USCP – Appendix J: Forest Land and Trees

Primary authors: Richard Birdsey, Nancy Harris, Donna Lee and Stephen Ogle

Contents

L.0 Introduction 3

L.0.1 Land-based GHGs and current coverage in this Appendix 3

L.0.2 Land-based GHGs and interactions with other sectors 4

L.1 Defining the Inventory Analysis Period 6

L.2 Representation of Land 7

L.2.1 Delineating the land base 7

L.2.2 Land use classification 8

L.2.3 Categories for reporting GHGs from land use 11

L.3 Estimating GHGs from Land Use (General Methods) 13

L.3.1 Carbon Pools 13

L.3.2 Non-CO₂ Greenhouse Gas Emissions 14

L.3.3 Calculating GHG emissions and removals from land use 15

L.4 Forest Land 17

L.5 Trees Outside Forest Land 31

Annex: General Methods for Estimating GHGs from Land Use 37

References 43

Figure 1: The forest sector in relation to land use, wood products, and energy 4

Figure 2. Schematic of approach to quantify HWP storage and emissions. 28

Table 1: Scope 1 to 3 in relation to land use 4

Table 2: Crosswalk among USCP, GPC, IPCC and the US GHG Inventory 5

Table 3: Land use change matrix 11

Table 4: Geospatial data sets available to estimate land cover and change for U.S. communities 12

Table 5: Relevant categories within the land use conversion matrix for reporting on forest land 18

Table 6: Data sources available for estimating emission and removal factors in U.S. forests. 19

Table 7: Percentage losses of C with forest land conversion to other land uses. 21

Table 8: Land use and change matrix 31

Box 1. Topics not covered in the Appendix 3

Box 2. Using the National Land Cover Database 6

Box 3. Managed vs. Unmanaged land 7

Box 4. Land Use Definitions in the US GHG Inventory 8

Box 5. Land use versus land cover 9

Box 6. Units of measurement and related terminology 15

Box 7. Developing data useful for Climate Action 18

Box 8.. Estimating carbon in harvested wood products. 29

Equation 1: General equation for net GHG flux from land use (simplified gain-loss method) 15

Equation 2: Change in C stocks for forest land remaining forest land 22

Equation 3: Change in C stocks for forest land converted to non-forest land 23

Equation 4: Change in C stocks for non-forest land converted to forest land 25

Equation 5: Total net GHG flux from forests, annualized 30

Equation 6: Change in C stock change for trees outside forests 33

Equation 7: Total net GHG flux from trees outside forests, annualized 35

Sample calculation 1. Change in carbon stocks for forest land remaining forest land. 22

Sample calculation 2. Change in carbon stocks over one year for forest land converted to other land. 24

Sample calculation 3. Change in carbon stocks for land converted to forest land. 26

Sample calculation 4. Change in carbon stocks for trees outside forests. 33

L.0 Introduction

L.0.1 Land-based GHGs and current coverage in this Appendix

Greenhouse gases (GHGs) exchanged between the atmosphere and land are the result of a range of processes and activities—from forest conservation to agricultural production. This Appendix currently1 covers a portion of emissions and removals from land in the U.S. by providing guidance only for estimating GHG emissions and/or removals from forest land and “trees outside forest”. For forest land, all categories are covered—forest converted to non-forest2, non-forest converted to forest, and forests remaining forests. Trees may also be present in other types of non-forested land in the community (cropland, settlement, grassland, wetland, other land); including the tree component of these lands is a preliminary step towards covering non-forest land categories more comprehensively in the future.

Box 1. Topics not covered in the Appendix

Consumption-based or demand-side emissions: Emissions and removals from lands can result from the consumption of and demand for commodities. For example, a recent study estimates that four commodities, beef, palm oil, soy and timber/pulp/paper, accounted for at least 40% of deforestation3 (mostly occurring in the tropics). The embodied emissions from consumption of such products—or from use of bioenergy—depends to a great deal on where, and how much of these commodities are produced.

Substitution effects of harvested wood products: Section FT L.4 covers methods to estimate the GHG emissions and removals from harvested wood products. What it does not include are the substitution effects when wood is used in place of, e.g. fossil fuels for energy use or cement and steel for buildings. Demand for, and use of, long-lived wood products, in particular, can have a positive climate benefit by reducing emissions associated with producing structural materials, and by providing economic incentives for reforestation that otherwise would not occur.

There are reasons why communities may want to take actions related to forests and trees that go beyond the carbon storage and sequestration they provide. These include (but are not limited to):

• Reducing the heat island effect: Increasing urban canopy can reduce the urban heat island effect by increasing shade and through evapotranspiration. Trees can decrease demand for energy and the costs of air conditioning (and reduce heat-related illness), resulting in lower air pollution and GHG emissions.

• Watershed management: Forests and trees help to protect important watersheds that are critical for, e.g. clean drinking water. They also intercept and store rainwater, reduce runoff volume and delay peak flows. Trees also help to remove pollution and sediment from stormwater.

• Improved human health: An extensive and healthy urban forest can significantly improve air quality by reducing the formation of ozone, or smog. Trees can also remove particulate matter from the air, and other pollutants such as carbon monoxide, sulfur dioxide, and nitrogen dioxide.

For more information on the benefits of urban trees, see Talking Trees: An Urban Forestry Toolkit for Local Governments, published in November 2006 by ICLEI.

L.0.2 Land-based GHGs and interactions with other sectors

The use of land, particularly forests, interacts with other sectors in a greenhouse gas inventory. This is particularly true for energy (e.g. through the burning of biomass or agricultural residue for energy) and waste (e.g. through biomass that goes to landfill). Biomass from forests is also used in various “harvested wood products” that store carbon for various periods of time, from short periods (e.g. paper) to very long periods (e.g. furniture or in building construction) and the substitution effects of using such products can also have impacts on other sectors—for example, using wood for buildings compared to energy-intensive materials such as cement or steel.

Full and accurate estimation of land use impacts on GHGs requires inclusion of all these linked systems. However, often such systems cross boundaries—for example, wood produced within a community may be used for energy or wood products outside the community boundary. Similarly, biomass waste may end up in a landfill within (or outside) the community boundary.

When covering land-related emissions and removals, some GHG inventory guidance applies the concept of “scopes” to help organize the inventory process (see Table 1). This appendix primarily involves Scope 1.

Table : Scope 1 to 3 in relation to land use

Figure 1: The forest sector in relation to land use, wood products, and energy4

The chart below provides a crosswalk between how GHGs from the forest sector, as well as “trees outside forest”, are treated in this Appendix compared to the Global Protocol for Community-Scale Greenhouse Gas Inventories (GPC), the Intergovernmental Panel on Climate Change 2006 Guidelines for National Greenhouse Gas Reporting (IPCC) and their representation in the US GHG inventory.

Table 2: Crosswalk among USCP, GPC, IPCC and the US GHG Inventory

*Note: In the case of biomass burning for energy, CO₂ emissions from biomass losses often occur in a different community than energy emissions (i.e. non-CO₂ emissions from bioenergy), i.e. the community where the harvesting takes place may see a net loss of biomass (i.e. an emission), whereas the community that is burning bioenergy may see a net benefit from use of that biomass as a substitute for fossil fuel use (i.e. the non-CO2 emissions are lower than the burning of fossil fuels).

L.1 Defining the Inventory Analysis Period

Because estimating GHG emissions and removals from land use requires knowledge of both land cover/use and change, information is required from at least two points in time (to compare if and how land cover/use has changed over a given period). Data from one point in time can only provide information on carbon stocks; however, a GHG inventory requires information on GHG fluxes, or changes in carbon stocks—which requires data from two points in time.

Therefore, GHG estimates from land use are a result of analyzing changes in carbon stocks over an inventory analysis period, i.e. spanning several years, and is typically defined by:

• The year(s) of interest to estimate GHGs from land use;

• The years in which data on land cover/use are available.

A community may wish to develop information for a baseline8—i.e. a reference against which the community can set climate goals. Unlike other sectors, annual data are often unavailable for land use change and several years of data may be needed before the fate of the change can be accurately determined. Therefore, to estimate baseline GHGs from land, a “reference period” may be chosen that straddles the base year used for other sectors (see example in Box below). The ‘reference period’ will typically be a range of years, and generally it is better—for the purposes of a baseline—to take an average over a few years for land-based emissions or removals because weather patterns can lead to anomalies in a specific year. GHGs can be calculated for the reference period and annualized.

After baseline GHGs from land use are estimated for the reference period, subsequent GHG estimates are similarly made using various analysis periods, based on the years in which a GHG inventory is compiled for other sectors, combined with available land cover/use data. Such periods need not be the same length as the reference period; data are averaged over each period and annualized.

To assess trends in GHGs, it is important to develop a consistent time series, i.e. the same land use definitions (classifications), methods and data sources should be used in each inventory. If changes are made, which is often the case as advances in monitoring are deployed, the time series should be recalculated back to the baseline period with the change applied for the entire time series or by using methods to promote consistency of the estimates across time (see accompanying guidance document).

Box 2. Using the National Land Cover Database

The National Land Cover Database (NLCD) is developed every two to three years and is currently available for the years [2001, 2004, 2006, 2008, 2011, 2013 and 2016 - TBD when protocol published]. Let’s say a community decided to use the NLCD for activity data to estimate land-based GHGs and uses a base year of 2005 to set goals for other sectors (e.g. 80% reduction from 2005 emissions by 2050). In this instance, the community may estimate the average annual GHGs from a reference period of 2001-2011, or 2004-2006, and use this as the baseline (for land-based GHGs), comparable to the 2005 base year for other sectors. In the future, it may then compare annualized data from 2011-2013, 2013-2016, and 2016 to the next NLCD dataset, as subsequent periods of analysis to develop a time series and understand trends in land-based GHGs.

Alternately, a community may have its own forest/tree cover data that can be used, or it may wish to develop such data at periodic intervals—and this would define the reference period and subsequent analysis periods.

L.2 Representation of Land

L.2.1 Delineating the land base

Information about various land use and land use change categories is needed to estimate associated GHG emissions and removals. Before gathering this information, a community must first decide the land area over which it wishes to estimate emissions and removals and that will be included in the inventory. Usually this will be the administrative boundaries of the community’s jurisdiction. Exceptions may include, for example:

• Areas owned/managed by the community that are outside the jurisdictional boundary;

• Areas outside the jurisdictional boundary (Scope 3) that are influenced by within-boundary planning and development decisions;

• Land base inside the jurisdictional boundary over which the community has authority (i.e. excluding federal or state lands), and/or excluding lands inside the jurisdictional boundary designated by the community as unmanaged (see Box).

Box 3. Managed vs. Unmanaged land

IPCC guidelines further separate land into two categories: managed and unmanaged. The concept of “managed lands” was developed to separate effects of anthropogenic (human caused) activities from non-anthropogenic (or natural) effects on GHGs. In practice, separating out natural from human-caused emissions or removals can be a challenge. For example, natural disturbances, such as wildfires, can lead to significant emissions on managed land. IPCC guidance is to report GHG emissions and removals that occur on “managed lands” as a proxy for estimating anthropogenic emissions and removals.

