A site carbon estimate starts as a quantity takeoff multiplied by emission factors. Expected storage in planting and soil is estimated separately. Report it beside the emissions, as well as in any net figure, because it accrues slowly and can be lost. What decides the result is which life-cycle stages are counted, over how many years, and which factor stands behind each line.
Below are the method, the factors we could read on November 13, 2026, a worked example and the limits. This is design-stage guidance. It is not a certified life-cycle assessment or engineering advice.
The stages, and what an EPD usually covers
The AIA and Carbon Leadership Forum toolkit groups a life cycle into product (A1–A3), construction (A4–A5), use (B1–B7) and end of life (C1–C4), following EN 15978 and ISO 21930. Its best practice for embodied carbon is to include at least A1–A5, B2–B4 and C1–C4. It warns against comparing material types, wood and concrete for example, without stages B and C. It advises reporting biogenic carbon storage and recycling benefits separately in Module D, because they fall outside the building life cycle. Product EPDs prepared under ISO 21930 handle biogenic carbon differently: uptake and release are declared within the life-cycle modules, as the lumber EPD below shows. Check which convention a source uses before you add its numbers. AIA-CLF Toolkit, Part II
The University of California’s guide defines A1 as raw material supply, A2 transport, A3 manufacturing, A4 transport to the construction site and A5 construction and installation. It notes that an environmental product declaration (EPD) most often provides A1–A3, and that such an LCA is most useful for relative comparison or order of magnitude. UC, Life Cycle Analysis The sources used below bear that out. The concrete and lumber EPDs declare A1–A3, and NAPA’s asphalt benchmark is built from A1–A3 data. The lumber EPD also reports biogenic carbon for A5 and C3/C4. The PVC pipe declaration is cradle-to-grave with an A1–A3 summary.
An EPD factor alone gives the product stage. Delivery, placement, repair and demolition are separate lines, and the estimate should say which it includes.
Time horizons do not match
The tools and studies a landscape architect will meet use different periods:
- Pathfinder: 60-year project life for replacements and operations. Pathfinder 3.3 User Guide
- SITES Pilot Credit 3: CO2 over a 50-year lifespan. The project team sets the baseline, either existing conditions or the initial design. SITES Pilot Credit 3
- EPA’s tree factor: the first 10 years after planting. EPA Equivalencies
- McPherson’s Los Angeles tree study: 50 years. McPherson et al. 2015
A slab’s product-stage emissions occur before opening day; a tree’s storage accrues over decades and can be lost. A comparison needs one stated period.
Material factors
Concrete. NRMCA’s industry-average EPD, Version 4 (October 9, 2026), replaces Version 3.2. It gives A1–A3 per cubic metre of fresh concrete, US average, on AR5 100-year values, from 1,238 facilities. Multiply by 0.7645549 for a US cubic yard. NRMCA EPD V4
| Mix (normal weight), kg CO2e/m³, A1–A3 | No SCM | 30% fly ash | 50% slag |
|---|---|---|---|
| 2,500 psi, air-entrained | 272.59 | 220.69 | 182.58 |
| 3,000 psi, non-air-entrained | 276.34 | 222.10 | 183.10 |
| 4,000 psi, non-air-entrained | 330.17 | 263.83 | 216.12 |
| 4,000 psi, air-entrained | 376.21 | 301.74 | 247.06 |
For the 3,000 psi non-air-entrained, no-SCM mix, facilities range from 242.21 to 328.94 kg CO2e/m³. Carbonation is not counted. The facility range for this one mix (242 to 329 kg/m³) is wider than the step from 3,000 to 4,000 psi (276 to 330), so a plant-specific EPD can matter as much as the strength class.
Asphalt. NAPA’s benchmark deliberately avoids a single national number. It gives national benchmarks regionalized by state, because A1 (materials) values vary significantly across states. For California mixes with nominal maximum aggregate size of 12.5 mm or less, the report’s example A1–A3 benchmarks are 47.24 (20th percentile), 53.55 (40th), 55.36 (median) and 58.42 (average) kg CO2e per tonne. These come from Appendix 6, “Example National GWP Benchmarks”, Table 26, where A1 is taken from published EPDs and A2 and A3 from benchmark data. Treat them as illustrative. The table’s subtitle reads “A1 from BM Mixes”, which does not match its title. NAPA SIP-108
Aggregate. The same NAPA report lists aggregate at 1.94 kg CO2e/tonne (USLCI, prescribed), with forthcoming values of 5.32 for explosive mining and 3.76 for non-explosive. That is a 2.7-fold spread within one document. We found no primary factor for decomposed granite.
