Loading Dashboard
Verulam Blue · Applied Strategic Analysis · March 2026
Data & Assumptions

Sources, methodology and modelling constraints

This page documents the data sources, transformation logic, unit conventions, and known limitations underlying the dashboard. The scenario outputs shown on Tab 2 of the Power BI dashboard are scenario-based estimates derived from explicit assumptions and should not be interpreted as forecasts. All source data is publicly available. This is an independent project with no affiliation to any government or defence organisation.

Source Datasets

Five datasets are loaded into a Microsoft Fabric Warehouse and transformed through a dbt pipeline before reaching the Power BI semantic model. All five reside in the schema USA_Global_Military_Presence_WH.dbt_dev_landing.

DatasetDescriptionCoverage
Base Structure Report FY25U.S. base inventory with plant replacement values and geo-coordinates. Site-level granularity.FY2025 snapshot
DMDC Personnel ReportDefense Manpower Data Center personnel counts by location. Country/state level granularity.Snapshot 2025-12-31
GEM Main (field-level)Global Oil and Gas Extraction Tracker - field identifiers, names, coordinates, and country attribution. March 2026 edition.Global · ~190 countries
GEM Production (field-level)Production volumes at field level in source units. Requires unit standardisation before aggregation.Global · March 2026
GEM Reserves (field-level)Reserve volumes at field level across multiple classification schemes. Dashboard defaults to 2P (Proved + Probable).Global · March 2026
Pipeline Architecture

Raw data is loaded into Fabric Warehouse via Get Data. A dbt pipeline then runs through three layers before the mart: staging (raw extraction and text standardisation), intermediate (canonical mapping, unit enrichment, region assignment), and mart (dimensions, facts, and the map locations model). The Power BI semantic model connects to the mart via an Import mode connection to the Fabric Warehouse SQL analytics endpoint.

LayerMaterialisationSchema
StagingViewsdbt_dev_staging
IntermediateViewsdbt_dev_intermediate
MartTablesdbt_dev_mart

The mart contains eleven models: five military (dim_geography, dim_site, dim_branch, fact_base_inventory, fact_personnel_assignment), five energy (dim_energy_field, dim_fuel_type, dim_reserves_classification, fact_energy_production, fact_energy_reserves), and one map model (map_locations). Military and energy schemas share no SQL join; cross-schema calculations are performed in DAX only.

Unit Conventions

Source production volumes arrive in mixed units and are standardised to two canonical forms before reaching the mart layer.

FuelProduction unitReserves unitConversion applied
Oilbbl/daybblMillion bbl/year × 1,000,000 ÷ 365
GasMWh/dayMWhMillion m³/year × 1,000,000 × 0.01056 ÷ 365
Gas heating value assumption
Gas is converted using a standard heating value of 38 MJ per m³ (1 MWh = 3.6 GJ, giving 38 ÷ 3,600 = 0.01056 MWh per m³). This is a market convention. Actual heating values vary by field and composition.

BOE records are retained in the pipeline for traceability but excluded from all value calculations, as they represent mixed hydrocarbon streams that cannot be priced with a single unambiguous market price. Gas and condensate, oil and gas, and unknown fuel types are similarly excluded. Their combined omission is estimated to affect approximately 5 to 10 percent of total gross value.

Price Assumptions & Gross Loss Methodology

Reference prices are set to closing market levels on 27 February 2026, the day before the conflict began: $70/bbl for oil (Brent) and $45/MWh for European TTF gas. These are pre-war baselines. The scenario calculator allows wartime prices to be applied separately.

Gross loss is computed as two distinct components valued on different price bases by design.

ComponentFormulaPrice basis
Lost RevenueProduction × disruption % × duration (days) × TWAPTime-weighted average disrupted price - barrels sold into the wartime market
Reserve ImpairmentReserves × impairment %Pre-war price - permanent asset loss measured against the pre-conflict baseline

Oil and gas markets are characterised by inelastic short-run supply and demand, meaning supply losses produce disproportionate price responses. A short-run price multiplier of approximately 6-7× per 1% of supply lost is applied for oil; approximately 10× for gas, reflecting the additional constraints of pipeline dependency and limited short-run switching. Duration multipliers moderate the average disruption price over time as strategic reserves, alternative supply, and demand adjustment take effect.

