Carbon Baseline Drift in Major Projects and Why It Matters
A major project rarely finishes as the project that was first modelled.
The plot plan changes. The concrete mix is revised. A modular package becomes stick-built because a yard slot disappears. A preferred supplier is replaced after commercial negotiation. Temporary works grow because ground conditions are worse than expected. A schedule recovery plan brings in air freight, night shifts, extra cranes, and a different construction sequence.
The cost baseline moves. The schedule baseline moves. The risk register moves. The technical basis moves.
Too often, the carbon baseline does not.
That gap is where carbon baseline drift begins. It is not usually a dramatic failure. It is a slow separation between the emissions baseline used for reporting, approvals, investment cases, reduction targets, and the real project being built or operated.
For capital-intensive sectors, this is not an academic ESG issue. It affects embodied carbon, construction emissions, lifecycle emissions, assurance confidence, procurement decisions, client claims, financing narratives, and corporate decarbonization commitments. If the baseline no longer reflects the project, decisions based on it become less reliable.
Carbon Baselines Are Decision Tools, Not Archive Documents
People often treat a carbon baseline as a fixed reference point. In one sense, it should be. Projects need a reference case to compare options, measure reductions, and explain performance. Without a baseline, any claim of avoided emissions or lower-carbon delivery is hard to prove.
The problem starts when the baseline becomes fixed in the wrong way.
A useful baseline is not just a spreadsheet of emissions factors and quantities. It is a record of the project assumptions at a point in time. It should reflect:
Scope boundaries
Design maturity
Material quantities
Supplier assumptions
Construction methods
Logistics routes
Energy sources
Temporary works
Commissioning plans
Operational duty cycles
Maintenance and replacement assumptions
End-of-life treatment, where relevant
In other words, the baseline is only as valid as the assumptions behind it.
In mature project controls, cost baselines are tied to scope, schedule, procurement, risk, and estimate class. If the design develops or execution strategy changes, the cost estimate is reviewed. Schedule logic is updated. Risk exposure is reassessed. The project does not keep reporting against an early estimate as if nothing changed.
Carbon should work the same way.
That does not mean recalculating the full lifecycle assessment every time a valve size changes or a delivery truck takes a different route. Major projects already carry enough administrative burden. The point is proportionate control. When a change affects emissions in a material way, the carbon baseline should be reviewed, updated, or formally revalidated.
Standards and frameworks already point in this direction, even if they use different language. The GHG Protocol relies on clear inventory boundaries, consistent methods, and recalculation where structural changes affect comparability. ISO 14064 focuses on greenhouse gas quantification and reporting principles such as relevance, completeness, consistency, accuracy, and transparency. ISO 14067 and lifecycle assessment practice require clear system boundaries and assumptions. PAS 2080, widely used in infrastructure carbon management, places strong emphasis on reducing carbon through the value chain and across the asset lifecycle.
The common thread is simple: a carbon number without its basis of estimate is weak evidence.
In high-risk industries, that weakness matters because carbon data is now used for more than sustainability reports. It can influence:
Tender evaluation
Design optioneering
Board approvals
Green finance claims
Joint venture commitments
Client reporting
Regulatory disclosures
Internal carbon budgets
Supplier performance reviews
Asset acquisition and divestment decisions
If the baseline is stale, those decisions may still look precise. They may even be audited for arithmetic accuracy. Yet they can still be wrong in substance.
How Carbon Baseline Drift Happens on Real Projects
Carbon baseline drift usually comes from ordinary project change. The danger is that each individual change can look manageable. The cumulative effect is where the exposure builds.
A 2 percent material quantity change here, a supplier substitution there, a logistics workaround during execution, a revised operating assumption during commissioning. None of these may trigger an alarm on its own. Together, they can shift the emissions profile enough to undermine reported reductions or investment assumptions.
Scope Growth Quietly Changes the Denominator
Scope change is the most obvious source of drift, but it is often poorly handled in carbon reporting.
A processing facility may add utility systems after the original baseline. A port expansion may need extra dredging, quay reinforcement, or temporary marine works. A mining project may extend haul roads, add accommodation facilities, or increase water infrastructure. A data centre or manufacturing plant may add redundancy, backup capacity, or cooling systems.
If the added scope increases emissions but sits outside the original carbon model, the baseline becomes incomplete. If the baseline is updated but the reduction target is not reinterpreted, the project may appear to perform worse or better for reasons unrelated to carbon management.
