Back
Methodologies
Technical specifications and calculation frameworks for evaluating avoided emissions through wood product substitution.
Technical specifications and calculation frameworks for evaluating avoided emissions through wood product substitution.
General
Forest Tools
Harvested Wood Product Tools
Version 1.0, updated September 2026
The Landscape Carbon Tool enables users to evaluate forest carbon dynamics and harvesting activity across user-defined geographic areas within the conterminous United States (CONUS). Using data from the USDA Forest Service Forest Inventory and Analysis (FIA) program’s Nationwide Forest Inventory (NFI), the tool provides consistent, population-representative estimates of forest carbon stocks, stock changes, and associated harvest removals.
The Landscape Carbon Tool computes a suite of complementary landscape-level metrics that provide broader context for interpreting forest carbon conditions and trends. These include net annual carbon stock change, carbon change per unit area, carbon change per unit of harvested wood, growth-to-removals ratios, annual harvest levels and harvest intensity, and mortality rates expressed both per unit area and as a percentage of standing carbon stocks. Together, these metrics provide a more complete characterization of forest carbon dynamics, allowing users to assess not only whether carbon stocks are increasing or decreasing, but also how those changes relate to forest growth, mortality, and harvesting activity (Prisley and Sonne Hall, 2024).
In the Landscape Carbon Tool, changes in carbon stocks represent estimated net change in carbon stored within the defined system (i.e., the forest stand) over a specified period, expressed as metric tonnes of carbon dioxide equivalent (t CO₂e). The results are presented using an atmosphere-oriented sign convention whereby negative values indicate a removal of carbon dioxide from the atmosphere (i.e., carbon sequestration or carbon removal). Positive values indicate a release of carbon from the ecosystem to the atmosphere and therefore represent emissions. Carbon stocks, in contrast, are presented as positive quantities because they represent the amount of carbon stored in a given pool at a specified point in time, rather than a transfer of carbon to or from the atmosphere.
To accomplish this, the Landscape Carbon Tool uses pre-processed FIA plot data that include both current and previous measurements for each plot. Carbon stocks are estimated across multiple ecosystem pools, and stock changes are annualized based on plot-specific remeasurement periods. These estimates are then combined with delivered wood volumes to derive associated metrics.
Importantly, the tool is intended to provide landscape-level, population-representative estimates rather than stand-level or project-specific carbon accounting. Results reflect averages across large areas and should be interpreted accordingly, particularly when applied to decision-making or comparisons across regions.
AGLIVE: Aboveground live tree carbon
AGDEAD: Aboveground dead tree carbon
ANN_CHNG_*: Annualized carbon stock change (CHNG divided by REMPER)
Annual Harvest: Carbon stocks in delivered wood volume per year (t CO₂e yr⁻¹)
BGLIVE: Belowground live tree carbon
BGDEAD: Belowground dead tree carbon
CDD: Carbon in down dead wood
CHNG_*: Change in carbon stock between current and previous measurements
CHNG_ACRES: Area represented by a plot for change calculations
CL: Carbon in litter (forest floor organic material)
CO₂e : Carbon dioxide equivalent
CONUS: Conterminous United States
COUNTYCD: County code (FIPS)
CSO: Carbon in soil organic matter
CU: Carbon in understory vegetation
CURR_* Prefix indicating carbon stock estimates for the most recent measurement
CURR_OWNGRPCD: Ownership group classification (e.g., National Forest, private)
d.b.h.: Diameter at Breast Height; tree diameter measurement
DELIVGRNWT: Delivered green weight of harvested wood (short tons); harvested biomass delivered to mills after accounting for logging residues
Evaluation Group (EVAL_GRP): A complete set of FIA plot measurements for a state, typically collected over a 5–10 year period
FIA: Forest Inventory and Analysis Program (program of USDA Forest Service)
FIPS: Federal Information Processing Standard (here, codes for identifying U.S. counties)
GROWC: Annual carbon increment from tree growth
HARVC: Annual carbon removed through harvest
IPCC: Intergovernmental Panel on Climate Change
LAT, LON Approximate latitude and longitude coordinates of plot locations
Metric tonnes (t): Unit of weight equal to 1,000 kilograms (2,204.6 pounds), or approximately 1.102 short tons.
MORTC: Annual carbon loss due to mortality (t CO₂e ac⁻¹ yr⁻¹)
Mortality per Acre: Annual mortality normalized by land area (t CO₂e ac⁻¹ yr⁻¹)
NFI : Nationwide Forest Inventory of the Forest Inventory and Analysis Program (program of USDA Forest Service)
PREV_*: Prefix indicating carbon stock estimates for the previous measurement
REMPER: Remeasurement period (decimal years between successive plot measurements)
Short ton (St): Short ton. .S. customary ton; 2,000 pounds (907.2 kg). Equal to 0.907 metric tonnes.
