Landfill Emissions Data: What the Numbers Can and Cannot Tell You
A synthetic documentary-style landscape shows dark gas collection pipes and wellheads on a grassy landfill slope, with an unbranded portable case near the foreground, service tracks, a distant waste area and vegetation beneath a partly cloudy sky. No legible measurement display, facility identifier, customer logo, annual inventory value or verification statement appears. Equipment presence does not quantify captured gas, prove emissions performance or demonstrate customer data ingestion. This is not an identified real site.
Diagram text description
Landfill emissions data is useful only when the number stays connected to its source, period, method and physical coverage. A facility-reported annual estimate, a landfill gas project record and a methane plume observed during a short visit describe different things. Putting them in one spreadsheet does not make them interchangeable measurements.
Start by asking what decision the evidence must support. Identifying sites for follow-up, estimating a reporting-year inventory and evaluating a customer's waste treatment require different information. A strong observation can reveal an important issue without providing a complete annual facility total. A technically detailed database can describe infrastructure without measuring the methane actually emitted.
The practical task is reconciliation: preserve each record, establish identities and comparable boundaries, and leave unresolved matches explicit. This guide uses invented records to demonstrate that process. It does not calculate real landfill methane, rank actual operators or convert a public facility total into a product or customer footprint.
A data guide, not another landfill calculation method
Data interpretation begins with an existing record and asks what it represents. Calculation begins with a defined method and the activity inputs needed to produce a result. Confusing the two encourages an analyst to treat any available landfill number as an emissions factor, even when its units or coverage make that use inappropriate.
For interpretation, inspect the publisher, dataset purpose, represented period, gas, unit, method, geographical footprint and known exclusions. Determine whether the field is a measured rate, an annual estimate, a stock of waste, project capacity or an operational status. These are different variables, not alternative formats for a single emissions total.
Only after those checks should a calculation method be selected for a new reporting task. A customer's waste estimate needs its own treatment evidence; a facility-level investigation needs a defensible site match and suitable observations. Do not fill gaps in one dataset with a convenient value from another merely because both rows contain the word “landfill.” Preserve the original meaning before deciding whether transformation or aggregation is justified.
Know which landfills are in each dataset
Check the population behind each dataset before comparing counts, averages or totals. EPA's Greenhouse Gas Reporting Program, or GHGRP, covers applicable reporting facilities; its landfill overview distinguishes municipal solid waste and industrial landfill categories with different eligibility conditions. A missing row can reflect dataset scope or reporting eligibility rather than an absence of emissions.
EPA's Landfill Methane Outreach Program, or LMOP, serves a different purpose. Its database concerns municipal solid waste landfills and landfill gas energy projects. EPA explicitly describes it as non-exhaustive and says it does not cover industrial or hazardous waste landfills. It combines several source types, and not every field is updated annually. It should not be treated as an identical reporting population or a complete national landfill census.
Read the represented year and release information on the actual download. A page accessed recently can still display an older reporting-year map or database release. Keep eligibility questions separate from the analytical coverage check: this guide does not determine an actual site's current legal reporting obligation. For an inventory of sites, document the included landfill types, geography, status filters and missing population before drawing conclusions about a larger region.
Reported models and measured methane are different evidence
Reported inventories and remote observations offer complementary evidence, but their outputs need to remain distinct. An annual reported estimate may combine modelling assumptions, waste information and gas collection or control records. It is not simply a continuous direct measurement of every emission source across a facility.
| Evidence family | Typical output | Interpretation limit |
|---|---|---|
| Annual facility reporting | A defined reporting-year estimate under specified methods and boundaries | Depends on inputs and treatment of generation, collection, control and other assumptions |
| Technical or project database | Facility attributes, waste stocks, project status or capacity | A project or infrastructure field is not necessarily an emissions measurement |
| Remote methane observation | Detection and sometimes a quantified rate for a sampled source footprint | Represents particular observation conditions and times, not automatically all sources or the full year |
Table 1, shown one row at a time. Every field is retained.
