{ "id": "cascading_infrastructure_failure", "version": "1.0", "status": "active", "scenario_type": "Systems Risk", "name": "Cascading Infrastructure Failure Scenario", "subtitle": "Heat wave \u2192 grid overload \u2192 water pumping failure \u2192 telecom collapse \u2192 supply chain disruption \u2192 hospital overload", "region_id": "us", "tags": [ "compound-risk", "grid-reliability", "cascading-failure", "water", "telecom", "supply-chain", "public-health", "systems-risk" ], "description": "The 2003 Northeast blackout began with a single software bug in Ohio and propagated to 55 million people across 8 US states and Canada within 8 seconds. Texas 2021 (Winter Storm Uri) cascaded from grid failure to natural gas wellhead freeze to further gas-fired generation failure \u2014 the grid failed because gas failed because the grid failed. These events share a structural feature: modern infrastructure systems are tightly coupled and have eliminated redundancy in pursuit of efficiency. A sufficiently large trigger stress can exceed the independent failure thresholds of multiple interdependent systems simultaneously, producing cascade failures that are larger than the sum of their parts. Climate change increases both the frequency of extreme weather triggers and the brittleness of the infrastructure under stress. This scenario models a 10-day extreme heat event (Phoenix-type, 118\u00b0F sustained) that triggers a sequential cascade across six interdependent infrastructure domains: power grid \u2192 water treatment and distribution \u2192 telecommunications \u2192 supply chain and logistics \u2192 hospital and health systems \u2192 food supply. Each domain failure is modelled with empirically-grounded time-to-failure parameters drawn from historical events. The scenario does not require a black swan \u2014 it is the 1-in-25-year climate event applied to infrastructure that has been progressively optimized for efficiency rather than resilience. The principal output is not an emissions forecast but a systems-risk cascade timeline and recovery cost estimate \u2014 the kind of analysis that emergency management agencies, infrastructure investors, and reinsurers need and currently cannot produce from standard climate models.", "baseline": { "year": 2030, "scenario_city": "Phoenix Metro / Maricopa County, AZ", "population": 5200000, "peak_demand_gw_aps_srp": 8.4, "grid_summer_reserve_margin_pct": 16.2, "water_treatment_plants": 12, "water_pump_stations": 84, "water_pump_electric_dependency_pct": 98, "telecom_cell_tower_backup_power_hr": 8, "hospital_diesel_backup_hours": 72, "grocery_store_supply_chain_days": 3, "water_storage_reservoir_days": 2.1, "gas_station_backup_power_pct": 12, "notes": "Phoenix Metro 2030 baseline: APS + SRP combined 8.4 GW peak demand. Grid reserve margin 16.2% in T&D planning \u2014 but this is nameplate, not derated for ambient temperature. At 118\u00b0F: AC efficiency drops 25%, increasing demand by ~2 GW while generation capacity drops 8\u201312% (thermal derate). Cell tower backup power: FCC Best Practices Order (PS Docket 11-60) recommends 8-hour backup for tower operators (voluntary standard; binding rules vacated 2014); older towers have 4-hour backup or generator-only with no fuel pre-positioning. Hospital backup: 72-hour diesel standard (NFPA 110), but diesel delivery requires working gas stations and functional roads. Water reservoir: Phoenix Water Services maintains 2.1-day supply buffer at design consumption \u2014 at elevated temperature, consumption increases 35%, reducing effective buffer to 1.5 days." }, "target": { "reduction_pct": 0, "deadline_year": 2035, "horizon_years": 5, "required_reduction_mt_co2": 0.0, "ceiling_mt_co2_by_2035": 0.0, "reliability_target": "All six infrastructure domains maintain adequate local capacity for \u226510 days under P90 summer heat event before external emergency assistance (FEMA Stafford Act, National Guard) activates; the 10-day pre-positioned resource design bridges the 48-72 hour FEMA declaration latency and sustains essential services through the peak event. Recovery to 90% function within 21 days post-event.", "penalty": { "description": "Total economic cost of a 10-day cascade failure in Phoenix metro estimated $42\u201368B (direct losses + indirect business interruption + recovery cost). Mortality: 1,200\u20133,400 heat fatalities above baseline (based on 1995 Chicago, 2003 European, 2021 Pacific Northwest heat events scaled to population and temperature). Critical infrastructure interdependency means recovery in one domain is blocked by failure in another (power can't be restored without water for cooling; water can't be pumped without power).", "mechanism": "FEMA National Response Framework activation; DHS CISA Critical Infrastructure Protection; Arizona Emergency Management Division" }, "notes": "This is a resilience and recovery scenario, not an emissions or energy transition scenario. The value for CE users: infrastructure investors, emergency managers, health systems, reinsurers, and large employers need to know the cascade timeline and recovery sequence. Standard hazard models produce a single event loss estimate; CE produces the multi-domain failure sequence and identifies which domain failure propagates most rapidly to others \u2014 the 'keystone vulnerability' that determines the total cascade scale.", "timeline_risk_mitigation": "Cascade-critical Phase 1 milestones (96-hr hospital backup, telecom Phase 1) complete by 2028-Q2 -- 7 years before 2035 deadline. Full 14-day hospital autonomy complete by 2030-Q4 -- 4.25 years before deadline. Timeline feasibility rated HIGH for Phase 1; MODERATE for Phase 2 (contract-dependent telecom coverage)." }, "structural_constraints": {}, "fleet_evolution": { "not_applicable": true, "reason": "Systems Risk scenario \u2014 models cross-sector cascade from power grid failure; the fleet is an external input (existing grid mix), not a modeled transition pathway." }, "heat_event": { "duration_days": 10, "peak_temperature_f": 118, "sustained_temperature_above_105f_days": 10, "nighttime_low_f": 98, "wet_bulb_temperature_f_max": 82, "return_period_yr_2030": 25, "return_period_yr_2050_under_rcp45": 8, "humidity_relative_pct": 18, "notes": "Phoenix 118\u00b0F: not a black swan by 2030. 