18
8.83
0.36
120.55
5.85
40.66
2.86
1937
29.44
1.67
9.23
0.40
127.86
7.31
43.66
3.01
1938
31.45
2.01
9.68
0.46
136.30
8.44
46.85
3.19
1939
32.76
1.31
10.16
0.47
143.10
6.80
50.18
3.33
41
Table 1G. Annual changes and cumulative stocks in products-in-use and solid waste disposal sites (SWDS) and
annual and cumulative emissions with and without energy capture, from Oregon timber harvested 1906-2017.
Inventory year
Products-in use
(end-use and recovered)
Products in SWDS
(landfills and dumps)
Emissions w/energy
capture
(fuelwood and discard
burning)
Emissions w/o energy
capture
(landfills, dumps, recovered
products, burning and
compost)
Cumulative
stock
(MMT C)
Annual stock
change
(MMT C/year)
Cumulative
stock
(MMT C)
Annual stock
change
(MMT C/year)
Cumulative
emissions
(MMT CO
2
e)
Annual
emissions
change
(MMT
CO
2
e/year)
Cumulative
emissions
(MMT CO
2
e)
Annual
emissions
change
(MMT
CO
2
e/year)
1940
34.71
1.95
10.67
0.51
151.74
8.64
53.69
3.51
1941
36.94
2.23
11.22
0.55
161.37
9.63
57.46
3.77
1942
40.40
3.46
11.88
0.66
174.60
13.23
61.54
4.08
1943
43.87
3.47
12.58
0.70
181.91
7.32
65.84
4.31
1944
47.31
3.44
13.52
0.94
189.74
7.82
70.71
4.87
1945
50.48
3.17
14.66
1.13
197.66
7.92
76.15
5.44
1946
52.45
1.96
15.88
1.23
204.25
6.59
82.09
5.94
1947
54.71
2.27
17.14
1.25
211.44
7.19
88.45
6.36
1948
58.05
3.34
18.46
1.32
220.38
8.95
95.31
6.86
1949
61.28
3.22
19.88
1.43
229.53
9.15
102.76
7.44
1950
63.12
1.85
21.34
1.46
237.02
7.49
110.70
7.94
1951
65.69
2.56
22.78
1.43
245.62
8.60
119.07
8.37
1952
69.20
3.52
24.27
1.49
255.78
10.17
127.93
8.87
1953
73.37
4.16
25.88
1.61
267.23
11.45
137.41
9.48
1954
76.24
2.87
27.58
1.70
277.26
10.03
147.49
10.08
1955
79.20
2.96
29.29
1.71
287.61
10.35
158.08
10.59
1956
82.73
3.52
31.03
1.74
298.96
11.35
169.19
11.11
1957
85.75
3.02
32.79
1.76
309.87
10.90
180.84
11.65
1958
87.22
1.47
34.50
1.71
318.73
8.86
192.89
12.06
1959
88.84
1.62
36.09
1.59
327.73
9.00
205.24
12.34
1960
91.49
2.64
37.61
1.52
338.17
10.44
217.91
12.67
1961
94.05
2.56
38.60
0.98
347.96
9.79
231.13
13.22
1962
96.14
2.09
39.57
0.97
357.29
9.33
244.88
13.75
1963
99.10
2.96
40.55
0.98
367.98
10.70
259.09
14.21
1964
103.04
3.94
41.64
1.09
374.96
6.97
273.86
14.77
1965
107.53
4.49
42.81
1.17
382.53
7.57
289.13
15.28
1966
111.77
4.24
44.07
1.26
390.08
7.55
304.92
15.79
1967
115.35
3.58
45.37
1.31
397.25
7.17
321.16
16.24
1968
118.24
2.89
46.68
1.31
403.97
6.72
337.76
16.60
1969
122.43
4.19
48.03
1.35
411.81
7.83
354.74
16.97
1970
126.37
3.94
49.46
1.43
417.03
5.22
372.16
17.42
1971
128.91
2.55
51.54
2.08
421.59
4.56
387.54
15.38
1972
133.34
4.43
53.65
2.11
427.17
5.58
403.17
15.63
1973
138.44
5.10
55.86
2.21
433.19
6.02
419.12
15.95
42
Table 1G. Annual changes and cumulative stocks in products-in-use and solid waste disposal sites (SWDS) and
annual and cumulative emissions with and without energy capture, from Oregon timber harvested 1906-2017.