Managed land is defined by the IPCC as land where human interventions and practices have been applied to perform production, ecological or social functions. Most lands in the United States are classified as managed by the federal government for reporting GHGs. In fact, all lands in the “lower 48” and Hawaii are considered managed. There are large areas within the state of Alaska that are considered unmanaged due to the lack of direct human intervention because they are largely inaccessible or are remote areas far from settlements and roads.

It is recommended that communities consider all lands as managed to remain consistent with federal (and potentially state) GHG reporting. If a portion of community land is designated as unmanaged and excluded from the inventory, justification should be provided on methods used to delineate them and geospatial boundaries provided in documentation as applicable.

L.2.2 Land use classification

It is recommended when developing a GHG inventory to define and classify all community lands into six land use classes9: forest land, cropland, grassland, wetlands, settlements and other land. This classification forms the basis to estimate and report greenhouse gas emissions and removals from all land use and land-use conversions. Additional subcategories may be included to further classify the broad land uses, as needed, for calculating emissions, informing local decision making, or reporting purposes.

When classifying land, care should be taken to avoid overlaps or omissions in the classification of the land base. For example, overlaps could occur if trees on farms were included as both forest land and cropland; this land should be represented as only one land use class. The classification should represent land-use categories consistently over time, so that artifacts from changes in data or methods and/or artificial discontinuities in the time-series data are not included as actual land use changes. Since the inventory analysis period is likely to span several years (or at most, an annual time step), seasonal changes in land use that may occur within a single year are not considered; all land included in the inventory should be representative of the dominant land use in a given year of estimation for each land parcel in the community.

Definitions for each land use class applied by the US federal government in compiling the national GHG inventory are provided below (see Box). It is recommended that communities align with definitions applied by the federal (or state) government, if possible, to ensure consistency with national (or state) reporting. A community may, however, decide to use a land use classification system that diverges from that used by the federal (or state) government for policy or other reasons. For example, a different classification system may be critical to improve the estimation of GHG emissions. Alternatively, a particular classification may be selected that is consistent with laws and policies relevant to land management that the community wishes to monitor as part of its climate action.

Box 4. Land Use Definitions in the US GHG Inventory

The US National GHG inventory uses the following definitions of the six IPCC land use classes10:

• Forest Land: An area at least 120 feet wide (36.6 meters), one acre (0.4 hectare) in size with at least 10% cover (or equivalent stocking) by live trees. Land with such tree area and cover is not classified as Forest if completely surrounded by urban or developed lands (rather, they are classified as Settlements); land that is predominantly under agricultural land use is also not considered Forest.

• Cropland: Areas used for the production of adapted crops for harvest; this category includes both cultivated (row crops, close-grown crops) and non-cultivated lands (e.g. hay, orchards).

• Grassland: Areas on which the plant cover is composed principally of grasses, grass-like plants (i.e., sedges and rushes), forbs, or shrubs suitable for grazing and browsing, and includes both pastures and native rangelands

• Wetland: Land covered or saturated by water for all or part of the year, in addition to the areas of lakes, reservoirs, and rivers.

• Settlement: Developed areas consisting of units of 0.25 acres (0.1 ha) or more that includes residential, industrial, commercial, and institutional land (including farm buildings and road networks). Also includes tracts of less than 10 acres (4.05 ha) that may meet the definitions for Forest Land, Cropland, Grassland, or Other Land but are completely surrounded by urban or built-up land.

• Other Land: Bare soil, rock, ice, and all land areas that do not fall into any of the other five land-use categories; C stock changes and non-CO2 emissions are not estimated for Other Land because these areas are largely devoid of biomass, litter and soil C pools11.

Both the US National GHG inventory and the GPC provide a hierarchy in the case where an area meets more than one category of the land use classification. In such cases, lands are assigned the following classes by priority: Settlements > Cropland > Forest land > Grassland > Wetlands > Other land. The US national inventory report provides the following underlying rationale for this ranking:

Settlements are given the highest assignment priority because they are extremely heterogeneous with a mosaic of patches that include buildings, infrastructure, and travel corridors, but also open grass areas, forest patches, riparian areas, and gardens. The latter examples could be classified as Grassland, Forest Land, Wetlands, and Cropland, respectively, but when located in close proximity to settlement areas, they tend to be managed in a unique manner compared to non-settlement areas. Consequently, these areas are assigned to the Settlements land-use category. Cropland is given the second assignment priority, because cropping practices tend to dominate management activities on areas used to produce food, forage, or fiber. The consequence of this ranking is that crops in rotation with pasture are classified as Cropland, and land with woody plant cover that is used to produce crops (e.g., orchards) is classified as Cropland, even though these areas may meet the definitions of Grassland or Forest Land, respectively. Similarly, Wetlands are considered Croplands if they are used for crop production, such as rice or cranberries. Forest Land occurs next in the priority assignment because traditional forestry practices tend to be the focus of the management activity in areas with woody plant cover that are not croplands (e.g., orchards) or settlements (e.g., housing subdivisions with significant tree cover). Grassland occurs next in the ranking, while Wetlands and then Other Land complete the list.

Once definitions are applied, they should be used consistently12 in future GHG inventories to be comparable. Or, if a definition is changed, the historical time series should be recalculated to reflect the change in definition.

Box 5. Land use versus land cover

According to the IPCC guidelines, reporting GHGs should be separated by land use categories (based on the six land use classification types). Land use refers to the way in which humans use or manage land (e.g. the arrangements, activities and inputs undertaken or it may also refer to the social and economic purposes for which land is managed13). Land use change, therefore, refers to a change in the use or management of land by humans. In contrast, land cover refers to the biophysical attributes of land (e.g. whether or not there are trees is a key attribute of forests) and may not always be consistent with the use of the land. In some cases, land use change and land cover change will be identical, e.g., conversion of forest land to cropland. In other cases, a change in land cover from forest to grassland may reflect a temporary disturbance where the forest is expected to regrow.

The IPCC suggests that land is not fully converted until the land use change took place >20 years ago14. If the conversion is <20 years, the land is placed into a conversion category, i.e. is considered in “transition”—this is largely to account for the slow change in dead organic matter and soil carbon that occurs when land is converted from one land use to another. Remote sensing technologies are increasingly used to monitor land cover (and change). Many communities may rely on remote sensing images to provide Activity Data for estimating GHGs, as a simplification compared to stratifying by land use. Without having additional information on land use, however, it is not possible to know whether a land cover change is temporary or permanent.

For community-scale inventories, it may be a reasonable simplification—in the absence of additional data and/or to simplify the GHG inventory for lands—to use land cover for activity data and take a “committed” approach to estimating GHGs, i.e. to assume full or partial loss or gain of C stock at the time of detection and to report the conversion in the year of detection. This will overestimate the actual emissions / removals in the year of detection; however, over time, the inventory will self-correct.

For example, in the case of a temporary disturbance—such as a forest fire in an area that remains forest land use but shows a land cover change from forest to grassland in the year of the fire—the GHG inventory may report full C stock loss for the conversion, which overestimates the actual loss in the year of the fire (as deadwood will decay over time). However, when the forest regenerates the C stock is returned to the reporting. Similarly, if a forest is converted to a developed area the full change in C stock may be reported in the year of conversion, even though in reality soil C stock may take years to decay. The benefit in the latter case is that the full emission is more closely related to the activity that caused it, rather than creating a ‘legacy’ emission over time.

L.2.3 Categories for reporting GHGs from land use

When developing a land use GHG inventory, it is recommended to develop a land use change matrix as illustrated in Table 3. Land included in the inventory analysis period is stratified into the six land use classes for the beginning and the end of each analysis period using consistent definitions (as above) and then all lands included in the inventory are assigned to one of the categories in Table 3. This process is repeated for each analysis period in the inventory, which will depend on the years in which land cover or use data are produced. It is recommended to estimate (or convert) values into hectares (see Box in Section L.3.3 “Units of Measurement”) and apply the same units consistently across categories and throughout the estimation process. Once area estimates are combined with appropriate emission/removal factors, values in the individual cells of the matrix represent GHG fluxes from each category.

Table 3: Land use change matrix

Table 3 reflects a 6x6 land use conversion matrix based on six land use categories. Such matrices can also be expanded to include more sub-categories within each broad land category, e.g. different forest and/or management types within the Forest Land Remaining Forest Land category, based on the categories that make sense in the context of both local geography and climate action plans. Note that partitioning the land base into smaller areas has implications for availability of data that represent the different land strata.

It is recommended that a community track all 36 possible land use and change categories (6x6 matrix), even if it is initially reporting only on a smaller subset of categories. For example, even though this appendix currently covers only Forest Lands (and trees outside forests), in the case of land use change—when forest lands are converted to another land use or when lands are returned to forest lands—knowledge of the pre- and/or post-forest land use will provide more accurate estimates of C stock changes. In addition, guidance may be developed for other land uses and so establishment of a well-defined land base can be useful in the future.

Currently available, “off the shelf” data sources for the U.S. to create a land use and change matrix are summarized in Table 4. These are elaborated further in the protocol’s associated guidance document.

Table 4: Geospatial data sets available to estimate land cover and change for U.S. communities

It is recommended to estimate land-based GHGs by land unit rather than by activity. In other words, GHGs are estimated across the entire land area regardless of the variety of actions and/or activities that may occur on such lands. This helps to avoid ‘double-counting’ of the impacts of a variety of activities that may occur on the same land.

L.3 Estimating GHGs from Land Use (General Methods)

This section provides general information applicable to estimate greenhouse gas emissions and removals for all land uses. Subsequent sections provide additional guidance for specific land uses. Currently this Appendix includes only Forests and Trees Outside Forest but may be updated later to provide guidance on other land use categories.

L.3.1 Carbon Pools

The seven main carbon pools generally defined by IPCC and adapted for reporting in the United States are:

• Biomass Carbon – Aboveground and belowground carbon in living plant mass.

  • Tree biomass – Live trees with diameter at breast height (d.b.h.) of at least 2.5 cm (1 inch), including carbon mass of coarse roots (greater than 0.2 to 0.5 cm, published distinctions between fine and coarse roots are not always clear), stems, branches, and foliage.

  • Understory vegetation - Live vegetation that includes the roots, stems, branches, and foliage of seedlings (trees less than 2.5 cm d.b.h.), shrubs, and bushes.

• Dead Organic Matter – This pool includes carbon in standing dead, downed dead wood, coarse dead roots and litter pools.

  • Standing dead trees - Standing dead trees with d.b.h. of at least 2.5 cm, including carbon mass of coarse roots, stems, and branches.

  • Down dead wood – Woody material that includes logging residue and other coarse dead wood on the ground and larger than 7.5 cm in diameter, and stumps and coarse roots of stumps.

  • Litter – Dead organic matter transferred from the aboveground or belowground live biomass pools on the soil surface or within the soil. In forest, the dead organic matter overlying the mineral soil is referred to as forest floor and includes fine woody debris up to 7.5 cm in diameter, tree litter, and fine roots in the organic forest floor layer above mineral soil.

• Soil organic carbon – Carbon in soil, i.e., soil organic matter, that is the decomposition products of microbial organisms or partially decomposed plant material that is smaller than 2 mm.