Steel. Buy Clean California sets maximum acceptable values, effective January 1, 2025, for unfabricated steel, A1–A3, per metric ton: reinforcing bar 755 kg CO2e, hot-rolled sections 1,010, hollow structural sections 1,710 and plate 1,490. These are procurement limits set at the industry average of facility EPDs. They cap what a compliant product may declare. They do not measure the steel on your job. Buy Clean California (read through a page summary)
Lumber and biogenic carbon. The North American softwood lumber EPD reports 63.12 kg CO2e/m³ for A1–A3. A cubic metre holds 230.09 kg of carbon, or 843.66 kg CO2e, and the EPD shows the biogenic uptake and release netting to zero within the declared modules. Do not subtract the 843.66 kg from a footprint. The EPD reports the same total with and without biogenic carbon. Under the PCR’s landfill addendum, with an assumed 100% landfill end of life, the EPD reports 708.33 kg CO2e/m³ as permanent sequestration net of landfill gas. That figure depends on the end of life, which a design-stage estimate cannot know, so report it separately and name the assumption. It expired in July 2025, and we did not read the regional replacements. AWC/CWC Lumber EPD
PVC pipe, as a proxy. The Uni-Bell EPD’s A1–A3 summary gives 9.10 × 10³ kg CO2e per 1,000 ft of 8-inch DR 18 pipe weighing 3,974 kg, which our arithmetic puts at about 2.3 kg CO2e per kg of pipe, A1–A3. These are municipal pipes. We found no EPD for Schedule 40, HDPE or drip tubing, so label this as a stand-in by weight. Uni-Bell PVC EPD
Two widely used datasets are left out. The ICE database is now an educational edition whose licence excludes project calculations, tools and consultancy work. Circular Ecology CLF’s material baselines sit behind a download form, so we did not read their values. CLF Baselines
Transport and fuel
EPA’s emission factors hub gives the numbers most estimates need for delivery and equipment: 8.78 kg CO2 per gallon of gasoline and 10.21 per gallon of diesel, combustion only; 0.186 kg CO2 per short ton-mile for medium- and heavy-duty trucks; and construction diesel equipment at 1.01 g CH4 and 0.94 g N2O per gallon. Installation (A5) is fuel gallons times these factors; we found no published per-area fuel rate for landscape work. EPA GHG Emission Factors Hub
Name the grid factor you use. Three EPA pages give three US average electricity factors: 771.5 lb CO2/MWh in the hub (eGRID2023, January 2025), 823.1 on the equivalencies page (eGRID 2022), and 767.2 on the current eGRID summary (eGRID2023, Revision 2, June 2025), which gives 428.5 for California (CAMX). For CAMX the hub gives 436.7 lb, which we use below. EPA GHG Emission Factors Hub · EPA Equivalencies · eGRID summary
Maintenance
Turf. The most-quoted lawn study, from Irvine, California, was corrected after publication. Corrected fuel emissions are 122 g CO2 m⁻² yr⁻¹; the paper first reported 1,469. Corrected fuel emissions equal about 24% of the soil carbon stored by ornamental lawns. With fuel, fertilizer and irrigation counted, ornamental lawns came to −108 g CO2 m⁻² yr⁻¹ at low fertilizer and +285 at high fertilizer. Athletic fields, which stored no soil carbon, came to +405 to +798. Townsend-Small and Czimczik, Correction Management sets the sign.
Fertilizer N2O. The IPCC’s aggregated default for direct emissions is 0.010 kg N2O–N per kg N applied (range 0.002–0.018). Disaggregated, synthetic fertilizer in wet climates is 0.016 and all inputs in dry climates 0.005. IPCC 2019 Refinement, Table 11.1 Convert with 44/28 and the AR5 GWP of 265 used in the EPA hub. EPA GHG Emission Factors Hub Indirect N2O from volatilisation and leaching is not included here. These are agricultural inventory defaults. Our interpretation, which the table does not state, is that an irrigated landscape in a dry region may behave more like the wet case. As a check, 10 g N/m² at the aggregated default gives about 42 g CO2e/m²; Irvine reported 45 (range about 20 to 153).