Inventory year
Products-in use
(end-use and recovered)
Products in SWDS
(landfills and dumps)
Emissions w/energy
capture
(fuelwood and discard
burning)
Emissions w/o energy
capture
(landfills, dumps, recovered
products, burning and
compost)
Cumulative
stock
(MMT C)
Annual stock
change
(MMT C/year)
Cumulative
stock
(MMT C)
Annual stock
change
(MMT C/year)
Cumulative
emissions
(MMT CO
2
e)
Annual
emissions
change
(MMT
CO
2
e/year)
Cumulative
emissions
(MMT CO
2
e)
Annual
emissions
change
(MMT
CO
2
e/year)
1974
143.30
4.85
58.13
2.27
437.69
4.51
435.39
16.28
1975
146.74
3.44
60.47
2.34
441.72
4.02
452.04
16.65
1976
148.92
2.18
62.77
2.30
445.26
3.55
468.97
16.93
1977
152.02
3.10
65.03
2.27
449.18
3.92
486.17
17.19
1978
154.29
2.26
67.30
2.26
454.55
5.37
503.63
17.46
1979
156.78
2.49
69.44
2.14
460.00
5.45
521.21
17.59
1980
158.99
2.21
71.48
2.04
465.24
5.24
538.89
17.68
1981
160.63
1.64
74.02
2.55
469.98
4.74
553.64
14.76
1982
161.22
0.59
76.47
2.45
474.05
4.07
568.31
14.67
1983
162.01
0.79
78.84
2.37
478.16
4.11
582.84
14.53
1984
164.73
2.72
81.27
2.42
484.11
5.95
597.23
14.39
1985
167.48
2.75
83.80
2.54
490.13
6.02
611.49
14.26
1986
170.83
3.35
86.47
2.66
496.60
6.48
625.65
14.16
1987
175.06
4.22
89.27
2.81
502.69
6.09
639.75
14.10
1988
178.42
3.36
92.28
3.01
508.42
5.72
653.87
14.12
1989
182.10
3.67
95.46
3.18
514.42
6.00
668.05
14.18
1990
185.64
3.55
98.76
3.30
519.21
4.79
682.33
14.27
1991
187.87
2.22
100.96
2.20
523.22
4.02
699.16
16.83
1992
189.75
1.89
103.22
2.26
527.15
3.93
716.25
17.09
1993
191.07
1.32
105.53
2.31
530.86
3.71
733.48
17.23
1994
192.27
1.20
107.83
2.30
532.45
1.60
750.71
17.23
1995
191.72
-0.55
110.15
2.32
533.71
1.26
767.92
17.21
1996
191.26
-0.46
112.45
2.30
535.72
2.01
785.02
17.09
1997
190.39
-0.87
114.65
2.20
537.55
1.83
801.85
16.84
1998
189.92
-0.47
116.78
2.13
539.45
1.90
818.38
16.52
1999
188.79
-1.13
118.83
2.06
541.10
1.65
834.56
16.18
2000
188.47
-0.32
120.86
2.03
541.53
0.43
850.39
15.84
2001
189.54
1.07
122.28
1.42
542.01
0.49
866.61
16.22
2002
189.95
0.42
123.62
1.34
542.45
0.44
882.80
16.19
2003
191.19
1.23
125.07
1.45
542.95
0.50
898.76
15.96
2004
192.56
1.37
126.49
1.42
543.46
0.51
914.83
16.06
2005
193.88
1.32
127.95
1.46
546.23
2.77
930.85
16.03
2006
195.12
1.23
129.44
1.49
548.97
2.74
946.85
16.00
2007
196.18
1.06
130.97
1.54
551.66
2.69
962.81
15.96
43
Table 1G. Annual changes and cumulative stocks in products-in-use and solid waste disposal sites (SWDS) and
annual and cumulative emissions with and without energy capture, from Oregon timber harvested 1906-2017.
Inventory year
Products-in use
(end-use and recovered)
Products in SWDS
(landfills and dumps)
Emissions w/energy
capture
(fuelwood and discard
burning)
Emissions w/o energy
capture
(landfills, dumps, recovered
products, burning and
compost)
Cumulative
stock
(MMT C)
Annual stock
change
(MMT C/year)
Cumulative
stock
(MMT C)
Annual stock
change
(MMT C/year)
Cumulative
emissions
(MMT CO
2
e)
Annual
emissions
change
(MMT
CO
2
e/year)
Cumulative
emissions
(MMT CO
2
e)
Annual
emissions
change
(MMT
CO
2
e/year)
2008
196.41
0.23
132.52
1.55
554.02
2.36
978.67
15.86
2009
196.11
-0.30
134.07
1.54
556.16
2.14
994.38
15.71
2010
195.14
-0.97
135.57
1.51
557.34
1.18
1009.84
15.46
2011
195.04
-0.10
137.10
1.53
558.72
1.38
1025.09
15.25
2012
195.65
0.61
138.69
1.59
560.28
1.56
1040.18
15.09
2013
196.40
0.75
140.34
1.65
561.89
1.60
1055.16
14.98
2014
197.86
1.45
142.08
1.74
563.68
1.80
1070.11
14.95
2015
199.10
1.25
143.89
1.81
565.45
1.76
1085.09
14.98
2016
199.71
0.61
145.74
1.85
567.07
1.62
1100.09
15.00
2017
200.45
0.74
147.63
1.89
568.73
1.66
1115.12
15.03
2018
201.09
0.64
149.56
1.92
570.38
1.65
1130.16
15.04
44
Table 1H. Cumulative storage of HWP carbon in products-in-use and in solid waste
disposal sites (SWDS), and cumulative emissions with and without energy capture
from Oregon timber harvests, 1906-2017.