• Carbon in harvested wood – Includes products in use and in landfills. “Products in use” includes end-use products that have not been discarded or otherwise destroyed. Examples include residential and nonresidential construction, wooden containers, and paper products. “Products in landfills” includes discarded wood and paper placed in landfills where most carbon is stored long-term and only a small portion of the material is assumed to degrade, at a slow rate.

Carbon is transferred among these 7 pools and between them and the atmosphere. The amount of carbon in each pool is commonly called a stock, and the transfers are changes in carbon stocks. A decrease in total carbon stock in the terrestrial ecosystem equates to an emission of carbon dioxide from the terrestrial ecosystem to the atmosphere, and an increase in total carbon stock equates to a removal of carbon dioxide from the atmosphere into the terrestrial ecosystem. Emissions and removals are often referred to as CO₂ fluxes.

It may not be necessary to calculate changes in C stocks for all 7 pools. For example, if no harvesting takes place there will be no additions of C to HWP pools, although there may be legacy effects of past harvests on GHG emissions that can last for many decades. Also, it may be acceptable to assume that there are no changes to the soil organic C pool for those activities that do not disturb the soil, such as a light forest thinning. Additional guidance is provided in Section L.4, Step 3 below (emission/removal factors).

L.3.2 Non-CO₂ Greenhouse Gas Emissions

Non-CO₂ GHG emissions refer to methane (CH₄) and nitrous oxide (N₂O) emissions that occur as a result of land use and management activity. These activities include biomass burning from prescribed or wildfires, nitrogen (N) management practices on managed soils, livestock management and rice cultivation.

In the case of prescribed fire or wildfires, oxidation of biomass releases CO₂, CH₄, N₂O, and other precursors that later form greenhouse gases, namely CO₂, CH₄, carbon monoxide (CO), non-methane volatile organic compounds (NMVOCS), N₂O, and nitrogen oxides (NO_x). The CO₂ emissions are factored into the calculation of carbon stock changes for the pools that are oxidized, and therefore do not need to be estimated as part of the non-CO₂ greenhouse gas emissions (i.e., this would lead to double counting of emissions). The amount of the other gases is estimated to determine the impact of biomass burning on anthropogenic GHG emissions if the fire occurs on managed land.

N management is common in many land uses, particularly croplands but also forest lands in some regions, and leads to direct and indirect soil N₂O emissions. Management activities include synthetic fertilizer additions, organic amendments such as manure, and N additions in dead organic matter. Enhancement of soil organic matter mineralization due to management activity is also considered a source of N additions that leads to anthropogenic emissions of N₂O.

Rice cultivation in flooded conditions can generate significant amounts of CH₄ emissions and should be classified as part of the land representation analysis but is often included in the reporting for the Agriculture sector. Communities will need to assess rice cultivation as part of their land representation regardless of the sector in which the emissions are reported. The calculation of emissions will depend on land use and land use change data as well as water management practices, organic amendment practices, soil type and cultivar selection. Livestock management is covered in the Agriculture Appendix and is not discussed here.

L.3.3 Calculating GHG emissions and removals from land use

When estimating GHG emissions for other sectors, a general equation is applied in which activity data is multiplied by an emission factor. For land use, unlike other sectors, both emissions and removals can occur, and so the net total GHG flux (which can be positive or negative) is represented as the summation of the GHG emissions and removals from all areas of land. The approach summarized below represents a simplification of the “gain-loss” method described by IPCC and often used in national GHG inventories. The full IPCC gain-loss method is presented in the annex, which communities may want to apply if the required data are available. For the simplified gain-loss method, both carbon gains (GHG removals) and losses (GHG emissions) are represented by activity data multiplied by emission factors.

Equation 1: General equation for net GHG flux from land use (simplified gain-loss method)

Net GHG Flux = GHG emissions + GHG removals

$$\text{GHG}\ \text{emissions} = \ \sum_{\text{land}}^{}{\text{AD} \times \text{EF}}$$

$$\text{GHG}\ \text{removals} = \sum_{\text{land}}^{}{\text{AD} \times \text{RF}}$$

In most (but not all) cases:

• Activity Data (AD) is expressed in units of land area (hectares) over which an activity has occurred;

• Emission Factors (EF) or Removal Factors (RF) are the annual change in CO₂eq per area, expressed in, e.g. tCO₂eq/hectare.

Each section below (currently including Forests and Trees Outside Forests) will go into greater detail on the development of Activity Data and Emission and Removal Factors for each land use type.

Box 6. Units of measurement and related terminology

GHGs that are relevant for forests include primarily CO₂. For cases where other GHGs are a significant part of the inventory, see section L.3.2.

All stock computations are performed in terms of mass of C in kilograms or metric tons per unit area in metric system units (C per hectare or C/ha). Rate data are reported in terms of change in C/ha over time, as in C per hectare per year (C/ha yr⁻¹). All C biomass is referenced to its dry weight basis and the fraction of biomass in C. For the purpose of this Protocol Appendix, the fraction of dry biomass that is C is 0.5. An example stock is 100 metric tons C ha⁻¹, and an example stock change is 1.0 metric tons C ha⁻¹ yr⁻¹. It is important to differentiate between units of C and Carbon Dioxide Equivalents (CO₂ eq), and report the appropriate units to the reporting entity. It is recommended to use metric units (e.g. a metric ton = 2,204.6 pounds) and to report GHGs in CO₂ eq. This convention places all C mass estimates into units of the emission gas, CO₂, which can be derived by multiplying the C mass by 44/12.

Terminology and number signs (plus or minus) can be confusing. Generally, these items are referenced from the perspective of the atmosphere according to the following table:

L.4 Forest Land

This method allows a community to develop GHG estimates for three forest categories: forest remaining forests (i.e. standing forests), forest converting to non-forest (forest loss or deforestation), and non-forest converting to forest (afforestation or reforestation). The method breaks down the process into seven steps and describes the basic elements of the inventory process for each step.

Recommended Approach: Estimate forest-related GHGs using the following steps:

Step 1 – Consider the need for further stratification

Step 2 – Determine areas of forest-related land use and change over the inventory analysis period (activity data, disaggregated by strata and sub-strata as appropriate).

Step 3 – Determine the appropriate emission/removal factors for forest land use and change categories (disaggregated by strata and sub-strata if appropriate).

Step 4 – Calculate CO₂ emissions and/or removals from forest land and forest-related land use transitions

Step 5 – Calculate non-CO₂ emissions from forest land if appropriate, and convert into units of CO₂ equivalents

Step 6 - Determine whether to include harvested wood products (HWPs) in the inventory and calculate C stock changes in the HWP pool if appropriate.

Step 7 – Estimate total net GHG flux from forests over the inventory analysis period and annualize the result into units of tCO₂e/yr

Agencies of the U.S. federal government and many states provide nationally or statewide consistent databases, data repositories and formats relevant to the steps identified above. The methods below and companion guidance document to this Appendix recommend a combination of data sources that will be most widely applicable to communities across the U.S., but if high quality and more locally appropriate data are available for a community, then the use of these data sources is encouraged. In all cases, the methodologies and data sources applied should be disclosed wherever the results are published.

Step 1: Consider the need for further stratification.

A community may wish to disaggregate the total area of forest land included in the inventory into smaller units, or strata, either for reporting purposes or because estimates of carbon density (carbon per unit area) or forest carbon increment values (carbon sequestered per unit area per unit time) across different strata may vary significantly, thus influencing the final emission or removal factor applied. The need for stratification is more likely to be necessary for forest land vs. trees outside forests and other non-forest land use categories (e.g., grassland, cropland) because these other non-forest land uses are likely to have much lower and/or less variable carbon stock densities.

Stratification of a community’s forest land can be achieved by spatially intersecting maps of forest cover with other relevant datasets. These can relate to ecological factors, such as climate, soil or vegetation type, and/or to management factors such as forest ownership or forest use. The advantage of additional stratification is to potentially reduce uncertainties in the resulting GHG flux estimate; this must be weighed against the disadvantage of adding complexity to the GHG estimation process. Communities with diverse types of forests and forest management practices within their inventory’s land base are encouraged to consider additional stratification of the forest land use category (e.g. where forest land is mountainous, subject to different management regimes such as plantation forestry vs. wilderness areas, forest stands that vary considerably in age or hydrological regime, etc.). On the other hand, communities with relatively homogenous forests and/or with limited information about how forest carbon density varies over the landscape may wish to apply “average” emission or removal factors for all forest land included in the inventory.

There is no single “correct” way to stratify land. Stratification procedures are designed primarily with the goal to minimize uncertainty/error within strata by assigning the most applicable emission and removal factors, or to focus attention on specific land areas of interest to the community. The simplest option (no stratification, i.e., all forest land is treated as a homogenous unit) may be the best option if data are lacking or until additional data become available to improve the precision of per-strata GHG estimates. However, even if data are lacking, it may still be important to stratify for reporting purposes; in these cases, the same emission or removal factors can be used for multiple strata.

Box 7. Developing data useful for Climate Action

For the purposes of using the data beyond a GHG inventory estimate—for example, to support the development of climate mitigation actions—it may be also be useful to consider stratifying forest land data in ways that inform intended actions. For example, land may be separated by authority (e.g. federal, state, and community managed lands or public versus private lands) or by management practice (e.g. conservation or protected areas versus forest plantations). Doing so can help to track the GHG impact of various activities within a community’s climate action plan.

Step 2. Determine activity data for estimation of forest-related emissions and removals

The next step in preparing the GHG inventory for forest land is to assemble the required activity data, i.e. forest land-use and change areas (disaggregated by strata and sub-strata as appropriate). Relevant activity data involves obtaining area estimates for the blue boxes highlighted in the matrix below. Once the decision has been made about which source of activity data to use, estimates can be generated by a qualified GIS technician following instructions provided in the protocol’s companion guidance document.

Table 5: Relevant categories within the land use conversion matrix for reporting on forest land

Activity data will need to be matched with appropriate emissions and removal factors (Step 3). As data are compiled and evaluated, it may be necessary to iterate across Steps 1, 2 and 3 before arriving at the most appropriate way to match land areas with emission and removal factors.

In some instances, a community preparing activity data for forest land may assess changes in forest land cover over the inventory analysis period and apply the assumption that any observed change in forest land cover corresponds to a change in land use. However, as discussed in Section L.2.2 (Box 5 on Land Use versus Land Cover), inferring land use from land cover could result in misplacing GHG estimates into the wrong land use reporting category. Therefore, it is recommended that supplementary data and information be used to first test the assumption that an observed change in forest cover during the inventory analysis period does, in fact, correspond to a change in land use. This can be done using a combination of local knowledge/expert judgement, site visits, and/or high-resolution satellite or aerial imagery from the year the change was observed. In addition, spatial overlays of forest land use areas (e.g. areas designated as national, state, or local forest or park areas, or managed forest areas), areas of natural disturbances, or planned urban expansion areas can support improving the attribution of land cover change to the correct reporting categories.

In the more complex case where a community stratifies its forest land into further sub-categories, it is worth noting that emission or removal factors (see Step 3) specific to additional strata may not be available. However, as mentioned above (in Step 1), a community may wish to track and report on such lands for policy or other purposes and may simply apply the same emission and removal factors to multiple strata.