STATECD: State code
t CO₂e: Metric tonnes of carbon dioxide equivalent
TOTC: Total ecosystem carbon; sum of all carbon pools
UNITCD: FIA survey unit code
U.S.: United States of America
USDA: United States Department of Agriculture
The Landscape Carbon Tool evaluates carbon stocks in the five forest carbon pools defined by the Intergovernmental Panel on Climate Change (IPCC): (1) aboveground biomass, (2) belowground biomass, (3) dead wood, (4) litter, and (5) soil organic carbon. Outputs of the Landscape Carbon Tool break down these pools into the following categories to support transparency around which pools or sub-pools are measured or modeled (see Box 1 for an explanation on how these designations are made):
Box 1: Measured v.s. Modeled Carbon Pool Delineation Pools are designated "measured" where the key variable for determining the carbon estimate is observed on every forested FIA plot (e.g., measurements of species, diameter, and height). Because these estimates derive from direct field measurement, they are generally assumed to carry lower uncertainty than modeled pools, though converting field measurements into carbon estimates still introduces uncertainty in all cases. Pools are designated as "modeled" where values are predicted for all conditions from statistical relationships rather than from observation of the pool itself. Down woody material, litter, and soil are measured on a subset of plots, but the values applied to any given condition are generally predicted rather than observed. Because these values are predicted rather than observed, they are assumed to carry higher uncertainty, and the degree of uncertainty varies with how closely the modeled relationship reflects conditions on a particular site. |
These pools are also included in estimates of changes in forest carbon stocks.
The Landscape Carbon Tool does estimate the carbon within harvested wood (as an emission from the forest pool) but not carbon transfers from the forest stand to the Harvested Wood Product (HWP) pool. Users who wish to do so may use the FACT Harvested Wood Product Carbon Tool which quantifies carbon storage and release in HWPs using user-entered harvested wood amounts.
The Landscape Carbon Tool includes sinks and sources within the Agriculture, Forestry, and Other Land Use (AFOLU) sector, as defined by the Intergovernmental Panel on Climate Change (IPCC), and specifically quantifies changes in forest carbon stocks associated with forest growth and mortality. The Landscape Carbon Tool produces estimates by drawing from a pre-calculated dataset derived from the USDA Forest Inventory and Analysis Program (FIA) Nationwide Forest Inventory (NFI) (herein after referred to as FIA).
These data include both current and previous measurements for each inventory plot, along with pre-computed carbon stocks, stock changes, and harvested wood estimates. Because this dataset was constructed using FIA plots classified as forest land at both measurement periods (i.e., forest remaining forest), estimated carbon stock changes reflect changes occurring within existing forest land and do not include carbon stock changes associated with transitions into or out of forest land, such as afforestation, reforestation following a nonforest land use, or deforestation.
The AFOLU-sector carbon accounting performed by the Landscape Carbon Tool does not constitute a full life-cycle assessment. Emissions associated with activities outside the AFOLU sector, such as fuel use in harvesting and transportation, wood-product manufacturing, construction, or avoided emissions from substitution of wood for more emissions-intensive materials or fuels, are outside the system boundary.
While FIA maintains a nationally consistent network of permanent forest inventory plots across all ownerships, plots are distributed at an approximate intensity of one plot per 6,000 acres. As a result, analyses conducted over small geographic areas may rely on relatively few FIA plots, and sampling error can increase further where forest conditions are highly heterogeneous. In these cases, FIA-based estimates should be interpreted with appropriate consideration of the associated sampling uncertainty. The Landscape Carbon Tool enforces this by implementing a requirement that any geographic area selection must have a minimum of 150 FIA plots.
In addition, the Landscape Carbon Tool’s underlying dataset was constructed using the publicly-available version of the FIA database, the latitude/longitude coordinates of plots are “fuzzed”, or randomly adjusted to protect landowner confidentiality. Therefore when the geographic selection depends on geographic overlay (National Forest, circular area, or shapefile) some plots may be erroneously included or excluded due to the fuzzed location. Nevertheless, this is likely to have a very minimal effect on results, especially for larger spatial domains.
Sources of uncertainty for the Landscape Carbon Tool are largely attributable to size and characteristics of the user-specified geographic area (i.e., size and degree of forest heterogeneity) as well the equations, assumptions, and data used to convert FIA field measurements into carbon stock estimates.
In general, measurement-derived carbon estimates carry lower uncertainty than those derived from predictive models. Aboveground live and standing dead biomass estimates are derived from measured diameter at breast height (d.b.h.) and height via allometric equations, and belowground live biomass as a function of those. In contrast, down dead wood, litter, and soil carbon are populated for most conditions by models predicted from live tree carbon density, stand age, forest type, and geographic area.
The Landscape Carbon Tool supports user interpretation of this relative uncertainty (i.e., measured vs. modeled) by labeling carbon pool outputs accordingly and allowing users to select which carbon pools to include.
The Landscape Carbon Tool relies on user-defined geographic selection and filtering options to subset pre-processed forest inventory data and generate landscape-scale estimates of forest carbon dynamics and harvesting activity.
Unlike tools that require user-supplied inventory data, the Landscape Carbon Tool uses a pre-calculated dataset derived from FIA. These data include both current and previous measurements for each inventory plot, along with pre-computed carbon stocks, stock changes, and harvested wood estimates. As a result, user inputs are limited to defining the geographic domain and selecting from a set of optional filters that determine how results are summarized.
Users first define an area of interest using predefined administrative boundaries (e.g., states, counties, or regions) or custom spatial inputs (shapefiles or user-defined coordinates). The tool then subsets FIA plots that fall within the selected domain and aggregates plot-level estimates to produce population-representative summaries of carbon stocks, carbon stock change, and harvesting activity.