This is especially relevant when teams report emissions intensity, such as tonnes of CO₂e per tonne of product, per square metre, per megawatt, or per unit of throughput. Scope growth may increase absolute emissions while improving or worsening intensity depending on the denominator. Both measures need careful explanation.
A project that has genuinely expanded is not necessarily failing on carbon. But the baseline must make the change visible.
Design Development Changes Quantities and Specifications
Early carbon estimates often use concept design quantities, benchmark factors, or generic material assumptions. That may be acceptable at feasibility stage. It is not acceptable if the same baseline is later used to defend detailed ESG claims after design maturity has moved on.
Design development can alter emissions through:
Structural steel tonnage
Concrete volume and cement content
Rebar density
Pipework quantities
Cable lengths
Equipment weight
Insulation type and thickness
Coating systems
Fire protection requirements
Redundancy levels
Process efficiency
Electrical load
Water demand
In construction, embodied carbon can be sensitive to material type, quantity, and manufacturing route. A change from one concrete specification to another may matter. A shift from local aggregate to imported material may matter. A revised steel grade, fabrication source, or corrosion allowance may matter.
In energy, utilities, oil and gas, and heavy industry, operational assumptions can dwarf construction emissions over the asset life. A design revision that affects efficiency, fuel use, fugitive emissions, power demand, or maintenance frequency can change the lifecycle picture far more than a material substitution.
This is why carbon baselines need version control. A concept-stage baseline and a detailed-design baseline are not interchangeable. Both can be valid, but they answer different questions.
Procurement Decisions Can Overwrite Theoretical Reductions
Many projects identify carbon reductions during design, then lose them during procurement.
A low-carbon material option may depend on a supplier with limited capacity. A product-specific environmental product declaration may be used in the baseline, but a generic supplier wins the package. A local fabrication route may be assumed, but the commercial award shifts work to another country. A framework supplier may change production sites. A nominated technology may be replaced because of lead time, warranty, operability, or cost.
None of this is unusual. Procurement teams must manage price, quality, delivery, risk, local content, contractual terms, and technical compliance. Carbon is one criterion among many, and often not the most heavily weighted.
The problem arises when the carbon baseline continues to assume the original supplier, original production route, or original material profile.
Supplier substitutions can affect emissions through:
Energy source used in manufacturing
Recycled content
Process efficiency
Transport distance and mode
Packaging
Waste rates
Installation method
Maintenance needs
Expected service life
A project may still meet the technical specification, cost target, and schedule need while missing the carbon basis of the approved case.
That is not necessarily a procurement failure. It is a control failure if the change never reaches the carbon model.
Value Engineering Can Cut Cost While Increasing Emissions
Value engineering is meant to improve value, not simply remove cost. In practice, some value engineering decisions reduce capital cost but increase carbon. Others reduce carbon and cost at the same time. The difference is rarely obvious unless carbon is part of the assessment.
A cheaper material may have a higher embodied carbon factor. A simplified construction method may require more temporary works. A reduced upfront specification may shorten service life and increase replacement emissions. A supplier with a lower unit price may require longer transport routes. A compressed installation package may need more plant, more shifts, and more rework risk.
The reverse is also true. Design rationalisation can reduce quantities. Standardisation can reduce waste. Modularisation can cut site emissions if transport and fabrication impacts are properly accounted for. Higher-quality components can reduce maintenance and replacement emissions.
Carbon-literate value engineering asks a practical question: what emissions move when the commercial decision changes?
If the answer is unknown, the project is making a trade-off without seeing all of its costs.
Schedule Acceleration Often Imports Hidden Construction Emissions
Acceleration is a familiar feature of major projects. Delay costs can be severe. Liquidated damages, production milestones, financing conditions, regulatory windows, shutdown dates, or seasonal constraints may justify a recovery plan.
Carbon impacts often follow.
Acceleration may require:
Air freight instead of sea freight
Additional cranes, generators, compressors, and lighting
Double handling of materials
Larger temporary facilities
Extended worker transport
Parallel work fronts
Lower labour productivity
More waste and rework risk
Overtime energy demand
Temporary power rather than grid connection
These emissions can sit outside the original construction baseline. They may be treated as execution noise rather than material carbon change.