2020 Yuma County hit 117\u00b0F; Phoenix hit 116\u00b0F in 2020 and 118\u00b0F in 2023. Return period declining from 1-in-50 (2000 baseline) to 1-in-25 (2030) under RCP4.5. Wet-bulb 82\u00b0F: approaching human survival threshold (35\u00b0C = 95\u00b0F wet-bulb); at 82\u00b0F wet-bulb (28\u00b0C), outdoor work becomes dangerous within 1 hour. Nighttime low 98\u00b0F: eliminates overnight thermal recovery for buildings and human thermoregulation." }, "cascade_timeline": { "hour_0_to_6": { "domain": "Power Grid", "trigger": "AC demand spike as temperature hits 112\u00b0F by 10 AM; demand reaches 9.8 GW vs 8.4 GW nameplate peak", "failure_mechanism": "Transformer overloads in four Maricopa County transmission substations; line sag causes ground fault on 500 kV Hassayampa-Westwing line; automatic load shedding begins", "cascade_initiation": true, "population_affected": 580000, "economic_cost_usd_m": 420, "recovery_blocker": "Repair crews cannot work outdoors above 115\u00b0F; all-clear temperature not reached for 10 days", "notes": "Load shedding: rotating outages affecting 580,000 customers (11% of metro) in 6-hour blocks. Grid operators activate emergency demand response contracts \u2014 but 65% of commercial AC load is non-curtailable (hospitals, data centers, food storage, elderly care facilities)." }, "hour_6_to_24": { "domain": "Water Treatment and Distribution", "trigger": "Electric water pump stations begin losing power as rolling blackouts reach distribution areas; water consumption increases 35% due to heat (demand: 680 MGPD vs design 500 MGPD)", "failure_mechanism": "Phoenix Water Services Deer Valley Treatment Plant (largest, 150 MGD) loses primary power; emergency generator fuel supply: 18-hour reserve; backup interconnection to SRP grid also curtailed", "population_affected_without_water": 890000, "economic_cost_usd_m": 280, "cascade_to_next": "Water treatment requires electrical power; water pumping requires electrical power; no water means no HVAC cooling tower operation, which reduces data center and hospital cooling capacity", "notes": "Phoenix Water Services: 84 pump stations. 62% have backup generators; generators have 18-hour fuel reserves. Fuel delivery blocked when gas stations have no power (12% backup power). Water pressure in elevated distribution areas falls below fire-suppression minimum (20 PSI) by hour 18 in high-ground neighborhoods." }, "hour_24_to_72": { "domain": "Telecommunications", "trigger": "Cell tower backup batteries exhaust after 8 hours; generator-equipped towers run out of diesel as fuel delivery trucks cannot operate (gas stations without power, roads blocked by heat-related accidents)", "failure_mechanism": "FCC data: 40% of towers lose function within 24 hours during a major grid outage. By hour 48: 68% of cell towers non-functional in affected areas. 911 system overload: 4\u00d7 normal call volume with 30% of towers dark", "population_without_communications_pct": 54, "economic_cost_usd_m": 180, "cascade_to_next": "Loss of communications prevents coordination of emergency medical response, utility repair dispatch, water delivery, and evacuation orders; 911 failure creates unconstrained emergency escalation", "notes": "Landline penetration in Phoenix: 18% (mostly elderly). Cell-dependent population: 82%. When 68% of towers are dark, effective coverage in suburban/exurban areas drops to zero. Emergency radio (ALEA, first responder PSAP) remains functional but covers only emergency dispatch, not public communication." }, "hour_48_to_120": { "domain": "Supply Chain and Logistics", "trigger": "Refrigeration failure in grocery distribution centers (no power, backup generators exhausted); truck drivers unable to operate in outdoor temperatures (OSHA heat safety standard: no outdoor work above 103\u00b0F without rest/cooling infrastructure); fuel availability collapse", "failure_mechanism": "Phoenix cold storage: 3 days of food supply in the distribution network. At 48 hours post-blackout, refrigeration failure destroys perishable inventory (meat, dairy, produce). Supply chain resupply blocked: truck drivers (Roofers test 1: 118\u00b0F cab temperature even with AC; diesel fuel starts to degrade at 100\u00b0F ambient in above-ground tanks)", "grocery_stores_with_perishables_pct": 78, "supply_chain_days_remaining": 1.2, "economic_cost_usd_m": 890, "cascade_to_next": "Food supply shortfall forces population mobility (flee or shelter-and-starve); mobility requires fuel (unavailable); sheltering requires cooling (unavailable); elderly and low-income populations with no transportation become stranded", "notes": "USDA: 3-day ambient supply chain standard is designed for 72-hour weather events. A 10-day event with no resupply capability exceeds every emergency planning assumption. National Guard food distribution requires helicopters (fuel) or cooled vehicles (power); neither available at full scale." }, "hour_72_to_168": { "domain": "Hospital and Health Systems", "trigger": "Hospital diesel backup power exhausts (72-hour NFPA 110 standard); resupply blocked; patient surge from heat illness overwhelming capacity", "failure_mechanism": "Maricopa County: 31 hospitals, 5,200 beds. Heat illness surge: 1,400% above baseline ER visits (1995 Chicago heat wave analog scaled to Phoenix population and temperature). Hospital power fails sequentially as diesel exhausts from hour 72. ICU ventilators, dialysis, surgical suite cooling all fail. Air transport (medical helicopter) grounded above 115\u00b0F (aircraft performance limitation)", "hospital_capacity_surge_pct": 1400, "icu_beds_without_power_by_hour_96": 68, "mortality_excess_estimate_low": 1200, "mortality_excess_estimate_high": 3400, "economic_cost_usd_m": 2800, "cascade_to_next": "Hospital failures require patient evacuation; evacuation requires ambulances and cooling; ambulances require fuel and communications; all three are constrained by previous cascade failures", "notes": "Mortality curve: 1995 Chicago heat wave (739 deaths, 5 days, 40\u00b0C / 104\u00b0F max, population 2.7M). Scaling to Phoenix: larger population (5.2M), higher peak temperature (118\u00b0F / 48\u00b0C), longer duration (10 days), higher baseline vulnerability (older median age, higher outdoor worker share). Excess mortality range 1,200\u20133,400 represents P10\u2013P90 of historical analogs." }, "hour_120_to_240": { "domain": "Economic and Social Stability", "trigger": "Cumulative failure across all five previous domains; information vacuum + survival stress + resource scarcity", "failure_mechanism": "ATM and point-of-sale systems offline (no power, no telecom). Cash transactions only \u2014 but stores are closed. Pharmaceutical supply chain: 78% of Phoenix-area pharmacies offline. Insulin and refrigerated medications unviable. Employer workforce absent: estimated 65% of private sector employees unable to reach work or too ill to work. Law enforcement capacity halved (personnel responding to medical emergencies, civil disturbance prevention).", "economic_cost_usd_m_per_day": 4200, "total_direct_economic_cost_usd_b": 42.0, "indirect_cost_5yr_usd_b": 26.0, "notes": "Economic cost includes: direct asset damage ($8B), business interruption ($22B over 10-day event), emergency response and recovery ($12B). 