In addition to geographic selection, users may refine their analysis by selecting ownership categories and specifying whether to include only measured carbon pools or all available carbon pools (measured and modeled). These selections influence both the scope of the data included in the analysis and the interpretation of results, particularly with respect to uncertainty.
Once inputs are defined, the tool processes the selected data to compute a suite of landscape-level metrics, including total carbon stocks, annualized carbon stock change, harvested wood volumes, growth/drain ratio, mortality, and the net carbon change per harvested unit.. Because the tool is based on FIA sampling data, all outputs represent statistically derived estimates for the selected population of forest land rather than exact measurements for specific locations.
To use the Landscape Carbon Tool, the user must first define the geographic area for which carbon stocks, stock changes, harvest estimates, and associated landscape metrics will be summarized. The forest land within this selected area becomes the analysis domain for the tool.
The Landscape Carbon Tool uses a pre-calculated dataset derived from FIA plots across the conterminous United States (CONUS). Forest carbon estimates for the Landscape Carbon Tool, as well as other tools in FACT, are drawn from FIA plots that match the FIA’s “unreserved forest” definition, which represents forest land not withdrawn from management for production of wood products through statute or administrative designation.
Once a user specifies an area of interest, the tool subsets the available FIA plot records to include only those plots that fall within the selected domain. Current and previous plot measurements associated with that domain are then used to compute the reported metrics. Depending on the selections, the tool identifies plots either by matching geographic codes (such as state or county identifiers) or by performing a spatial intersection between the user-defined area and FIA plot locations.
Because FIA plot data underpin all Landscape Carbon Tool analyses, the tool enforces a minimum plot threshold to ensure sufficient sample size and reduce sampling uncertainty in the results. It therefore requires that any selected analysis domain include a minimum of 150 FIA plots. Selections below this threshold are flagged as insufficient for reliable results, and users are prompted to expand the analysis area.
The tool provides the following options for defining the area of interest:
Figure 1: Map of broad CONUS regions applied in FACT Forest tools. Adapted from USDA Entity Guidelines, Quantifying greenhouse gas fluxes in agriculture and forestry: Methods for entity-scale inventory, Technical Bulletin 1939, 2nd ed., Chapter 5, Figure 5-4 (2024).
Note that because the tool uses the publicly-available version of the FIA database, the latitude/longitude coordinates of plots are “fuzzed”, or randomly adjusted to protect landowner confidentiality. Therefore when the geographic selection depends on geographic overlay (National Forest, circular area, or shapefile) some plots may be erroneously included or excluded due to the fuzzed location. This is likely to have a very minimal effect on results, especially for larger spatial domains.
After the geographic area is defined, the user initiates data retrieval within the tool. The tool then subsets the plot-level dataset for the specified domain and prepares the records needed for summary calculations.
Because all subsequent calculations are based on the selected geographic domain, the area selection step is one of the most important user inputs in the Landscape Carbon Tool. Smaller or highly customized areas may better reflect a user’s area of interest, but they may also reduce the number of FIA plots available for estimation. Larger areas generally provide more stable estimates because they include more plots and therefore more information from the underlying FIA sample.
In addition to defining a geographic area of interest, users may optionally filter results by ownership group. Ownership classifications are based on FIA ownership groups and represent broad groupings of forest land ownership across the United States. The available ownership categories include:
These categories correspond to the ownership group codes included in the underlying dataset and are applied to the current plot measurement.
When an ownership filter is applied, the tool subsets the selected geographic domain to include only FIA plots that currently match the specified ownership category or categories. All subsequent calculations, including carbon stocks, stock changes, harvest estimates, and derived landscape metrics, are then computed using only the filtered subset of plots. If no ownership selection is made, all ownership groups are included by default. This results in estimates that represent the full population of unreserved forest land within the selected geographic area.
Ownership selection can have a meaningful influence on results, as forest structure, management intensity, and harvesting patterns often differ across ownership types. For example, privately owned forests may exhibit different growth, harvest, and mortality dynamics compared to federally managed lands. As a result, filtering by ownership allows users to isolate and evaluate carbon dynamics within specific management or ownership contexts.
Because ownership filtering reduces the number of FIA plots included in the analysis, it may also affect the statistical robustness of estimates, particularly when combined with smaller geographic selections. Users should consider the tradeoff between specificity and sample size when applying ownership filters. Where ownership filtering results in fewer than 150 representative FIA plots, users will be directed to adjust selections to meet the minimum threshold.
The Landscape Carbon Tool allows users to control which forest carbon pools are included in the graphical presentation. This selection determines how total carbon stocks and carbon stock change are calculated and interpreted for some of the outputs. The metrics at the top of the results are computed from the aboveground live tree pool only. For the graphs and tabular outputs, users may choose to include only measured carbon pools or to include all carbon pools. As outlined in Box 1, the measured carbon pool option includes those components of forest carbon that are either directly measured by FIA or estimated using well-established allometric relationships applied to measured tree attributes. These pools consist of aboveground live tree carbon, belowground live tree carbon, aboveground dead tree carbon, and belowground dead tree carbon. Because these estimates are closely tied to field measurements, are based on factors for individual tree species, and developed using widely accepted estimation approaches, they are generally considered to have lower uncertainty.
