Scope 1 and Scope 2 emissions: the difference, and why it decides your data collection

Scope 1 and Scope 2 emissions are the two halves of a company's own energy story. Scope 1 covers direct emissions from sources the company owns or controls: the gas it burns, the fuel in its vehicles, the refrigerant that leaks from its chillers. Scope 2 covers the indirect emissions from the electricity, steam, heat and cooling it purchases. Together they are the part of a greenhouse gas inventory a company can measure almost entirely from its own records, which is why every reporting regime expects them first, and expects them to be accurate.

That one-paragraph definition hides most of the decisions that make a first inventory slow. This guide walks through the boundary between the two scopes, the places where classification goes wrong, the two ways Scope 2 is calculated, and what an auditor will eventually expect from both. If you want the full three-scope picture first, start with our overview of scope 1 2 3 emissions and the deep dive into Scope 3, which is where most of a footprint usually lives.

Scope 1: what you burn, leak and drive

The GHG Protocol Corporate Standard defines Scope 1 as direct GHG emissions from sources owned or controlled by the reporting company. In practice, four buckets cover nearly every mid-market company:

• Stationary combustion. Natural gas in boilers and furnaces, heating oil, propane for process heat, diesel in backup generators. If fuel is delivered to your site and burned in your equipment, it is Scope 1.
• Mobile combustion. Fuel in vehicles the company owns or leases and operates: delivery vans, sales cars, forklifts, yard trucks. The fuel-card statement is usually the best source document.
• Fugitive emissions. Refrigerant leaks from HVAC and cold storage. These look tiny in dollars and are huge in CO2e: common refrigerants have global warming potentials in the thousands, so a 6 kg top-up of R-404A is roughly 23 tonnes of CO2e. The service invoice that says "recharge" is the tell.
• Process emissions. Emissions from chemistry rather than combustion: cement calcination, certain metal and chemical processes. Most companies outside heavy industry have none.

The measurement rule for Scope 1 is simple and strict: quantity times factor. Liters of diesel times the combustion factor for diesel, therms of gas times the factor for natural gas, kilograms of refrigerant times its global warming potential. Combustion chemistry does not vary by country, so these factors are stable and public. There is no defensible reason for a mature inventory to estimate Scope 1 from spend; the quantities are on the invoices.

Scope 2: the energy someone else generates for you

Scope 2 is purchased energy: electricity above all, plus district heat, steam and chilled water where they exist. The emissions happen at the power plant, not at your site, but they exist because of your demand, so the Protocol gives them their own scope rather than folding them into the general value chain. That separation matters because purchased energy is one of the few emission sources a company can change quickly, by contract, without touching its operations.

Scope 2 has a wrinkle no other scope has: it is calculated twice, under two methods that both appear in a compliant disclosure.

• Location-based. Your kWh times the average emission intensity of the grid you draw from. This answers: what does the physical grid emit on your behalf? It ignores any green contracts you hold.
• Market-based. Your kWh times the factor of the electricity you contractually purchased: renewable power purchase agreements, green tariffs, unbundled certificates, or, absent any of those, the residual mix. This answers: what does your procurement choice emit?

The dual reporting exists because each method alone can mislead. A company on a coal-heavy grid with genuine renewable contracts looks bad under location-based alone; a company waving cheap certificates over the same coal grid looks misleadingly clean under market-based alone. CSRD's ESRS E1 asks for both, and an auditor will ask to see the contractual instruments behind any market-based claim.

The boundary questions that cause real arguments

Almost every classification dispute between Scope 1, Scope 2 and Scope 3 reduces to two questions: who owns or controls the emitting asset, and who holds the energy contract. Some recurring cases:

• The leased office. You lease a floor and the landlord holds the utility contract, with energy bundled into the service charge. Your electricity is not Scope 2, because you did not purchase it; it is Scope 3, category 8, upstream leased assets. The day you take over the meter contract, it moves to Scope 2. Same office, same electrons, different scope.
• The leased vehicle. An operating-leased van that your drivers fuel with your fuel card: the fuel is your Scope 1 under the operational-control approach, because you operate the asset. The van's manufacture is the lessor's problem and your category 8.
• The contractor's truck. A carrier hauls your freight in their truck with their fuel: their Scope 1, your Scope 3 category 4. The moment you buy the truck and hire the driver, it becomes your Scope 1. Outsourcing moves emissions between scopes without removing a single tonne.
• Employee cars. An employee driving their own car to the office is category 7, commuting. The same employee driving the same car to a client with mileage reimbursement is category 6, business travel. Neither is ever Scope 1, because the company does not control the asset.
• Rooftop solar. Power you generate and consume from your own panels is not Scope 2 (you purchased nothing) and produces no Scope 1 combustion. It simply reduces purchased kWh, and it only supports a market-based renewable claim if you retain the certificates.

Worked line-by-line versions of these calls, for three different company types, are in our post on scope 1 2 3 emissions examples.

Where the data comes from

Scope 1 and 2 are the accounts-payable-plus-meters part of the inventory. The practical source documents: twelve months of utility invoices per meter (kWh, therms, kWh of district heat), fuel-card statements by vehicle, bulk fuel delivery notes, HVAC service invoices for refrigerant quantities, and the lease register to settle who holds which energy contract. None of this requires a questionnaire; all of it already exists in finance systems.

That is also why Scope 1 and 2 are where auditors start. The evidence is documentary, the factors are public, and the calculation is arithmetic. What assurance providers look for is not cleverness but completeness and lineage: every meter accounted for, every figure traceable to an invoice, every factor cited with its source and vintage. A spreadsheet can hold the numbers; what it cannot easily hold, a year and a staff change later, is the trail. Keeping that trail is the entire argument for treating this as accounting, which is the design premise of our carbon accounting software.