Irrigation energy. The California Energy Commission’s 2005 report gives 12,700 kWh per million gallons in Southern California and 3,950 in the north, including 2,500 for wastewater treatment. Outdoor irrigation produces no wastewater, so we use 10,200 and 1,450. At 436.7 lb/MWh, Southern California comes to about 0.53 kg CO2 per m³ of water. CEC 2005 The energy data are 21 years old. McPherson and colleagues report 0.56 kg CO2 per kL for the Los Angeles Department of Water and Power, which agrees closely. McPherson et al. 2015
Equipment. We found no primary per-hour factor for a mower or blower. Estimate gallons of fuel per year and multiply by the hub factors; for small gasoline engines the hub also gives CH4 and N2O per gallon. EPA GHG Emission Factors Hub
Sequestration and storage
Storage is a stock; sequestration is a rate. Storage is lost when a tree dies or decays.
Per unit of canopy. From field data in 28 cities and 6 states, Nowak and colleagues found US urban trees store on average 7.69 kg C per m² of tree cover, with gross sequestration of 0.277 and net sequestration of 0.205 kg C m⁻² yr⁻¹. Storage density ranged from 3.14 to 14.1 kg C/m² across cities and states. These are carbon units; multiply by 44/12 for CO2. Nowak et al. 2013
Single trees. The urban tree database, from more than 14,000 trees in 17 cities, converts above-ground dry weight to CO2 as weight × 1.28 (roots) × 0.5 (carbon) × 3.67. Its worked example is a sweetgum in the Northern California Coast region, 33 years after planting, with a d.b.h. of 42.2 cm estimated from the regional growth equation, storing 1,220.89 kg CO2. The text gives the height as 15.3 m and the printed formula uses 15.1; an errata sheet exists that we did not open. Regional differences are large: silver maple at age 40 holds 378.1 kg C in the Northeast and 4,505.6 in the Midwest. Sweetgum on the Southern California Coast reaches 27.4 cm d.b.h. at age 30, against 38.9 cm on the Northern California Coast, and holds 238.6 kg C at age 60. McPherson et al. 2016
Per planted tree. EPA uses 0.060 metric ton CO2 per tree per year, averaged over the first 10 years for a 15-gallon tree, with survival of 68% at 5 years and 59% at 10 built in. EPA Equivalencies
A full tree life cycle. McPherson, Kendall and Albers followed one California sycamore in Los Angeles over 50 years. It stored 7.798 t CO2, with a 95% prediction interval of 1.5 to 21.1 t. Irrigation over 50 years cost 0.82 to 1.34 t CO2, or 10.6–17.2% of storage. The fate of the wood at removal cost 24.5–78.5%; their mulch scenario treats all stored above-ground CO2 as released. Scenarios ranged from a net removal of 3.768 t to a net emission of 1.204 t. Nursery production of a #15 tree was 15.3 kg CO2. McPherson et al. 2015
Soil. Golf turf near Denver and Fort Collins accumulated up to about 0.9 t C/ha/yr on fairways, continuing for up to 45 years on greens. USDA ARS summary Development lowered urban soil carbon in some cities and raised it in others. Pouyat et al. 2006 We read summaries only. EPA’s compost factor is a waste-diversion estimate whose modelled soil storage falls from about 0.30 to 0.07 t CO2e per wet ton over 24 years; it is not a factor for purchased compost on a site. EPA WARM v16 We found no primary per-area rate for shrubs, perennials, meadows or green roofs.
The tools
- Pathfinder 3.3 is a free LCA tool for exterior projects with Materials, Plants, Operations and Benefits tabs. Totals include transport, installation and 60 years of replacements; custom materials can come from an EPD or EC3. The guide reports the Climate Positive Design Challenge targets, set in 2019, as years to offset a project’s own emissions: 5 for parks, gardens and campuses; 20 for plazas and streetscapes; 25 for infrastructure. Its methodology report is behind a form, so we cannot say what factors it uses. Pathfinder 3.3 User Guide ASLA’s climate plan asks Pathfinder-logged projects for 50–65% emission reductions and doubled sequestration by 2030; the passages we read do not state the baseline. ASLA Climate Action Plan
- i-Tree Eco takes field tree data and reports carbon storage and sequestration; i-Tree Canopy uses the Nowak 2013 multipliers. Both i-Tree pages state that a USDA Forest Service policy change has indefinitely halted traditional i-Tree funding. i-Tree Eco · i-Tree methods
- EC3 is a free EPD database listing more than 50,000 EPDs, for comparing specific products. EC3
A worked example: the scale of a patio and of three trees
This compares two design options for one area. It is not a like-for-like material comparison: a patio and three trees do not do the same job, and the patio is counted for stages A1–A4 only. AIA-CLF advises against comparing material types without stages B and C, so read the result as a check on scale only.