Conflict duration - 60-day baseline
The 60-day baseline reflects the midpoint of the Defence Intelligence Agency's assessed Strait of Hormuz closure range of one to six months, consistent with the Dallas Fed's base case modelling. As of the report date the conflict is already 25 days old with no ceasefire in effect.
Baseline KPIs - Tab 1
2.81M
Total U.S. military personnel - all locations, snapshot 2025-12-31
$2.47T
U.S. base replacement value - plant replacement value, FY25, all sites
70.76M
Total oil production (bbl/day, global, all fields)
100.64M
Total gas production (MWh/day, global, all fields)
60.11bn
Total oil reserves (bbl, 2P classification)
37.75bn
Total gas reserves (MWh, 2P classification)
Known Analytical Constraints
Granularity mismatch - military
fact_base_inventory is at site level; fact_personnel_assignment is at country/state level. Personnel cannot be attributed to a specific base. Base replacement value and personnel counts are therefore independent figures.
Reserves classification double-count
dim_reserves_classification contains multiple schemes (1P, 2P, 3P, contingent, prospective). Summing across schemes overstates reserves materially. The dashboard filters to 2P (Proved + Probable) throughout.
Semi-additive personnel
personnel_count is semi-additive across time. All personnel figures are filtered to the single snapshot date of 2025-12-31 to prevent double-counting across periods.
Georgia double-row
dim_geography contains two rows for Georgia - one for the U.S. state (us domestic) and one for the country (europe). Surrogate keys are generated from country_or_state concatenated with region to ensure uniqueness.
Import mode semantic model
The Power BI semantic model connects to the Fabric Warehouse via the SQL analytics endpoint in Import mode. Data reflects the state of the mart layer at the time of the last scheduled refresh.
Data, methodology, and limitations: All data used in this project is drawn from publicly available sources and processed through a reproducible data pipeline. Scenario outputs are estimates generated by applying user-defined parameters to baseline figures and should be interpreted as indicative rather than predictive. This case study is intended to demonstrate modelling structure, data integration, and scenario design. It does not claim completeness, operational accuracy, or real-time situational awareness.
Verulam Blue · Matthew Barr · Analytics Engineering (Microsoft Fabric · dbt · Power BI)
Analysis Overview

Key findings and scenario interpretation

Situation Summary

On 28 February 2026, the United States and Israel launched coordinated strikes against Iranian military and leadership targets, initiating the 2026 Iran war. Iran responded with missile and drone attacks targeting U.S. military installations, Gulf energy infrastructure, and commercial shipping in the Strait of Hormuz. As of the date of this report, the conflict is in its third week with no ceasefire in effect.

This report estimates the gross economic loss and exposure associated with damage to U.S. military infrastructure and physical destruction of regional energy supply across the Middle East. The analysis evaluates a baseline scenario and presents scenario comparisons illustrating how gross economic loss scales under alternative conflict assumptions. It does not attempt to forecast the duration, escalation, or outcome of the conflict.

Summary of Findings
FindingHeadline Result
Price RedistributionThe same price shock that costs China $6.22bn net earns the USA $72.44bn and Russia $28.64bn - with no additional physical damage to either country.
Symmetric EscalationUnder matched escalation severity, the GCC absorbs $111.35bn in gross energy losses vs $69.84bn for Iran - approximately 1.6× - due to greater production scale.
Dual Chokepoint ClosureSimultaneous Hormuz and Red Sea closure multiplies gross exposure 4.4× over a Hormuz-only baseline, from $22.29bn to $97.20bn.
Defence Alignment GapSaudi Arabia ranks first for energy exposure and last for U.S. military presence - a rank delta of −7. Oman has the highest exposure-to-military ratio at 103:1.
Country ConcentrationSaudi Arabia ($6.06bn) and Iran ($6.05bn) are the two largest single-country energy exposures. Kuwait is the primary co-location risk.
Duration vs IntensityA 180-day moderate conflict (Scenario B, $95bn) generates ~30% higher total exposure than a 45-day high-intensity conflict (Scenario A, $73bn).
Binding Escalation LeverDuration is the dominant single lever: extending conflict to 180 days produces $63.51bn vs $42.90bn for intensity escalation and $38.94bn for price escalation.
$1 Trillion Stress CaseTrillion-scale exposure ($1.04T) requires simultaneous extreme assumptions across all parameters. Reserve impairment alone cannot reach this threshold.
Key Findings
FINDING 01 - Who Benefits from Disruption Without Bearing Physical Damage?
Price redistribution delivers a $72.44bn windfall to the USA and $28.64bn to Russia. China is the net loser at −$6.22bn.