Step 3 – Determine the appropriate emission/removal factors for all land use and change categories (disaggregated by sub-strata if appropriate)

There are many sources of data for calculating emissions and removals factors, as well as sources for pre-compiled emissions and removal factors that can be readily applied to the inventory. A summary of data sources is provided in Table 6, and more details about these sources are contained in the companion guidance document. Generally, using prepared lookup tables, on-line models and calculation tools will simplify the process. Calculating emissions and removals factors from data that is specific to the inventory area is preferred.

Table 6: Data sources available for estimating emission and removal factors in U.S. forests.

Emission and removal factors are highly dependent on the nature of the activity. For example, a deforestation event will have a large initial pulse of emissions that occurs at the time of the event, whereas afforestation or forest growth within undisturbed stands will have a much smaller removal factor that can be applied over many years or decades as the forest grows. For each type of activity, communities will need to choose a time frame over which the factor applies. The time frame can be annual or multiple years, and the factor may be used for tracking changes in carbon stocks annually or as described earlier in the section on land cover, it may be simpler to take a “committed” approach to estimating changes in carbon stocks, i.e. to assume full or partial loss or gain of C stock at the time of detection and to estimate the whole future change in the year of detection. As noted, this will overestimate the actual emissions / removals in the year of detection; however, over time, the inventory will self-correct.

Emissions and removals factors should represent the changes in carbon stocks for all of the seven main carbon pools if they are significantly affected by the activity. However, it is common practice to ignore very small changes in selected carbon pools (e.g., less than 3% of the total change in all C pools) because such small changes can be hard to measure or estimate. In this section we describe an approach where data from changes in all the carbon pools are summed into either a removal factor associated with land that has been undisturbed (no stand-replacing disturbance) during the inventory analysis period, or an emission factor in the case of land that has had a significant disturbance. It may be the case that within either of these broad categories, some C pools may increase and some may decrease simultaneously, according to the typical transfers between C pools that occur in both disturbed and undisturbed forests. For advanced users who are interested in delving deeper into the changes in individual C pools, equations are provided in the annex that represent the complete “gain-loss” approach as defined by the IPCC, rather than the simplified version used in this section.

Ideally, emissions and removals factors are calculated with data that represent the forest strata defined for the GHG inventory, if applicable (see Step 1 above). Because data may be lacking for each and every stratum in the area of the inventory, combining strata or extending the area beyond that of the GHG inventory area (see Section L.2.1 Delineating the Land Base) may be necessary for calculating emissions and removals factors. For example, a county may be interested in separate GHG reporting for three different ownership groups (e.g. federal, state and locally owned forests), but only able to calculate emissions and removals factors for all groups combined. For another example, there may be interest in a specific forest type, but if there is insufficient data about that forest type within the inventory area, it may be necessary to include data about that forest type from a larger geographic area that closely resembles the inventory area. Note that the availability of data for calculating emissions and removal factors is likely to be different for different activity categories and carbon pools. Local data representing conversion of forest land to other land-use categories is often lacking; therefore, we highlight here a set of assumptions that may be used to help estimate the changes in carbon stocks for this activity (Table 7). For additional examples of how to handle these types of situations, refer to the companion guidance document.

Table 7: Defaults for estimating percentage losses of C with forest land conversion to other land uses.

Biomass, dead organic matter and soil C losses may be restored with conversion back to forest land, but the time dynamics are typically longer than 20 years. These estimated percentage losses should be applied to estimates of C stocks on the land prior to conversion.

¹Biomass C loss for forest land converted to grassland based on assumptions developed for the US National Greenhouse Gas Inventory.

² Soil organic C loss based on factors used in the US National Greenhouse Gas Inventory. The estimate for cropland based on aggregated value across all climate in the United States.

Step 4 – Calculate change in C stocks (CO₂ emissions and/or removals) from forest land and forest-related land use transitions

This step covers the estimation of changes in C stocks (CO₂ emissions and removals) for forest land remaining forest land as well as land that converts between forest and non-forest during the inventory analysis period. Each equation below estimates the change in carbon stock as the product of the appropriate activity data paired with the appropriate emission/removal factor.

Step 4.1 Calculate change in C stocks from forest land remaining forest land (∆C_FRF)

This is forest land that did not change land use during the period of analysis. This category includes areas that may have been subject to tree harvest or other type of disturbance during the inventory period, but that are still considered forest land remaining forest land from a land-use perspective. The GHG flux reflects the net balance of emissions and removals, and both can occur simultaneously in forest land remaining forest land in different areas and/or at different times during the inventory analysis period. In the case of forest land remaining forest land, it is most transparent to account for carbon gains and carbon losses separately if possible, and then calculate the net balance between the two fluxes

With the simplified gain-loss method, two general cases of forest land remaining forest land are defined for making estimates – undisturbed and disturbed. For relatively undisturbed forest land in the community (including land that may have very small or low severity disturbances), carbon stock changes (CO₂ removals) are estimated as the area of undisturbed forest land in each strata (as appropriate) over the inventory period multiplied by the appropriate removal factor for that strata and the number of years in the inventory analysis period. Note that some “undisturbed” old forests may have emissions from dead or dying trees that exceed removals from live trees; in this case, the factor would be a net emission rather than a removal. For disturbed forests, carbon stock change (CO₂ emissions) are estimated as the area of disturbance in each strata over the inventory period multiplied by the appropriate emission factor for that strata. All estimates should be converted to average annual estimate as a final calculation step, if necessary (see Step 7).

Equation 2: Change in C stocks for forest land remaining forest land

ΔC_FRF = ΔC_undisturbed + ΔC_disturbed

$${\mathrm{\Delta}C}_{\text{undisturbed}} = \sum_{i = 1}^{n}{\text{AD}_{i} \times}\text{RF}_{i} \times T$$

$${\mathrm{\Delta}C}_{\text{disturbed}} = \sum_{i = 1}^{n}{\sum_{j = 1}^{J}{\text{AD}_{\text{ij}} \times \text{EF}_{\text{ij}}}}$$

Where:

ΔC_FRF= change in carbon stocks in forest land remaining forest land over the inventory period; tons C

ΔC_undisturbed= change in carbon stocks in undisturbed forest land remaining forest land over the inventory period; tons C

ΔC_disturbed= change in carbon stocks in disturbed forest land remaining forest land over the inventory period; tons C

AD_ij= area of forest land in stratum i (of disturbance type j, if applicable); ha

i = 1, 2, 3, …, n forest strata

j = 1, 2, 3, … , J disturbance types

T= number of years in inventory analysis period; years

EF_ij = emission factor for each disturbance type j in stratum i; t C/ha

RF_i = removal factor for each stratum i; t C/ha/yr (average annual removal factor)

There is a significant amount of data available for estimating disturbance impacts and thus emission factors for different forest and disturbance types (EF_ij). The U.S. Forest Service has calculated the effects of main disturbances on carbon pools for the main forest and disturbance types for National Forests, by region (Birdsey et al. 2019; Raymond et al. 2015). For more information on estimating disturbance impacts, see the accompanying guidance document.

An example of calculating the change in carbon stocks for two different forest types in a community, one that has been disturbed during the inventory analysis period and one that has not, is presented in the box below.

BOX 9. The "stock change” method

Most communities will use the approach outlined above (i.e. the simplified gain-loss method) to estimate GHGs from forest land remaining forest land. Another method is called the stock-change approach and can be used in instances where there are sufficient permanent, remeasured forest inventory sample plots available in or near the community to characterize the strata of interest. As a general rule of thumb, roughly 20 remeasured sample plots are needed per stratum to use the stock-change method. Calculations would be repeated for each additional stratum of interest as defined by the activity data and any other classification variable of interest as determined by activity data, a map, or statistical reference, such as forest type. Note that the stock-change method does not explicitly account for disturbances. Additional calculations are needed to explicitly estimate effects of disturbances if using the stock-change method (see later section).

Step 4.2 Calculate change in C stocks from forest land converted to non-forest land (∆C_F>NF)

This is forest land that changed land use during the period of analysis; if it occurs, forest clearing is likely to be one of the largest sources of GHG emissions from forests during the inventory period. For forest land converted to non-forest land, carbon stock change (CO₂ emissions) are estimated as the area of conversion in each strata over the inventory analysis period multiplied by the appropriate emission factor for that strata.

Equation 3: Change in C stocks for forest land converted to non-forest land

$${\mathrm{\Delta}C}_{F > NF} = \sum_{i = 1}^{n}{\sum_{k = 1}^{K}{\text{AD}_{\text{ik}} \times \text{EF}_{\text{ik}}}}$$

Where:

ΔC_{F > NF}= change in carbon stocks in forest land converted to non-forest land over the inventory period; t C

AD_ik= area of forest strata i converted to non-forest category k; ha

EF_ik= emission factor for each forest strata i converted to non-forest category k; t C/ha

i = 1, 2, 3…n forest strata

k = 1, 2, 3…K non-forest land categories (i.e., cropland, settlement, grassland, wetland, other land)

For estimating emission factors, the conversion of forests to other land uses immediately reduces the stock of carbon in both living biomass and dead organic matter. Over time, soil carbon may be reduced in some cases, particularly with conversion to croplands or settlements, but other conversions may lead to limited change in soil carbon stock or even an increase, such as conversion to grassland or wetlands. Regardless, the conversion of forest land to other land uses is likely to reduce the overall long-term carbon storage potential of the land. Some of the biomass can also be removed from the site and converted to forest products such as lumber, paper, pulp, and other products that have longer term but variable decomposition rates—and hence longer term and variable emissions over time (see Step 6 below on HWPs). All of these changes in C stocks for various pools should be estimated when determining the changes in carbon stocks (i.e., the emission factor) due to forest land converted to non-forest land.

The most important activity data to collect are the area and rates of forest clearing for each stratum or parcel in the inventory area and, if possible, the land category forest land was converted to, which will help to identify the appropriate emission factors to apply to the activity data (see Table 7 above). To estimate emissions, it is necessary to estimate also the characteristics of the stratum prior to clearing, including the biomass, dead organic matter and soil organic carbon of the site. The default emission factors (Table 7) may be modified with additional information. For example, if some live trees were left standing on the site after conversion, the loss of biomass would be less than 100%. It may also be significant to estimate the fraction of the aboveground biomass that was burned, the fraction that was removed in the form of wood products, and the fraction that was removed in the form of other products. This is the same information needed about harvesting for forest land remaining forest land. If this information is not available for all transitions from forest to other land uses in all areas, then it is recommended to use regional averages for harvested wood products, which are available from published sources (refer to the guidance document).

An example of calculating the change in carbon stock for forest land converted to other land is presented in the box below. This calculation would be repeated for each forest stratum of interest i converted to non-forest land category k as defined by the activity data and any other classification variable determined by a map or statistical reference, such as forest type. Note that the calculation for carbon in harvested wood products is described in Step 6.

Step 4.3 Calculate change in C stocks from non-forest land converted to forest land (∆C_NF>F)

This is non-forest land that changed to forest land over the period of analysis. For non-forest land converted to forest land, carbon stock change (CO₂ removals) are estimated as the area of forest gain over the inventory period multiplied by the appropriate removal factor.

