District Heating & Cooling Emissions
How to account for purchased heat, steam, and cooling under Scope 2 — the factor hierarchy that replaces the missing “grid average,” the CHP/cogeneration allocation problem that drives most double-counting, and the reason district cooling is an electricity problem in disguise. Aligned to the GHG Protocol Scope 2 Guidance, UK DEFRA 2025 factors, EUROSTAT renewable-share data, and ASHRAE plant-efficiency benchmarks. MasterBrain v2026.284.
Most Scope 2 guidance assumes a clean grid-average emission factor sits ready for every kWh you buy. For purchased heat, steam, and cooling, that factor does not exist — and the workarounds practitioners reach for quietly under- or over-count emissions by margins large enough to fail validation.
District energy is the part of Scope 2 where the easy assumptions break, and the methodology has to be built from the network up.
Purchased district heating, steam and cooling are Scope 2. Multiply metered delivered energy by a supplier-specific emission factor where available, or a residual-derived factor where it is not. District cooling is calculated from plant electricity intensity.
District heating
Hot water or low-pressure steam piped from a central plant to buildings for space heating and hot water. Factor basis is the plant fuel mix at the generation point.
kg CO₂e / kWh delivered heatPurchased steam
High-pressure or process steam from a third-party plant, common on industrial estates and campuses. Accounted on an enthalpy (delivered-energy) basis, often from a CHP host.
kg CO₂e / kWh or / tonne steamDistrict cooling
Chilled water from a central plant for air conditioning and process cooling. The emissions are driven almost entirely by the electricity the chillers consume.
kW electric / ton cooling × grid factorWhat District Energy Is, and Why It’s Scope 2
District energy systems generate heat, steam, or cooling at a central plant and distribute it through a network of insulated pipes to multiple buildings. The consuming organisation buys a finished thermal product — a kWh of delivered heat or a ton-hour of cooling — without operating any combustion equipment itself. That ownership boundary is what places district energy in Scope 2: the emissions occur at a plant the reporting organisation does not own or control, but are a direct consequence of energy it purchased and consumed.
The distinction from Scope 1 is the same one that separates purchased electricity from an on-site boiler. If your organisation runs the boiler, the natural-gas combustion is Scope 1 and you follow the natural-gas combustion methodology. If a utility runs the plant and sells you the heat, the same physical combustion becomes your Scope 2. The GHG Protocol Corporate Standard classifies purchased steam, heating, and cooling alongside purchased electricity for exactly this reason — all four are energy generated elsewhere and consumed by the reporter.
Scope 2 has four energy carriers, not one. The GHG Protocol names purchased electricity, steam, heating, and cooling together. Inventories that report electricity rigorously but treat district heat as an estimate, or omit district cooling because “it’s just electricity,” leave a material and auditable gap.
Provider directory
Need help keeping this up all year?
See who does this work. Every listing names the standards it works to, and paid placements are labelled. Including KERAMIDA and Small World Consulting.
Browse 6 carbon accounting & inventory providers →Do this work? A listing is US$390 a year. Get listed →
Where Heating & Cooling Sit in the GHG Protocol
The Scope 2 Guidance was written principally for electricity, then extended to thermal carriers by analogy. Two structural decisions carry over directly; one does not.
3.1 Scope 2 by default; the on-site-generation exception
Purchased thermal energy is Scope 2 whenever the generating plant is outside the organisation’s operational or financial control boundary. The exception is on-site generation under the reporter’s control: a CHP unit the organisation owns and operates produces Scope 1 emissions for the fuel it burns, even if some of the heat is sold to neighbours. Where a third party owns and operates a CHP plant on the reporter’s site and sells back heat, the purchased heat is Scope 2 and the boundary question turns on control, not physical location.
3.2 Location-based vs market-based for a thermal carrier
The Scope 2 Guidance requires dual reporting — a location-based figure and a market-based figure — for electricity. The same logic applies in principle to heat and cooling, but the instruments differ. For electricity, the market-based method draws on a deep ecosystem of energy attribute certificates and a published residual mix. For heat, contractual instruments are thinner: a supplier-specific factor disclosed by the network operator is usually the strongest market-based signal available, and a published heat residual mix exists in very few jurisdictions. This asymmetry is the single most consequential difference between electricity Scope 2 and thermal Scope 2, and it shapes the entire factor hierarchy in §5.
Routing pre-check — which path applies to your thermal energy?
