Ask what climate technology careers look like and most people describe inventing something. The actual bottleneck is counting.
The unglamorous centre of the field
Every emissions target, every reduction claim, every piece of climate regulation and every green financial instrument rests on a number. Somebody has to produce that number, defend it, and produce a comparable one next year.
That work is emissions accounting, monitoring, reporting and verification, and it has four properties that make it harder than it sounds.
It requires domain knowledge to do properly. Calculating emissions from an industrial process, a logistics network or an agricultural operation demands understanding of what actually happens there. Applying a generic emission factor to a spend figure produces a number, not a measurement.
Data quality is usually poor. Activity data lives in systems built for other purposes, in inconsistent units, with gaps, across entities that record things differently.
Methodology choices are consequential and contested. Boundary definitions, allocation methods and factor selection can move a result substantially, which is exactly why standards and assurance exist.
It is becoming legally binding. Several jurisdictions have introduced or are introducing mandatory climate disclosure, with requirements for reported figures and, in some cases, external assurance. Specific requirements differ by market and continue to change, but the direction is towards reporting that carries the same weight as financial reporting.
Once a number carries legal weight, the people who produce it need to be competent rather than enthusiastic.
What the direction covers
The scope: climate science, emissions measurement, climate modelling, adaptation technology and environmental monitoring.
Four areas.
Measurement and accounting. Inventories, emission factors, boundaries, verification and assurance.
Monitoring technology. Sensors, satellite observation, continuous emissions monitoring and leak detection.
Climate science and modelling. Projections, downscaling and the honest communication of uncertainty.
Adaptation. Designing infrastructure, operations and systems for conditions that differ from the historical record.
Scope three, where the numbers get soft
The distinction that explains most disagreement about corporate emissions figures.
Direct emissions from owned sources are measurable with reasonable confidence. Emissions from purchased electricity are calculable from consumption and grid factors, with some methodological argument about how electricity is attributed. Both are tractable.
The third category, everything else in the value chain, is usually the largest share of a company's total footprint and the least reliable part of the figure.
Three reasons.
It depends on other organisations' data. Which is frequently unavailable, inconsistent or estimated, and which the reporting company cannot audit.
Spend-based estimation is weak. Multiplying money spent by an average emission factor for a sector produces a figure that responds to price changes and supplier mix rather than to anything physical, and it cannot detect genuine improvement by a supplier.
Double counting is structural. One company's downstream emissions are another's upstream emissions, which is appropriate for management purposes and makes aggregation across companies meaningless.
The practical skill, and the one organisations are short of, is knowing which parts of an inventory are solid, which are estimates, and how to improve the estimates where it matters rather than uniformly. That is judgement built on domain knowledge, and it is the difference between a report that survives assurance and one that does not.
Where this sits in the domain
Climate science and climate technology is the seventh of nine directions in Astra Trainer's energy, climate and nuclear domain. It underpins the others in a practical sense: carbon management and industrial decarbonisation depends on knowing where emissions actually are, and energy economics, markets and policy depends on measurement that regulators and investors accept.
It also connects outward to AI and computing for monitoring data and modelling, to engineering and built world for adaptation design, and to advanced manufacturing for process emissions accounting. You can see the nine directions here.
Why offsets damaged the credibility of the whole field
Worth addressing directly, because it shapes how climate claims are now received and because avoiding it would be dishonest.
Carbon offsetting allows an organisation to fund a reduction elsewhere instead of reducing its own emissions. The logic is sound in principle: a tonne avoided is a tonne avoided regardless of location.
In practice, several categories of credit were found through independent investigation and academic analysis to represent far less real reduction than claimed. The recurring failures were in additionality, whether the reduction would have happened anyway; in baselines, which were sometimes set to overstate what was avoided; in permanence, where carbon stored in forests can be released by fire or logging; and in leakage, where protecting one area displaced activity to another.
Avoided deforestation credits attracted particularly critical findings, and the consequences were substantial. Buyers withdrew, some corporate claims were challenged as misleading, and regulators in several jurisdictions began scrutinising carbon neutrality claims in advertising.
Two things follow for anyone working in this area.
The quality distinction is now the whole game. Credits differ enormously, and the ability to assess methodology, permanence and verification is a genuine specialism with real demand.