Equation 4: Change in C stocks for non-forest land converted to forest land

$${\mathrm{\Delta}C}_{NF > F} = \sum_{i = 1}^{n}{\sum_{k = 1}^{K}{\text{AD}_{\text{ki}} \times \text{RF}_{\text{ki}} \times T}}$$

Where:

∆C_NF>F= change in carbon stocks in non-forest land converted to forest land over the inventory period; t C

i = 1, 2, 3…n forest strata

k = 1, 2, 3…K non-forest land categories

AD_ki = area of non-forest category k converted to forest strata i ha

RF_ki = removal factor for each non-forest category k to forest category I; t C/ha/yr (annual average removal factor over the inventory analysis period)

T= number of years of the conversion; if unknown (i.e. when in the inventory analysis period the conversion took place), then it is suggested to use the average (i.e. T = ½ * length of the inventory analysis period); years

A parcel of non-forest land can be converted to forest, plantation, or other treed landscape either through intentional planting or natural regeneration. Note that this applies to land that is not currently in forest land but does not include forest land that is regenerated after harvest as part of forest management (although the estimation approach would be the same for both). As noted above, it is good practice when calculating activity data to distinguish between a land-use change and a land-cover change that may more appropriately be classified as regenerating a forest after harvesting or natural disturbance, which is associated with forest land remaining forest land (see Box 11).

Box 11. Defining land use change from non-forest to forest

Afforestation is the term used to define instances of land use change from non-forest to forest, i.e., the area has not been classified as a forest for an extended period (e.g. 20 years) and is planted or seeded to establish forest land. This is defined as a land use change from non-forest to forest. In other cases, a land area may be defined as a forest, even though it may not be permanently covered by trees, i.e. it may be “temporarily destocked” particularly if it is a ‘working forest’ or plantation managed for periodic harvesting. Such destocked areas are often subsequently reforested, i.e. canopy cover is reestablished either naturally or through deliberate planting or seeding. A forest area may also experience a disturbance event that is followed by (natural) regeneration, such as after an insect outbreak. In both cases of reforestation and natural regeneration, these areas should be considered as forest land remaining forest land and not counted as a land use change from non-forest to forest. In practice, if remote sensing is used to classify lands, detection of tree canopy cover associated with afforestation, reforestation or regeneration may occur some years after establishment (e.g. 5-6 years) and so reporting of the C stock changes may be delayed.

Initially, non-forest land converted to forest land can be identified as a new stratum, or multiple strata if different practices or tree species are established. These additional strata may be tracked separately over time or merged with other forest strata with similar characteristics.

For land that converts from non-forest to forest, generally speaking, the stock of carbon in biomass and dead organic matter will increase over time but at varying rates as the new forest becomes established. Biomass increases predictably as trees and other vegetation are established on the site. Soil organic carbon also changes, but in less predictable ways. For instance, the establishment of a forest plantation on grassland in cool temperate regions may result in a temporary loss of carbon in soil before it builds up again after the plantation is fully established. Removal factors should reflect the variability of changes in stocks of carbon over time. For example, one may choose to apply one factor representing the first 10 years of growth when growth rates are relatively slow, a different factor representing years 10-30 when trees are experiencing fast growth, and a third factor that represents the slower growth of the maturing forest. It is also possible to use an average removal factor for the lifetime of the new forest, recognizing that this selection would mask the variability of actual C removal over time. Additional recommendations for selecting removal factors for re-growing forest land is provided in the guidance document.

For plantations, basic information on site preparation, species selection, and densities of plantings can be used with a projection of the long-term plan for the site to make a reasonable calculation of changes in C stocks. If natural regeneration is the primary means of establishment, then estimates of seedling counts can be used to develop a growth projection, or regional yield tables may be used to estimate projected stocks. The prior use and management of the stratum or land use should also be documented, since the historical use of the land influences carbon stock and stock changes estimates, particularly for soil C. For instance, establishment of a forest stand on a grassland will have a different result in terms of carbon than establishment on a previously row crop agricultural field. For practical reasons, this Protocol considers the land use stratum to be a forest land when the observed characteristics of the stand meet the definition of a forest. Most often this will be when the site is well stocked with trees to the definitional crown cover or tree count.

An example of calculating the change in carbon stock for non-forest land converted to forest land is presented in the box below. This calculation would be repeated for each non-forest land category of interest I converted to forest stratum of interest i as defined by the activity data and any other classification variable determined by a map or statistical reference, such as forest type.

Step 5 – Calculate non-CO₂ emissions (GHG_nonCO2) if appropriate

The main sources of non-CO₂ emissions from forest lands include a) CH₄ and N₂O from biomass burning during prescribed or wild fires and b) soil N₂O emissions with mineral fertilization and organic amendments. Other emissions may also occur, such CH₄ from forested wetlands, but guidance is not provided here because they are typically minor sources in most communities.

Step 5.1: Estimate non-CO₂ emissions from biomass burning using Equation A.11 in the Appendix.

The area of forest lands that are burned may be available from remote sensing products or local mapping of fire extents. The biomass stock for the area could be based on data compiled to estimate biomass C stocks for disturbances (See Step 3). The emission factors, EF_CH4 and EF_N2O, may be based on the values in the 2006 IPCC guidelines (IPCC 2006). The values are 4.7±1.9 kg CH₄ per metric tonne of biomass burned and 0.26±0.07 kg N₂O per metric tonne of biomass burned for temperate forest lands.

Step 5b. Estimate direct and indirect soil N₂O emissions using Equations A.12 through A.16 in the Appendix. Synthetic fertilizer application data may be available from sales data in the community or could be compiled with a survey. Application of organic amendments may be more challenging to compile, particularly if there is large proportion of manure amendments from livestock farms in the area. IPCC (2006) provides methods for estimating manure production that could be used to approximate the amount of manure N available for application to soils. These methods require data on the livestock population in the county and the manure management systems. The N content of fertilizers is available from fertilizer handbooks or manufacturing data. Manure N content depends on the livestock type, and values are provided in the Agricultural Waste Management Field Handbook (USDA 1996).

The emission factors can be based on values provided in the 2006 IPCC guidelines (IPCC 2006). Emission factors could also be based on state-level estimates of implied emission factors from the US National Greenhouse Gas Inventory (US Environmental Protection Agency).

Step 6 – Calculate C stock changes in the HWP pool (if appropriate) (∆C_HWP)

Harvesting wood products removes carbon from the forest ecosystem where it goes through a series of production processes and end-uses, with eventual disposal (and emission to the atmosphere). The harvested C is therefore temporarily stored over a time period, which is a net benefit to the atmosphere (i.e. delayed emissions). Referring to Figure 1 (Section L.0.2), GHG estimation and reporting typically occurs in the “services used by society” section, and involves (1) estimating the carbon that is temporarily stored in wood products; (2) estimating the carbon temporarily stored in solid waste disposal (SWD) sites, mainly landfills; and (3) estimating the GHG benefits that may occur from using wood instead of other energy sources or materials (the “substitution effect” which is mentioned here for reference but note that it is outside the scope of this appendix).

In this Step, we describe the approaches for estimating C additions to and losses from the stock of harvested wood products (∆C_HWP), though we do not provide detailed calculation methods because of their complexity and availability in other references. The companion guidance document describes several estimation tools that may be used for these calculations, and the required input data about harvested wood and type of product that is often readily available. Biomass that enters into the waste stream are included in Appendix F.

The most common method used to estimate GHGs from HWPs is called the Production Approach20, which tracks C in wood that was harvested in the area of interest (i.e. boundaries used for the GHG inventory, Section L.2.1 Delineating the Land Base) regardless of where the wood is consumed.

Guidance here is intended to estimate changes in C in harvested wood using the production approach for two optional accounting methods: (1) estimate the average amount of carbon from the current year’s harvest that remains stored in end uses and landfills over the subsequent 100 years (Hoover et al. 2014), and (2) estimate annual changes in the carbon stored in HWP, accounting for prior year harvests through the current year’s harvest (Stockman et al. 2014). The intent of option 1 is to approximate the average annual climate benefit of withholding carbon from the atmosphere by a certain amount each year for 100 years as described by a “decay” curve. This average benefit is one that can be credited in the year of harvest, thus avoiding the need to keep track of each year’s (or period’s) harvested C indefinitely. The intent of option 2 is to credit cumulative changes in HWP from past harvests, accounting for the fact that it takes a long time for the C in harvested wood products to return to the atmosphere. However, option 2 requires much more detailed and continuous accounting over time, as well as data on harvest quantities and types of wood products produced over at least several decades in the past.

The basic set of calculations that quantify carbon in harvested wood and then through 4 different “fates” is shown in Figure 2, which tracks carbon through the product life cycle from harvest to timber products to primary wood products to end use to disposal, applying best estimates for product ratios and half-lives at each stage. Harvest records are used to distribute annual cut volumes among specific timber product classes (e.g., softwood ties, softwood sawlogs, softwood pulpwood, softwood poles, softwood fuel wood, softwood non-saw, etc.). Timber products are further distributed to specific primary wood products (e.g. softwood lumber, softwood plywood, softwood mill residue used for non-structural panels, etc.) using default average primary product ratios from national level accounting that describe primary products output according to regional forest industry structure (Smith et al. 2006, Appendix A), or alternately, local data. Readers interested in calculating carbon in HWP can refer to Box 8 for a listing of the data requirements, and the link to an estimation tool that follows the calculations outlined in Figure 2.

Box 8.. Estimating carbon in harvested wood products.

Following the schematic shown in Figure 2, calculations require the following data for the production approach. This example is described using the annual accounting method. Similar data is necessary if using the average accounting method, except that only the current year’s data is needed.

• Historical annual harvest data as far back in time as records are available. Units are cubic feet. If data are in other units, conversion factors are available in Stockman et al. (2014).

• Historical timber product data by softwood and hardwood classes -- the proportion of total harvest that went into each timber product class: sawtimber, fuelwood, pulpwood, and other. Regional averages are available if specific data is not.

• Historical end-use data by softwood and hardwood classes – the proportion of total harvest that went into end uses: fuelwood, lumber, non-structural panels, oriented strandboard (OSB), other industrial products, plywood, wood pulp. Regional averages are available if specific data is not.

This data may be used with one of several harvested wood product calculators. More information on calculators and tools available are included in the companion guidance document.

Step 7 – Estimate total net GHG flux from forests over the inventory analysis period and annualize the result into units of t CO₂e/yr

After calculating the total net GHG flux from forests over the inventory period using Equation 2, the estimate is annualized into units of tCO2/yr based on the number of years in the inventory analysis period using the Equation 5 below.