All physical damage parameters are held constant. Only energy prices change, reflecting today's market prices versus pre-conflict levels - Brent crude $102/bbl, TTF gas ~$57/MWh.

MetricUSARussiaChina
Oil Production (bbl/day)12,614,9925,690,8332,875,767
Oil Windfall (60 days)$24.22bn$10.93bn$5.52bn
Gas Windfall (60 days)$48.22bn$17.71bn$7.34bn
Import Cost (60 days)N/AN/A−$19.08bn
Net Position (60 days)+$72.44bn+$28.64bn−$6.22bn
The USA captures the largest windfall, driven primarily by its dominant gas production position. China's import burden of −$19.08bn overwhelms its domestic production windfall of +$12.86bn. The temporary lifting of Iranian sanctions provides China only partial relief - Iranian oil is still sold above pre-war baseline prices. Energy price shocks do not distribute economic pain evenly; they redistribute it according to whether a country produces or consumes.
FINDING 02 - Symmetric Infrastructure Destruction: Iran vs GCC
The GCC absorbs $111.35bn in gross energy losses vs $69.84bn for Iran - approximately 1.6× - under matched escalation severity.

Parameters are held symmetric across both sides. The asymmetry in outcomes is structural, not parametric - the GCC has a larger production base and more energy assets at risk.

MetricBaseline IranBaseline GCCEscalation IranEscalation GCC
Total Energy Exposure$6.05bn$11.60bn$69.84bn$111.35bn
Oil Destroyed Value$3.38bn$8.14bn$32.61bn$78.45bn
Gas Destroyed Value$2.67bn$1.94bn$37.23bn$27.02bn
U.S. Base Damage (Gross)N/A$11.01bnN/A$27.52bn
Iran's losses are oil-dominated at baseline but shift toward gas-led destruction under escalation. GCC losses exceed Iran's primarily due to scale and price amplification, not disproportionate severity. U.S. base damage costs in the GCC reach $27.52bn under escalation - reinforcing that Gulf states combine the highest energy exposure with the greatest direct U.S. military replacement risk. Rhetoric around mutual obliteration masks a structural imbalance: the side with the largest export-oriented energy system stands to lose more economically under sustained infrastructure warfare.
FINDING 03 - Dual Chokepoint Closure: Strait of Hormuz and Red Sea
Simultaneous closure multiplies gross exposure 4.4× over the Hormuz-only baseline - from $22.29bn to $97.20bn.

The Houthi Red Sea closure is treated as a direct extension of the Iran war scenario, not an independent event.

MetricBaseline (Hormuz)Dual ChokepointIncremental
Total Energy Exposure$22.29bn$97.20bn+$74.91bn
Oil Destroyed Value$15.59bn$66.81bn+$51.22bn
Gas Destroyed Value$5.02bn$27.97bn+$22.95bn
Oil Reserve Exposure$1.65bn$2.36bn+$0.71bn
Oil is the dominant incremental risk at +$51.22bn. Total destroyed values increase by more than threefold despite output destruction rates only doubling, reflecting the combined effect of higher prices, greater destruction intensity, and extended duration. Reserve exposure rises modestly, driven by higher prices rather than changes in impairment rates.
FINDING 04 & 05 - Defence Alignment Gap and Country Concentration
Saudi Arabia ranks first for energy exposure and last for U.S. military presence (rank delta −7). Oman has the highest exposure-to-military ratio at 103:1.

A negative rank delta indicates military presence ranks lower than energy exposure - a structural defence gap.