Inventory year
Products-in-use
Products in
SWDS
Emissions w/
energy capture
from:
Emissions w/o energy capture from:
End-use
Recovered
In landfills
In dumps
Fuelwood
Burning
(discarded)
Landfills
Dumps
Recovered
products
Burning
Compost
Cumulative storage (MMT C)
Cumulative emissions (MMTCO
2
e)
1907 0.7 0.0 0.0 0.0 1.9 0.0 0.0 0.0 0.0 0.1 0.0
1908 1.4 0.0 0.0 0.1 4.0 0.0 0.0 0.0 0.0 0.2 0.0
1909 2.1 0.0 0.0 0.2 6.1 0.0 0.0 0.0 0.0 0.3 0.0
1910 2.8 0.0 0.0 0.3 8.5 0.0 0.0 0.1 0.0 0.5 0.0
1911 3.6 0.0 0.0 0.5 11.0 0.0 0.0 0.1 0.0 0.8 0.0
1912 4.3 0.0 0.0 0.7 13.6 0.0 0.0 0.2 0.0 1.1 0.0
1913 5.1 0.0 0.0 0.9 16.3 0.0 0.0 0.4 0.0 1.5 0.0
1914 5.8 0.0 0.0 1.1 18.9 0.0 0.0 0.6 0.0 1.8 0.0
1915 6.4 0.0 0.0 1.3 21.4 0.0 0.0 0.8 0.0 2.2 0.0
1916 7.0 0.0 0.0 1.5 23.9 0.0 0.0 1.0 0.0 2.7 0.0
1917 7.6 0.0 0.0 1.7 26.7 0.0 0.0 1.4 0.0 3.1 0.0
1918 8.5 0.0 0.0 2.0 30.1 0.0 0.0 1.7 0.0 3.6 0.0
1919 9.4 0.0 0.0 2.2 33.6 0.0 0.0 2.1 0.0 4.2 0.0
1920 10.4 0.0 0.0 2.5 37.6 0.0 0.0 2.6 0.0 4.8 0.0
1921 11.2 0.0 0.0 2.8 41.2 0.0 0.0 3.1 0.0 5.4 0.0
1922 12.2 0.0 0.0 3.1 45.2 0.0 0.0 3.6 0.0 6.1 0.0
1923 13.2 0.0 0.0 3.4 49.5 0.0 0.0 4.2 0.0 6.8 0.0
1924 14.5 0.0 0.0 3.7 54.7 0.0 0.0 4.9 0.0 7.6 0.0
1925 15.9 0.0 0.0 4.1 60.2 0.0 0.0 5.7 0.0 8.4 0.0
1926 17.6 0.0 0.0 4.5 66.7 0.0 0.0 6.5 0.0 9.4 0.0
1927 19.2 0.0 0.0 4.9 73.2 0.0 0.0 7.4 0.0 10.4 0.0
1928 20.7 0.0 0.0 5.4 79.6 0.0 0.0 8.4 0.0 11.5 0.0
1929 22.3 0.0 0.0 5.9 86.4 0.0 0.0 9.4 0.0 12.7 0.0
1930 24.1 0.0 0.0 6.4 94.1 0.0 0.0 10.6 0.0 13.9 0.0
1931 25.0 0.0 0.0 6.9 99.3 0.0 0.0 11.9 0.0 15.2 0.0
45
Table 1H. Cumulative storage of HWP carbon in products-in-use and in solid waste
disposal sites (SWDS), and cumulative emissions with and without energy capture
from Oregon timber harvests, 1906-2017.
Inventory year
Products-in-use
Products in
SWDS
Emissions w/
energy capture
from:
Emissions w/o energy capture from:
End
-use
Recovered
In landfills
In dumps
Fuelwood
Burning
(discarded)
Landfills
Dumps
Recovered
products
Burning
Compost
Cumulative storage (MMT C) Cumulative emissions (MMTCO
2
e)
1932 25.4 0.0 0.0 7.4 103.4 0.0 0.0 13.2 0.0 16.5 0.0
1933 25.3 0.0 0.0 7.8 106.0 0.0 0.0 14.7 0.0 17.7 0.0
1934 25.8 0.0 0.0 8.1 110.0 0.0 0.0 16.2 0.0 18.8 0.0
1935 26.6 0.0 0.0 8.5 114.7 0.0 0.0 17.8 0.0 20.0 0.0
1936 27.8 0.0 0.0 8.8 120.6 0.0 0.0 19.5 0.0 21.2 0.0
1937 29.4 0.0 0.0 9.2 127.9 0.0 0.0 21.2 0.0 22.5 0.0
1938 31.5 0.0 0.0 9.7 136.3 0.0 0.0 23.0 0.0 23.9 0.0
1939 32.8 0.0 0.0 10.2 143.1 0.0 0.0 24.8 0.0 25.4 0.0
1940 34.7 0.0 0.0 10.7 151.7 0.0 0.0 26.8 0.0 26.9 0.0
1941 36.9 0.0 0.0 11.2 161.4 0.0 0.0 28.8 0.0 28.7 0.0
1942 40.4 0.0 0.1 11.8 174.6 0.0 0.0 30.9 0.0 30.6 0.0
1943 43.9 0.0 0.2 12.4 181.9 0.0 0.0 33.2 0.0 32.7 0.0
1944 47.3 0.0 0.2 13.3 189.7 0.0 0.0 35.5 0.0 35.2 0.0
1945 50.5 0.0 0.3 14.4 197.7 0.0 0.0 38.1 0.0 38.0 0.0
1946 52.4 0.0 0.4 15.5 204.2 0.0 0.0 40.9 0.0 41.2 0.0
1947 54.7 0.0 0.5 16.7 211.4 0.0 0.1 43.9 0.0 44.5 0.0
1948 58.1 0.0 0.6 17.9 220.4 0.0 0.1 47.3 0.0 48.0 0.0
1949 61.3 0.0 0.6 19.2 229.5 0.0 0.1 50.9 0.0 51.8 0.0
1950 63.1 0.0 0.7 20.6 237.0 0.0 0.1 54.8 0.0 55.8 0.0
1951 65.7 0.0 1.0 21.8 245.6 0.0 0.2 59.0 0.0 59.8 0.0
1952 69.2 0.0 1.3 23.0 255.8 0.0 0.2 63.6 0.0 64.1 0.0
1953 73.4 0.0 1.5 24.3 267.2 0.0 0.3 68.4 0.0 68.8 0.0
1954 76.2 0.0 1.8 25.8 277.3 0.0 0.4 73.5 0.0 73.7 0.0
1955 79.2 0.0 2.1 27.2 287.6 0.0 0.5 78.9 0.0 78.7 0.0
1956 82.7 0.0 2.4 28.6 299.0 0.0 0.6 84.6 0.0 84.0 0.0
1957 85.7 0.0 2.7 30.1 309.9 0.0 0.7 90.7 0.0 89.4 0.0
46
Table 1H. Cumulative storage of HWP carbon in products-in-use and in solid waste
disposal sites (SWDS), and cumulative emissions with and without energy capture
from Oregon timber harvests, 1906-2017.