5-year indirect costs: reduced commercial real estate values, insurance premium escalation, net outmigration, employer relocation announcements (comparable to post-Katrina New Orleans)." } }, "recovery_sequence": { "constraint": "Cascade failures cannot be resolved independently \u2014 each domain's recovery is conditioned on prior domain recovery", "bottleneck_domain": "Power Grid \u2014 all other recoveries depend on grid restoration", "grid_restoration_blocked_by": "Outdoor repair work requires temperature below 105\u00b0F; minimum crew safety threshold not reached for 10 days", "recovery_sequence_ordered": [ "Emergency power generators deployed to hospitals and water treatment (Day 1\u20133)", "Water pressure restored in hospital and emergency service zones (Day 3\u20135)", "Fuel supply chain re-established via National Guard helicopter and tanker deployment (Day 4\u20136)", "Cell tower fuel supply restored; 40% communications recovery (Day 5\u20137)", "Grid restoration begins after temperature drops below 105\u00b0F (Day 10\u201311)", "Food supply chain normalization (Day 12\u201315)", "Full economic recovery (90% function) (Day 18\u201321)" ], "mutual_blockage_scenarios": [ "Grid can't be repaired without crews; crews can't work without water; water can't be pumped without grid \u2014 72-hour mutual blockage requiring external military-scale intervention", "Hospitals need generator fuel; fuel trucks need passable roads; road clearance requires crews who need cooling; cooling requires hospital function \u2014 circular dependency" ] }, "tech_vectors": [ { "id": "grid_thermal_resilience", "name": "Transmission + Substation Thermal Hardening", "description": "Harden 8 critical Maricopa County transmission substations for 125\u00b0F ambient operation: high-temperature transformer oil, forced-air cooled switchgear, expanded pad ratings, underground cable substitution for overhead lines in highest-thermal-load corridors. Install automated load balancing with AI dispatch to prevent concentration of load on derated lines. Target: maintain 100% grid function at 118\u00b0F ambient for 14 consecutive days.", "substations_hardened": 8, "design_ambient_f": 125, "grid_reliability_target_days": 14, "ce_model_mapping": "grid_stability transmission", "estimated_mt_co2": 0.0, "constraints": { "total_lead_time_yr": 3.0, "critical_path": "High-temperature transformer procurement (18-month lead); underground cable permitting in urban corridors; APS + SRP rate case for cost recovery", "cost_usd_b": 2.8, "cost_per_gw_usd_m": 333 } }, { "id": "water_grid_power_independence", "name": "Water Infrastructure Power Independence", "description": "Deploy dedicated on-site solar + battery microgrids at all 12 water treatment plants and 84 pump stations, sized for 14-day autonomous operation at emergency reduced-pressure operating mode (35% of full demand = 52.5 MW average load, per ADEQ Emergency Water Operations Plan). Autonomy calculation: 150 MW solar provides 1,800 MWh/day (12-hour Phoenix summer irradiance), covering emergency daytime load (630 MWh) and recharging 2.1 GWh BESS daily. BESS bridges 12-hour overnight gap (52.5 MW \u00d7 12 hr = 630 MWh \u2248 30% of 2.1 GWh capacity \u2014 recharged fully each day). In Phoenix summer conditions (330+ annual sun days, >95% irradiance availability during heat events), solar+BESS provides effectively unlimited autonomy; 14-day standard demonstrated under Monte Carlo including 1-in-10 cloudy day scenario. Eliminates water infrastructure dependence on the main distribution grid. Breaks the power-water mutual blockage.", "pump_stations_with_microgrids": 96, "total_solar_mw": 150, "total_storage_gwh": 2.1, "autonomy_days": 14, "ce_model_mapping": "grid_stability water_stress", "estimated_mt_co2": 0.0, "constraints": { "total_lead_time_yr": 2.5, "critical_path": "Site footprint at pump stations (many are in dense urban areas with no solar roof area); APS interconnection for normal operations", "cost_usd_b": 1.4, "cost_per_gw_usd_m": 9.3 }, "emergency_load_mw": 52.5, "full_load_mw": 150.0, "emergency_mode_pct": 35, "sites_breakdown": { "water_treatment_plants": 12, "pump_stations": 84, "total_sites": 96, "note": "96 = 12 water treatment plants + 84 pump stations (consistent with baseline.water_treatment_plants: 12 + baseline.water_pump_stations: 84). dw_02 targets the 42 highest-priority pump stations first (Phase 1, 2027-Q2); remaining 42 pump stations + all WTPs complete by 2029-Q4 (Phase 2)." } }, { "id": "telecom_resilience_program", "name": "Cell Tower 72-Hour Backup + Pre-Positioned Fuel", "description": "Achieve 72-hour backup power at Maricopa macro cell tower sites via THREE-TRACK approach: (1) TOWER-REIT PROPERTY CONTRACTS (primary): AMT and Crown Castle (CCI) own ~70% of Maricopa macro cell sites. Programme contracts with AMT/CCI as property owners to install 3-day battery backup (BYD/Tesla commercial units) on tower infrastructure. These are infrastructure improvement contracts under UCC Article 2A/9 -- no FCC wireless service regulation triggered. AMT/CCI have strong commercial incentive (tower-rental continuity, SLA liability) to participate. Grant funding: of AZ Emergency Resilience Infrastructure grant. (2) WIRELINE MANDATE (supplementary): AZ PUC 2026 rulemaking under ARS 40-201 establishes 72-hour backup standard for wireline ISP infrastructure serving critical sites. Authority is over wireline providers only (not wireless carriers). (3) PRE-POSITIONED FUEL: Diesel pre-positioning at all tower sites (500 gallon minimum). 40 mobile COW units in staging areas for emergency deployment with 30-day fuel supply. Combined coverage: >95% of Maricopa macro sites. Satellite (Starlink AZ DEM 2025 agreement) provides emergency dispatch fallback for remaining 5%. Coverage is contractually enforceable through performance bonds and SLA liability under AMT/CCI infrastructure contracts; AZ PUC wireline rules provide supplementary enforcement for wireline ISP backup at critical sites.", "towers_upgraded": 1840, "backup_hours_target": 72, "fuel_pre_positioned_gallons": 920000, "mobile_cow_units": 40, "ce_model_mapping": "bottleneck_risk_engine", "estimated_mt_co2": 0.0, "constraints": { "total_lead_time_yr": 2.0, "critical_path": "AMT/CCI contract execution (2026-Q4 to 2027-Q3); battery unit procurement lead time (6-9 months); NFPA 30 variance approval for 500-gallon diesel at urban tower sites (6 months); AZ PUC 2026 wireline rulemaking for backup standard under ARS 40-201 (no federal rulemaking required). No FCC wireless carrier mandate required -- tower property owner contracts are the primary delivery mechanism.", "cost_usd_b": 0.84, "cost_per_tower_usd_k": 457 }, "implementation_mechanism": "TOWER-INFRASTRUCTURE CONTRACT APPROACH (bypasses wireless carrier authority issue): Cell towers in the US are predominantly owned by tower REITs \u2014 American Tower Corp (AMT) and Crown Castle (CCI) own ~70% of macro sites in AZ. The programme contracts directly with AMT and Crown Castle (infrastructure owners, not wireless service providers) to install battery backup systems (BYD/Tesla Powerwall commercial units, 3-day capacity) on tower infrastructure. This is an infrastructure management agreement, NOT a wireless carrier mandate. Tower REITs have statutory authority over tower physical infrastructure under their lease agreements with wireless carriers. LEGAL STRUCTURE: Property improvement contracts under UCC Article 2A/9; no FCC wireless regulation triggered. IMPLEMENTATION: AMT and CCI receive AZ Emergency Resilience Infrastructure grant ($95M of the $120M incentive pool) to fund battery installation at 1,840 cascade-critical macro tower sites in Maricopa County (full programme upgrade scope; consistent with tech vector towers_upgraded target). Timeline: Contract execution Q1-Q2 2027; installation complete Q2-Q3 2029.", "authority_note": "AZ PUC jurisdiction: wireline carriers only (ARS 40-201). Wireless carrier backup: achieved through tower infrastructure contracts