Establish the boundary and carrier before reaching for a factor. The route determines whether you are in this methodology at all.
kW/ton against the grid factor.The Emission Factor Problem: No Grid Average for Heat
Electricity Scope 2 rests on a published national grid factor — a single defensible number per country, updated annually. District heating has no equivalent. Two networks in the same city can differ by an order of magnitude: one fired by gas boilers, the next recovering waste heat from an incinerator or a data centre. There is no “national heat grid,” so there is no national heat factor. The methodology is therefore a hierarchy of fallbacks, ordered by data quality, that substitutes for the missing grid average.
The network operator publishes the actual kg CO₂e/kWh of delivered heat for the reporting year, derived from its fuel mix and plant efficiency. This is the market-based signal and the highest-quality input. Always request it first.
Where a jurisdiction publishes a representative district-heat factor — the UK is one — it serves as the location-based default. Surfaced live via 0.17529 for UK reporting.
Where no headline factor exists, derive one from the fossil baseline scaled by the network’s renewable + waste-heat share — the EU framework method in §6.3. Country renewable shares are surfaced live (e.g. 72).
The EU fossil district-heat baseline of 0.215 kg CO₂e/kWh is a conservative screening floor when nothing network-specific is available. Replace it the moment supplier data arrives.
Do not borrow another network’s published factor because it is in the same country. A waste-heat network at near-zero carbon and a gas-boiler network at the fossil baseline are both “district heat in the UK,” and substituting one for the other can misstate the line item by 5–10×. The factor belongs to the network, not the postcode.
Calculating District Heating Emissions
Once the factor is selected from the hierarchy, the heating calculation is a single multiplication against metered delivered energy. The discipline is entirely in the factor choice and in keeping upstream losses in the right scope.
6.1 The core formula
0.17529
The metering boundary matters. The factor is defined per kWh delivered to your substation, so distribution losses between plant and meter are already excluded from your Scope 2 figure — they are the network’s loss, accounted upstream. Reporting against generated rather than delivered energy double-counts the losses.
6.2 The DEFRA combined heat factor (UK)
For UK reporting, DEFRA 2025 publishes a district-heat conversion factor that is surfaced live as 0.17529 kg CO₂e/kWh, with a well-to-tank companion at 0.03341 kg CO₂e/kWh for the upstream Scope 3 element. These are DEFRA-convention factors and carry an AR5 GWP-100 basis by design — do not convert them to AR6 or blend them with AR6-basis factors inside a single total. The on-site equivalent (heat generated and consumed within the boundary) reads from 0.17529 and belongs to Scope 1, not here.
6.3 The renewable-share derivation (EU framework)
Where no headline factor is published, the EU framework method derives one. Start from the fossil district-heat baseline, then scale it down by the share of the network’s final energy that comes from renewables and recovered waste heat:
0.215 kg CO₂e/kWh (HRE4 / DEFRA-derived)
34.527%
The renewable share varies enormously by country, which is exactly why a single regional factor would mislead. Denmark’s district-heat networks run at roughly 72% renewable and waste heat, and Spain at around 82%, while Albania sits near 3.368%. A derived factor anchored to the actual network is always preferable to a regional average; the derivation is a screening tool, not a substitute for supplier disclosure.
The renewable-share method gives a location-based estimate, not a market-based claim. A high renewable share lowers your derived location-based factor, but you can only report a market-based reduction if the supplier contractually attributes that renewable heat to you. Statistics about the network are not a contractual instrument.
6.4 Well-to-tank (Scope 3 Category 3) for purchased heat
The delivered-heat factor captures combustion at the plant. The upstream emissions — extracting and refining the fuel, plus distribution losses across the network — are Scope 3 Category 3, not Scope 2. Use the WTT companion factor and follow the well-to-tank methodology. Keeping these in Category 3 rather than folding them into Scope 2 preserves comparability with electricity, where the same split applies.
Need to run delivered-heat figures with the live DEFRA factor and a WTT split? Use the Scope 2 calculator for the electricity side and the well-to-tank calculator for the upstream element. (The district heating & cooling calculator is now live.)
Calculating District Cooling Emissions
District cooling looks like a separate carrier but behaves like electricity. A central plant runs electric chillers to produce chilled water; the emissions are overwhelmingly the electricity those chillers draw. So the cooling calculation is a two-step translation: convert delivered cooling into the electricity it required, then apply the grid factor.