In marine operations, offshore construction, remote mining, pipeline work, and island infrastructure, logistics choices can be especially important. Vessel days, helicopter use, temporary camps, barge movements, and fuel supply routes can materially change the project footprint.
Schedule recovery may still be the right commercial decision. Carbon change control should not be a veto by default. It should make the emissions consequence visible, costed where possible, and approved at the right level.
Construction Methodology and Temporary Works Are Easy to Undercount
Temporary works are often underestimated in carbon baselines because early models focus on the permanent asset. Yet temporary works can be significant, especially in complex ground conditions, constrained sites, marine environments, tunnelling, major lifting, heavy civil works, and brownfield modifications.
Examples include:
Cofferdams
Temporary access roads
Working platforms
Scaffolding
Falsework and formwork
Temporary bridges
Dewatering systems
Temporary power
Site accommodation
Traffic diversions
Temporary marine structures
Bypass systems during shutdowns
Construction methodology also matters. A cast-in-situ solution, precast approach, modular build, heavy lift strategy, or offshore campaign each carries different material, plant, logistics, waste, safety, schedule, and quality implications.
Many early carbon baselines do not yet know the final construction method. That is understandable. What is less defensible is failing to revisit the baseline when the method is selected.
Why Drift Creates Commercial, Reporting, and Assurance Risk
A stale carbon baseline is not just a technical imperfection. It can create real exposure across governance, reporting, finance, delivery, and stakeholder trust.
Embodied Carbon Claims Can Become Unreliable
Embodied carbon claims often rely on comparison. A project may claim that it reduced embodied carbon against a reference design, used lower-carbon material, avoided emissions through design changes, or delivered below a benchmark.
Those claims need a stable and credible comparison point. If the baseline does not reflect the scope, specification, or procurement route that would otherwise have happened, the claim becomes fragile.
The risk is not only overclaiming. Underclaiming can also occur. A project may make strong carbon reductions during detailed design, but the reduction is invisible because the baseline was never updated from an early model with different quantities or boundaries.
Both outcomes are poor management. Good carbon accounting should show where performance improved, where it worsened, and why.
Construction Emissions Can Fall Between Control Systems
Many corporate inventories capture fuel and electricity used during construction, but the treatment can vary depending on ownership, contracting model, operational control, and reporting boundary. Contractor fuel, subcontractor plant, temporary power, logistics, and third-party fabrication may sit in different Scope 1, Scope 2, or Scope 3 categories depending on the organisation and the accounting approach.
This is where project carbon and corporate carbon can diverge.
A project team may consider construction emissions part of delivery performance. A corporate reporting team may only include certain categories. A contractor may report fuel use to its own inventory. A client may report embodied carbon in Scope 3 purchased goods and capital goods. If the baseline does not clearly define boundaries, double counting and omissions both become possible.
The accounting treatment must be transparent. So must the management treatment. Even if an emission source sits outside a company’s direct inventory boundary, it may still sit inside its project decision boundary.
Lifecycle Emissions Can Be Distorted by Late Operational Assumptions
For assets with long operating lives, operational assumptions can dominate lifecycle emissions.
Consider a processing facility, utility plant, LNG facility, desalination asset, fleet asset, smelter, refinery unit, marine terminal, or manufacturing line. Changes in load factor, fuel mix, electricity source, process efficiency, methane leakage, refrigerant use, maintenance strategy, product slate, or asset life can reshape the lifecycle model.
A baseline developed during feasibility may assume one operating profile. The business case may later assume another. Commissioning may reveal performance constraints. Grid emissions factors may change over time. A future electrification pathway may be delayed. A carbon capture interface may remain “future ready” but not installed.
If operational carbon assumptions are not revalidated, the project may carry a lifecycle claim that no longer matches expected reality.
This matters for investment decisions. A design option with higher capital carbon may be justified if it reduces operating emissions over the asset life. But if the operating assumption changes, that trade-off needs to be revisited. The reverse can also happen. A lower-capital-carbon option may lock in higher operating emissions for decades.
ESG Reporting Needs Comparability and Traceability
ESG reporting is moving towards more disciplined disclosure. Investors, lenders, clients, and regulators expect clearer links between targets, actions, data, assumptions, and governance. The International Sustainability Standards Board’s IFRS S2 has increased focus on climate-related risks, transition plans, metrics, and governance in many markets. Local adoption and application differ, but the direction is clear.