Equation 5: Total net GHG flux from forests, annualized

$$Net\ GHG\ Flux = \frac{\left\lbrack \left( \frac{44}{12} \right) \times {(\mathrm{\Delta}C}_{\text{FRF}} + {\mathrm{\Delta}C}_{F > NF} + {\mathrm{\Delta}C}_{NF > F} + {\mathrm{\Delta}C}_{\text{HWP}}) \right\rbrack + \text{GHG}_{nonCO2}}{T}$$

Where:

Net GHG Flux = net GHG flux from forests over inventory analysis period T (t CO₂e/yr); reflects the net balance of emissions and removals;

ΔC_FRF= change in carbon stocks in forest land remaining forest land over the inventory period; tons C (see Step 4.1)

ΔC_{F > NF}= change in carbon stocks in forest land converted to non-forest land over the inventory period; t C (see Step 4.2)

∆C_NF>F= change in carbon stocks in non-forest land converted to forest land over the inventory period; t C (see Step 4.3)

∆C_HWP = C additions to and losses from the stock of harvested wood products over the inventory period; t C (See Step 6)

GHG_nonCO2 = CH₄ and N₂O emissions from biomass burning during prescribed or wild fires and soil N₂O emissions with mineral fertilization and organic amendments on forest land; t CO2e (See Step 5)

T = total number of years of the inventory analysis period (years)

44/12 = conversion factor to convert units of carbon to carbon dioxide; unitless

This estimate reflects the average net GHG flux over the inventory analysis period and can be compared to estimates generated for previous periods (e.g., the baseline) provided each estimate was developed using consistent methods, data and approaches to ensure comparability.

L.5 Trees Outside Forest Land

This section covers “trees outside forest”, or any trees or areas of tree cover that occupy land not defined as “forest” in the IPCC land use matrix above (section 4.2.3). The trees may therefore fall within any of the yellow boxes in the Table 8 below. For cities, the dark yellow box will likely be the focus since most of the land area with tree cover in cities will be classified as “settlements remaining settlements”.

Table 8: Land use and change matrix

Trees outside forests may be individual trees or trees in small patches embedded in a non-forest land category. Such trees may not be defined in sub-strata the same way as forest trees (e.g. by forest type) and may have only a negligible effect on some carbon pools in the landscape, such as soil carbon. Unlike forests (where the US Forest Service regularly monitors permanent sample plots and makes such data available), data sources may be limited or lacking for calculating emission and removal factors.

It is important to ensure that GHGs are not reported in more than one category. For example, if a community is reporting and accounting for grassland then it should ensure that, when estimating GHGs for “trees outside forest”, these are not counted more than once.

Both woody and herbaceous vegetation are present in land uses outside of forests, but for the purposes of this protocol, “trees outside forests” are considered as the woody perennial vegetation in all non-forest land use classes (croplands, grasslands, wetlands, settlements, other lands). The simplification of including only the woody component of vegetation is because the carbon stored in the woody components of trees makes up the largest compartment of standing biomass stocks and annual biomass increment in non-forest land uses.

Recommended Approach: Estimate GHGs from Trees Outside Forests using the following steps:

Step 1 – Consider the need for further stratification

Trees outside forests occur in different landscapes within a community that could affect the estimation of activity data and/or emission/removal factors; therefore, additional stratification may be desirable for the purposes of setting and monitoring specific policies (see Box 7 above). In addition to estimating GHG fluxes associated with changes in tree cover within the broad IPCC land-use categories typically used for reporting (see Table 8 above), one might consider creating additional strata to track GHG fluxes associated with, e.g. distinct suburban areas or different community park and recreation areas within the settlements class, or different agroforestry systems within the croplands or grassland class.

That said, it is more likely that communities would collapse categories (e.g. combine yellow squares in Table 8 into a single strata), rather than estimate GHGs for all 25 strata or create additional stratification beyond these strata, which would add to the complexity of the GHG estimation. Unlike forest land, additional stratification of non-forest land is unlikely to significantly improve the accuracy or precision of emission/removal factors related to changes in biomass carbon.

Step 2 – Estimate tree cover outside forests / urban canopy (Activity Data)

Sources of activity data for calculating the net GHG balance of trees outside forests included in a community’s GHG inventory are likely to be somewhat different than for forests.

There are two ways to estimate activity data for trees/perennial vegetation outside of forests:

1. on the basis of the number of trees/perennial plants present in the inventory area;

2. on the basis of tree crown or canopy cover

For urban areas classified as settlements, inventory data on individual trees (e.g., street trees) disaggregated into species or broad species classes may be collected by the municipal agencies caring for urban vegetation, either on the ground through field data collection or remotely using very high-resolution aerial or satellite imagery. Tree lists are often developed by inventorying either all trees or a sample of trees in the community, with information collected for each measured tree typically including tree species, diameter at breast height (DBH), tree height, and information about tree health/mortality (if applicable).

For other non-forest land use classes outside of urban areas, or for areas where inventory data are not available for individual trees (this is likely to be a common situation for most communities), canopy cover data may be available from aerial photographs, three-dimensional lidar imagery, or other satellite imagery. In these cases, the area of tree canopy outside of forest lands in the community is estimated. If a community has this type of spatial data and is interested in knowing how canopy is changing within a specific non-forest class (or change class) k, then tree canopy data can be disaggregated accordingly to the separate yellow boxes of the land use and change matrix above (Table 8). Otherwise, it is recommended to track these as a single stratum to simplify the estimation and reporting.

To ensure that community lands are represented consistently in the inventory and to ensure against double counting of trees into more than one land use reporting category (see Section L.2.1), imagery used to delineate tree crown or canopy cover within the community should be overlaid with the map used to delineate IPCC land use and change classes. After this overlay is performed, GHG fluxes associated with all categories of forest and forest-related changes should be addressed using methods in Section L.4 above for Forest Land. GHG fluxes associated with changes in tree canopy that occurred within each remaining non-forest category k should be evaluated using Steps 3-7 below.

Step 3 – Calculate emission and removal factors for trees outside forests

Methods to calculate C stocks and changes for trees are well developed for urban forestry in the U.S., and calculation tools are available to assist the process. See Hoover et al. (2014) for detailed guidance which is summarized here. Usually, only the carbon in trees (including trees roots) is counted, since other carbon pools such as soil C are difficult to associate with individual trees within a non-forest land use category.

Two methods are described here, which can be applied, respectively, to the two methods for estimating activity data above.

The first method applies when activity data are available at the individual tree level through local inventories, where a tree list is developed by inventorying all trees (or a sample of trees) within each defined stratum. Models calibrated to the sampling area are then applied that convert data on tree species, diameter at breast height (DBH), tree height, dieback, crown light exposure, and distance and direction to buildings into estimates of storage, gross annual C sequestration, and net annual C sequestration (which accounts for tree mortality or death), including sampling errors.

The second method applies when activity data are available as estimates of the area of tree canopy outside forests within a community. In this case, removal factors, or average rates of C sequestration per hectare of tree canopy per year, are estimated from inventories of representative areas or by using regional averages. Such estimates can be found in lookup tables for different regions of the U.S. (see Guidance Document).

Step 4 – Calculate C stock change from trees outside forest

Procedures for calculating C stock changes from trees outside forest are slightly different than the simplified gain-loss method for forests: changes in carbon stocks reflect the net balance of the sequestration that occurs where tree canopy is maintained and/or increased and the emissions that occur where loss of tree canopy occurs during the inventory analysis period.

Equation 6: Change in C stock for trees outside forests

ΔC_TOF = ΔC_trees + ΔC_treeloss

$${\mathrm{\Delta}C}_{\text{trees}} = \sum_{k = 1}^{K}{\text{AD}_{trees\_ k} \times \text{RF}_{k} \times T}$$

$${\mathrm{\Delta}C}_{\text{tre}e_{\text{loss}}} = \sum_{k = 1}^{K}{\text{AD}_{treeloss\_ k} \times \text{EF}_{k}}$$

Where:

∆C_TOF = = net GHG flux from trees outside of forests over inventory analysis period T (t CO₂e); reflects the net balance of emissions and removals

AD_{trees_k} = average area of land with tree canopy cover over the inventory analysis period; ha

AD_{treeloss_k} = area of tree cover loss over the inventory analysis period; ha

RF_k = average sequestration rate of trees in non-forest category k, t C/ha/yr

EF_k = average emission factor from loss of trees in non-forest category k; t C/ha

k = 1,2,3, …, K non-forest strata

T= number of years in the inventory period; years

In some cases, activity data may be available only for one point in time, such that changes in tree canopy cannot be calculated. In these cases, only ∆Ctrees can be calculated for the current inventory analysis period based on the average tree canopy cover; emissions associated with losses of tree cover should be considered during the next inventory period once these data are available.

An example of calculating the change in carbon stocks for trees outside forest is presented in the box below.

Step 5 – Calculate non-CO₂ emissions. if appropriate

The main sources of non-CO₂ emissions from non-forest lands include a) CH₄ and N₂O from biomass burning during prescribed or wild fires, and b) soil N₂O emissions with mineral fertilization and organic amendments. Other emissions may also occur, such CH₄ from rice cultivation, but guidance is not provided here because they are typically minor sources in most communities.

Step 5a: Estimate non-CO₂ emissions from biomass burning using Equation A.11 in the Appendix.

The area of non-forest lands that are burned may be available from remote sensing products or local mapping of fire extents. The biomass stock for the area could be based on data compiled to estimate biomass C stocks for disturbances for trees outside of forest (See Step 3). Biomass stocks for cropland residue burning may be derived with local data and methods in the US National Greenhouse Gas Inventory (US-EPA 2017), and grasslands could be based on defaults provided in the 2006 IPCC guidelines (IPCC 2006). The emission factors, EF_CH4 and EF_N2O, may be based on the values in the 2006 IPCC guidelines (IPCC 2006). The values are 4.7±1.9 kg CH4 per metric tonne of biomass burned and 0.26±0.07 kg N2O per metric tonne of biomass burned for temperate forest lands.

Step 5b. Estimate direct and indirect soil N₂O emissions using equations A.13 through A.17 in the Appendix.

Synthetic fertilizer application data may be available from sales data in the community or could be compiled with a survey. Application of organic amendments may be more challenging to compile, particularly if there is large amount of manure amendments from livestock farms in the area. IPCC (2006) provides methods for estimating manure production that could be used to approximate the amount of manure N available for application to soils. These methods require data on the livestock population in the county and the manure management systems.

N content of fertilizers is available from fertilizer handbooks or manufacturing data. Manure N content depends on the livestock type, and values are provided in the Agricultural Waste Management Field Handbook (USDA 1996).

The emission factors may be based on values provided in the 2006 IPCC guidelines (IPCC 2006). Emission factors could also be based on state-level estimates of implied emission factors from the US National Greenhouse Gas Inventory (US-EPA 2017).

Step 6 - Calculate C stock changes in the HWP pool, if appropriate

Urban and suburban areas are sources of wood for various uses including fuelwood and mulch, often associated with tree maintenance. Roundwood products may also be provided from harvested trees outside forests. Therefore, it is likely to be important to keep track of the carbon sequestered or emitted in harvested wood products, similar to forests.

Methods to estimate carbon in harvested wood products and landfills from trees outside forests are nearly identical to methods described for forests in step 6 of section L.4. One significant difference, particularly for urban trees, is that some of the harvested wood will be in the form of yard waste and trimmings from tree maintenance. Some of this material may be mulched or directly deposited in landfills or dumps rather than following one of the more lengthy disposition pathways typical of harvests from forests. For assessing harvests of trees outside forests, different sources of data for estimating retention times for the stored carbon will be needed. Referring to Figure 2 in section L.4, yard waste and trimmings not used for roundwood products would enter the calculations in the boxes labelled landfills, dumps, or compost, as appropriate.