For 100 m² (about 1,076 ft²) in Southern California, how does a cast-in-place concrete patio compare with three trees? Field Method assumptions, not sourced: a 100 mm (about 4 in) slab of 3,000 psi non-air-entrained concrete, which suits a frost-free site (use the strength and air content your specification and jurisdiction require), a 150 mm aggregate base at 2.0 t/m³, a 20-mile (32 km) haul, no reinforcement, 150 m² of eventual canopy, and 30 years of potable irrigation. We use a 30-year period from planting. Where a source gives another period, the line says so.
Patio, stages A1–A4 only
| Line | Arithmetic | kg CO2e |
|---|---|---|
| Concrete, 3,000 psi, no SCM | 10 m³ (13.1 yd³) × 276.34 (NRMCA V4) | 2,763 |
| Same, 30% fly ash | 10 × 222.10 | 2,221 |
| Aggregate base, A1 | 30 t × 1.94 to 5.32 (NAPA Table 7) | 58 to 160 |
| Concrete delivery, A4 | 10 m³ × 2,340 kg/m³ (NRMCA V4 fresh density) = 23.4 t = 25.8 short tons × 20 mi × 0.186 kg CO2/short ton-mile (EPA Hub Table 8) | 96 |
| Total, no SCM | about 2,920 to 3,020 | |
| Total, 30% fly ash | about 2,380 to 2,480 |
Not counted: placement (A5), reinforcement, formwork, sealer, base delivery, repair and replacement (B) and demolition (C). The facility range for the mix moves the concrete line by about −12% to +19%. The 4,000 psi air-entrained mix (376.21 kg/m³) would raise the concrete line to about 3,760 kg. We use only the CO2 term of EPA’s fleet-average medium- and heavy-duty truck factor; CH4 and N2O add little.
Three trees
| Line | Arithmetic | t CO2 |
|---|---|---|
| EPA, first 10 years, survival included | 3 × 0.060 × 10 | 1.8 |
| Storage 33 years after planting, no mortality, Northern California Coast growth | 3 × 1,220.89 kg (sweetgum, McPherson 2016) | 3.66 |
| Storage at age 60, no mortality, Southern California Coast growth | 3 × 238.6 kg C × 44/12 (sweetgum, McPherson 2016) | 2.6 |
| Canopy, gross, per year once mature | 150 m² × 0.277 kg C × 44/12 (Nowak 2013) | 0.152 |
| Canopy, net, per year once mature | 150 × 0.205 × 44/12 | 0.113 |
| Nursery production | 3 × 15.3 kg (range 10.45–25.75) | 0.046 |
| Irrigation, 30 years | 3 × 0.82 to 1.34 t × 30/50 (McPherson 2015) | 1.5 to 2.4 |
The 3.66 t line is the report’s worked example for the Northern California Coast, at 33 years, 3 years beyond our period. It is not a Southern California value. The same report shows sweetgum on the Southern California Coast growing more slowly: 27.4 cm d.b.h. at age 30 against 38.9 cm, and 238.6 kg C (about 875 kg CO2) per tree at age 60. For a Southern California site, treat 3.66 t as an upper value and use the regional equations for the species you specify. The EPA line covers 10 years and the canopy lines are annual rates, so neither adds up to the 30-year period. They are shown as checks on scale. The 30/50 irrigation proration is our simplification, borrowed from a Los Angeles sycamore, and is the weakest line. We show no 30-year canopy total: canopy starts near zero and 150 m² is assumed.
Optional lawn, 100 m² for 30 years (Irvine data). Net −108 to +285 g m⁻² yr⁻¹ × 100 m² × 30 years gives −324 to +855 kg CO2e. Fuel, 366 kg, is inside that figure. It assumes soil carbon accrues for all 30 years.
The lines are not on one basis. Concrete is GWP-total on AR5 100-year values. The EPA truck factor as used here and the tree studies count CO2 only. The lumber EPD uses TRACI 2.1. In the lawn figure a negative value is a net removal. In the tree table storage is shown as a positive number. State these differences when you report a total.