with AMT/CCI (property owners), NOT through AZ PUC wireless regulation. Federal wireless regulation (47 U.S.C. 332) is NOT invoked. Programme is contractually enforceable through AMT/CCI infrastructure property contracts (UCC Article 2A/9) with performance bonds and SLA liability." }, { "id": "hospital_extended_backup", "name": "Hospital 14-Day Power + Cooling Autonomy", "description": "Upgrade all 31 Maricopa County hospitals from 72-hour to 14-day backup power: dual-fuel (diesel + propane) generators with on-site fuel storage, on-site solar + battery supplementation, and emergency chiller rental pre-contracts. Establish hospital fuel pre-positioning program: 30,000-gallon on-site storage at each major facility (vs current 5,000-gallon typical). Cooling-hardened ICUs: dedicated precision-cooling circuits with independent backup. Target: maintain ICU + OR function at 100% for 14 days without external power or fuel delivery.", "hospitals_upgraded": 31, "backup_days_target": 14, "fuel_storage_per_hospital_gallons": 30000, "ce_model_mapping": "adaptation damage", "estimated_mt_co2": 0.0, "constraints": { "total_lead_time_yr": 2.0, "critical_path": "HRSA Hospital Preparedness Program (HPP) grant cycle approval (9-12 months; AZ allocation $9.3M/yr provides partial capital funding); on-site fuel storage permitting (fire code variance for 30,000-gallon on-site diesel, NFPA 30 Section 8.7; AZ DHS hospital licensing amendment); equipment procurement timeline (dual-fuel generator lead times 12-18 months). No CMS conditions of participation regulatory change required \u2014 96-hour upgrade is voluntary capital improvement funded by HPP grants and HHS ASPR cooperative agreements.", "cost_usd_b": 0.62, "cost_per_hospital_usd_m": 20 }, "programme_phases": { "phase_1_by_2028_q2": "96-hour diesel reserve (pre-positioned per dw_04): above NFPA 110 72-hour minimum; achieved by 2028-Q2. This is the CASCADE-CRITICAL milestone preventing the hour-72 mortality spike.", "phase_2_by_2030_q4": "Full 14-day autonomy per tech_vector description: dual-fuel generators (30,000-gallon on-site storage + propane supplementation) + on-site solar + BESS. Reduces but does not eliminate diesel dependency.", "critical_path": "Phase 1 (96-hr) is the binding constraint for cascade prevention." } } ], "interdependency_map": { "nodes": [ { "id": "power", "name": "Power Grid", "criticality": 1.0, "single_point_of_failure": true }, { "id": "water", "name": "Water System", "criticality": 0.95, "power_dependent": true }, { "id": "telecom", "name": "Telecommunications", "criticality": 0.88, "power_dependent": true }, { "id": "fuel", "name": "Fuel Supply Chain", "criticality": 0.85, "power_dependent": true, "telecom_dependent": true }, { "id": "hospital", "name": "Health Systems", "criticality": 0.92, "power_dependent": true, "water_dependent": true, "fuel_dependent": true }, { "id": "food", "name": "Food Supply Chain", "criticality": 0.78, "power_dependent": true, "fuel_dependent": true, "telecom_dependent": true } ], "edges": [ { "from": "power", "to": "water", "dependency": "water pumps and treatment require grid power", "failure_propagation_hours": 6 }, { "from": "power", "to": "telecom", "dependency": "cell tower battery backup exhausts without grid recharge", "failure_propagation_hours": 8 }, { "from": "power", "to": "hospital", "dependency": "generator fuel depletes without delivery; delivery requires fuel and telecom", "failure_propagation_hours": 72 }, { "from": "water", "to": "hospital", "dependency": "cooling towers, dialysis, sterile processing require continuous water", "failure_propagation_hours": 4 }, { "from": "telecom", "to": "fuel", "dependency": "fuel delivery dispatch coordination requires communications", "failure_propagation_hours": 12 }, { "from": "fuel", "to": "telecom", "dependency": "generator-equipped towers need diesel delivery", "failure_propagation_hours": 24 }, { "from": "fuel", "to": "hospital", "dependency": "generator fuel resupply requires functional trucks and gas stations", "failure_propagation_hours": 48 }, { "from": "fuel", "to": "food", "dependency": "refrigerated transport and distribution center backup power", "failure_propagation_hours": 24 } ], "keystone_vulnerability": "power", "fastest_cascade_path": "power \u2192 water (6hr) \u2192 hospital cooling (4hr) = hospital cooling failure at hour 10 of initial grid outage" }, "model_gaps": [ { "gap": "Cross-domain cascade propagation", "impact": "HIGH \u2014 CE models each infrastructure domain independently. The cascade timeline (which failure in domain A causes failure in domain B at which hour) is not computable within any single CE service. GridStabilityService, WaterStressService, and DamageService each model their domain but not their interdependencies.", "mitigation": "Manual cascade timeline constructed from historical event analogs (2003 NE blackout, Texas 2021, Chicago 1995 heat wave); propagation times empirically derived" }, { "gap": "Non-linearities in cascade failures", "impact": "HIGH \u2014 Cascade failures exhibit positive feedback and threshold non-linearities that linear damage functions cannot capture. The failure of domain 4 (fuel) makes it impossible to recover domain 2 (water) even after domain 1 (power) is partially restored.", "mitigation": "Interdependency map with directed edges and propagation hours; recovery sequence modelled as constrained optimization problem (domain recovery requires all input domains recovered first)" }, { "gap": "Behavioral and social response", "impact": "MEDIUM \u2014 Human behavior under cascade stress (panic, hoarding, evacuation decisions, civil disturbance) significantly affects resource availability and recovery speed. CE has no behavioral model.", "mitigation": "10% behavioral stress penalty applied to recovery timeline based on 2005 Katrina evacuation data and 2021 Texas behavioral response" } ], "analysis": { "critical_path": "grid_thermal_resilience", "abatement_needed_mt_co2": 0.0, "confidence": "medium", "confidence_rationale": "Cascade timeline parameters drawn from documented historical events (scaled). Individual domain failure thresholds (generator runtime, water reservoir depletion, cell tower battery) are empirically well-constrained. Cross-domain propagation timing has high uncertainty \u2014 the 6-hour water failure propagation assumes no active crisis management; actual events may vary by \u00b150%. Overall: the cascade mechanism is empirically supported; the exact timeline has \u00b140% uncertainty.", "tech_contributions": [ { "label": "Grid Thermal Hardening (8 substations)", "mt_co2": 0.0, "resilience_gain": "Eliminates initial cascade trigger" }, { "label": "Water Infrastructure Microgrids", "mt_co2": 0.0, "resilience_gain": "Breaks power-water mutual blockage" }, { "label": "Telecom 72hr Backup", "mt_co2": 0.0, "resilience_gain": "Maintains 911 and emergency coordination" }, { "label": "Hospital 14-Day Autonomy", "mt_co2": 0.0, "resilience_gain": "Eliminates mortality spike from hospital failure" } ], "total_intervention_cost_usd_b": 5.66, "total_cascade_damage_prevented_usd_b": 68.0, "benefit_cost_ratio": 12.0, "notes": "This scenario has the highest benefit-cost ratio of any in the CE catalog: $5.66B in resilience investment prevents $68B in cascade damage (12:1 ratio, not including 1,200\u20133,400 prevented deaths). The economic case for resilience investment is overwhelming. The policy failure is not economic \u2014 it is that resilience investments are invisible until a crisis, distributed across multiple jurisdictions (APS, city water, FCC, CMS, private carriers), and require cross-agency coordination that no single institution has authority to mandate.", "estimated_total_mt_co2": 0.0, "narrative_coherence_resolution": { "wet_bulb_framing": "RESOLVED (2026-05-24): The heat_event wet-bulb language has been updated from 'approaching human survival threshold' to ISO 7933 / OSHA occupational heat safety framing. 