Step 7 – Estimate total net GHG flux from Trees Outside Forests over the inventory analysis period and annualize the result into units of tCO₂e/yr

After calculating the total net GHG flux from trees over the inventory period, the estimate is annualized into units of tCO₂/yr based on the number of years in the inventory analysis period as:

Equation 7: Total net GHG flux from trees outside forests, annualized

$$Net\ GHG\ Flux = \frac{\left\lbrack \left( \frac{44}{12} \right) \times {(\mathrm{\Delta}C}_{\text{trees}} + {\mathrm{\Delta}C}_{\text{treeloss}}) \right\rbrack + \text{GHG}_{nonCO2}}{T}$$

Where:

Net GHG Flux = net GHG flux from trees outside forests over inventory analysis period T (tCO₂e/yr); reflects the net balance of emissions and removals;

T = total number of years of the inventory analysis period (years)

44/12 = conversion factor to convert units of carbon to carbon dioxide; unitless

This estimate reflects the average net GHG flux over the inventory analysis period and can be compared to estimates generated for previous periods (e.g., the baseline) provided each estimate was developed using consistent methods, data and approaches to ensure comparability.

Annex: IPCC Gain-Loss Method for Estimating GHGs from Forests

The IPCC gain-loss method is described in this annex (IPCC 2006). This method is based on estimating the gains in C depending on the age of forests in a community, and the losses from harvest, fuelwood gathering and stand-replacing disturbance. A community may decide to use this method if there is sufficient data on these activities along with the age distribution for forests in the community.

The overall net GHG emissions or removals for all land uses is expressed as the sum of fluxes occurring within the six individual IPCC land use classes identified above (settlements, forest lands, croplands, grasslands, wetlands, other lands). However, in this annex, the equations are only focused on the gains and losses associated with forest lands and conversions to/from forest lands, which is consistent with the methods in the appendix.

For each forest land use category i (Forest Land Remaining Forest Land, Land Converted to Forest Land and Forest Land Converted to other land uses), net GHG emissions or removals are estimated with the following equation.

GHG_i= ΔCS_i + NonCO2_i (Equation A.1)

Where:

GHG_i = total greenhouse gas emissions for land use i, metric tonnes CO₂ eq.

ΔCS_i = net carbon stock change in pools for land use i, metric tonnes CO₂ eq.

NonCO2_i = non-CO₂ greenhouse gas emissions for land use i, i.e., CH₄ or N₂O, metric tonnes CO₂ eq.

I = settlements, forest lands, croplands, grasslands, wetland, and other lands

Carbon Stock Change

Change in carbon stocks for each land use (ΔCS_i) is estimated with the following equation.

ΔCS_i = ΔBC_i + ΔDOM_i + ΔSOC_i+ ΔHWP_i (Equation A.2)

Where:

ΔBC_i = change in biomass carbon stocks for land use i, metric tonnes CO₂ eq.

ΔDOM_i = change in dead organic matter carbon stocks for land use i, metric tonnes CO₂ eq.

ΔSOC_i = change in soil organic carbon stocks for land use i, metric tonnes CO₂ eq.

ΔHWP_i = change in carbon stocks in harvested wood products for land use i, metric tonnes CO₂ eq.

For Forest Land Remaining Forest Land and Land Converted to Forest Land, change in biomass carbon stocks for each land use (ΔBC_i) is estimated using the gain-loss method with the following equation.

ΔBC_i = (CG_i*A_i) − (CL_H+CL_F+CL_D) * 44/12 (Equation A.3)

Where:

CG_i = increase in biomass carbon stocks for land use i, metric tonnes C per ha

A_i = area for corresponding increase in biomass carbon stock for land use i, ha

CL_H = carbon losses from harvest of wood for land use i, metric tonnes C

CL_F = carbon losses from fuelwood gathering for land use i, metric tonnes C

CL_D = carbon losses from disturbance for land use i, metric tonnes C

44/12 is a conversion from C to CO₂

The carbon gain in biomass (CG_i) is estimated with the following equation.

CG_i = AGI_i * (1+R_i) * CF (Equation A.4)

Where:

AGI_i = growth increment in the aboveground biomass, metric tonnes C per ha

R_i = ratio of belowground biomass to aboveground biomass, metric tonnes of aboveground biomass per metric tonne of belowground biomass

CF = carbon fraction of the biomass, metric tonnes of C per metric tonne of biomass

The carbon loss from harvest (CL_H) is estimated with the following equation.

CL_H = H_t * D_t * BEF_t * (1+R_i) * CF (Equation A.5)

Where:

H_t = volume of harvest for tree type t associated with land use i, m³

D_t = density of the wood for tree type t, dry weight metric tonnes of biomass per m³

BEF_t = biomass expansion factor from the harvested portion of the tree to the entire tree including branches and leaves for tree type t associated with land use i, m³

R_i = ratio of belowground biomass to aboveground biomass, metric tonnes of aboveground biomass per metric tonne of belowground biomass

CF = carbon fraction of the biomass, metric tonnes of C per metric tonne of biomass

The carbon loss from fuelwood gathering (CL_F) is estimated with the following equation.

CL_F = {[W_t*D_t*BEF_t*(1+R_i)]+[P_t*D_t]} * CF (Equation A.6)

Where:

FG_t = volume of whole tree harvesting for fuelwood by tree type t associated with land use i, m³

D_t = density of the wood for tree type t, dry weight metric tonnes of biomass per m³

BEF_t = biomass expansion factor from the harvested portion of the tree to the entire tree including branches and leaves for tree type t associated with land use i, m³

R_i = ratio of belowground biomass to aboveground biomass, metric tonnes of aboveground biomass per metric tonne of belowground biomass

P_t = volume of partial tree harvesting for fuelwood by tree type t associated with land use i in which the entire tree is not killed, m³

CF = carbon fraction of the biomass, metric tonnes of C per metric tonne of biomass

The carbon loss from disturbances (CL_D), such as fires, diseases, pest outbreaks, hurricanes, other events that destroy a majority of the trees, is estimated using the following equation.

CL_D = S_i * (1+R_i) * CF * FL_i * A_i (Equation A.7)

Where:

S_i = standing stock of biomass in the area of the disturbance for land use i, metric tonnes per ha

R_i = ratio of belowground biomass to aboveground biomass for land use i, metric tonnes of aboveground biomass per metric tonne of belowground biomass

CF = carbon fraction of the biomass, metric tonnes of C per metric tonne of biomass

FL_i = fraction of biomass lost due to the disturbance for land use i, dimensionless

A_i = area affected by disturbances for land use i, ha

Change in biomass carbon stocks for each land use (ΔHWP_i) is estimated using the production approach, as discussed in the main section of the appendix.

For Forest Land Converted to other land uses (i.e., deforestation), change in biomass carbon stocks for each land use (ΔBC_i) is estimated with the following equation.

ΔBC_i = [ΔCcon_i+ (CG_i*A_i) − (CL_H+CL_F+CL_D)] * 44/12 (Equation A.8)

Where:

ΔCcon_i = initial change in C for land use conversion i, metric tonnes C

CG_i = increase in biomass carbon stocks for land use conversion i, metric tonnes C per ha

A_i = area for corresponding increase in biomass carbon stock for land use conversion i, ha

CL_H = carbon losses from harvest of wood for land use conversion i, metric tonnes C

CL_F = carbon losses from fuelwood gathering for land use conversion i, metric tonnes C

CL_D = carbon losses from disturbance for land use conversion i, metric tonnes C

44/12 is a conversion from C to CO₂

The initial change in C during the year of conversion for land use conversion I (ΔCON_i) is estimated with the following equation:

ΔCcon_i = [(Bafter_i−Bbefore_i) *A_i] * CF (Equation A.9)

Where:

Bafter_i = biomass stock after conversion for land use conversion i, metric tonnes C

Bbefore_i = biomass stock before conversion for land use conversion i, metric tonnes C

A_i = area for conversion during the inventory year or an average area over the analysis time period for land use conversion i, ha

Estimate CG_i, CL_H, CL_F, and CL_D, using equations 5, 6, 7, and 8, respectively.

Non-CO₂ Emissions

Non-CO₂ greenhouse gas emissions (NonCO2_i) are estimated for soil N₂O emissions associated with synthetic fertilization and organic amendments, and the CH₄ and N₂O emissions from burning of biomass in fires. There are other sources of N that impact soil N₂O emissions, but this guidance has focused on the two main practices with the goal of providing methods that are more tractable for communities. Communities may refer to 2006 IPCC guidelines for additional guidance if they decide to include additional practices (IPCC 2006). Similarly, the method associated with biomass burning has been simplified. Trace gases are released as biomass burns, referred to as precursors, that undergo other reactions in the atmosphere to form greenhouse gases. The estimation precursors has not been included in this guidance, but may be estimated using the 2006 IPCC guidelines (IPCC 2006).

Total Non-CO₂ emissions (NonCO2_i) are estimated using the following equation:

NonCO2_i = BB_i + NM_i (Equation A.10)

Where:

BB_i = total biomass burning emissions for land use i, metric tonnes CO₂ eq.

NM_i = total soil nitrous oxide emissions from soil nitrogen management for land use i, metric tonnes CO₂ eq.

Total emissions from biomass burning (BB_i) are estimated using the following equation.

BB_i= A_i * M_i * CB_i * [(EF_CH4*GWP_CH4)+(EF_N2O*GWP_N2O)] * 10 ^{− 3} (Equation A.11)

Where:

A_i = area affected by disturbances for land use i, ha

M_i= biomass stock in the area affected by the disturbance for land use i, metric tonnes dry matter per ha

CB_i = proportion of biomass oxidized in the area affected by the disturbance for land use i, dimensionless

EF_CH4 = emission factor for CH₄, kg CH₄ emitted per metric tonne dry matter burned

GWP_CH4 = global warming potential of CH₄ over a 100-year time horizon, kg CO₂ eq. per kg CH₄

EF_N2O = emission factor for N₂O, kg N₂O emitted per metric tonne dry matter burned

GWP_N2O = global warming potential of N₂O over a 100-year time horizon, kg CO₂ eq. per kg N₂O

The multiplier of 10 ^{− 3} is a conversion from kg CO₂ eq. to metric tonnes CO₂ eq.

Total emissions from soil nitrogen management (NM_i) are estimated for direct and indirect soil N₂O using the following equation.

NM_i= (DE_i+IE_i) * GWP_N2O (Equation A.12)

Where:

DE_i = direct soil N₂O emissions for land use i, metric tonnes N₂O

IE_i = total indirect soil N₂O emissions for land use i, metric tonnes N₂O

GWP_N2O = global warming potential of N₂O over a 100-year time horizon, metric tonne CO₂ eq. per metric tonne N₂O

Direct soil N₂O emissions (DE_i) are estimated with the following equation.

DE_i= [(SN_i*N_SN*EF_SN)+(OA_i*N_OA*EF_OA)] * 44/28 (Equation A.13)

EF_SN = N₂O emission factor for synthetic fertilizer applied to soils, kg N2O-N per kg N applied

OA_i = amount of organic amendments applied to soils for land use i, kg dry matter applied

N_OA = proportion of N in organic amendments, kg N per kg dry matter applied

EF_OA = N₂O emission factor for organic amendments applied to soils, kg N2O-N per kg N applied

44/28 is a conversion from N₂O-N to N₂O

Total indirect soil N2O emissions (IE_i) are estimated for volatilized losses of N (NO_x and NH₃) at the site of application plus the losses of N through water flows, leaching and runoff (NO₃^- and organic forms of N), at the site of application using the following equation.