Inventory year
Products-in-use
Products in
SWDS
Emissions w/
energy capture
from:
Emissions w/o energy capture from:
End
-use
Recovered
In landfills
In dumps
Fuelwood
Burning
(discarded)
Landfills
Dumps
Recovered
products
Burning
Compost
Cumulative storage (MMT C) Cumulative emissions (MMTCO
2
e)
1958 87.2 0.0 3.0 31.5 318.7 0.0 0.9 97.1 0.0 94.9 0.0
1959 88.8 0.0 3.3 32.7 327.7 0.0 1.0 103.9 0.0 100.3 0.0
1960 91.5 0.0 3.6 34.0 338.2 0.0 1.2 110.9 0.0 105.8 0.0
1961 93.6 0.5 4.1 34.5 348.0 0.0 1.4 118.2 0.0 111.6 0.0
1962 95.3 0.8 4.5 35.0 357.3 0.0 1.6 125.5 0.4 117.3 0.0
1963 98.0 1.1 5.0 35.6 368.0 0.0 1.8 133.0 1.1 123.2 0.0
1964 101.7 1.3 5.4 36.2 375.0 0.0 2.1 140.5 2.1 129.2 0.0
1965 106.0 1.5 5.9 36.9 382.5 0.0 2.4 148.1 3.2 135.5 0.0
1966 110.1 1.7 6.4 37.7 390.1 0.0 2.7 155.8 4.5 142.0 0.0
1967 113.6 1.8 6.9 38.5 397.3 0.0 3.0 163.6 5.9 148.7 0.0
1968 116.3 1.9 7.4 39.3 404.0 0.0 3.3 171.6 7.4 155.4 0.0
1969 120.5 2.0 7.9 40.1 411.8 0.0 3.6 179.7 9.0 162.3 0.0
1970 124.3 2.0 8.5 41.0 417.0 0.0 4.0 188.0 10.7 169.4 0.0
1971 126.9 2.0 9.8 41.7 421.6 0.0 4.4 196.4 12.5 174.3 0.0
1972 131.3 2.0 11.2 42.5 427.2 0.0 4.8 205.0 14.2 179.1 0.0
1973 136.4 2.0 12.6 43.3 433.2 0.0 5.3 213.6 16.0 184.2 0.0
1974 141.3 2.0 14.0 44.1 437.7 0.0 5.9 222.4 17.7 189.4 0.0
1975 144.7 2.1 15.5 45.0 441.7 0.0 6.5 231.4 19.5 194.7 0.0
1976 146.8 2.1 16.9 45.8 445.3 0.0 7.2 240.5 21.2 200.1 0.0
1977 149.9 2.1 18.4 46.7 449.2 0.0 8.0 249.7 23.1 205.4 0.0
1978 152.1 2.1 19.8 47.5 454.5 0.0 8.8 259.1 24.9 210.8 0.0
1979 154.6 2.1 21.2 48.2 460.0 0.0 9.6 268.7 26.7 216.2 0.0
1980 156.9 2.1 22.6 48.9 465.2 0.0 10.5 278.4 28.5 221.4 0.0
1981 158.4 2.3 26.4 47.6 470.0 0.0 11.5 288.1 30.4 223.7 0.0
1982 158.9 2.3 30.1 46.3 474.1 0.0 12.6 297.6 32.3 225.8 0.0
1983 159.7 2.4 33.7 45.1 478.2 0.0 13.9 306.8 34.3 227.9 0.0
47
Table 1H. Cumulative storage of HWP carbon in products-in-use and in solid waste
disposal sites (SWDS), and cumulative emissions with and without energy capture
from Oregon timber harvests, 1906-2017.
Inventory year
Products-in-use
Products in
SWDS
Emissions w/
energy capture
from:
Emissions w/o energy capture from:
End
-use
Recovered
In landfills
In dumps
Fuelwood
Burning
(discarded)
Landfills
Dumps
Recovered
products
Burning
Compost
Cumulative storage (MMT C) Cumulative emissions (MMTCO
2
e)
1984 162.4 2.3 37.3 44.0 484.1 0.0 15.3 315.6 36.3 230.0 0.0
1985 165.1 2.3 40.8 43.0 490.1 0.0 16.8 324.2 38.3 232.1 0.0
1986 168.5 2.3 44.4 42.1 496.6 0.0 18.5 332.6 40.3 234.3 0.0
1987 172.7 2.4 48.1 41.2 502.7 0.0 20.2 340.7 42.3 236.5 0.0
1988 176.0 2.4 51.8 40.5 508.4 0.0 22.1 348.6 44.4 238.8 0.0
1989 179.6 2.5 55.6 39.8 514.4 0.0 24.1 356.4 46.4 241.1 0.0
1990 183.1 2.6 59.5 39.2 519.2 0.0 26.3 364.0 48.6 243.5 0.0
1991 184.7 3.2 63.7 37.3 523.2 0.0 28.5 371.4 50.8 247.9 0.5
1992 186.1 3.7 67.7 35.5 527.2 0.0 30.9 378.5 53.6 252.2 1.1
1993 187.0 4.0 71.7 33.8 530.9 0.0 33.5 385.1 56.7 256.5 1.6
1994 188.0 4.3 75.6 32.2 532.5 0.0 36.2 391.4 60.2 260.8 2.2
1995 187.2 4.5 79.5 30.7 533.7 0.0 39.0 397.4 63.9 265.0 2.7
1996 186.6 4.6 83.2 29.2 535.7 0.0 41.9 403.0 67.7 269.2 3.2
1997 185.7 4.6 86.8 27.9 537.5 0.0 44.9 408.4 71.7 273.2 3.7
1998 185.3 4.6 90.2 26.6 539.5 0.0 48.0 413.5 75.7 277.0 4.2