CountryTotal Energy Exp.Energy RankBase Damage CostMilitary RankRank Delta
Saudi Arabia$6.06bn1$08−7
Qatar$2.20bn2$1.31bn20
UAE$2.19bn3$228.81M4−1
Oman$1.13bn4$10.94M7−3
Kuwait$1.02bn5$8.84bn1+4
Bahrain$122.83M7$626.82M3+4
Oman has the highest calculable exposure-to-military ratio at 103.29 - more than $103 of gross energy exposure for every $1 of U.S. military replacement value. Saudi Arabia's rank delta of −7 is driven by the absence of publicly disclosed U.S. base replacement values in the dataset; it reflects a measurement limitation, not a confirmed absence of U.S. presence. Kuwait and Bahrain both show strong positive alignment (+4), with military replacement value materially exceeding energy exposure. Within this scope, U.S. military presence does not systematically track energy exposure.
FINDING 06 - Trade-offs Between War Intensity and Duration
A 180-day moderate conflict generates ~30% higher total exposure than a 45-day high-intensity conflict - $95.04bn vs $73.37bn.
MetricReference CaseScenario A (Short, Hard)Scenario B (Long, Moderate)
Energy Output Destroyed20%50%20%
Days of Destruction6045180
Oil Price ($/bbl)$70$120$100
Total Energy Exposure$22.29bn$73.37bn$95.04bn
Oil Destroyed Value$15.59bn$50.11bn$66.81bn
Gas Destroyed Value$5.02bn$17.48bn$25.82bn
Duration and price persistence dominate cumulative impact. Sustained production losses over 180 days, combined with elevated prices, outweigh the effects of a shorter period of extreme output destruction. Higher intensity increases daily losses but does not guarantee higher total damage. Reserve impairment acts as a secondary amplifier rather than a primary driver. Economic risk is more sensitive to the duration of energy-market disruption than to peak destruction intensity.
FINDING 07 - Which Escalation Lever Becomes Binding?
Duration is the dominant single lever. Extending from 60 to 180 days produces $63.51bn - materially higher than price escalation ($38.94bn) or intensity escalation ($42.90bn).

Single-variable design: all but one parameter held at baseline, isolating each lever and ruling out interaction effects.

MetricBaselineDuration ×3Price EscalationIntensity ×2
Days / Price / Output60d / $70 / 20%180d / $70 / 20%60d / $120 / 20%60d / $70 / 40%
Total Energy Exposure$22.29bn$63.51bn$38.94bn$42.90bn
Oil Destroyed Value$15.59bn$46.77bn$26.72bn$31.18bn
Gas Destroyed Value$5.02bn$15.06bn$9.32bn$10.04bn
Intensity narrowly outperforms price escalation at these parameter levels. Reserve impairment contributes modestly only under price escalation, where higher prices mechanically increase the dollar value of impaired reserves. This single-variable design confirms the finding from Finding 06: duration is the dominant driver of cumulative exposure. Prolonging disruption is the most powerful way to raise cumulative economic exposure - strategic risk depends less on how destructive a conflict becomes on any given day, and more on how long energy markets remain disrupted.
FINDING 08 - Under What Conditions Does Gross Exposure Exceed $1 Trillion?
Trillion-scale exposure ($1.04T) requires simultaneous extreme assumptions across all parameters. Reserve impairment alone cannot reach this threshold.
ParameterTest ATest BTest CTest DTest E
Output Destroyed20%20%20%20%50%
Days60606060365
Reserve Impairment0.1%0.5%1%2%2.9%
Oil Price ($/bbl)$70$70$70$70$200
Total Energy Exposure$22.29bn$29.03bn$37.46bn$54.31bn$1.04T
Exceeds $1 Trillion?NNNNY
Tests A-D confirm that reserve impairment alone cannot reach $1 trillion at any parameter level within the model - even at the maximum impairment rate of 2.9%, total exposure reaches only $54.31bn with all other parameters at baseline. Test E breaches the threshold only under a combined extreme scenario: maximum reserve impairment coinciding with 50% output destruction, a 365-day disruption, and extreme price escalation. In that scenario, destroyed production value ($895.51bn) is the dominant contributor; reserve impairment ($139.62bn) is significant but secondary. The $1 trillion outcome should be interpreted as a theoretical upper-bound stress case, not a central or likely scenario.
Model Linearity Note