7.1 Why cooling is an electricity problem
A district cooling plant’s carbon intensity is governed by its efficiency — how many kilowatts of electricity it spends to deliver one ton of cooling. ASHRAE benchmarks this in kW per ton of cooling (lower is better). Once you know the plant’s kW/ton and the local grid emission factor, the carbon per unit of cooling follows directly. There is no separate “cooling emission factor” to look up — the grid factor does the work.
| Plant / benchmark | Efficiency (kW/ton) | Interpretation |
|---|---|---|
| Texas Medical Center, Houston | 0.89 | High-efficiency large-scale plant — among the best-performing benchmarks |
| District Energy St. Paul, Minnesota | 0.9 | Efficient cold-climate plant |
| Typical-plant framework benchmark | 1 | Reasonable default where the operator’s actual figure is unavailable |
| Business Bay Executive Towers, Dubai | 1.065 | Hot-climate plant — higher kW/ton reflects ambient cooling load |
Plant benchmarks: ASHRAE District Cooling Guide (2013), surfaced via the GreenCalculus MasterBrain.
7.2 The cooling formula
0.497 for Singapore (lifecycle intensity), 0.131 for the UK (DEFRA location-based)
Because the grid factor carries the carbon, district cooling decarbonises automatically as the local grid cleans up — no plant retrofit required. It also means the same chilled-water contract has very different emissions in different geographies: a plant at the framework benchmark of 1 kW/ton emits far more in a high-carbon grid than in a low-carbon one. Where the plant uses thermal storage, free cooling, or absorption chillers driven by waste heat, the kW/ton falls and the operator-specific figure should always replace the benchmark.
For a market-based cooling figure, apply the contractual electricity factor the cooling operator can demonstrate — for example, a plant running on a renewable PPA. For the location-based figure, always use the published grid average. The dual-reporting requirement applies to cooling exactly as it does to electricity.
CHP / Cogeneration Allocation
Combined heat and power plants are the backbone of efficient district heating — and the single hardest accounting problem in this methodology. A CHP plant burns one stream of fuel and produces two useful outputs, heat and electricity. The plant’s total emissions have to be split between those outputs, because the heat buyer and the power buyer cannot both claim 100% of them.
8.1 Why allocation is unavoidable
If the heat buyer and the power buyer each applied the plant’s total emissions to their own purchase, the same kilogram of CO₂e would be counted twice across the two inventories. Allocation prevents that double count by dividing the plant’s emissions into a heat share and a power share whose total equals the plant’s actual emissions. The supplier-specific heat factor a good network operator publishes is, under the hood, the result of one of these allocation methods.
8.2 The three allocation methods
| Method | Basis | Effect on heat factor | When to use |
|---|---|---|---|
| Energy (proportional) method | Splits emissions by the energy content of each output — heat kWh vs electricity kWh | Assigns a large share to heat, because heat output usually exceeds power output in energy terms | Simple, transparent, widely used as the default where no method is mandated |
| Efficiency (exergy) method | Weights outputs by thermodynamic quality — electricity is higher-grade energy than low-temperature heat | Shifts share toward electricity, lowering the heat factor relative to the energy method | Most defensible thermodynamically; preferred where regulators or verifiers expect it |
| Alternative-generation method | Credits each output against the emissions of producing it separately (a reference boiler + the grid) | Highly sensitive to the chosen reference efficiencies; can favour heat or power depending on benchmarks | Where a defensible counterfactual reference exists and is disclosed |
The allocation method is not interchangeable between buyer and seller. If the network operator allocates by the exergy method and the heat buyer back-calculates an implied factor using the energy method, the two inventories will not reconcile and the difference is pure methodological artefact. Always use the operator’s stated allocation method, and disclose which one was applied.
Location-Based vs Market-Based for District Energy
The dual-reporting requirement is the same shape as electricity, but the available instruments are thinner. The location vs market comparison holds for thermal carriers with one practical caveat: the market-based column is often constrained by what the network operator can contractually offer.
Average grid / network intensity
For heat: the national default factor or the renewable-share-derived network factor. For cooling: the published grid average via the plant’s kW/ton. Reflects the physical energy system the reporter operates within, regardless of contracts.
Contractual / supplier-specific intensity
For heat: the supplier-specific factor the operator discloses, or a heat-attribute instrument where one exists. For cooling: the contractual electricity factor the plant can demonstrate (e.g. a renewable PPA). Reflects the energy the reporter chose to procure.