1999 184.2 4.5 93.4 25.4 541.1 0.0 51.1 418.3 79.6 280.8 4.7
2000 184.0 4.4 96.6 24.2 541.5 0.0 54.4 422.9 83.5 284.5 5.1
2001 184.8 4.7 99.1 23.2 542.0 0.0 57.7 427.2 87.4 287.8 6.5
2002 184.9 5.0 101.5 22.1 542.5 0.0 61.0 431.4 91.4 291.2 7.9
2003 186.0 5.2 103.9 21.2 542.9 0.0 64.3 435.3 95.7 294.2 9.3
2004 187.2 5.4 106.3 20.2 543.5 0.0 67.5 439.0 100.2 297.3 10.8
2005 188.4 5.5 108.6 19.4 546.2 0.0 70.8 442.6 104.8 300.4 12.3
2006 189.4 5.7 110.9 18.5 549.0 0.0 74.1 446.0 109.5 303.4 13.8
2007 190.4 5.8 113.2 17.7 551.7 0.0 77.3 449.2 114.4 306.5 15.3
2008 190.5 5.9 115.5 17.0 554.0 0.0 80.6 452.3 119.4 309.6 16.8
2009 190.2 5.9 117.8 16.3 556.2 0.0 83.8 455.2 124.4 312.6 18.3
48
Table 1H. Cumulative storage of HWP carbon in products-in-use and in solid waste
disposal sites (SWDS), and cumulative emissions with and without energy capture
from Oregon timber harvests, 1906-2017.
Inventory year
Products-in-use
Products in
SWDS
Emissions w/
energy capture
from:
Emissions w/o energy capture from:
End
-use
Recovered
In landfills
In dumps
Fuelwood
Burning
(discarded)
Landfills
Dumps
Recovered
products
Burning
Compost
Cumulative storage (MMT C) Cumulative emissions (MMTCO
2
e)
2010 189.3 5.9 119.9 15.6 557.3 0.0 87.0 458.0 129.5 315.5 19.8
2011 189.2 5.8 122.1 15.0 558.7 0.0 90.2 460.7 134.6 318.4 21.2
2012 189.9 5.8 124.3 14.4 560.3 0.0 93.4 463.3 139.6 321.3 22.6
2013 190.6 5.8 126.5 13.8 561.9 0.0 96.6 465.7 144.5 324.3 24.1
2014 192.0 5.8 128.8 13.3 563.7 0.0 99.7 468.1 149.5 327.3 25.6
2015 193.2 5.9 131.1 12.8 565.4 0.0 102.9 470.3 154.5 330.3 27.1
2016 193.7 6.0 133.5 12.3 567.1 0.0 106.0 472.4 159.5 333.4 28.6
2017 194.3 6.1 135.8 11.8 568.7 0.0 109.2 474.5 164.7 336.6 30.2
2018 194.9 6.2 138.2 11.4 570.4 0.0 112.3 476.5 170.0 339.7 31.7
Note: HWP storage and emissions resulting from 2017 and prior harvests are reported in 2018.
49
Table 1J. Cumulative storage of HWP carbon in products-in-use and SWDS by
ownership in MMT C, from Oregon timber harvests, 1962-2017.
Inventory year
Private USFS BLM
State and
other
Tribal
Products
-in
-
use
SWDS
Products
-in
-
use
SWDS
Products
-in
-
use
SWDS
Products
-in
-
use
SWDS
Products
-in
-
use
SWDS
1963
3.6
0.1
2.7
0.1
1.0
0.0
0.1
0.0
0.0
0.0
1964
7.0
0.5
5.6
0.4
2.2
0.1
0.3
0.0
0.1
0.0
1965
10.3
1.0
8.5
0.8
3.6
0.3
0.5
0.0
0.2
0.0
1966
13.4
1.7
11.5
1.4
4.5
0.6
0.8
0.1
0.2
0.0
1967
16.4
2.4
13.8
2.0
5.4
0.8
0.9
0.1
0.3
0.0
1968
18.8
3.2
15.8
2.7
6.0
1.1
1.0
0.2
0.4
0.1
1969
21.7
4.1
18.2
3.4
7.0
1.3
1.1
0.2
0.4
0.1
1970
24.4
5.0
20.5
4.2
7.8
1.6
1.2
0.3
0.5
0.1
1971
26.7
6.2
21.9
5.2
8.3
2.0
1.3
0.3
0.5
0.1
1972
29.6
7.4
24.0
6.1
9.2
2.4
1.4
0.4
0.6
0.1
1973
32.0
8.7
26.9
7.2
10.3
2.8
1.6
0.4
0.7
0.2
1974
34.1
9.9
29.6
8.3
11.4
3.2
1.9
0.5
0.7
0.2
1975
36.4
11.1
31.4
9.4
11.9
3.6
2.1
0.6
0.8
0.2
1976
38.6
12.4
32.5
10.5
11.9
4.0
2.1
0.7
0.9
0.3
1977
40.4
13.7
34.2
11.5
12.4
4.3
2.3
0.7
1.0
0.3
1978
42.0
14.9
35.4
12.6
12.9
4.7
2.4
0.8
1.0
0.3
1979
43.5
16.1
37.0
13.6
13.1
5.0
2.5
0.9
1.1
0.4
1980
44.7
17.3
38.5
14.6
13.4
5.4
2.7
1.0
1.2
0.4
1981
46.1
18.6
39.3
15.7
13.7
5.7
2.8
1.0
1.2
0.4
1982
47.0
19.8
39.7
16.8
13.8
6.1
2.9
1.1
1.3
0.5
1983
48.7
21.1
39.7
17.8
13.6
6.4
3.0
1.2
1.3
0.5
1984
50.3
22.4
41.2
18.8
13.9