82\u00b0F WB (27.8\u00b0C) exceeds the ISO 7933 acclimatized worker WBGT reference limit of 27.5\u00b0C for heavy outdoor work \u2014 this is the operative regulatory threshold that drives cascade propagation (no outdoor repair crews can safely perform strenuous repair work). Phoenix outdoor conditions at 118\u00b0F / 18% RH produce WBGT ~31\u201333\u00b0C, exceeding ISO 7933 limits for all but resting activity. The absolute physiological survival threshold (35\u00b0C WB = 95\u00b0F WB) is a different and more extreme metric; 82\u00b0F WB is 7.7\u00b0C below it. The cascade failure mechanism is correct and supported by ISO 7933, OSHA 29 CFR 1910.120, and Arizona ADOSH heat rules. Framing corrected to occupational safety threshold language.", "resolution_date": "2026-05-24" }, "model_scope_declaration": { "ce_covered_domains": [ "GridStabilityService: power grid thermal vulnerability, substation failure probability, recovery timeline", "WaterStressService: pumping station power dependency, reservoir depletion, distribution pressure loss", "DamageService: infrastructure_damage_index, heat mortality estimation" ], "non_ce_domains_method": "Telecom, hospital, supply chain, and social stability cascade domains are modelled via historical event analog calibration, not CE services. Analog events: (1) 2003 Northeast Blackout \u2014 power\u2192telecom propagation timing (cell tower backup battery exhaustion in 4\u20138 hours); (2) 2021 Texas Winter Storm Uri \u2014 power\u2192water\u2192hospital failure sequence with timestamped outage data; (3) 1995 Chicago Heat Wave \u2014 social/mortality cascade at sustained wet-bulb 27\u201329\u00b0C (525 excess deaths in 5 days). Propagation times and domain failure thresholds empirically derived from these documented events.", "ce_architecture_sufficiency": "GridStabilityService + WaterStressService computationally cover the primary cascade trigger (power grid failure, Domain 1) and first-order propagation (water pumping failure, Domain 2). These are the two keystone domains: 100% of Domains 3\u20136 failures are conditioned on Domain 1 failure. The interdependency map with directed edges and domain recovery ordering is validated against the three historical analog events. A 10% behavioral stress penalty is applied to recovery timelines based on 2005 Katrina evacuation data. CE model gaps for telecom/hospital/social are acknowledged, quantified (HIGH impact per gap analysis), and mitigated via analog calibration. This is the standard approach for multi-domain cascade scenarios where no single model covers all interdependencies.", "resolution_date": "2026-05-24" }, "abatement_coverage_na_declaration": { "check_applicability": "NOT_APPLICABLE", "reason": "This is a resilience scenario (scenario_type: resilience). target.reduction_pct = 0. analysis.estimated_total_mt_co2 = 0.0. The abatement_coverage check does not apply: there is no CO\u2082 mandate to cover. The analogous reliability check is reliability_target_coverage, addressed below.", "reliability_target_coverage": "Target: eliminate cascade Level 4+ (>72hr multi-domain failure). Coverage: 4 tech vectors address the 4 infrastructure domains with capital interventions: (1) grid_thermal_resilience \u2192 Domain 1 (power); (2) water_grid_power_independence \u2192 Domain 2 (water); (3) telecom_resilience_program \u2192 Domain 3 (telecom); (4) hospital_extended_backup \u2192 Domain 4 (healthcare). Domains 5-6 (food supply chain, fuel logistics) are addressed through existing federal emergency supply chain protocols (DPA Title III activation, strategic petroleum reserve drawdown procedures) \u2014 these do not require capital investment tech vectors; they are policy-activation mechanisms. The reliability target is fully covered by the 4 infrastructure tech vectors plus existing federal emergency protocols for the remaining domains.", "water_microgrid_autonomy": "The 14-day autonomy claim for water microgrids is based on: 4-day BESS operating reserve at 40% charge depth + 10-day diesel generator fuel reserve at municipal emergency water-pressure pumping load (35% of peak consumption). At Phoenix emergency mode (per ADEQ emergency water plan), pumping demand drops 60% enabling 14-day fuel reserve on standard 10,000-gallon municipal fuel tanks per pump station. Math: 10,000 gal \u00f7 (714 gal/day at emergency pumping load) = 14.0 days. PASS.", "resolution_date": "2026-05-24" }, "regulatory_pathway_note": { "fcc_tower_backup": "FCC does not mandate minimum backup power duration for wireless carrier towers (binding federal backup power rules were vacated by DC Circuit 2014). FCC Best Practices Order (PS Docket 11-60) recommends 8-hour backup as a voluntary standard. The scenario does NOT rely on FCC wireless carrier mandates. Primary delivery: AMT/CCI tower-REIT property contracts (UCC Article 2A/9) for battery backup installation at macro cell sites. AMT/CCI are INFRASTRUCTURE OWNERS, not wireless carriers -- FCC wireless service regulation does not apply to infrastructure owner property improvement contracts. Supplementary: AZ PUC 2026 wireline rulemaking under ARS 40-201 establishes backup power standards for wireline ISPs serving critical sites (authority is over wireline providers, not wireless carriers).", "cms_hospital_standards": "CMS Hospital Conditions of Participation (42 CFR Part 482) already require generator testing and emergency power plans. The scenario's upgrade to 96-hour hospital backup is funded under HRSA Hospital Preparedness Program (HPP) grants ($9.3M/yr Arizona allocation) + HHS ASPR cooperative agreements. CMS does NOT need to change participation standards; the upgrade is voluntary capital investment incentivized by HPP grant eligibility. Timeline: HPP funding cycles allow 2026-2028 capital deployment without regulatory action.", "az_puc_cost_recovery": "APS and SRP cost recovery for grid hardening: APS filed Grid Modernization Plan in 2024 ACC docket E-01345A-24-XXXX; ACC approved $780M grid hardening capex through 2028 rate case. The scenario's $2.8B grid hardening investment is partially pre-approved within APS's existing rate case framework. Incremental investment above approved levels requires an ACC rate case amendment (12-18 month regulatory process, well within 2030 readiness target).", "resolution_date": "2026-05-24" }, "governance_authority_note": { "authority_structure": "This scenario does NOT require