What the example shows. With Northern California Coast growth rates and no losses, three trees could store about 3.7 t CO2 by 33 years after planting, the same order as the patio’s 2.4 to 3.0 t. With Southern California Coast growth, the same report gives about 2.6 t for three trees even at age 60, below the patio. That assumes every tree survives; EPA’s factor assumes 59% survival at 10 years. It leaves out the ground treatment under the trees. Storage without irrigation is not supported by these sources: in our interpretation, the growth data come from urban California trees that are typically irrigated. With 30 years of potable irrigation, net storage falls to roughly 1.2 to 2.1 t using Northern California growth, and could be near zero or negative using Southern California growth. If the wood is chipped for mulch at removal, the Los Angeles study treats the above-ground storage as released. The patio’s emissions happen before opening day; the trees’ storage builds over decades and can be reversed. In our interpretation, that timing difference is a further reason not to treat the two totals as equal. Mix design moves the patio by about 0.5 t, and strength and air content can move it by about 1 t. Region, survival, irrigation and the fate of the wood decide whether the trees ever match it.
Popular claims we could not support
- “A mature tree absorbs 48 lb (22 kg) of CO2 a year.” No primary source found. EPA’s figure is 0.060 t per tree per year, first 10 years.
- “About 75% of a landscape project’s emissions are embodied in materials.” Not in the ASLA plan or Pathfinder guide.
- “Lawns are net emitters.” The corrected Irvine study shows the sign depends on fertilizer and use.
- “Wood is carbon negative.” Within A1–A3 the lumber EPD treats biogenic carbon as net neutral and reports 63.12 kg CO2e/m³ of fossil emissions. A net removal appears only under an assumed landfill end of life (708.33 kg CO2e/m³), so the claim depends on what happens to the wood.
- Per-area sequestration for shrubs, meadows and green roofs, and factors for pavers, stone and decomposed granite: none found.
Open questions
- What factors does Pathfinder use, and over what stages, once its methodology report is read?
- What do the CLF 2026 baselines say for pavers and brick?
- Is there a defensible irrigation rate per tree outside Los Angeles?
Using this in Field Method
Material Studio measures areas from a model, multiplies them by the thickness you enter for volumes, and traces each quantity. It does not calculate carbon. Pair its quantities with the factors above, and record the source, stage and year of each.
Sources
- AIA and Carbon Leadership Forum — Embodied Carbon Toolkit for Architects, Part II: Measuring Embodied Carbon
- University of California — Life Cycle Analysis, Phases, and EPDs
- Sustainable SITES Initiative — Pilot Credit 3: Assess and Improve Site Carbon Performance
- ASLA — Climate Action Plan 2022–2025
- NRMCA — Member Industry-Average Ready Mixed Concrete EPD, Version 4 (2026)
- NAPA — SIP-108, EPD Benchmark for Asphalt Mixtures, Version 2.0 (2024)
- California Department of General Services — Buy Clean California Act
- American Wood Council and Canadian Wood Council — North American Softwood Lumber EPD (2020)
- Uni-Bell PVC Pipe Association — EPD for PVC and PVCO Pressure Pipe and PVC Non-Pressure Pipe (2023)
- US EPA — Emission Factors for Greenhouse Gas Inventories (2025)
- US EPA — Greenhouse Gas Equivalencies Calculator: Calculations and References
- US EPA — eGRID Summary Data
- Townsend-Small and Czimczik (2010) — Correction to Carbon sequestration and greenhouse gas emissions in urban turf
- IPCC — 2019 Refinement to the 2006 Guidelines, Volume 4, Chapter 11: N2O Emissions from Managed Soils
- California Energy Commission — California's Water–Energy Relationship, CEC-700-2005-011-SF (2005)
- McPherson, Kendall and Albers (2015) — Life cycle assessment of carbon dioxide for different arboricultural practices in Los Angeles, CA
- Nowak, Greenfield, Hoehn and Lapoint (2013) — Carbon storage and sequestration by trees in urban and community areas of the United States
- McPherson, van Doorn and Peper (2016) — Urban tree database and allometric equations, PSW-GTR-253
- US EPA — Documentation for GHG Emission and Energy Factors Used in WARM, Management Practices Chapters, Version 16
- USDA ARS — Assessing soil carbon sequestration in turfgrass systems using long-term soil testing data
- Pouyat, Yesilonis and Nowak (2006) — Carbon storage by urban soils in the United States
- Climate Positive Design — Pathfinder 3.3 User Guide
- i-Tree — i-Tree Eco
- i-Tree — Methods and Files
- Building Transparency — EC3
- Circular Ecology — Embodied Carbon: The ICE Database
- Carbon Leadership Forum — North American Material Baselines
Try a relevant tool
These tools support site review and design studies. Read each tool’s stated capabilities and limits before use.
Explore Material Studio