6.6
3.2
1.3
1.4
0.6
1985
51.4
23.7
42.9
19.9
14.3
7.0
3.3
1.4
1.5
0.6
1986
52.9
25.1
45.0
21.0
14.7
7.3
3.5
1.5
1.5
0.7
1987
54.6
26.4
47.5
22.2
15.3
7.6
3.6
1.6
1.6
0.7
1988
56.0
27.8
49.3
23.6
16.0
8.0
3.8
1.7
1.6
0.8
1989
57.4
29.2
51.2
25.0
17.0
8.4
3.9
1.9
1.7
0.8
1990
59.3
30.7
52.8
26.4
17.4
8.9
4.0
2.0
1.8
0.9
1991
61.3
31.8
53.3
27.4
17.7
9.2
4.1
2.1
1.8
0.9
1992
63.3
33.0
53.9
28.4
17.6
9.5
4.1
2.1
1.9
0.9
1993
65.5
34.2
53.5
29.3
17.6
9.8
4.2
2.2
1.9
1.0
1994
68.0
35.5
52.9
30.1
17.4
10.0
4.2
2.3
2.0
1.0
1995
69.8
37.0
51.7
30.9
16.8
10.3
4.2
2.4
2.0
1.1
1996
71.5
38.5
50.4
31.6
16.4
10.5
4.3
2.4
2.0
1.1
1997
72.6
40.0
49.1
32.2
16.2
10.6
4.3
2.5
2.0
1.1
50
Table 1J. Cumulative storage of HWP carbon in products-in-use and SWDS by
ownership in MMT C, from Oregon timber harvests, 1962-2017.
Inventory year
Private USFS BLM
State and
other
Tribal
Products
-in
-
use
SWDS
Products
-in
-
use
SWDS
Products
-in
-
use
SWDS
Products
-in
-
use
SWDS
Products
-in
-
use
SWDS
1998
73.9
41.6
48.2
32.7
15.9
10.8
4.4
2.6
2.0
1.2
1999
74.7
43.1
47.0
33.1
15.5
11.0
4.5
2.7
2.1
1.2
2000
76.1
44.6
45.9
33.5
15.3
11.1
4.7
2.8
2.1
1.3
2001
78.5
45.8
45.0
33.8
15.0
11.2
5.1
2.9
2.1
1.3
2002
80.4
46.9
44.0
34.0
14.6
11.2
5.4
2.9
2.1
1.3
2003
83.0
48.1
43.1
34.1
14.3
11.3
5.7
3.0
2.2
1.3
2004
85.4
49.3
42.3
34.3
14.1
11.3
6.0
3.1
2.2
1.4
2005
87.7
50.6
41.7
34.5
13.9
11.4
6.3
3.2
2.2
1.4
2006
89.7
51.8
41.2
34.6
13.7
11.4
6.7
3.4
2.2
1.4
2007
91.8
53.1
40.6
34.8
13.6
11.5
6.9
3.5
2.3
1.4
2008
93.0
54.4
39.9
34.9
13.5
11.5
7.1
3.6
2.3
1.5
2009
93.6
55.7
39.3
35.0
13.4
11.6
7.3
3.7
2.3
1.5
2010
93.6
57.0
38.8
35.2
13.3
11.7
7.4
3.8
2.3
1.5
2011
94.1
58.2
38.3
35.3
13.2
11.7
7.6
4.0
2.4
1.6
2012
95.2
59.5
38.1
35.5
13.2
11.8
7.9
4.1
2.4
1.6
2013
96.4
60.8
37.9
35.6
13.1
11.9
8.0
4.2
2.4
1.6
2014
98.3
62.2
37.7
35.8
13.1
11.9
8.2
4.4
2.4
1.6
2015
100.0
63.6
37.5
36.0
13.1
12.0
8.3
4.5
2.4
1.7
2016
101.0
65.0
37.3
36.2
13.1
12.1
8.5
4.7
2.4
1.7
2017
102.1
66.5
37.0
36.4
13.1
12.2
8.7
4.8
2.4
1.7
2018
103.3
68.0
36.7
36.6
13.0
12.3
8.9
5.0
2.4
1.8
Note: HWP storage and emissions resulting from 2017 and prior harvests are reported in 2018.
51
APPENDIX 2
A Oregon timber product ratios (embedded .pdf file)
ORTimberProductR
atios.pdf
B Oregon primary product ratios (embedded .pdf file)
ORPrimaryProductR
atios.pdf
C End use product ratios (embedded .pdf file)
OREndUseProductR
atios.pdf
D Harvested wood product end-use half-lives (embedded .pdf file)
OREndUseHalfLifeD
ata.pdf
E Discarded product disposition ratios (embedded .pdf file)
DiscardDisposition
Data.pdf
F
Discarded harvested wood and paper half-lives and landfill fixed
ratios (embedded .pdf file)
DiscardedWood&Pa
perHalfLives&FixedRa
52
Appendix 3
MONTE CARLO SIMULATION
Simulation Specifications
The goal of the MC simulation was to produce estimates and confidence intervals for overall
values for SWDS, PIU, EWOEC, EEC, and their combinations. To achieve this goal the Monte
Carlo (MC) simulation directly alters 15 different parameters, listed in Table 3a below. The
columns Min Value, Peak Value, and Max Value describe the desired 90% confidence interval
for the range of random proportions against which the parameters of interest were adjusted.