Where no contractual instrument exists for heat, the market-based figure defaults to the same value as the location-based figure — the operator’s published average — and the inventory should state that explicitly rather than leave the market-based column blank. Organisations targeting RE100-style claims should note that heat attribute certificates are recognised in far fewer markets than electricity certificates; verify the instrument is accepted in the relevant jurisdiction before claiming a market-based reduction.
Metering, Data Quality & the Factor Hierarchy
Thermal Scope 2 is only as good as its metering and its factor provenance. Three data-quality decisions govern the result.
- Metered vs estimated consumption. A heat or cooling meter at the building substation gives delivered energy directly. Where no meter exists, floor-area or degree-day estimation introduces uncertainty that should be disclosed and improved at the next reporting cycle.
- Factor provenance. Record which rung of the §5 hierarchy the factor came from — supplier-specific, national default, renewable-share-derived, or screening baseline — and the reporting year of the source. A factor from the wrong year drifts as the network decarbonises.
- GWP basis consistency. DEFRA-sourced heat factors carry an AR5 GWP-100 basis; most corporate-reporting frameworks default to AR6. Do not mix bases inside a single total. The asymmetry is by design — flag the basis per line item rather than silently converting.
The factor hierarchy maps cleanly onto the data-quality language used elsewhere in GHG accounting: supplier-specific is primary data, the national default and renewable-share derivation are secondary, and the fossil baseline is a screening estimate to be replaced. Plan the progression from screening to supplier-specific at the same time as the rest of the inventory, not after.
Worked Example: A Mixed-Carrier Campus
A complete worked example computed end-to-end with all arithmetic shown. Factors are fixed illustrative values at the snapshot date of this methodology — this is an audit record, not a live read. Substitute your own metered data and supplier-disclosed factors.
The two figures tell different true stories: the location-based 950.34 tCO₂e reflects the physical grid CampusCo sits on, while the market-based 663.60 tCO₂e reflects the renewable PPA it procured for cooling. Both are reported. The heat figure is identical in both columns because the gas-CHP network offers no contractual renewable instrument — the supplier-specific factor is the strongest available market signal, and it equals the location-based input here.
The upstream well-to-tank emissions on both carriers (fuel extraction plus network distribution losses) are not in these Scope 2 totals. They belong in Scope 3 Category 3 and are reported separately — see the WTT methodology. Folding them into Scope 2 overstates the line and breaks comparability with electricity.
Compliance & Disclosure
Thermal Scope 2 feeds the same disclosure regimes as electricity Scope 2, with a few carrier-specific notes.
- CSRD ESRS E1. Scope 2 must be disclosed both location-based and market-based; the standard expects purchased heat, steam, and cooling to be inside the Scope 2 boundary, not omitted. Disclose the factor provenance and GWP basis for each carrier.
- UK SECR. Purchased heat is reportable under Streamlined Energy and Carbon Reporting using the DEFRA district-heat factor; the UK grid factor covers the electricity behind any district cooling. The SECR calculator handles the aggregation.
- CDP Climate. CDP’s Scope 2 questions accept both methods and reward supplier-specific factors over national defaults — another reason to climb the §5 hierarchy where the data exists.
- GHG Protocol dual reporting. Both figures appear in the inventory with the methods and instruments disclosed, consistent with ISO 14064-1 transparency expectations.
Error Traps with Calculable Magnitudes
Each trap below produces a specific, quantifiable distortion. Magnitudes use CampusCo’s worked-example figures where applicable.
| Error | What happens | Magnitude | How to avoid |
|---|---|---|---|
| Borrowing another network’s factor | Applying a waste-heat network’s near-zero factor to a gas-boiler network (or vice versa) because both are “national district heat” | 5–10× misstatement on the heat line | Use the network’s own published factor; the factor belongs to the network, not the country |
| Treating cooling as a bespoke factor | Looking for a “cooling emission factor” instead of computing kW/ton × grid factor; usually defaults to an over-conservative guess | Variable; commonly 20–50% off | Compute from plant efficiency and the live grid factor — the grid factor carries the carbon |
| Double-counting distribution losses | Reporting against generated energy and then adding network losses again in Scope 3 | Losses counted twice (often 5–15%) | Report delivered energy at your meter; keep losses in Scope 3 Cat 3 only |
| CHP allocation mismatch | Buyer back-calculates with a different allocation method than the operator used | Inventories fail to reconcile | Use and disclose the operator’s stated allocation method |
| Omitting district cooling entirely | Cooling left out because “it’s just electricity that’s already in Scope 2 elsewhere” | CampusCo: 286.74 tCO₂e missing (30% of Scope 2) | Cooling electricity is consumed by the plant, not your meters — it must be added explicitly |
| Mixing AR5 and AR6 GWP | DEFRA heat factor (AR5) summed with AR6-basis factors without disclosure | Small but a greenwashing-risk finding | Keep DEFRA factors on AR5; disclose the basis per carrier; never blend inside one total |
| Leaving the market-based column blank | No contractual heat instrument exists, so the market-based figure is omitted entirely | Incomplete dual report | Default market-based to the supplier-specific average and state it explicitly |
What the Calculator Handles vs What You Decide
A calculator automates the mechanical multiplication and the dual-report aggregation. The upstream decisions — boundary, factor provenance, allocation method — stay with the reporter.