The all carbon pools option includes the above-mentioned measured pools as well as additional ecosystem components that are not directly measured at the same level of detail in FIA and are therefore estimated using modeling approaches. These additional pools include carbon in down dead wood, litter, understory vegetation, and soil organic matter. Inclusion of these pools provides a more complete accounting of total ecosystem carbon by capturing carbon stored outside of tree biomass, but these estimates may carry greater uncertainty due to the assumptions and models used in their derivation.
The selected carbon pool option directly affects graphic and tabular outputs. Total carbon stocks are calculated as the sum of the included pools, and carbon stock change is computed using those same pools across current and previous measurements. Because derived metrics such as net carbon change is based on these totals, their values will differ depending on whether measured-only or all pools are included.
Users should consider the intended application of results when selecting carbon pools. Analyses focused on comparability, reporting consistency, or lower-uncertainty estimates may benefit from using measured carbon pools only. In contrast, analyses seeking a more complete representation of ecosystem carbon should include all carbon pools, recognizing the additional uncertainty associated with modeled components.
All outputs are derived from aggregated FIA plot-level data and represent population-level estimates for the selected geographic area. Carbon stocks are summarized for both the current and previous measurement periods, and carbon stock change is annualized using the remeasurement period associated with each plot.
Results are presented through a combination of summary metric panels, tabular outputs, and visualizations. Together, these outputs provide a comprehensive view of forest carbon dynamics and the relationship between carbon change and forest product removals.
The Landscape Carbon Tool results page first presents a geographic representation of the selected analysis area, along with a summary of the underlying data used in the calculations (see Figure 2) . The map displays the user-defined area of interest and the spatial distribution of FIA plots included in the analysis. Plot locations are shown within the selected boundary, providing a visual indication of the sample used to generate estimates. The density and distribution of plots can vary depending on the size and location of the selected area.
Accompanying the map is a summary description of the analysis domain. This includes the number of FIA plots used in the calculations and the total acreage of unreserved forest land represented by those plots. These values provide important context for interpreting results, as they reflect the sample size and spatial extent underlying all reported metrics.
Ownership information may also be displayed alongside the map, showing the distribution of plots across ownership categories within the selected area. This helps users understand how different ownership groups contribute to the overall analysis.
Because all subsequent outputs are derived from the subset of plots shown in the map, this section provides a critical link between user inputs and reported results. Users should consider both the number of plots and their spatial distribution when evaluating the reliability and representativeness of the outputs.
Figure 2: Landscape Carbon Tool geographic output summary
Upon processing of the user-supplied data, several summary metrics and graphical outputs are rendered, as detailed below. Summary metrics are derived by aggregating plot-level values across the selected geographic domain and are intended to provide interpretable indicators of forest condition and carbon dynamics (Prisley and Sonne Hall 2024). Graphical and tabular outputs allow users to explore carbon stocks and carbon dynamics in greater detail.
Additional summary statistics are displayed at the top of the user interface, including the total number of FIA plots, the area, and the time range across which forest carbon change estimates have been calculated.
Summary metrics:
Graphical Outputs:
Figure 3: Carbon by pool and ownership stacked bar chart
Figure 4: Carbon by pool bar chart
Figure 5: Carbon by pool table
Together, these graphical and tabular outputs complement the summary metrics by providing additional detail on the composition and distribution of forest carbon. They allow users to move beyond aggregate values and better understand the underlying structure of carbon stocks and changes across pools and ownerships.
The Landscape Carbon Tool is built on a pre-processed dataset derived from FIA which provides a nationally consistent, plot-based inventory of forest conditions, including tree measurements, land use, ownership, and associated carbon estimates. These data are collected on a rotating basis across the United States, with individual plots remeasured at intervals that typically range from approximately five to ten years.
Rather than querying the FIA database directly, the Landscape Carbon Tool relies on a single, integrated dataset (to be updated at least annually) that combines current and previous plot measurements, carbon pool estimates, and forest change variables. Each record in this dataset represents a plot/condition combination, where a condition corresponds to a distinct forested portion of a plot. Because a single FIA plot may contain multiple forest conditions, multiple records may share the same plot identifier but differ in condition identifier and proportional area. The proportion of each plot represented by a given condition is explicitly tracked and used in all subsequent aggregation steps.
The dataset includes only unreserved forest land, meaning forest areas that are not legally restricted from timber harvest. This ensures that estimates of carbon change and harvested wood are aligned with actively managed forest systems and avoids inclusion of lands where harvest is not permitted.
The workflow underlying the dataset can be summarized as:
All outputs in the tool are derived from these plot-level quantities.
The estimation approach for Landscape Carbon Tool outputs is summarized in Figure 6 and described below.
Figure 6: Conceptual Workflow for the Landscape Accounting Factor Tool from FIA data inputs to landscape-level carbon metrics.