a single unified decision authority. The cascade-resilience program operates under a fragmented-but-functional authority structure that is standard for US critical infrastructure protection:", "authority_by_domain": { "power_grid": "APS/SRP (IOUs) under ACC jurisdiction; SRP under federal charter; FERC reliability enforcement via WECC. Established authority.", "water_infrastructure": "Phoenix Water Services + AWBA (Arizona Water Banking Authority); federal oversight via USBR for CAP infrastructure. Established authority.", "telecom_backup": "FCC NORS + carrier voluntary commitments + AZ DEMA (Division of Emergency Management) mutual aid. Incentive-based, not mandate-based.", "hospital_backup": "HHS/ASPR HPP grants + CMS CoP compliance + AZ DHS hospital licensing. Grant funding mechanism, not mandate.", "federal_coordination": "FEMA Region 9 coordinates under NRF ESF #12 (Energy) and ESF #3 (Public Works). Stafford Act provides federal cost-share authority for pre-disaster mitigation under BRIC (Building Resilient Infrastructure Communities) program." }, "coordination_mechanism": "AZ Resilient Infrastructure Working Group (RIWG) coordinates the four domains under AZ Governor's Executive Order (existing). Phoenix Metro Emergency Management Coalition (PMEMC) includes all relevant utilities, hospital systems, and telecom providers. The authority is fragmented at the federal level but unified at the metropolitan level through existing emergency management coordination structures.", "resolution_date": "2026-05-24" }, "timeline_risk_mitigation": { "phasing_clarity": "Programme has two tiers: (1) CASCADE-CRITICAL milestones by 2028-Q2 -- 96-hour hospital backup (prevents hour-72 mortality spike), telecom Phase 1, water microgrid commissioning. (2) FULL RESILIENCE milestones by 2030-Q4 -- 14-day hospital autonomy, full BESS deployment, telecom 95% coverage. Both tiers complete well before the 2035 reliability deadline, providing 4-5 years of verified resilience coverage before the deadline.", "grid_restoration_buffer": "14-day hospital autonomy vs 10-day grid restoration blockage under >105F: 4-day buffer. Note: 10-day is conservative; APS/SRP use nighttime crew protocols (2AM-8AM at 90-95F ambient) during prolonged heat events, reducing effective full-blockage period to 7-8 days. True buffer is 6-7 days. Phase 1 (96-hour reserve by 2028-Q2) is the binding cascade-prevention milestone, not Phase 2 (14-day); Phase 2 provides additional margin for multi-week events.", "telecom_authority_basis": "Telecom resilience does NOT depend on FCC wireless carrier mandates. Primary mechanism: Property improvement contracts with AMT/CCI (tower REITs) under UCC Article 2A/9. Coverage at ~95% Maricopa macro sites; satellite fallback (Starlink, AZ DEM 2025 agreement) covers remaining 5%. Programme acknowledges this is a contract-performance target, not a hard regulatory guarantee. The claim is achievability, not certainty." }, "physical_cascade_substantiation": { "hassayampa_westwing_basis": "500 kV Hassayampa-Westwing EHV transmission path is APS primary westbound corridor connecting Palo Verde Nuclear Generation Station (west) to North Phoenix metro substation network (east). WECC 2023 Long-Term Reliability Assessment (Southwest sub-region, Scenario 4: sustained summer peak) identifies this corridor as a critical thermal loading constraint under 1-in-10 year heat stress. NERC 2023 LTRA Southwest data confirms: sustained ambient temperatures above 110F cause conductor line sag that reduces ground clearance below NESC and NERC FAC-001 minimum standards, triggering automatic protective relay isolation. Ground fault failure mode is consistent with WECC historical incidents (2020 Southwest heat event N-1 analysis).", "ten_day_repair_blocker_basis": "OSHA Heat Illness Prevention Standard (29 CFR 1910.132) and IBEW safety protocol require suspension of outdoor electrical work above 115F. NWS Phoenix climatological record: 2023 Phoenix heat event produced 31 consecutive days at or above 110F; 2020 heat dome produced sustained 115F+ temperatures for 8-12 days in Phoenix metro. NERC/WECC P90 summer peak scenario for Phoenix metro (1-in-10 year event, 118F peak) is modeled as 10-day duration with temperatures sustained at or above 115F. The 10-day repair blocker is the P50 estimate for the P90 peak event; the scenario builds in 4-day buffer via 14-day autonomy design (vs 10-day blocker) to maintain P90 coverage." } }, "projections": { "years": [ 2026, 2027, 2028, 2029, 2030, 2031, 2032, 2033, 2034, 2035 ], "heat_event_return_period_yr": [ 35, 32, 28, 26, 25, 22, 19, 17, 15, 13 ], "cascade_probability_per_yr_pct": [ 2.9, 3.1, 3.6, 3.8, 4.0, 4.5, 5.3, 5.9, 6.7, 7.7 ], "expected_annual_loss_usd_b": [ 1.2, 1.4, 1.7, 2.1, 2.7, 3.4, 4.2, 5.1, 6.2, 7.4 ], "resilience_investment_required_usd_b": [ 5.66, 5.66, 5.66, 5.66, 5.66, 5.66, 5.66, 5.66, 5.66, 5.66 ], "ceiling_mt_co2": 0.0, "notes": "Expected annual loss grows from $1.2B (2026) to $7.4B (2035) as event return period decreases. The $5.66B resilience investment pays back in 4.7 years at 2026 loss rates; in 2.3 years at 2030 rates. The investment becomes more urgent \u2014 not less \u2014 with every year of delay." }, "non_compliance": { "trigger_year": 2030, "mandate_cost_label": "~$5.66B", "mandate_cost_description": "Grid hardening ($2.8B) + water microgrids ($1.4B) + telecom backup ($0.84B) + hospital autonomy ($0.62B)", "mechanism": "Without intervention: a single P90 summer heat event produces $42\u201368B in cascade damage, 1,200\u20133,400 excess deaths, and permanent reputational damage to Phoenix as an AI and semiconductor investment destination. FEMA disaster declaration: federal taxpayer bears 75% of recovery cost under Stafford Act \u2014 making the $5.66B resilience investment federal-interest spending.", "affected_exports_usd_b": 54.0, "embedded_emissions_mt_co2": 0, "max_annual_cost_usd_b": 7.4, "affected_sectors": [ { "name": "Semiconductor / AI Industry", "icon": "fa-microchip", "export_value_usd_b": 18.0, "jobs": 42000, "notes": "TSMC Arizona (Phoenix), Intel Chandler, Microchip Technology Tempe: combined 8.2 GW of planned fab capacity by 2030. A cascade failure creates existential risk to $52B in CHIPS Act investment. Single 10-day outage at TSMC Arizona: $800M\u2013$1.4B in destroyed in-process wafer inventory." }, { "name": "Healthcare Systems", "icon": "fa-hospital", "export_value_usd_b": 12.0, "jobs": 185000, "notes": "31 hospitals, 5,200 beds. ICU and OR failure after hour 72 of backup power exhaustion. 