⚙ Calculator handles automatically
- Delivered-energy × factor for heat and steam
- kW/ton × grid factor for cooling
- Location-based and market-based dual totals
- Live DEFRA / EUROSTAT / grid factor reads
- AR5/AR6 basis tagging per carrier
- Scope 2 vs Scope 3 WTT split
- Unit conversion (kWh, ton-hours, GJ) with audit trail
✎ You must decide before using it
- Boundary — purchased (Scope 2) vs owned-plant (Scope 1)
- Factor rung — supplier-specific, national default, renewable-share, or screening baseline
- Allocation method for any CHP-sourced heat (use the operator’s)
- Metered vs estimated consumption, and the uncertainty that implies
- Market-based instruments — which contracts are recognised in your jurisdiction
- Reporting year of every factor, so it tracks network decarbonisation
Frequently Asked Questions
Scope 2, when the generating plant sits outside your operational or financial control boundary. The combustion happens at a plant you do not own, but the emissions are a direct consequence of energy you bought and consumed — the same logic that makes purchased electricity Scope 2. It becomes Scope 1 only if you own and operate the plant yourself, in which case you follow the combustion methodology and account for the fuel directly.
Because there is no “national heat grid.” Two networks in the same city can have wildly different carbon intensities — one fired by gas boilers, the next recovering waste heat from an incinerator or data centre. Electricity has a single national grid factor because the grid is physically interconnected; heat networks are isolated and local. That is why the methodology is a hierarchy of fallbacks ordered by data quality, with the supplier’s own published factor at the top.
Treat it as an electricity calculation. Multiply delivered cooling (ton-hours) by the plant’s electricity intensity (kW/ton) to get the electricity the chillers drew, then multiply by the local grid emission factor. There is no separate cooling factor to look up — the grid factor carries the carbon. Use the operator’s actual kW/ton where disclosed; an ASHRAE benchmark of around 0.9–1.0 kW/ton is a reasonable default otherwise.
A combined heat and power plant burns one fuel stream and produces two useful outputs — heat and electricity. Allocation splits the plant’s total emissions between those outputs so the heat buyer and the power buyer don’t both claim 100% of them. The three common methods are the energy method (split by energy content), the exergy method (weight by thermodynamic quality), and the alternative-generation method (credit against separate production). The method changes the heat factor materially, so always use and disclose the operator’s stated method.
Yes. The GHG Protocol Scope 2 dual-reporting requirement applies to all four Scope 2 carriers. The location-based figure uses the average network or grid intensity; the market-based figure uses contractual instruments — a supplier-specific heat factor, or a renewable PPA behind a cooling plant. Where no contractual instrument exists for heat, default the market-based figure to the supplier’s published average and state that explicitly rather than leaving the column blank.
In Scope 3 Category 3, not Scope 2. Your Scope 2 factor is defined per kWh delivered to your substation, so the energy lost in the network between plant and meter is already excluded from it — that loss is the network’s, accounted upstream alongside the fuel’s well-to-tank emissions. Reporting against generated rather than delivered energy, then adding losses again in Scope 3, double-counts them.
DEFRA’s conversion factors are published on an AR5 GWP-100 basis by convention, and that basis is part of the factor’s definition — it is pre-baked, not an error. Most corporate-reporting frameworks default to AR6 for line items they compute themselves. The two coexist: keep DEFRA-sourced factors on AR5, keep your AR6-basis factors on AR6, and never sum the two inside a single total without disclosing the mixed basis. Mixing them silently is a CSRD greenwashing exposure.
Methodology Metadata — for GHG Inventory Documentation
Copy into your GHG inventory methodology statement for ISO 14064-1 transparency compliance. Adjust the factor-provenance lines to your actual sources.