FIA measures the proportion of each unique condition on each plot. Conditions are formed due to differences in variables such as forest type or ownership class. A plot's contribution to the population total estimate of area variables (such as area of forest change) for a given domain is calculated by summing the condition proportions represented by that domain to the plot level , expanding the plot's domain estimate to the population level using a population expansion factor, and adjusting for partial nonresponse. Further details about FIA data and estimation procedures are available in the FIADB User Guide (Burrill, et al. 2024) and the FIADB population estimation user guide (Pugh, et al. 2018). Hereafter, we refer to the amount of the domain of interest on a condition as the plot-condition amount. The contribution of plot-condition j on plot i to the population estimate of the total area changed, for example, is therefore calculated as:
Where:
C = the total number of unique conditions c=1..C on the plot,
EXPCHNGi = the plot-level population expansion factor for change estimates associated with plot i,
CONDPROPcij = the proportion of the plot represented by condition c on plot i containing plot-condition j, and
ADJi = the partial plot correction factor associated with that plot.
Carbon stocks and carbon stock change in the Landscape Carbon Tool are calculated at the plot-condition level using a combination of FIA tree-level and condition-level data. The workflow follows a structured sequence: (1) estimation of carbon stocks for current and previous measurements, (2) expansion of those values to represent landscape area, and (3) calculation and annualization of carbon stock change.
Carbon stock estimation by pool
Carbon is estimated for eight ecosystem pools. These are derived using two distinct approaches depending on how the data are represented in FIA. For tree-based pools (aboveground and belowground live and dead carbon), values are calculated from individual tree records. FIA provides tree-level carbon estimates, which are expanded to the plot-condition level using trees-per-acre values. For a given type of tree (e.g., trees of a given species) on a given plot-condition j on plot i, the contribution to the population total estimate of carbon is calculated as:
Where:
CARBONAG,k = aboveground carbon for tree k
TPAUNADJ,k = trees per acre represented by that tree
FACTORk = partial plot correction factor
EXPVOLj = population area expansion factor for estimates of volume
Division by 2000 converts pounds to short tons
Equivalent calculation principles are used for belowground live (CURR_BGLIVEij) aboveground dead (CURR_AGDEADij), and belowground dead (CURR_BGDEADij) pools.
For non-tree pools (down dead wood, litter, understory vegetation, and soil organic carbon), FIA provides carbon values on a per-acre basis at the condition level. These values are expanded to totals using the area represented by the plot-condition:
The variable CHNG_ACRESj represents the area associated with each plot-condition in the change estimation framework as described above.
Total carbon stocks and carbon stock change
Total carbon for each plot-condition is calculated as the sum of all eight pools. For the current and previous measurements:
Where:
Ct,j = total ecosystem carbon (t CO₂e) for plot-condition j at time t (current or previous)
CDDt,j = carbon in down dead wood (t CO₂e per acre)
CLt,j = carbon in litter (forest floor) (t CO₂e per acre)
CSOt,j = carbon in soil organic matter (t CO₂e per acre)
CUt,j = carbon in understory vegetation (t CO₂e per acre)
CHNG_ACRESj = area represented by plot-condition j (acres)
AGLIVEt,j = aboveground live tree carbon (t CO₂e)
BGLIVEt,j = belowground live tree carbon (t CO₂e)
AGDEADt,j = aboveground dead tree carbon (t CO₂e)
BGLIVEt,j = belowground dead tree carbon (t CO₂e)
t = time period (current or previous)
j indexes the plot-condition
Carbon stock change is calculated directly from the difference between current and previous total carbon. This calculation is performed at the plot-condition level and ensures that both time periods are aligned for the same area. Only plot-conditions classified as forest in both measurements are included, meaning that estimates reflect changes within forest land rather than land-use transitions. Because FIA plots are remeasured at different time intervals, carbon stock change must be standardized to an annual basis to allow for consistent comparison across plots and regions. This is done by dividing the change in carbon for each plot-condition by the length of time between its current and previous measurements (REMPER). The resulting value represents the average annual rate of carbon change for that plot-condition. Standardizing change in this way ensures that estimates are comparable across regions with different inventory cycles and remeasurement periods.
While carbon stock change captures the net effect of all processes influencing forest carbon dynamics, it does not distinguish among the underlying drivers of change. To provide additional context, the dataset also includes estimates of growth, mortality, and harvest derived from FIA growth-removals-mortality (GRM) data. These variables allow carbon stock changes to be interpreted in terms of biological accumulation, natural losses, and removals associated with wood utilization, and form the basis for several landscape metrics reported by the Landscape Carbon Tool.
In addition to carbon stocks and carbon stock change, the Landscape Carbon Tool’s dataset includes estimates of growth, mortality, harvest removals, and delivered wood. These variables describe the processes driving changes in forest carbon over time and are derived from FIA growth, removals, and mortality (GRM) data. All values are calculated at the plot-condition level and expressed on an annual basis.
Growth, mortality, and harvest represent distinct components of forest carbon dynamics. Growth (more precisely, net growth; equal to gross growth minus mortality) reflects the net accumulation of carbon in live trees between measurements, including both survivor growth and ingrowth. Mortality represents carbon transferred from live trees to dead pools due to natural or anthropogenic causes. Harvest removals represent carbon removed from the forest through management activities. These quantities are estimated using FIA GRM components, which track changes in tree biomass between successive measurements. For each plot-condition, these components are calculated using tree-level information and then aggregated to the plot-condition level, consistent with the carbon stock calculations described in Section 7.2.