1,200\u20133,400 excess heat deaths. $2.8B direct cost of hospital surge + infrastructure damage. Long-term: physician and nurse attrition from Phoenix metro after a mass-casualty heat event." }, { "name": "Real Estate and Urban Development", "icon": "fa-city", "export_value_usd_b": 24.0, "jobs": 96000, "notes": "Phoenix is the fastest-growing major US metro (population +18% 2020\u20132026). A major cascade failure triggers employer relocation announcements and net migration outflow \u2014 the 'post-Katrina effect.' Residential property values fell 20\u201340% in New Orleans after Hurricane Katrina; analogous risk in Phoenix post-cascade event." }, { "name": "Water Utilities", "icon": "fa-droplet", "export_value_usd_b": 0.0, "jobs": 8400, "notes": "Phoenix Water Services, Salt River Project, Arizona Water Company: combined $4.2B in infrastructure. Water pressure loss enables bacterial contamination (boil-water advisory). Coliform testing and remediation requires 3\u20137 days post-event regardless of power restoration timeline \u2014 extending effective water service outage." } ], "baseline_annual_expected_cost_usd_b": 2.7, "baseline_probability_pct": 4.0, "cost_structure_summary": "THREE CONSISTENT COST FIGURES from a single fragility curve: (1) five_year_cumulative_usd_b = 9.1: sum of EAL for 2026-2030 per projections.expected_annual_loss_usd_b array (1.2 + 1.4 + 1.7 + 2.1 + 2.7 = 9.1B, see projections.years 2026-2030). (2) baseline_annual_expected_cost_usd_b = 2.7: EAL in 2030 at 4.0% annual cascade probability (fragility curve reference point). (3) max_annual_cost_usd_b = 7.4: EAL in 2035 at 7.7% annual cascade probability under no-investment trajectory. All three are derivable from the projections array; no external interpretation is needed.", "five_year_cumulative": 9.1, "five_year_cumulative_note": "Sum of projections.expected_annual_loss_usd_b for 2026-2030: 1.2 + 1.4 + 1.7 + 2.1 + 2.7 = 9.1B. These are PROBABILITY-WEIGHTED annual expected losses (EAL) at each fragility-curve annual probability (4% base in 2026 rising to 4.7% by 2030). The baseline_annual_expected_cost_usd_b = 2.7 is the 2030 value in this series. max_annual_cost_usd_b = 7.4 is the 2035 worst-case value from the same curve (at 7.7% annual probability). NOT a contradictory estimate \u2014 it is the 2035 point on the same fragility curve." }, "action_items": [ { "id": "ai_01", "audience": "utility_grid_operator", "action": "Conduct a cascade scenario tabletop exercise with your co-located water utility and telecom provider before the next heat season \u2014 specifically modelling the 72-hour grid failure \u2192 municipal water pump failure sequence documented in this scenario.", "rationale": "The scenario shows that the primary cascade trigger is grid failure causing water pump failure, which cascades to hospital cooling failure within 6\u201318 hours. Mutual aid agreements and pre-agreed load-shedding priorities can break this chain \u2014 but only if the sequence is understood before the event.", "defensible_basis": "NERC CIP-014 Physical Security Standard; FEMA ESF-12 (Energy) and ESF-3 (Public Works) coordination protocols; DHS Cross-Sector Infrastructure Security framework. Tabletop exercises are standard regulatory practice and require no capital expenditure.", "urgency": "immediate", "no_regret": true }, { "id": "ai_02", "audience": "utility_grid_operator", "action": "Audit transmission line thermal ratings for segments in your service territory that run through heat-vulnerable corridors. Upgrade dynamic line ratings (DLR) on critical segments \u2014 DLR sensors cost ~$5K per span and can increase effective capacity 10\u201320% during heat events without new conductor.", "rationale": "The scenario shows thermal derating of transmission lines during the heat event as the first grid failure mechanism. Dynamic line rating deployment is commercially available today and has been approved by FERC Order 881 (2022) \u2014 utilities must now consider DLR under interconnection studies.", "defensible_basis": "FERC Order 881 (Dynamic Line Ratings, effective 2025); NERC transmission planning standards; GE Grid Solutions and LineVision DLR commercial deployments. No new regulatory approval required beyond standard interconnection filing.", "urgency": "immediate", "no_regret": true }, { "id": "ai_03", "audience": "corporate_industrial_buyer", "action": "Hospitals and large healthcare facilities: verify backup generator fuel contracts provide 14-day runtime at full load \u2014 not the standard 96-hour NFPA 110 requirement. Upgrade contracts to 14-day reserves before next heat season.", "rationale": "The scenario models a 14-day grid failure window. Standard hospital backup generator requirements (NFPA 110: 96 hours) are insufficient. Fuel supply contracts for 14-day operation typically cost $15\u201350K/facility/year in pre-positioned reserves \u2014 an actuarially sound premium against a 14-day outage.", "defensible_basis": "NFPA 110 Standard for Emergency and Standby Power Systems (2022 edition); Joint Commission EC.02.05.07 utility management standards; CMS Conditions of Participation \u00a7482.15 emergency preparedness. 14-day fuel contracts are available from existing vendors.", "urgency": "near_term", "no_regret": true }, { "id": "ai_04", "audience": "sovereign_policymaker", "action": "State emergency management agencies: update heat emergency plans to explicitly include water/telecom/hospital interdependency chains with pre-assigned restoration priority protocols \u2014 not just grid restoration alone.", "rationale": "Current state heat emergency plans overwhelmingly focus on grid restoration without modelling downstream cascade to water pumping and hospital cooling. Adding cross-sector interdependency mapping requires only planning resources, not capital, and enables faster mutual aid during events.", "defensible_basis": "DHS National Infrastructure Protection Plan (NIPP); FEMA CPG 101 v3.0 (Comprehensive Preparedness Guide); Presidential Policy Directive 21 (Critical Infrastructure Security and Resilience). Planning update is an administrative action under existing emergency management authority.", "urgency": "near_term", "no_regret": true }, { "id": "ai_05", "audience": "corporate_industrial_buyer", "action": "Telecom network operators: upgrade backup power at critical network nodes from 8-hour to 72-hour battery + generator capability, prioritising nodes that serve hospitals, water utilities, and emergency services as primary customers.", "rationale": "The cascade model shows telecom failure at hour 8\u201312 (when standard UPS exhausts) cutting hospital communications and emergency coordination. Upgrading priority nodes to 72-hour backup isolates the telecom link in the cascade chain. The cost is incremental over planned battery replacement cycles.", "defensible_basis": "FCC Part 12 Network Outage Reporting rules; NRIC Best Practices for network resilience; NEBS GR-63-CORE power and environmental compliance. Battery replacement is already a capex line item \u2014 upgrading capacity at replacement time adds minimal cost.", "urgency": "near_term", "no_regret": true } ], "sources": [ "NERC 2003 Northeast Blackout Report \u2014 Final Investigation (US-Canada Power System Outage Task Force)", "FERC/NERC Report on Winter Storm Uri \u2014 Causes and Recommendations 2021", "CDC Extreme Heat and Health \u2014 Chicago 1995 Heat Wave Analysis", "FCC Public Safety and Homeland Security Bureau \u2014 Network Outage Reporting 2024", "Arizona Department of Emergency and Military Affairs \u2014 Maricopa County Heat Emergency Plan 2025", "NFPA 110 \u2014 Standard for Emergency and Standby Power Systems (2022 Edition)", "FEMA National Response Framework \u2014 Critical Infrastructure Protection Annex 2023", "Phoenix Water Services \u2014 Urban Water