Row 1
- Evidence family
- Annual facility reporting
- Typical output
- A defined reporting-year estimate under specified methods and boundaries
- Interpretation limit
- Depends on inputs and treatment of generation, collection, control and other assumptions
Row 2
- Evidence family
- Technical or project database
- Typical output
- Facility attributes, waste stocks, project status or capacity
- Interpretation limit
- A project or infrastructure field is not necessarily an emissions measurement
Row 3
- Evidence family
- Remote methane observation
- Typical output
- Detection and sometimes a quantified rate for a sampled source footprint
- Interpretation limit
- Represents particular observation conditions and times, not automatically all sources or the full year
Carbon Mapper's landfill study summary discusses repeated airborne observations and differences from traditional reporting estimates. It supports investigating those differences, not replacing every annual record with one observed plume or applying a universal correction ratio to all sites. A remote estimate also has methodological uncertainty. Retain the observation's detection conditions and spatial attribution so a reviewer can judge what additional evidence it contributes.
Why a detected plume is not an annual inventory
A detected plume shows that an observable source was emitting under the conditions of that observation. It does not establish how long the source operated, whether the measured rate stayed constant or what other parts of the landfill emitted. Conversely, a nondetection is not proof of zero emissions: sensitivity, wind, coverage and the timing of a visit affect what can be detected.
Separate four steps: detection, repeat visits, evidence about persistence, and justified integration over time. Repeated observations can strengthen evidence of a recurring source, but a schedule of visits still samples rather than observes every hour. Estimating annual emissions requires an explicit method for temporal coverage, uncertainty and sources outside the detected plume footprint.
Do not multiply one short-visit rate by every hour in a year and label the result a measured annual inventory. That calculation assumes continuous persistence and representative coverage that the visit has not established. Instead, retain the rate and observation window, request revisits or supporting operational records, and distinguish the observed source from an inferred facility-wide total. The appropriate follow-up depends on the question, not on which extrapolation produces the preferred answer.
Match reporting year, unit and facility identity
Fictional reconciliation example: an analyst holds three teaching records for a place described as Rivermoor. All names, identifiers, dates and quantities below are invented; shorthand keys are not official EPA or Carbon Mapper identifiers. No actual public dataset rows were joined. The first task is to identify each field's meaning before comparing a number.
| Record and identity | Represented period and revision | Value and unit | Method and coverage |
|---|---|---|---|
| Annual record A; reporting key G104; Rivermoor site | Calendar year 2024; extract A-v1 issued in 2025 | 1,200 metric tonnes CH4 for that year | Fictional annual reporting model; specified facility reporting boundary, not a plume measurement |
| Technical/project record B; landfill L77 with project P21; River Moor East | Waste stock dated 31 December 2024; project metadata dated June 2025; extract B-v2 | 1,500,000 metric tonnes of waste in place | Fictional technical attribute repeated in a project composite; not a methane quantity or rate |
| Observation C; source S51; candidate location near G104 | One twenty-minute pass on 16 June 2025; observation C-v1 | Estimated 400 kg CH4 per hour during the pass | Fictional remotely observed point-source rate; whole-facility coverage and persistence unestablished |
Table 2, shown one row at a time. Every field is retained.
Row 1
- Record and identity
- Annual record A; reporting key G104; Rivermoor site
- Represented period and revision
- Calendar year 2024; extract A-v1 issued in 2025
- Value and unit
- 1,200 metric tonnes CH4 for that year
- Method and coverage
- Fictional annual reporting model; specified facility reporting boundary, not a plume measurement
Row 2
- Record and identity
- Technical/project record B; landfill L77 with project P21; River Moor East
- Represented period and revision
- Waste stock dated 31 December 2024; project metadata dated June 2025; extract B-v2
- Value and unit
- 1,500,000 metric tonnes of waste in place
- Method and coverage
- Fictional technical attribute repeated in a project composite; not a methane quantity or rate
Row 3
- Record and identity
- Observation C; source S51; candidate location near G104
- Represented period and revision
- One twenty-minute pass on 16 June 2025; observation C-v1
- Value and unit
- Estimated 400 kg CH4 per hour during the pass
- Method and coverage
- Fictional remotely observed point-source rate; whole-facility coverage and persistence unestablished
Keep the represented period separate from publication or revision. A-v1's issue in 2025 does not make its annual value a 2025 measurement. B-v2 mixes a dated stock attribute with later project metadata; neither is an annual methane observation. C-v1 estimates a rate during a short 2025 pass, not the emissions mass over that calendar year.
These records cannot be summed: waste mass is not methane mass, and a short-period rate is not an annual emissions total. Converting A's tonnes to kilograms would change its unit scale, not resolve its different period, method or footprint relative to C. If a separate source reports CO2e, retain its gases, warming-potential basis and time horizon rather than silently treating it as CH4.