For major projects, this creates a practical question: can the organisation trace reported carbon performance back to the project decisions that created it?
If the answer is no, assurance becomes difficult. A carbon baseline that lacks change history, approvals, assumptions, and evidence can undermine confidence even where the numbers were prepared in good faith.
Good reporting does not require perfect foresight. It requires a clear audit trail.
Decarbonisation Commitments Depend on Project Controls
Corporate decarbonisation commitments often assume that future projects will be designed and delivered differently from past projects. That may include lower-carbon materials, electrification, renewable power, methane reduction, energy efficiency, circularity, modular construction, reduced flaring, alternative fuels, or supplier engagement.
Those commitments can fail one project change at a time.
A target may be set at corporate level, but delivery happens through engineering decisions, procurement awards, construction methods, and operational handover. If carbon is not integrated into change control, the corporate target relies on goodwill and after-the-fact reporting.
That is too weak for large capital programmes.
When a Carbon Baseline Should Be Reviewed or Revalidated
Not every change needs a full recalculation. A mature process separates routine updates from formal revalidation.
A review checks whether a change may affect the carbon baseline. A revalidation confirms whether the baseline remains fit for purpose after a material change. A rebaseline creates a new approved carbon baseline, with version history and governance.
The right trigger depends on project type, maturity, carbon intensity, contractual requirements, and reporting use. A small building refurbishment and a multi-billion-dirham energy project should not use the same thresholds. Still, the main triggers are common.
Major Scope Changes
A baseline should be reviewed when the project adds, removes, or transfers scope. This includes permanent works, enabling works, utilities, offsite infrastructure, marine works, camps, access roads, grid connections, pipelines, storage, or supporting facilities.
The review should ask:
Does the changed scope sit inside the original project boundary?
Does it affect absolute emissions, emissions intensity, or both?
Does it change the comparison basis for any reduction claim?
Does it affect corporate inventory categories?
Does it change lifecycle assumptions?
If the answer is yes to any of these, formal revalidation may be needed.
Design Revisions and Specification Changes
Revalidation is often needed at major design gates, such as concept selection, FEED completion, detailed design freeze, construction issue, or approved-for-construction stage.
Design revision triggers may include:
Material quantity changes above an agreed threshold
Specification changes for concrete, steel, asphalt, insulation, cables, coatings, or equipment
Changes to process efficiency or energy demand
Changes to capacity, redundancy, or operating mode
Changes to design life or maintenance interval
Changes to resilience requirements that add material or energy demand
A good rule is to align carbon reviews with existing engineering gates. Do not create a parallel bureaucracy if the project already has a disciplined stage-gate process. Add carbon questions to the gates that already matter.
Procurement Decisions and Supplier Substitutions
Procurement is one of the most important control points because it converts a design assumption into a real supply chain.
A baseline should be reviewed when:
The awarded supplier differs from the baseline assumption
The manufacturing location changes
A product-specific emissions factor is replaced by a generic one
Transport mode or route changes materially
A lower-carbon option is removed during commercial negotiation
A package is split or combined in a way that changes logistics
A material substitution affects service life, maintenance, or replacement
For high-carbon packages, the project should define carbon-relevant procurement hold points. These may apply to cement and concrete, steel, aluminium, asphalt, major equipment, modules, electrical systems, vessels, offshore structures, heavy fabrication, and long-lead items.
This is not about asking procurement to solve carbon alone. It is about giving procurement clear rules on when a commercial decision changes the approved carbon case.
Value Engineering
Value engineering should include carbon impact as a standard decision criterion where emissions are material.
The project should review the baseline when a value engineering proposal:
Changes material type or quantity
Changes installation method
Changes fabrication location
Changes asset life or maintenance profile
Changes energy efficiency
Shifts emissions from construction to operation, or the reverse
Reduces capital cost by increasing replacement or operating emissions
This is especially important where project teams are under pressure to cut capital expenditure. Carbon can be unintentionally traded away if nobody is required to show the emissions consequence.
Schedule Acceleration and Logistics Changes
A schedule recovery plan should trigger a carbon review if it changes construction resources, transport mode, working hours, temporary power, vessel days, aviation use, or plant utilisation.
The review should be pragmatic. The project does not need to count every litre of diesel before approving a recovery plan. It does need to understand whether the emissions impact is likely to be material and whether lower-carbon recovery options exist.