IE_i= IEV_i + IEW_i (Equation A.14)

IEV_i = indirect soil N₂O emissions from volatilization of N at the site of application for land use i, metric tonnes N₂O

IEW_i = indirect soil N₂O emissions with N losses in water flows (leaching and runoff) at the site of application for land use i, metric tonnes N₂O

Indirect soil N₂O emissions (IEV_i) from volatilized losses of N are estimated with the following equation.

IEV_i= [(SN_i*N_SN*FV_SN * EFV_SN)+(OA_i*N_OA*FV_OA*EFV_OA)] * 44/28 (Equation A.15)

Where:

SN_i = amount of synthetic fertilizer applied to soils for land use i, kg fertilizer applied

N_SN = proportion of N in synthetic fertilizer, kg N per kg synthetic fertilizer

FV_SN = fraction of synthetic N applied to soils that is volatilized from the site of application, dimensionless

EFV_SN = indirect N₂O emission factor for synthetic fertilizer N applied to soils that is lost through volatilization from the site of application, kg N₂O-N per kg N applied

OA_i = amount of organic amendments applied to soils for land use i, kg dry matter applied

N_OA = proportion of N in organic amendments, kg N per kg dry matter applied

FV_OA = fraction of N in organic amendments applied to soils that is volatilized from the site of application, dimensionless

EFV_OA = indirect N₂O emission factor for N in organic amendments applied to soils that is lost through volatilization from the site of application, kg N₂O-N per kg N applied

44/28 is a conversion from N₂O-N to N₂O

Indirect soil N₂O emissions (IEW_i) from losses of N in water flows (leaching and runoff) are estimated with the following equation.

IEW_i= [(SN_i*N_SN*FW_SN * EFW_SN)+(OA_i*N_OA*FV_OA*EFW_OA)] * 44/28 (Equation A.16)

Where:

SN_i = amount of synthetic fertilizer applied to soils for land use i, kg fertilizer applied

N_SN = proportion of N in synthetic fertilizer, kg N per kg synthetic fertilizer

FV_SN = fraction of synthetic N applied to soils that is lost in water flows (runoff and leaching) from the site of application, dimensionless

EFW_SN = indirect N₂O emission factor for synthetic fertilizer N applied to soils that is lost in water flows (runoff and leaching) from the site of application, kg N₂O-N per kg N applied

OA_i = amount of organic amendments applied to soils for land use i, kg dry matter applied

N_OA = proportion of N in organic amendments, kg N per kg dry matter applied

FV_OA = fraction of N in organic amendments applied to soils that is volatilized from the site of application, dimensionless

EFW_OA = N₂O emission factor for N in organic amendments applied to soils that is lost in water flows (runoff and leaching) from the site of application, kg N₂O-N per kg N applied

44/28 is a conversion from N₂O-N to N₂O

References

Birdsey, Richard, Philip Duffy, Carolyn Smyth, Werner A. Kurz, Alexa J. Dugan, Richard Houghton. 2018. Climate, Economic, and Environmental Impacts of Producing Wood for Bioenergy. Environmental Research Letters 13, 050201. https://doi.org/10.1088/1748-9326/aab9d5

Birdsey, Richard A. 2006. Carbon accounting rules and guidelines for the United States forest sector. Journal of Environmental Quality. 35[Not issued]: 1518-1524.

Climate Action Reserve. 2018. Forest Project Protocol. Available at: http://www.climateactionreserve.org/how/protocols/forest/

DOE. 1992. Technical Guidelines for Voluntary Reporting of Greenhouse Gas Program Chapter 1, Emission Inventories, Part I, Appendix: Forestry. Washington DC: Office of Policy and International Affairs United States Department of Energy

Department of Energy. 2005. Enhancing DOE’s voluntary reporting of greenhouse gases (1605(b)) program. http://www.pi.energy.gov/enhancingGHGregistry/.

Gillenwater, Michael. 2012. What is additionality? Greenhouse Gas Institute, Discussion Paper No. 001. Available at: http://ghginstitute.org/wp-content/uploads/2015/04/AdditionalityPaper_Part-1ver3FINAL.pdf

Goetz S, Dubayah R: Advances in remote sensing technology and implications for measuring and monitoring forest carbon stocks and change. Carbon Manag 2(3), 231-244 (2011).

Hairiah, Kurniatun, SM Sitompul, Meine van Noordwijk and Cheryl Palm. 2001. Methods for sampling carbon stocks above and below ground. International Centre for Research in Agroforestry. Available at: http://www.worldagroforestry.org/downloads/Publications/PDFS/LN01035.pdf

Heath, L.S., J.E. Smith, K.E. Skog, D.J. Nowak, et al. 2011. Managed forest carbon estimates for the US greenhouse gas inventory, 1990-2008. Journal of Forestry 109 (3):167-173.

Hoover, C.M., ed. 2008. Field Measurements for Forest Carbon Monitoring: A Landscape-Scale Approach. New York, NY: Springer.

Hoover, C., R. Birdsey, B. Goines, P. Lahm, Gregg Marland, D. Nowak, S. Prisley, E. Reinhardt, K. Skog, D. Skole, J. Smith, C. Trettin, C. Woodall, 2014. Chapter 6: Quantifying Greenhouse Gas Sources and Sinks in Managed Forest Systems. In Quantifying Greenhouse Gas Fluxes in Agriculture and Forestry: Methods for Entity‐Scale Inventory. Technical Bulletin Number 1939, Office of the Chief Economist, U.S. Department of Agriculture, Washington, DC. 606 pages. July 2014. Eve, M., D. Pape, M. Flugge, R. Steele, D. Man, M. Riley‐Gilbert, and S. Biggar, Eds.

G. Hurtt, M. Zhao, R. Sahajpal, A. Armstrong, R. Birdsey, E. Campbell, K. Dolan, R. Dubayah, J. Fisk, S. Flanagan, C. Huang, W. Huang, K. Johnson, R. Lamb, L. Ma, R. Marks, D. O’Leary III, J. O’Neil-Dunne, A. Swatantran, H. Tang. Beyond MRV: High-resolution forest carbon modeling for climate mitigation planning over MD, USA.

IPCC. 2003. Good Practice Guidance for Land Use, Land-Use Change and Forestry. Kanagawa, Japan.: Intergovernmental Panel on Climate Change. IPCC National Greenhouse Gas Inventories Programme. http://www.ipcc-nggip.iges.or.jp/public/gpglulucf/gpglulucf.html.

IPCC. 2006. 2006 IPCC Guidelines for National Greenhouse Gas Inventories, Prepared by the National Greenhouse Gas Inventories Programme. Edited by H. S. Eggleston, L. Buendia, K. Miwa, T. Ngara and K. Tanabe. Japan: IGES. http://www.ipcc-nggip.iges.or.jp/public/2006gl/vol4.html

Jenkins, J.C., D.C. Chojnacky, L.S. Heath, and R.A. Birdsey. 2003a. Comprehensive database of diameter-based biomass regressions for North American tree species. Newtown Square, PA: U. S. Department of Agriculture, Forest Service, Northeast Research Station.

Jenkins, J.C., D.C. Chojnacky, L.S. Heath, and R.A. Birdsey. 2003b. National-Scale Biomass Estimators for United States Tree Species. Forest Science 49 (1):12-35

McPherson, E. Gregory; van Doorn, Natalie S.; Peper, Paula J. 2016. Urban Tree Database and Allometric Equations. Gen. Tech. Rep. PSW-GTR-253. Albany, CA: U.S. Department of Agriculture, Forest Service, Pacific Southwest Research Station. 86 p.

McPherson, E.G., and P.J. Peper. 2012. Urban tree growth modeling. Arboriculture & Urban Forestry 38:175–183.

Nowak, D.J., E.J. Greenfield, R. Hoehn, and E. LaPoint. 2013. Carbon storage and sequestration by trees in urban and community areas of the United States. Environmental Pollution, 178:229‐236.

Nowak, David J.; Walton, Jeffrey T.; Baldwin, James; Bond, Jerry. 2015. Simple street tree sampling. Arboriculture and Urban Forestry. 41(6): 346-354.

Pearson, T.R.H., S.L. Brown, and R.A. Birdsey. 2007. Measurement guidelines for the sequestration of forest carbon. Newtown Square, PA: US Department of Agriculture, Forest Service, Northern Research Station.

Prisley, S.P., and M. J. Mortimer. 2004. A synthesis of literature on evaluation of models for policy applications, with implications for carbon accounting. For. Ecol. And Mgt., 198(1-3): 89-103.

Sathre R, O’Connor J. Meta-analysis of greenhouse gas displacement factors of wood product substitution Environ. Sci. Policy. 2010;13:104–114.

Skog, K.E. 2008. Sequestration of carbon in harvested wood products for the United States. Forest Products Journal 58 (6):56-72.

Smith, J.E., L.S. Heath, and J.C. Jenkins. 2003. Forest volume-to-biomass models and estimates of mass for live and standing dead trees of U.S. forests. Newtown Square, PA: U.S. Department of Agriculture, Forest Service, Northeastern Research Station.

Smith, J.E., L.S. Heath, K.E. Skog, and R.A. Birdsey. 2006. Methods for calculating forest ecosystem and harvested carbon with standard estimates for forest types of the United States. Newtown Square, PA: US Department of Agriculture, Forest Service, Northern Research Station.

Smyth CE, Stinson G, Neilson E, Lemprière, TC, Hafer M, Rampley GJ, Kurz WA. Quantifying the biophysical climate change mitigation potential of Canada’s forest sector. Biogeosciences. 2014;3515-3529.

Stockmann, K., Anderson, N., Young, J., Skog, K., Healey, S., Loeffler, D., Butler, E., Jones, J. G. & Morrison, J. 2014. Estimates of carbon stored in harvested wood products from United States Forest Service Northern Region, 1906-2012. Missoula, MT: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Forestry Sciences Laboratory. 27 p.

USDA (1996) Agricultural Waste Management Field Handbook, National Engineering Handbook (NEH), Part 651. Natural Resources Conservation Service, U.S. Department of Agriculture. July 1996.

USDA Forest Service. 2010a. Forest Inventory and Analysis field methods for phase 3 measurements. Version 5.0: U.S. Department of Agriculture, Forest Service. Unpublished information on file at http://www.fia.fs.fed.us/library/field-guides-methods-proc/.

USDA Forest Service. 2010b. Forest Inventory and Analysis National Core Field Guide: field data collection procedures for phase 2 plots. Version 5.0: Unpublished information on file at http://fia.fs.fed.us/library/field-guides-methods-proc/

US-EPA (2017). Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2015. EPA 430-P-17-001, US-Environmental Protection Agency, Washington, D.C.

Craig A. Wayson, Kris D. Johnson, Jason A. Cole, Marcela I. Olguín, Oswaldo I. Carrillo, et al.. Estimating uncertainty of allometric biomass equations with incomplete fit error information using a pseudo-data approach: methods. Annals of Forest Science, Springer Verlag/EDP Sciences, 2015, 72 (6), pp.825-834.