Methodology - Asset-Specific Vulnerability Framework (2/2)

An asset's financial exposure to a climate hazard is a function of its specific physical characteristics and operational purpose. For example, a sustained period of extreme heat will have a profoundly different financial consequence for a data center, which relies on constant, energy-intensive cooling, than it will for a transportation bridge. To quantify these differences, we use an Asset Sensitivity Matrix.

The coefficients in the matrix are assigned based on a 1-to-5 ordinal scale. This scale represents the relative degree to which an asset's financial performance is sensitive to the underlying climate drivers. Each score corresponds directly to a multiplier, which is then applied to the baseline geographic risk.

Score	Sensitivity Level	Multiplier	Description
1	Minimal Impact	0.0x	The asset is not impacted by the climate drivers. This effectively zeroes out the baseline risk for this metric.
2	Some Impact	0.5x	A noticeable but minor impact. The effect is measurable but unlikely to be a primary driver of financial performance.
3	Normal Impact	1.0x	A significant and clear impact. The asset's sensitivity is aligned with the baseline geographic risk.
4	Elevated Impact	1.5x	A primary and critical impact. The baseline risk is amplified and is a major driver of financial performance.
5	Extremely Elevated Impact	2.0x	An existential or transformative impact. The baseline risk is severely amplified and could alter the asset's economic viability.

Asset Type	Maintenance Increase	Operational Downtime	Efficiency Loss	Additional CapEx	Climate Discount Rate	Insurance Premium Increase	Insurability Index	Utility Demand Increase
Residential Buildings	3	2	1	3	2	4	3	5
Commercial Buildings	4	3	2	4	3	4	3	5
Power Plants	4	4	5	4	4	4	3	1
Electricity T&D	4	4	4	4	4	4	3	1
Water & Wastewater	4	5	3	5	3	3	3	1
Transport (Road & Rail)	5	3	1	4	2	2	1	1
Airports	4	4	3	5	4	4	4	2
Seaports	5	5	2	5	4	5	5	1
Data Centers	4	5	5	4	5	5	4	5

Illustrative Example: Data Center Maintenance Increase

This matrix is used to scale the baseline, geography-only metrics. This two-layer process provides a more accurate and actionable financial risk estimate.

• Step 1: Calculate Baseline Metric. Based on climate projections for a location, the geography-based model calculates a baseline Maint_Increase of 3.0%.

• Step 2: Identify Asset Type and Sensitivity Score. The asset is a "Data Center." We consult the matrix and find the sensitivity score for "Maintenance Increase" is 4.

• Step 3: Apply Sensitivity Multiplier. Based on the sensitivity scale, a score of 4 (Elevated Impact) corresponds directly to a multiplier of 1.5x.

• Step 4: Calculate Final Asset-Specific Impact. The baseline metric is multiplied by the sensitivity-derived multiplier: Final_Maint_Increase = 3.0% × 1.5 = 4.5%.

Methodological Limitations

This framework is designed to balance scientific rigor with scalability and transparency. The most significant methodological choice is applying Sensitivity Coefficients at the aggregate metric level, rather than at the level of each underlying climate feature.

For example, the Maintenance Increase metric is calculated from five climate features (heat, humidity, etc.). Our framework calculates a single, blended Maintenance Increase score for a location and then applies one sensitivity coefficient to that total score for a given asset.

While a more granular, component-level model appears more precise, we have chosen the aggregate approach for three key reasons:

• Scalability and Transparency: A component-level approach would create a combinatorial explosion of coefficients (hundreds), rendering the model opaque and impractical to scale. The current 9x8 matrix is transparent and manageable.

• Avoiding False Precision: Defining and defending hundreds of micro-level coefficients could introduce more error and uncertainty than it resolves, as the required data often does not exist.

• Maintaining Clear Directionality: The current framework provides clear, defensible insights (e.g., "Data Centers have an extreme sensitivity to efficiency loss"). A component-level approach risks obscuring this primary insight in a sea of complexity.

The primary limitation is that this simplification may mask nuance in cases where an asset has extreme and opposing sensitivities to different components of a single metric. We contend this is a justifiable trade-off for a model that is robust, explainable, and fit for strategic financial decision-making.