Examples include using sea freight instead of air freight where time allows, connecting temporary facilities to grid power sooner, using efficient generators, consolidating deliveries, avoiding double handling, or sequencing works to reduce standby time for heavy plant.
Construction Methodology Changes
A construction method statement can carry major carbon consequences. Revalidation may be needed when the project changes from:
Precast to cast-in-situ construction
Modular to stick-built delivery
Onshore fabrication to offshore installation
Mechanical excavation to blasting
Temporary bridge to haul road
Grid power to temporary generation
Planned campaign execution to multiple remobilisations
Method changes can also affect safety, quality, programme, cost, and community impact. Carbon should be assessed alongside those factors, not after the method has already been locked in.
Operational Assumptions
For assets where operating emissions matter, the baseline should be revalidated when core assumptions change. These include throughput, utilisation, fuel type, electricity source, efficiency, flaring, venting, refrigerants, maintenance strategy, product mix, asset life, and decommissioning approach.
Operational assumptions deserve particular care because they often sit between project development and asset operations. A project team may hand over a plant with one emissions expectation. The operator may then run it under different commercial or technical conditions.
Carbon governance should bridge that handover.
How Carbon Change Control Can Work in Practice
Carbon change control should sit beside cost, schedule, scope, risk, quality, and technical change management. It should not be a separate sustainability workflow that people bypass because it slows delivery.
The practical model is simple. Every material project change should answer one extra question: does this change affect the approved carbon baseline or carbon commitments?
If the answer is no, record the screening decision. If the answer is yes or uncertain, assess the impact at the right level of detail.
Set Clear Thresholds Before the Pressure Starts
Thresholds prevent carbon change control from becoming personal judgement in a crisis. They also stop teams from escalating minor changes that do not matter.
Thresholds can be based on:
Threshold types | How it can work | Where it helps |
Absolute emissions | Escalate if a change is expected to add or remove more than an agreed tonnes CO₂e value | Large projects with carbon budgets |
Percentage change | Escalate if the change moves a package, phase, or total baseline by more than an agreed percentage | Projects with good baseline maturity |
Carbon cost | Escalate if emissions carry a material internal carbon price or offset exposure | Organisations using shadow carbon pricing |
High-carbon categories | Always review changes to defined materials, packages, activities, or suppliers | Early-stage projects with limited data |
Reporting sensitivity | Escalate if the change affects external claims, green finance, permits, or client commitments | Projects with public or contractual commitments |
Lifecycle impact | Escalate if the change affects operating emissions, asset life, maintenance, or replacement | Long-life industrial assets |
The exact numbers should be set by the organization and project context. A percentage that is sensible for a small package may be meaningless for a megaproject. A low absolute threshold may be appropriate for a material covered by a public reduction claim. A higher threshold may suit early design when uncertainty is already wide.
The key is to define thresholds early, record them, and apply them consistently.
Assign Ownership Where Decisions Happen
Carbon change control fails when everyone assumes someone else owns it.
The sustainability team may maintain the method and reporting. But engineering, procurement, construction, project controls, commercial, risk, and operations all create carbon outcomes.
A workable ownership model might look like this:
Engineering owns design quantities, specifications, and technical alternatives.
Procurement owns supplier data, award decisions, and logistics assumptions.
Construction owns methodology, temporary works, plant, fuels, and sequencing.
Project controls owns integration with cost, schedule, change registers, and reporting cycles.
Commercial owns contract incentives, claims, and approved trade-offs.
Operations owns operating assumptions, duty cycles, maintenance, and handover data.
Sustainability or carbon management owns methodology, factors, assurance rules, and reporting integrity.
This division avoids a common trap, where carbon becomes a reporting function rather than a management function.
Use the Existing Change Process
The best carbon process is usually an addition to existing project controls.
A change request can include a carbon screening field. A technical query can flag carbon impact. A procurement award recommendation can compare supplier carbon assumptions. A value engineering form can include lifecycle emissions. A schedule recovery paper can include construction emissions. A management of change process can require carbon review for defined triggers.
The workflow can be light:
Screen the change against carbon triggers.
Estimate the likely emissions impact using an agreed level of accuracy.
Identify options to avoid, reduce, or mitigate the increase.
Record the decision, including trade-offs and approvals.