Identity is also unresolved. Similar names and nearby coordinates do not establish that L77 and G104 have the same permitted footprint or reporting boundary. The observation needs a defensible source-to-facility attribution too. Keep candidate relationships separate from confirmed keys, preserving all original record versions.
Open decision: obtain the source cross-reference, operator or permit details and mapped facility boundary needed to establish whether L77, G104 and S51 describe the same site and covered sources. Consequence: do not join them into one facility total, annualise C-v1 or choose an apparent disagreement as proof of under-reporting while that match and coverage remain unresolved.
Avoid double counting project and landfill rows
Fictional duplicate-row example: continue the invented Rivermoor records, not actual LMOP downloads. Two project rows both name landfill L77, while an annual model and a later observation remain candidate cross-program matches. Their distinct periods and units must survive the identity check; repeated facility attributes do not become additional physical quantities.
| Fictional record | Period and method | Quantity or status | Aggregation treatment |
|---|---|---|---|
| L77 / project P21 | Waste stock at 31 December 2024; project metadata June 2025 | 1,500,000 metric tonnes waste in place; project operational | One facility attribute plus one project record |
| L77 / project P22 | Same dated waste-stock attribute; project metadata June 2025 | Same 1,500,000 metric tonnes; different project planned for 2026 | Do not add the repeated waste stock or count another landfill |
| G104 / annual A-v1 | Calendar 2024 reporting-model estimate | 1,200 metric tonnes CH4/year | Separate inventory variable; L77 crosswalk unconfirmed |
| S51 / observation C-v1 | Twenty-minute pass on 16 June 2025; point-source estimate | 400 kg CH4/hour; persistence unknown | Separate observation, not an extra annual total |
Table 3, shown one row at a time. Every field is retained.
Row 1
- Fictional record
- L77 / project P21
- Period and method
- Waste stock at 31 December 2024; project metadata June 2025
- Quantity or status
- 1,500,000 metric tonnes waste in place; project operational
- Aggregation treatment
- One facility attribute plus one project record
Row 2
- Fictional record
- L77 / project P22
- Period and method
- Same dated waste-stock attribute; project metadata June 2025
- Quantity or status
- Same 1,500,000 metric tonnes; different project planned for 2026
- Aggregation treatment
- Do not add the repeated waste stock or count another landfill
Row 3
- Fictional record
- G104 / annual A-v1
- Period and method
- Calendar 2024 reporting-model estimate
- Quantity or status
- 1,200 metric tonnes CH4/year
- Aggregation treatment
- Separate inventory variable; L77 crosswalk unconfirmed
Row 4
- Fictional record
- S51 / observation C-v1
- Period and method
- Twenty-minute pass on 16 June 2025; point-source estimate
- Quantity or status
- 400 kg CH4/hour; persistence unknown
- Aggregation treatment
- Separate observation, not an extra annual total
Within the fictional L77 rows, the shared confirmed landfill key identifies one landfill with two distinct projects. Adding the repeated stock would incorrectly produce 3,000,000 tonnes, while counting rows would produce two landfills. Keep project-specific status separate: one operational project and one planned project are not two currently operating projects. The diagram shows the one-to-many relationship, not a flow of gas or additive emissions.
One enclosing outline labelled Landfill L77 contains two separate boxes labelled Project P21 and Project P22. Containment represents one-to-many record identity, not a chronological process, physical gas flow or additive emissions. A facility-level waste-stock field repeated in both project rows remains one facility attribute. Distinct project status must still be retained. This diagram does not establish a match between L77, reporting key G104 and observed source S51; that cross-program relationship remains unresolved. All identifiers and records are fictional teaching examples, not actual EPA data or verified methane measurements.
Diagram text description
EPA warns against summing its composite file directly because landfill or project data can repeat. For real work, select the publisher's appropriate landfill-only or operational-project aggregate where available, and inspect the key and field definitions before aggregation. Do not merely remove duplicate names, because distinct facilities can share names and different projects can share a landfill.
Open decision: confirm the actual source identifier cross-reference for G104 and L77, checking the publisher's facility key, address, operator, permit and mapped footprint where needed. Consequence: leave the cross-program match pending; preserve A-v1 and C-v1 separately and do not select a convenient value or combine their incompatible quantities. A confirmed within-file landfill/project relationship does not settle that wider identity question.