Harvest removals are further processed to estimate delivered wood, which represents the portion of harvested biomass that is transported to mills. This calculation involves converting harvested dry biomass to green weight using species-specific conversion factors and then adjusting for logging residues using region-specific residue factors. The resulting variable represents delivered wood in green tons per year at the plot-condition level. Delivered wood is used in the calculation of the net change per unit of harvest, linking forest carbon change to wood utilization. This factor, carbon stock change per short ton of delivered wood, can be used to allocate the net change across the numerous entities that may be responsible for a share of the total harvest. For example, if Mill X purchased 1M tons of wood annually, and the total annual harvest is 5M tons, then their harvest activity accounts for 20% of the total, and their “share” of the total net change would be 20%.
These variables provide the basis for several key metrics reported by the Landscape Carbon Tool. Growth and harvest are used to calculate the growth-to-removals ratio, which indicates whether forests are accumulating or losing biomass relative to removals. Mortality is used to quantify carbon losses due to primarily natural processes and is reported both as a per-acre value and relative to carbon stock. Because all variables are calculated at the plot-condition level and expressed on an annual basis, they can be aggregated consistently across user-defined geographic areas.
All variables described in Sections 7.1 through 7.3 are calculated at the plot-condition level and represent expanded, population-level estimates. When a user defines a geographic area, the Landscape Carbon Tool Tool identifies all plot-conditions within that area and aggregates their values to produce landscape-level estimates.
Carbon stocks, carbon stock change, growth, mortality, harvest, and delivered wood are calculated as the sum of plot-condition values within the selected region. Because carbon stock change and GRM variables are expressed on an annual basis, aggregated results represent average annual rates of change. Per-acre metrics are derived by dividing aggregated totals by the total area represented, and ratio-based metrics are calculated using the corresponding aggregated components (e.g., growth divided by harvest).
This aggregation approach ensures that all reported outputs are consistent with FIA’s design-based sampling framework while allowing users to evaluate carbon dynamics across flexible, user-defined geographic areas.
Burrill, Elizabeth A.; DiTommaso, Andrea M.; Turner, Jeffery A.; Pugh, Scott A.; Christensen, Glenn; Kralicek, Karin M.; Perry, Carol J.; Lepine, Lucie C.; Walker, David M.; Conkling, Barbara L. 2024. The Forest Inventory and Analysis Database, FIADB user guides, volume: database description (version 9.3), nationwide forest inventory (NFI). U.S. Department of Agriculture, Forest Service. 1026 p. [Online]. Available at web address: https://research.fs.usda.gov/understory/forest-inventory-and-analysis-database-user-guide-nfi
Murray, L.T., C. Woodall, A. Lister, K. Stockmann, H. Gu, S. Urbanski, K. Riley, E. Greenfield, et al. 2024. Chapter 5: Quantifying greenhouse gas sources and sinks in managed forest systems. In Hanson, W.L., C. Itle, K. Edquist. (eds.). Quantifying greenhouse gas fluxes in agriculture and forestry: Methods for entity‐scale inventory. Technical Bulletin Number 1939, 2nd edition. Washington, DC: U.S. Department of Agriculture, Office of the Chief Economist.
Prisley, S. and E. Sonne Hall. 2024. Calculating a Land Carbon Accounting Factor in the United States: an Example and Implications. Journal of Forestry, 2024, 122, 1–12 https://doi.org/10.1093/jofore/fvad037
Pugh, Scott A.; Turner, Jeffery A.; Burrill, Elizabeth A.; David, Winnie. 2018. The Forest Inventory and Analysis Database: population estimation user guide (Edition: November, 2018). U.S. Department of Agriculture, Forest Service. 166 p. [Online]. Available at web address: http://www.fia.fs.fed.us/library/database-documentation/.
Terms specific to this methodology are defined below. See the main FACT Glossary for general terms and for the sources from which definitions are adapted.
Aboveground dead tree carbon pool (Live AG): A component of the deadwood pool that represents carbon in the aboveground portions of standing dead trees at least 5 inches d.b.h. including stems and branches for both timber and woodland species. Standing dead trees are those that have not fallen and do not lean more than 45 degrees from vertical; material beyond that threshold is assigned to the downed dead wood carbon pool. (Adapted from FIA EVALIDator and FIADB Database Description)
Aboveground live tree carbon pool (Dead AG): A component of the aboveground live biomass pool that represents carbon in the aboveground portion of living trees, including stems and branches. Estimates are derived from tree measurements using established allometric relationships and conversion models. (Adapted from FIA EVALIDator and FIADB Database Description)
Administrative unit: A user-defined analysis boundary based on political or jurisdictional divisions (e.g., states or counties) used to group FIA plots.
Annualized Carbon Stock Change: The average annual net change in the mass of carbon within a defined carbon pool over a specified time period. It is calculated by dividing the total carbon stock change by the number of years across that period and is expressed in metric tons of carbon dioxide equivalents per year (t CO2e/yr-1).
Belowground dead tree carbon pool (Dead BG): A component of the deadwood carbon pool that represents carbon in the coarse roots of standing dead trees at least 5 inches d.b.h. for both timber and woodland species. Estimates are derived from aboveground tree measurements through allometric models, with adjustments for decay and structural loss. (Adapted from FIA EVALIDator and FIADB Database Description)
Belowground live tree carbon pool (Live BG): A component of the belowground live biomass pool that represents carbon in the coarse roots of living trees. Estimates are derived from tree measurements using established allometric relationships and conversion models applied by FIA. (Adapted from FIA EVALIDator and FIADB Database Description)
Biomass: The mass of organic material in trees, including stems, bark and branches, typically expressed as oven-dry or green weight. Biomass may refer to live or dead components and can be converted to carbon using standard conversion factors.
