Building world’s leading asset-level climate scenario model

Starting point

By 2020, climate risk had moved from a specialist concern to a board and supervisor priority. Institutional investors, banks, and large corporates were all being asked the same questions: how exposed are you to climate risk, how robust are your transition plans, and how credible are your net-zero targets.

Existing tools tended to split into two camps. On one side were engineering-led models that captured physical risk in detail but did not translate easily into financial decisions. On the other were financial models that produced portfolio scores or simple alignment metrics, often with thin links back to actual assets, technologies, and locations.

The consulting firm wanted to build something different. The ambition was to create a climate scenario model that could:

• Link real assets to corporate structures and portfolios.
• Provide a consistent emissions history for thousands of companies.
• Project trajectories under multiple climate pathways.
• Combine physical and transition risk in one framework.
• Help clients judge whether corporate climate targets were realistic or simply aspirational.

This model would need to operate at asset level where data existed, scale to more than two thousand public companies, and produce decision-ready outputs that could be understood by investment committees, risk officers, and regulators. It also had to be robust enough to withstand scrutiny from supervisors and, eventually, potential acquirers.

‍

Approach

The firm stood up a cross-functional build team spanning energy systems, financial risk, data engineering, and climate analytics. The core group included Bhuvan Maingi, who now leads Strathen Group’s work on this space, as a co-lead responsible for architecture, analytics design, target realism metrics, and client readiness. The mandate was clear: design and build a world-class climate scenario engine that was technically sound, transparent, and commercially usable.

The work unfolded in four main streams.

I. Architecture and data model

The team first defined the architecture and data model. They needed a structure that could connect assets, companies, sectors, and regions in a way that supported both bottom-up and top-down views.

They designed an entity graph that linked:

• Individual assets and facilities, tagged with technology, location, and key operational attributes.
• Legal entities and listed companies that owned or controlled those assets.
• Sector and regional taxonomies aligned to how clients thought about risk and allocation.

Data came from a combination of Bloomberg, Trucost, public filings, and vetted public sources. The data model was then structured so that each data point had clear lineage, versioning, and validation rules. Outliers were flagged, assumptions were recorded, and refresh cycles were defined so that clients and auditors could see how the model evolved over time.

II. Historical emissions and projections

With the architecture in place, the team built the emissions history and projection engine.

For history, they assembled emissions data from 2000 to 2020, focusing on Scope 1 and 2 as a minimum and adding Scope 3 where data was reliable. Where direct data was missing, they used conservative, documented methods to estimate values based on technology, energy use, and sector benchmarks. All estimation logic was transparent so that users could understand where the model was interpolating.

For projections, they designed a climate pathway engine covering roughly 2021 to 2050. Climate scenarios representing approximate 1.0-, 1.5-, and 2.0-degrees trajectories were defined in collaboration with energy system and policy specialists. These pathways translated global and regional climate trajectories into asset-level expectations: technology shifts, utilization changes, fuel mix evolution, and policy and pricing impacts.

Under the hood, assets with different technologies and locations responded differently to each pathway. A coal plant in one jurisdiction might see accelerated phase-out under a 1.5 degree-like scenario, while a renewable asset might see increased utilization and expansion. These asset-level responses were then rolled up to company, sector, and portfolio views, with the ability to toggle scopes and scenarios.

III. Physical and transition risk integration

The third stream integrated physical and transition risk into a single workflow.

On the physical side, the model drew on hazard data for heat, flood, storm, drought, and wildfire, mapped to asset locations. For each asset, the team assessed how hazard frequency and severity changed under different climate pathways and what that meant for operational risk, downtime, and potential damage.

On the transition side, they modeled policy, technology, market, and liability drivers. This included carbon pricing trajectories, likely policy measures in key jurisdictions, technology cost curves, and demand shifts by sector. The goal was not to predict exact policies, but to provide structured, scenario-consistent pressure on emissions-intensive assets and business models.

These elements fed into exposure metrics and illustrative financial impacts, such as potential margin pressure, stranded asset risk, or capex requirements. The integration allowed users to see, for a given company or portfolio, which parts of the business were more exposed to physical risk, which were more exposed to transition risk, and how that varied across scenarios.

IV. Target realism and decision-ready outputs

A distinctive feature of the model was the introduction of a target realism coefficient, R.

For any company with stated climate targets, the model compared the target trajectory against:

• The company’s historical emissions and starting point.
• Sector- and region-specific decarbonization pathways drawn from the scenario engine.
• Feasible rates of reduction given technology, capital intensity, and regulatory context.

The R coefficient did not pass moral judgment. Instead, it provided a transparent measure of how ambitious or conservative a target was relative to pathway-consistent reductions, and how much “stretch” remained. Targets that implied reductions far beyond feasible pathways would be flagged as potentially unrealistic, while flat or weakly declining trajectories in hard-to-abate sectors would be flagged as under-ambitious.

Once the analytics were in place, the team focused on making outputs usable. They designed issuer and asset scorecards, exposure heatmaps, and portfolio views that could support:

• Board and investment committee discussions.
• TCFD-style disclosures on governance, strategy, risk, metrics, and targets.
• Engagement conversations with issuers on the credibility and sequencing of their plans.

Instead of another opaque climate score, the model exposed the wiring from asset-level physics and locations through to company trajectories, target realism, and portfolio risk.

Outcome

The climate scenario model gave the consulting firm and its clients something that had been largely missing from the market: a single, transparent engine that connected engineering reality with financial decision-making at scale.

For institutional investors, the tool provided:

• Asset-to-portfolio emissions trajectories from 2000 to 2050, with clear scope toggles.
  ‍
• Side-by-side comparisons of companies and sectors under different temperature pathways.
  ‍
• Integrated physical and transition risk views that highlighted where value was most at risk.
  ‍
• A realism coefficient that allowed them to benchmark corporate targets within and across sectors.

For banks and corporates, the model supported analysis of lending books, capital plans, and decarbonization strategies. They could see which assets and business units were most exposed, which investments or closures would matter most for achieving stated goals, and how their plans compared to peers and plausible pathways.

Internally, the project also changed how the consulting firm approached climate analytics. The data model, scenario library, and documentation standards became a reference for subsequent offerings. Teams across energy, financial services, and risk could reuse components rather than building bespoke tools from scratch.

Externally, the model’s credibility was tested in commercial and regulatory environments. The transparent architecture, documented methods, and decision-ready outputs made it easier to explain to supervisors, board members, and risk committees. The value of the engine was ultimately validated by its acquisition and integration into a global asset manager’s climate platform, where it became part of a broader ecosystem of analytics used by asset owners and managers worldwide.

Clients who used the platform could now, for example, run a portfolio through multiple climate pathways, see how physical and transition risk accumulated, and test how portfolio choices or issuer engagements might shift exposures. They could also interrogate issuer-level R coefficients to distinguish between marketing-heavy climate commitments and plans that were aligned with plausible transition paths.

The work demonstrated how rigorous modeling and thoughtful product design can translate a complex topic like climate risk into tools that real decision-makers will use. The project required balancing scientific and financial perspectives, handling imperfect data with care, and designing outputs that support governance rather than overwhelm it.

For risk and investment leaders, the real shift was moving from climate as a disclosure obligation to climate as a quantified, scenario-based lens for capital allocation and strategy.

The climate scenario model continues to inform how Strathen Group thinks about climate analytics today: linking asset-level detail to board-level decisions, insisting on transparency in methods and data, and treating target realism as a first-order question, not an afterthought.