Parameters for harvest, timber product ratios and primary product ratios each have a set of
three confidence intervals based on the years for which the estimates were provided. These
parameters require a separate MC random variable for each year set. Finally, parameters for
timber, product, and end use product ratios have a sum-to-one characteristic. For instance,
timber product ratios (TPRs) represent the proportion of total harvest for a given year that is
allocated to each of the 40 timber products. While the proportions change during the time series
the full set of 40 TPR proportions will always sum to one.
Table 3a. Monte Carlo simulation target parameters and ranges.
Parameter
ID
Parameter Name
First
Year
Last
Year
Min
Value
Peak
Value
Max
Value
Sum-
to-One
0
CCF to MgC conversion factors
n/a
n/a
0.95
1
1.05
No
1
Harvest
1906
1945
0.7
1
1.3
No
1
Harvest
1946
1979
0.8
1
1.2
No
1
Harvest
1980
2100
0.85
1
1.15
No
2
Timber product ratios
1906
1945
0.7
1
1.3
Yes
2
Timber product ratios
1946
1979
0.8
1
1.2
Yes
2
Timber product ratios
1980
2100
0.85
1
1.15
Yes
3
Primary product ratios
1906
1949
0.7
1
1.3
Yes
3
Primary product ratios
1950
1979
0.8
1
1.2
Yes
3
Primary product ratios
1980
2100
0.85
1
1.15
Yes
4
End use product ratios
n/a
n/a
0.85
1
1.15
Yes
5
Product half lives
n/a
n/a
0.85
1
1.15
No
6
Discarded disposition ratios (paper)
n/a
n/a
0.85
1
1.15
Yes
7
Discarded disposition ratios (wood)
n/a
n/a
0.85
1
1.15
Yes
8
Landfill decay limits (paper)
n/a
n/a
0.85
1
1.15
No
9
Landfill decay limits (wood)
n/a
n/a
0.85
1
1.15
No
53
10
Landfill half-lives (paper)
n/a
n/a
0.85
1
1.15
No
11
Landfill half-lives (wood)
n/a
n/a
0.85
1
1.15
No
12
Dump half-lives (paper)
n/a
n/a
0.85
1
1.15
No
13
Dump half-lives (wood)
n/a
n/a
0.85
1
1.15
No
14
Recovered half-lives (paper)
n/a
n/a
0.85
1
1.15
No
15
Recovered half-lives (wood)
n/a
n/a
0.85
1
1.15
No
Simulation Sampling
As in other HWP publications such as Stockmann et al. (2012) and Anderson et al. (2013), the
random variables in the Monte Carlo simulations are drawn from triangular distributions. The
distributions all have a peak value of 1.0 (Table 3a) and symmetrically taper to given 90%
confidence interval bounds. The values from the triangular distribution are used as proportions
for adjusting parameter values.
Drawing random variables from triangular distributions is a multi-step process. The first step
involves drawing random variables from a standard uniform (0, 1) distribution using a Latin
Hypercube Sampling (LHS) process. The standard uniform distribution is an excellent starting
place for randomly selecting points along other distributions because random uniform points can
be translated as locations along some other distribution’s Cumulative Distribution Function
(CDF). However, purely random sampling may be inefficient because with unorganized, truly
random draws of random uniform variables for more than one distribution there is a strong
probability that samples will, by chance, fall in the same local region for several of the 22
distributions. Many iterations are needed to obtain a full suite of the possible MC values across
distributions. LHS offers a method for partitioning the uniform distributions across the range of
possible values from which the random selections can be drawn, thus preventing clumping. With
LHS many fewer iterations are needed to achieve stable estimates from the MC simulation.
The function “randomLHS” in the R package “lhs” was used to conduct a random LHS sample.
The lhs package divides the uniform distribution into as many partitions as there are iterations
and then randomly selects a value from the range of values within each partition. The order of
the selected values was also randomized.
Once the matrix from the LHS (rows = number of iterations, columns = desired number of
distributions) has been created the cell values are transformed into draws from triangular
distributions (Wikipedia: Triangular Distribution).
=
0 < < () , +
(
)
()
,
(
1
)(
)
()
In this equation U is the random uniform variable (the LHS draw),
(
)
= ()/(), a is
the minimum value for the triangular distribution, b is the maximum value, and c is the mode.
54
Note that the specifications require 90% confidence intervals and do not state what the
endpoints for the triangular distributions should be. Endpoint values that corresponded with the
desired 90% confidence intervals (Table 3b) were derived through simulation and trial and error.
Table 3b. Translation of 90% Confidence Intervals to endpoints for triangular distributions
90% Confidence Interval Range Endpoints Used
0.7, 1.3
0.57, 1.43
0.8, 1.2
0.71, 1.29
0.85, 1.15
0.78, 1.22
0.95, 1.05
0.93, 1.07
Each random draw from a triangular distribution was used to adjust an MC parameter for all
years in a single iteration. The next iteration would use the subsequent random draw. The only
exception occurs when the 90% confidence intervals changed within the year set for an
iteration. In those instances (i.e., for Harvest, timber production ratios, and primary production
ratios) we simulate random draws for three random variables. For each iteration we use the
appropriate random triangular variable draw for the given year set (e.g., 1906 to 1945).
Some random variables should be correlated with one another (e.g., Stockmann et al. 2012).
For instance, there are three parameters for Harvest each to account for three changes in
confidence interval values during the time series. There is a separate random variable for each
of those three sets of years because the intervals are assumed to have been constructed in
similar ways and therefore would have been correlated. Therefore, a correlation among the
three Harvest variables was created. Timber Product Ratios and End Use Product Ratios each
have sets of three correlated variables and pairs of correlated random triangular variables for all
paper/wood discard disposition ratios and decay limits were created.