The scenario calculator is linear by construction. Every variable acts independently, scales linearly, and has no thresholds, feedbacks, or state dependence. Single-parameter sensitivity tests are therefore tautological - they confirm that more of X produces proportionally more Y, and are retained for coefficient transparency only. The analytical value in this report derives from findings that test combinations, reveal structural asymmetry, and generate conclusions that would not be obvious without the model: the simultaneity questions, the threshold analysis, and the geographic concentration findings. The model is a scenario accumulator, not a dynamic system, and it is used accordingly.

Scope & Baseline Assumptions

The baseline parameter set uses 20% base damage, 20% energy output destroyed, 60 days of disruption, 0.10% reserve impairment, $70/bbl oil, and $35/MWh gas. The 60-day baseline reflects the midpoint of the Defence Intelligence Agency's assessed Strait of Hormuz closure range of one to six months, consistent with Dallas Fed base case modelling. Oil and gas prices are pre-war references set to closing market levels on 27 February 2026, the day before the conflict began.

The geographic scope covers the Middle East: Bahrain, Iran, Iraq, Israel, Kuwait, Oman, Qatar, Saudi Arabia, UAE, and Yemen for energy assets; Bahrain, Israel, Jordan, Kuwait, Oman, Qatar, Saudi Arabia, and UAE for U.S. military presence. Egypt and Turkey are excluded from the military scope; Iran, Iraq, and Yemen carry no U.S. base replacement value by design. Iran reserve impairment is modelled at zero by design; total Iranian losses may be understated as a result.

The scope of oil production in this analysis includes "oil" and "crude oil" fields only, as classified in the Global Oil and Gas Extraction Tracker (GEM). Fields categorised as oil and gas or other mixed hydrocarbon streams are excluded by design, as these categories cannot be priced consistently against a single oil benchmark without additional allocation assumptions. Widely cited global oil production figures of around 100 million barrels per day typically refer to aggregated “all liquids” measures, which include a range of mixed hydrocarbon streams. As a result, reported oil volumes in this analysis represent a conservative, price‑consistent subset of commonly reported headline figures and should not be interpreted as “all liquids” production.

Analyst Notes

This scenario calculator treats all fields within a selected country as uniformly affected by the damage percentage. In practice, damage would be geographically concentrated and affect fields differently based on proximity to conflict zones, hardening, and redundancy. A more granular framework would assign damage probabilities at field level.

Reserve impairment represents permanent destruction or inaccessibility of reserves through wellhead destruction, reservoir pressure loss, or long-term field abandonment. The 0.10% baseline reflects the upper realistic bound based on historical precedent - Kuwait 1991, the most destructive documented conflict, produced less than 0.10% permanent reserve loss. Analysts should carefully distinguish between temporary production loss and permanent reserve impairment when interpreting outputs.

All dollar figures represent gross economic loss or gross economic exposure valued at reference prices, prior to adjustment for costs, substitution, market response, or recovery. They do not represent net, realised, discounted, or compensable economic losses. Production loss and reserve exposure represent different time horizons and should not be summed as a single realised loss figure.

About this analysis: This analysis is a scenario-based economic impact assessment built using publicly available data. It is presented as a case study demonstrating Verulam Blue’s analytical, data engineering, and modelling capabilities. The scenarios shown are illustrative stress cases derived from explicit assumptions and do not constitute forecasts, intelligence assessments, or policy advice. All conclusions are analytical in nature and intended to support understanding of how economic exposure scales under different conflict conditions.

Methodology: All scenario figures are computed in Power BI using DAX and publicly available source data. The data pipeline is built using dbt-fabric 1.9.8 on Microsoft Fabric Warehouse. Validation queries were run on 26 March 2026, and all baseline metrics reconcile exactly via SQL.
Verulam Blue · Matthew Barr · Analytics Engineering (Microsoft Fabric · dbt · Power BI)