Offsetting is not a substitute for reduction, and organisations that treated it as one are the ones now facing challenge. This is the clearest example in the domain of why measurement credibility matters more than ambition.
Adaptation, the half that gets less funding
Most attention and most money go to reducing emissions. Adaptation, preparing for conditions that are already changing, is consistently under-resourced and is mostly an engineering discipline rather than a scientific one.
The work includes flood risk assessment and defence design, heat resilience in buildings and cities, water resource planning under altered rainfall, infrastructure designed for temperature and load conditions outside the historical record, agricultural adaptation, and resilience planning for supply chains and operations.
One conceptual point does most of the work in this area. Engineering design has traditionally relied on historical statistics: a structure designed for the flood level expected once in a hundred years, based on recorded history. If the underlying distribution is shifting, that reference is no longer a reliable guide to the future, and design standards built on it understate the load.
Updating that practice requires engineers who can work with climate projections and their uncertainty, rather than climate scientists. The translation layer between projection and design decision is where the shortage sits.
The roles, named
Carbon accountants and emissions analysts. The largest growing group.
Sustainability reporting specialists, working to mandatory frameworks.
Emissions verification and assurance professionals.
Climate risk analysts, for physical and transition risk in financial and corporate settings.
Environmental monitoring engineers, including continuous monitoring and leak detection.
Climate data scientists, working with satellite and sensor data.
Adaptation and resilience engineers.
Life cycle assessment practitioners.
Carbon market analysts, assessing credit quality.
Who can be trained into it
Accountants and auditors. A strong and underused conversion. Emissions inventories are accounting problems with unfamiliar units, and audit discipline is exactly what the field lacks.
Process engineers. Into industrial emissions accounting, where knowing what the plant actually does is what separates a real inventory from a spend-based estimate.
Environmental scientists and consultants. Already close, needing the accounting framework and the reporting standards.
Data analysts. Into emissions data management and monitoring analytics.
Civil and structural engineers. Into adaptation and resilience, where the design implications sit.
Supply chain professionals. Into value chain emissions, holding the supplier relationships and data access that make the third category tractable.
Risk professionals. Into climate risk analysis, where existing methodology applies to a new hazard class.
Reporting obligations are legal and jurisdiction specific. Climate disclosure, emissions reporting and environmental permitting requirements are set by law and regulation, differ by market and entity, and continue to change. Some regimes require external assurance by accredited verifiers, and claims about carbon neutrality are subject to consumer protection and advertising law in several jurisdictions. Astra Trainer builds technical understanding of measurement and accounting. It is not legal, regulatory or assurance advice, and it does not confer verifier accreditation or any professional status.
What to take from this
The shortage in climate work is in measurement, accounting and verification rather than in technology development.
Disclosure is becoming a reporting obligation with legal weight, which raises the competence bar for the people producing the numbers.
Value chain emissions are usually the largest and least reliable part of a footprint, and the useful skill is knowing which parts of an inventory to improve.
Offsets damaged the credibility of corporate climate claims because a large share of some credit categories did not represent real reductions, which made quality assessment a specialism in itself.
And adaptation is an engineering problem, under-resourced, needing people who can translate a projection into a design decision.
Why is emissions measurement the bottleneck?
Because every target, claim and regulation rests on a number that must be produced, defended and repeated. Doing it properly requires domain knowledge, and disclosure is increasingly a legal obligation rather than a communications exercise.
Why are value chain emissions unreliable?
Because they depend on other organisations' data that cannot be audited, spend-based estimation responds to prices rather than physical change, and one company's downstream emissions are another's upstream emissions.
What went wrong with carbon offsets?
Independent investigation and academic analysis found that a large share of credits in some categories, particularly avoided deforestation, represented far less real reduction than claimed, through failures of additionality, baseline setting, permanence and leakage.
What does adaptation work involve?
Flood risk and defence design, heat resilience, water planning, and infrastructure designed for conditions outside the historical record. The central change is that historical statistics are no longer a reliable design reference when the distribution is shifting.
Who converts well into climate roles?
Accountants and auditors into emissions accounting and verification, process engineers into industrial inventories, environmental scientists into reporting, data analysts into monitoring, and civil engineers into adaptation.