Pearson’s correlation of 0.5 were used for all correlations. This value was also used by
Anderson et al. (2013) and Stockmann et al. (2012). A Pearson’s correlation matrix was
created, with the number of columns and rows corresponding to the number of random
variables to correlate. For a three-variable correlation the following matrix was used and a
Choleski decomposition was created from it:
=
1 0.5 0.5
0.5 1 0.5
0.5 0.5 1
=
1 0.5 0.5
0 0.866 0.289
0 0 0.816
Linear multiplication of a vector containing the three variables against the Choleski
decomposition matrix alters the second and third variables such that they are correlated to the
first. The correlated triangular random variables were generated by first transforming an
iteration’s three LHS random uniform variables into values from a standard normal distribution
by treating the random uniform variables as probability quantiles from a standard normal
55
distribution. The three standard normal values are then altered and become correlated by
linearly multiplying them by the Choleski decomposition. The values return to those from a
standard uniform distribution by finding the cumulative distribution function values of the three
(now normally distributed and correlated) random variables. These random values are then
ready to be processed as described above into draws from a triangular distribution and remain
correlated.
Sampling and the HWP Program
The goal of the MC simulation was to produce estimates and confidence intervals for overall
values for SWDS, PIU, EWOEC, EEC, and their combinations. Before entering the MC loop, the
program performs all LHS draws, forces correlations among specific variables, and transforms
the random draws into draws from triangular distributions.
A description of how each parameter was altered is provided below.
CCF to MgC Conversion Factors
Within a single iteration, one random triangular variable value is multiplied across the CCF to
MgC conversion factor values for all years.
Harvest
For Harvest three correlated random variables for every MC iteration was used. Each random
variable is used to adjust its corresponding range of years (Table 3a). Subsequent iterations use
a different draw of three correlated random variables. The matrix of random triangular variables,
with years = rows, iterations = columns, is multiplied by a vector of the harvest volume per year.
For each iteration in the MC loop a different vector of these altered values is used.
Timber Product Ratios
This parameter is more complicated than Harvest because the values within a year must sum to
one. It relies on three correlated random variables which are processed as described above. For
each iteration, the code examines all 40 proportion values and determines which one, for a
given year, is the maximum value. It finds the product of the random triangular variable value
and the maximum proportion. If the resulting value is ≥ 1 then the maximum proportion is set to
1.0 and all of the other 39 values are set to 0.0. If it is < 1 then the code calculates the ratio of (1
– new maximum proportion) / (1 – old maximum proportion) and multiplies this proportion
against all of the remaining proportions. All 40 altered values should sum to 1.0.
The result is that for each year there are 40 altered TPR values. When the MTC variable is
calculated for each iteration, the altered values are effectively propagated across all PPR and
EUR categories per year. This MC sum-to-one approach does have some substantial
drawbacks. For instance, Timber Product Ratio 2 is Softwood Sawtimber. It accounts for 77% to
97% of harvest in a given year. Multiplying the triangular distribution value by Timber Product
Ratio 2 frequently results in values that are ≥ 1.0. Therefore, the alteration of this variable
frequently results in the same value, and the distribution of Timber Product Ratio 2 is triangular
until the distribution crosses the 1.0 boundary, at which point the remainder of the distribution is
stacked at 1.0.
56
Primary Product Ratios
Three correlated random variables were used to alter PPR for each MC iteration, as was done
for TPR. The 64 Primary Product Ratios (PPRs) differ from the TPR values in that they sum to
40, instead of 1, for each year in the time series. However, subsets of PPR ratios sum to 1 but
36 of primary products are not part of a subset of a timber product. The PPRs were altered in
the same fashion as TPR values with one difference. The code found, for each PPR set with
more than one value, the maximum proportion and then adjusted that proportion with the
random triangular parameter value. It adjusted the remainder of the values in the PPR set as
described for TPR. If there was only one value in the set, it was always equal to 1.0 and the
code never adjusted it. These 1.0 values allowed the TPR value to be carried through PPR.
Similarly, if the maximum ratio in a set was already 1.0, no alteration occurred. These values
are propagated across all EUR categories per year so that they may be multiplied against other
values in the eu_ratio matrix to create the altered MTC values.
End Use Product Ratios
End Use Product ratios were altered as described for PPRs. One difference is that the EUR
alterations only used one random triangular variable value for all years in an iteration (i.e., not
three correlated variables) because the triangular distribution confidence intervals were held
constant across years. EUR values for a given year summed to 64. The code therefore
examined sets of EUR values for each PPR category.
Product Half Lives
There are 224 half-life values, one per EUR and the same random triangular random variable
value was multiplied by all 224 values per iteration. Each iteration of the MC loop used a
different vector of those 224 values in the decay function that is used to determine PIU over
time.
Decay and Half-Life Limits for Landfills, Dumps, and Recovered Pools
Description of these parameters is combined because they relied on the same process. Each
one has two correlated random triangular variables, for paper and wood. A function inputs the
original 224 ratios for each, where most of the 224 ratios are for wood and a few are for paper.
The function also obtains the two vectors of correlated random variables. Each iteration has a
single random variable value multiplied by the paper ratio and another for the wood ratio. The
altered vectors are then combined into a matrix. When the MC loop runs, each iteration obtains
a differently altered vector of 224 values to enter into the decay function.
Discarded Disposition Ratios for Wood and Paper
These ratios control what fraction of waste ends up as burned, burned with energy capture,
recycled, composted, placed in a landfill, or sent to a dump (the six fates). The values can differ
by year and a function was created that operated on the ratios for wood and paper separately.
The function performed the sum-to-one approach on the six ratios for either wood or paper.