Carbon (C): The mass of carbon contained within a defined pool (e.g., biomass, dead organic matter or harvested wood products), typically expressed in units of mass (e.g., metric tons of carbon, t C) and derived from biomass using established conversion factors.
Carbon Density: The amount of carbon stored per unit of land area, typically expressed as mass per area.
CO2e (Carbon dioxide equivalents): A common unit that allows the climate effects of different greenhouse gases to be compared and summed. The quantity of a greenhouse gas is multiplied by its global warming potential (GWP), which expresses its warming effect timber relative to carbon dioxide over a specified time horizon, conventionally 100 years. The result is reported as the mass of CO₂e that would produce an equivalent climate effect. (Adapted from ISO 14067)
Carbon pools: Discrete, mutually exclusive reservoirs in which carbon is stored, defined so that stocks can be summed without double counting and transfers between pools can be tracked. FACT follows the five ecosystem pools used in IPCC greenhouse gas reporting and implemented by FIA: aboveground live, belowground live, dead wood, litter and soil organic carbon. In the Landscape Tool, some pools are divided into subpools.
Carbon stock: The total mass of carbon contained within a defined pool, expressed as t CO₂e in FACT. Stocks are always reported as positive values because they represent carbon held in a pool, not movement of carbon to or from the atmosphere.
Change in carbon stocks: The estimated net change in carbon stored within the defined system over a specified period, expressed as t CO₂e. Values are reported from the atmosphere's perspective: negative when carbon moves from the atmosphere into forest or harvested wood product pools (sequestration), positive when it moves from those pools into other carbon pools or the atmosphere (emission).
CONUS (Conterminous United States): The 48 states within the contiguous United States, excluding Alaska and Hawaii.
Diameter at breast height (d.b.h.): The diameter of a tree stem, outside bark, measured at 4.5 feet above the ground on the uphill side of the tree. For multi-stemmed woodland species, diameter is instead measured at the root collar (d.r.c.). Diameter at breast height is the primary field measurement from which tree volume, biomass, and carbon are estimated.
Downed dead carbon pool (CDD): A component of the deadwood pool that represents carbon in all nonliving woody biomass at least 3 inches (7.6 cm) in diameter at the point of transect intersection, lying on the ground. Sometimes referred to as down woody material (DWM) or coarse woody debris. (Adapted from FIA EVALIDator and FIADB Database Description)
EVALIDator: An FIA web application (https://apps.fs.usda.gov/fiadb-api) that allows users to generate population-level estimates and associated sampling errors from the Forest Inventory and Analysis Database (FIADB). Users can select geographic areas, forest attributes, ownerships, forest types and other parameters to estimate metrics such as forest area, tree volume, biomass, growth, removals and mortality.
Forest Inventory and Analysis Program (FIA): A national program of the U.S. Forest Service that conducts continuous, standardized inventories of forest resources across the United States and provides authoritative information on forest extent, composition, structure, growth, removals, and mortality. FIA conducts a variety of inventories, surveys, and reporting activities, including the Nationwide Forest Inventory (NFI), National Resource Use Monitoring (NRUM), National Woodland Owner Survey (NWOS), and Urban Forest Inventory and Analysis (Urban FIA).
Forest Inventory and Analysis Database (FIADB): The publicly available database maintained by the U.S. Forest Service that stores and distributes data collected through FIA, including plot- and tree-level measurements, population estimates and derived forest metrics.
Federal Information Processing Standards (FIPS) code: A standardized numeric code used to identify geographic areas such as states and counties.
Forest type group: A classification used by FIA that aggregates individual forest types into broader categories based on dominant tree species and ecological similarity, enabling consistent summarization and reporting of forest attributes across regions. See Forest Type Group metadata: https://data.fs.usda.gov/geodata/rastergateway/forest_type. (Adapted from FIA Glossary: Standard Terminology)
Greenhouse gases (GHGs): A gaseous constituent of the atmosphere, both natural and anthropogenic, that absorbs and emits radiation at specific wavelengths within the spectrum of infrared radiation emitted by the Earth's surface, the atmosphere, and clouds, thereby trapping heat in the atmosphere. The greenhouse gases relevant to forest management are carbon dioxide (CO₂), methane (CH₄) and nitrous oxide (N₂O). (Adapted from ISO 14050:2020, 3.9.1)
Growing-stock tree: All live trees of commercial species 5.0 inches d.b.h. and larger that meet minimum merchantability standards. In general, these trees have at least one solid 8-foot section, are reasonably free of form defects on the merchantable bole, and at least 34 percent or more of the volume is merchantable. Excludes rough and rotten cull trees.
Growing-stock tree: All live trees of commercial species 5.0 inches d.b.h. and larger that meet minimum merchantability standards. In general, these trees have at least one solid 8-foot section, are reasonably free of form defects on the merchantable bole and at least 34 percent or more of the volume is merchantable. Excludes rough and rotten cull trees. (Adapted from the FIA Glossary: Standard Terminology)
Growing stock removals/harvest removals: The growing stock volume removed from poletimber and sawtimber trees. Includes volume removed for roundwood products, logging residues and other removals. (Adapted from FIA Glossary: Standard Terminology)
Harvest: A timber harvest is the process of cutting and removing trees from a forest for commercial, ecological or management purposes. Timber harvesting is a central activity in forestry that goes beyond simply cutting trees to transport them for use as roundwood.