Typical shares, and the trap in them

For an office-based company, Scopes 1 and 2 together are commonly 5 to 15% of the total footprint. For light manufacturing, 15 to 40%. For a logistics operator running its own fleet, Scope 1 alone can be the single biggest line. The trap is reading those shares as importance rankings and stopping there. Regulators and customers expect Scope 1 and 2 to be solid precisely because they are measurable, and a company that gets them wrong forfeits credibility on everything else. Meanwhile the other 70 to 90% sits in Scope 3, where the methods are coarser and the data is other people's. A sane first-year plan: measure Scopes 1 and 2 from meters completely, screen Scope 3 from spend completely, and let the screening tell you where activity data is worth collecting next.

A worked micro-example

Take a 120-person company with one office and a small fleet. Twelve months of records show: 82,400 kWh of electricity, 6,200 therms of gas, 4,100 gallons of gasoline across six cars, and one 4 kg R-410A recharge. The arithmetic: 82,400 kWh × 0.38 kg/kWh (US grid average) is about 31.3 tonnes, Scope 2 location-based. 6,200 therms × 5.31 kg/therm is about 32.9 tonnes, Scope 1. 4,100 gallons × 8.89 kg/gallon is about 36.4 tonnes, Scope 1. The recharge: 4 kg × a GWP of 2,088 is about 8.4 tonnes, Scope 1, from one line item on a maintenance invoice most inventories miss. Total: roughly 109 tonnes, of which the refrigerant line, invisible in dollar terms, is nearly 8%.

That is the pattern to internalise: Scope 1 and 2 accuracy is not about better estimation, it is about not missing lines. The misses hide in maintenance invoices, bulk deliveries and meters nobody owns. A classification pass over the whole AP export, the thing our demo does in about a minute, is the cheapest way to find them, because the export is complete even when your mental model is not.

Refrigerants: the small print that moves the total

The worked example above put one refrigerant recharge at nearly 8% of a company's footprint, which is reason enough to give the line its own section. The driver is global warming potential. Emission factors for fuels sit in single digits of kilograms of CO2e per unit; GWPs for common refrigerants sit in the thousands. R-410A, the standard charge in most commercial air conditioning installed over the last two decades, carries a GWP of roughly 2,088. R-404A, common in cold storage and supermarket refrigeration, sits near 3,922: a single 10 kg top-up is about 39 tonnes of CO2e, more than many small companies' entire vehicle fleet emits in a year. Newer low-GWP alternatives such as R-32, at roughly a third of R-410A, are exactly that, alternatives: they appear only where equipment has been replaced recently.

The accounting mechanics are forgiving once you know where to look. What you report is the refrigerant that escaped, and the practical proxy is the refrigerant that was added, because a recharge replaces what leaked. The source documents are HVAC and refrigeration service invoices, which state the refrigerant type and the kilograms added. Nobody circulates these invoices for carbon reasons; they arrive as routine maintenance costs, often a few hundred dollars, and no finance process flags them. That mismatch between dollar size and tonne size is why fugitive emissions are the most commonly missed Scope 1 line, and why a classification pass over the full AP export, the kind the Scope Classifier demo runs, catches them where a source-by-source mental checklist does not.

There is also a direction of travel worth knowing. High-GWP HFCs are being phased down internationally under the Kigali Amendment, and regional rules such as the EU F-gas regulation restrict them further. In practice that means the refrigerants in older equipment become scarcer and more expensive to service, replacement equipment ships with lower-GWP charges, and a refrigerant line that is material today should shrink over the coming decade for reasons partly outside your control. Until then, read every "recharge" invoice as a Scope 1 event.

Grid factors and why your Scope 2 moves without you doing anything

Location-based Scope 2 is your kWh times an average grid factor, and the second term is not under your control. As grids retire coal plants and add renewables, the average intensity of most grids falls year over year. The consequence is odd for accountants used to stable units of measure: a company can consume exactly the same electricity two years running and report a lower location-based figure the second year, having changed nothing. The reverse happens too; a dry year in a hydro-heavy grid, or a temporary return to fossil generation, can push a factor up.

This is why factor vintage is part of the disclosure rather than a footnote. Grid factors are published with a lag, so the factor available when you close an inventory typically describes the grid of one or two years earlier. Two otherwise identical inventories can differ simply because one used a newer factor set. The discipline that keeps this auditable is the same as for any accounting estimate: cite the factor source and its year against every figure, apply the same vintage policy consistently across sites, and when you move to a newer factor set, say so, so a year-over-year change can be decomposed into what you did and what the grid did.

That decomposition is the whole point. If your location-based Scope 2 fell 6% and the grid factor fell 6%, your consumption is flat and the reduction story belongs to the grid, not to you. Market-based Scope 2 exists partly to isolate the portion you actually chose through procurement. The clean way to hold all of this is kWh, factor, source and vintage on every row, which is the record format our carbon accounting software is being built around. For how the two methods sit inside the full inventory, see the overview of scope 1 2 3 emissions.

What to do with this

If you are starting a first inventory this quarter: pull the twelve-month AP export, the utility invoices and the fuel-card statements. Classify everything, meters to Scope 1 and 2, the rest to Scope 3 categories. Compute Scope 1 and 2 activity-based from day one, and let Scope 3 start spend-based with the method labeled per line. You can watch that exact workflow run on realistic sample data, or paste your own lines, in the Scope Classifier demo: it is free, ungated, and honest about being a screening estimate.