Update the carbon model or revalidate the baseline if thresholds are met.
Carry the change into ESG reporting, forecasts, and commitments.
This gives carbon the same discipline as other project variables without pretending it is the only thing that matters.
Keep Baselines Version-Controlled
A carbon baseline should have a clear version history. Each version should state:
Project scope and boundaries
Design maturity
Quantity source
Emissions factors and data hierarchy
Supplier assumptions
Construction methodology assumptions
Logistics assumptions
Operational assumptions
Exclusions and limitations
Uncertainty level
Approval date
Trigger for issue
Link to project change records
This is basic assurance hygiene. It helps internal reviewers, external assurers, clients, lenders, and future project teams understand what the numbers mean.
A baseline marked “Version 3, detailed design, post-procurement award for major steel and concrete packages” carries far more credibility than a single static file called “carbon baseline final”.
The word “final” is especially dangerous on a live project.
The Governance Question Is Whether Carbon Has Authority
Most major projects already know how to manage change. The gap is not process literacy. It is authority.
If a cost increase crosses a threshold, it goes to the right approval level. If a schedule delay threatens a milestone, management pays attention. If a technical change affects safety or operability, it receives formal review. Carbon needs similar authority where it affects approved commitments, investment decisions, or external claims.
That does not mean every carbon increase must be rejected. Some increases are justified. A heavier design may improve safety or resilience. A temporary works package may reduce construction risk. A supplier substitution may avoid severe schedule delay. An operational change may support reliability or output.
The issue is not that carbon can never be traded off. The issue is that carbon should not be lost invisibly.
A disciplined project can approve a higher-emissions option if the decision is transparent, evidence-based, and aligned with governance. It can also require compensating reductions elsewhere, revise forecasts, update disclosures, or adjust claims.
This is commercially realistic. It recognizes that capital projects live with constraints. It also recognizes that unmanaged carbon drift creates avoidable risk.
What Good Practice Looks Like
A mature approach to carbon baseline management has a few visible features.
The baseline has a stated basis
People can see what is included, what is excluded, what data was used, and which assumptions matter most.
Carbon is linked to project controls
Change requests, design gates, procurement approvals, and construction method changes include carbon screening where relevant.
Thresholds are proportionate
The project does not chase trivial changes, but it does escalate material ones.
Reduction claims are traceable
If the project says it avoided emissions, the baseline, option analysis, approval record, and final outcome support the claim.
Procurement data improves over time
The model moves from generic factors to supplier-specific or product-specific data where available and credible.
Lifecycle thinking is applied where it matters
Operational emissions, maintenance, replacement, and end-of-life assumptions are reviewed for assets where those factors drive total impact.
Assurance is planned rather than rescued
Evidence is collected during delivery, not reconstructed at the end when people have moved on and decisions are buried in email trails.
What Weak Practice Looks Like
Weak practice is also easy to recognise.
The carbon baseline sits outside the project controls system. It is updated only for annual ESG reporting. It has no clear link to approved scope. Procurement awards do not test carbon assumptions. Construction changes are invisible to the model. Operational assumptions are owned by nobody. External claims use old numbers because they are the only numbers available.
This can continue for a while. Then a lender, client, board committee, regulator, or assurance provider asks a simple question: what changed between the baseline and the asset we built?
If the project cannot answer, the carbon story becomes a credibility problem.
The Takeaway for Major Projects
Carbon baseline drift is a predictable outcome of real project delivery. Scope changes. Designs mature. Suppliers shift. Logistics plans break. Construction methods change. Operating assumptions evolve.
The answer is not to freeze projects or bury teams in carbon paperwork. The answer is to treat carbon as a controlled project variable.
That means setting a clear baseline, linking it to assumptions, reviewing it at the right gates, and revalidating it when material changes occur. It means giving procurement, engineering, construction, project controls, commercial teams, and operations a shared process for carbon change control. It means deciding trade-offs openly rather than discovering them during ESG reporting.
For high-risk and capital-intensive industries, this is where sustainability becomes operationally serious. Not in the slogan, not in the target deck, and not in a polished year-end chart. It happens in the change request, the supplier award, the method statement, the recovery plan, and the commissioning assumption.
A carbon baseline is only useful while it still describes the project. Once the project has moved and the baseline has not, the numbers may still add up, but the governance has already started to drift.