What to do when reported and observed numbers disagree
Investigate disagreement in a fixed order. First establish that the records refer to the same facility footprint and a relevant common period. Then check gas, unit, method, temporal coverage and revision. Resolve simple identity or dimensional mismatches before interpreting an apparent difference as a physical change.
Next inspect the reporting assumptions and observation conditions. Waste age and composition, collection or control treatment and the spatial sources represented can affect an annual estimate. A remotely observed plume may relate to a particular location or operating condition. Review attribution, weather information, detection limits, revisit coverage and available operational records without asserting a cause the evidence has not established.
Choose follow-up that can distinguish plausible explanations: corrected identifiers, a revised source record, a model-input review, repeat observation or a site investigation by the appropriate specialists. Preserve both the original values and the reconciliation outcome. Do not average incompatible records to make the disagreement disappear, and do not choose the higher or lower value solely because it supports a reporting or commercial objective.
Build a reusable landfill data record
Create a reusable record with stable identity and explicit provenance. Retain the original download or observation reference, then store normalised analytical fields separately. That allows later corrections without silently rewriting what the publisher supplied.
| Record group | Fields to preserve |
|---|---|
| Identity | Publisher's facility and project keys, names, address, coordinates, mapped footprint and match confidence |
| Provenance | Publisher, programme, source release, file or observation reference and original field definitions |
| Time and method | Represented period, observation window, publication/revision date, model or measurement method |
| Quantity and coverage | Gas, unit, warming-potential basis when relevant, spatial/source boundary, detection limits and exclusions |
| Review and change | Reviewer, unresolved questions, approved analytical use, previous version and update trigger |
Table 4, shown one row at a time. Every field is retained.
Row 1
- Record group
- Identity
- Fields to preserve
- Publisher's facility and project keys, names, address, coordinates, mapped footprint and match confidence
Row 2
- Record group
- Provenance
- Fields to preserve
- Publisher, programme, source release, file or observation reference and original field definitions
Row 3
- Record group
- Time and method
- Fields to preserve
- Represented period, observation window, publication/revision date, model or measurement method
Row 4
- Record group
- Quantity and coverage
- Fields to preserve
- Gas, unit, warming-potential basis when relevant, spatial/source boundary, detection limits and exclusions
Row 5
- Record group
- Review and change
- Fields to preserve
- Reviewer, unresolved questions, approved analytical use, previous version and update trigger
For the fictional records, store L77's project relationships separately from the pending G104 match. A reviewer should be able to see why two project rows became one facility attribute and why S51 remained an observation. Refresh the record when a publisher revises data or an identity check changes, preserving the previous release and explaining which conclusions must be reconsidered.
Keep facility totals separate from waste-stream estimates
A facility total cannot automatically become one customer's waste footprint. A landfill receives multiple waste streams over many years, and methane generated in a reporting year can reflect historical deposits. Dividing a facility's annual methane by one current customer's delivered tonnage would mix populations and time boundaries unless a suitable method and evidence justify that allocation.
For a customer inventory, establish waste material, quantity, collection and treatment route, destination, relevant period and the accounting method. Confirm whether a factor or operator result represents the appropriate treatment and gas-management conditions. A transfer-station receipt alone may not establish the final landfill, and a nearby remotely detected plume is not an emissions factor for every delivery there.
Use public facility data to frame evidence requests or investigate inconsistencies, not to manufacture customer-specific precision. Where route or allocation evidence is missing, retain the gap and select an appropriately qualified estimation approach under the chosen inventory method. Keep any wider claims about avoided emissions or waste-prevention benefits separate from attributing the facility's reported methane.
Prepare a traceable waste-data handoff
Connect product and supplier records to the questions an analyst must answer about waste data. Material descriptions, SKU revisions and source documents can help establish what waste was generated and which records belong together. They do not prove a landfill identity, validate a methane measurement or establish a live facility-data feed.
Prepare a neutral handoff containing the relevant material and waste references, available treatment documents, candidate facility identifiers, source versions and unresolved questions. Ask who can confirm final destination and who is responsible for reviewing the selected accounting method. Keep that preparation role distinct from specialist site investigation, remote-sensing interpretation and independent assurance.
Before relying on any tool or service, confirm its actual functions and supported data sources. A structured product record is not evidence of landfill telemetry or measured-methane ingestion. The useful endpoint is a traceable, qualified decision: which record can support this purpose, which relationship is confirmed and what evidence is still required before the number is reused.
Waste & wastewater