Litter carbon pool: Carbon in organic material on the forest floor above the mineral soil, corresponding to IPCC forest carbon pool 4. Comprises the litter, fulvic, and humic layers, together with fine woody debris less than 3 inches in diameter. (Adapted from FIA EVALIDator and FIADB Database Description)
Merchantability specification: The set of criteria determining which trees, and which portion of a tree, are considered usable for wood products, including minimum diameter at breast height, top diameter, minimum sawlog length and limits on form defect and cull. FIA applies regionally specific merchantability standards. (Adapted from FIA Glossary: Standard Terminology)
Merchantable volume: The volume of sound wood in the portion of a tree meeting merchantability criteria, measured on the central stem from a 1-foot stump to a specified top diameter, and expressed in cubic units. (Adapted from FIA Glossary: Standard Terminology)
Mortality: Mortality in the Landscape Carbon Tool represents the proportion of aboveground live carbon transferred to the dead wood pool each year, expressed as a percentage of the aboveground carbon pool from the previous inventory. Forest area is grouped into three mortality categories: healthy, the 70% of area with the lowest mortality; stressed, the next 20%; and disturbed, the 10% with the highest mortality.
Nationwide Forest Inventory (NFI): The ongoing inventory of U.S. forest resources conducted by FIA. The inventory provides consistent estimates of forest area, condition, composition, growth, removals and mortality across ownerships and forest lands in the United States. It is based on a network of permanent field plots distributed at approximately one plot per 6,000 acres, with additional measurements of attributes such as down woody material, soils and understory vegetation collected on a subset of plots. FIA measurements, together with models, are also used by the Forest Service to estimate forest carbon stocks and changes over time and form the primary data foundation for U.S. forest carbon reporting, including the national greenhouse gas inventory.
Ownership group (FIA Owner Group): A broad classification defined by FIA of forest land by the type of entity holding title or managing federal lands. FIA recognizes four groups:
(Adapted from FIA Glossary: Standard Terminology)
Product class: A category used to classify roundwood according to its intended product or processing pathway. Timber product categories often distinguish wood type and intended product, such as softwood sawtimber, hardwood pulpwood, posts, poles, or fuelwood.
Region: A spatial unit defined by administrative or analytical boundaries that groups states, or portions of states, for the purpose of summarizing and comparing forest attributes. In FACT, regions may be delineated in more than one way depending on how a given dataset is organized, including U.S. Forest Service administrative regions or aggregations of states. In general, regions reflect the finest spatial resolution at which the underlying data in FACT can support reliable estimates. See Figure 1 for the regions as defined in the Landscape Carbon Tool.
Removals: The term “removals” is associated with two separate concepts.
In FACT, harvesting removals (cutting trees and transporting them) are referred to as “harvest removals.”
In the Landscape Carbon Tool, the term “removals” is used primarily in the context of harvest removals. Also see growing stock removals/harvest removals glossary term.
Sequestration: The process by which greenhouse gases are taken up from the atmosphere and stored in a carbon pool. In forests this occurs primarily through photosynthesis, which converts atmospheric CO₂ into carbon stored in plant biomass. Used interchangeably with removals in greenhouse gas accounting.
Soil carbon pool (SOC): Carbon in fine organic material below the soil surface to a depth of 1 meter (100 centimeters), corresponding to IPCC forest carbon pool 5. Excludes roots, which are assigned to the live and dead tree pools. (Adapted from FIA EVALIDator and FIADB Database Description)
Species group: A classification used by FIA that aggregates individual tree species into broader groups based on taxonomic and ecological similarity for consistent summarization and analysis of forest attributes. (Adapted from FIA Glossary: Standard Terminology)
Stand: A community of trees that can be distinguished from adjacent communities due to similarities and uniformity in tree and site characteristics, such as age-class distribution, species composition, spatial arrangement, and structure.
Stand key: A unique identifier used to distinguish and track a specific forest stand or condition within a dataset, enabling linkage of plot-level observations and attributes across measurements.
Survey unit: A sub-state geographic division used by the USDA Forest Service to organize and report forest inventory data, typically grouping counties with similar forest characteristics for sampling and estimation purposes.
Uncertainty: Lack of knowledge of the true value of a variable that can be described as a probability density function characterizing the range and likelihood of possible values. Uncertainty depends on the analyst’s state of knowledge, which in turn depends on the quality and quantity of applicable data as well as knowledge of underlying processes and inference methods.
Understory carbon pool: A component of the aboveground live biomass carbon pool that represents carbon in non-tree vegetation beneath the forest canopy, including woody shrubs, tree seedlings below 1 inch (2.5 cm) d.b.h. and herbaceous plants. (Adapted from FIA EVALIDator and FIADB Database Description)
Unreserved forest land: Forest land not withdrawn from management by statute or administrative designation, and therefore generally available for multiple uses, including timber production. (Adapted from FIA Glossary: Standard Terminology)