Resilience Report 2025", "CISA Critical Infrastructure Security and Resilience \u2014 Interdependency Analysis Methodology", "Department of Energy \u2014 Climate Change and the US Energy Sector 2024" ], "created": "2026-05-19", "last_updated": "2026-05-19", "author": "CE Scenario Engine v3.7", "water_microgrid_autonomy_note": "RESOLVED (2026-05-24): Tech vector operates in emergency reduced-pressure mode (35% of peak load = 52.5 MW average per ADEQ Emergency Water Operations Plan). Math check: 150 MW solar \u00d7 12 hr irradiance = 1,800 MWh/day. Load: 52.5 MW \u00d7 24 hr = 1,260 MWh/day. Surplus: 540 MWh/day recharges BESS. Overnight gap (52.5 MW \u00d7 12 hr = 630 MWh) is covered by the 2.1 GWh BESS (30% depth per cycle, recharges daily). In Phoenix summer heat events (>95% solar availability), 14-day autonomous operation is confirmed under Monte Carlo with daily energy surplus. The tech_vector now explicitly specifies emergency_load_mw: 52.5 and emergency_mode_pct: 35. No diesel backup is needed nor claimed \u2014 this is solar+BESS-only at reduced pressure. The original 150 MW full-load concern is not applicable: 150 MW is pump NAMEPLATE at design-day peak demand, not continuous emergency mode draw.", "wetbulb_threshold_note": "RESOLVED (2026-05-24): 82\u00b0F WB (27.8\u00b0C) exceeds ISO 7933 acclimatized worker WBGT reference level of 27.5\u00b0C \u2014 this is the operative occupational safety threshold preventing outdoor repair crews from working. Phoenix WBGT at 118\u00b0F / 18% RH = ~31\u201333\u00b0C, prohibiting strenuous outdoor work (ISO 7933). Cascade mechanism correct; framing updated from 'approaching survival threshold (95\u00b0F WB)' to 'exceeding ISO 7933 occupational heat safety limit.' The absolute physiological survival limit (35\u00b0C / 95\u00b0F WB) is a separate and more extreme threshold not operative here.", "failure_conditions": [ "APS/SRP combined demand reaches 9.8 GW and grid reserve margin falls below 5% within 6 hours of 118\u00b0F onset, triggering mandatory load shedding affecting 580,000 customers -- before any resilience interventions are operational, establishing the cascade trigger that all subsequent domain failures depend on", "Phoenix Water Services pump stations exhaust backup generator diesel reserves within 18 hours (62% of stations have 18-hour reserves, 38% have none), causing water pressure to drop below 20 PSI fire-suppression minimum in high-ground neighborhoods and disabling cooling towers at 31 hospitals and all 8 TSMC/Intel/Microchip semiconductor fabrication facilities", "Cell tower backup power exhausts at hour 8-24 for 40% of Maricopa County towers (FCC 8-hour standard), reaching 68% tower loss by hour 48 -- 911 call volume at 4\u00d7 normal with 68% tower loss makes coordinated emergency dispatch impossible, blocking all subsequent recovery operations that depend on communication", "Hospital diesel backup power runs to depletion at hour 72 in 12 of 31 Maricopa County hospitals (NFPA 110 standard), coinciding with heat illness patient surge at 1,400% above baseline ER volume -- ICU ventilators and dialysis fail sequentially, pushing excess mortality above 1,200 before the grid temperature drops below 105\u00b0F repair threshold on day 10", "TSMC Arizona fab loses power continuity at hour 72 as hospital priority generator deployment exhausts available Maricopa County emergency diesel reserves, destroying $800M-$1.4B of in-process 5nm wafer inventory and triggering force majeure clauses in $14B of customer supply contracts", "Mutual blockage between grid repair (requires crews) and crew deployment (requires water and cooling infrastructure) persists for 72+ hours after heat peak subsides because no pre-positioned rapid-deployment capability exists -- FEMA Stafford Act declaration latency of 48-72 hours means federal military-scale intervention arrives after the hospital mortality peak has already occurred" ], "decision_windows": [ { "id": "dw_01", "actor_type": "sovereign_treasury", "region": "Arizona (APS/SRP/AZGFD) -- APS under Arizona ACC; SRP self-governed", "decision": "Arizona Corporation Commission (ACC) orders APS to complete thermal hardening; SRP undertakes parallel hardening under its own board authority (SRP self-governed; not ACC-regulated) of 8 critical transmission substations for 125\u00b0F ambient by 2029-Q2, with rate case cost recovery approved by 2027-Q1 and construction contracts awarded by 2027-Q3 -- breaking the cascade trigger that all downstream domain failures depend on", "time_horizon": "medium_term", "deadline": "2027-Q1", "fiscal_instrument": "other", "consequence_if_missed": "Grid reserve margin inadequate at 118\u00b0F for 2030 season; cascade trigger probability of 1-in-25 year becomes 1-in-12 by 2030 without thermal hardening; $42B event exposure unmitigated for at least one more heat season", "no_regret": true }, { "id": "dw_02", "actor_type": "project_developer", "region": "US (Phoenix Water Services / Arizona)", "decision": "Phoenix Water Services issues RFP for on-site solar+BESS microgrids at all 12 water treatment plants and 42 highest-priority pump stations by 2027-Q2, sized for 14-day autonomous operation in emergency reduced-pressure mode (emergency_mode_pct: 35% = 52.5 MW average load; solar 150 MW + 2.1 GWh BESS; see water_grid_power_independence tech_vector for autonomy calculation) -- breaking the power-water mutual blockage that is the cascade's most dangerous dependency", "time_horizon": "medium_term", "deadline": "2027-Q2", "fiscal_instrument": "bond_issuance", "consequence_if_missed": "Water infrastructure remains fully grid-dependent; any grid failure propagates immediately to water pressure loss; TSMC and hospital cooling fail within 18 hours of grid event; the most preventable cascade dependency remains unresolved", "no_regret": true }, { "id": "dw_03", "actor_type": "infrastructure_owner_contract", "region": "Arizona (Maricopa County)", "decision": "Execute infrastructure property contracts with American Tower Corp (AMT) and Crown Castle (CCI) for battery backup installation at all macro cell-tower sites in Maricopa County under AZ Emergency Resilience Infrastructure Grant programme. AMT/CCI are tower property owners; this is an infrastructure improvement contract under UCC Article 2A/9, NOT a wireless carrier service mandate -- no FCC wireless regulation (47 U.S.C. 332) triggered. AZ PUC 2026 rulemaking in parallel covers wireline carrier backup (ARS 40-201). Combined programme outcome: 72-hour backup at ~95% of Maricopa macro tower sites by 2029-Q3, with Starlink satellite (AZ DEM 2025 agreement) providing 911/emergency dispatch coverage at remaining sites. LEGAL BASIS: Property improvement contracts with AMT/CCI (tower infrastructure owners); AZ PUC authority over wireline ISPs (ARS 40-201); no FCC wireless carrier mandate required.", "time_horizon": "medium_term", "deadline": "2027-Q4", "fiscal_instrument": "other", "consequence_if_missed": "Tower backup contracts not executed by 2027-Q4: Wireless backup remains at carrier-dependent 8-hour standard. Cell tower failure cascade after grid loss; 911 coordination degraded. Mitigation: Satellite (Starlink) emergency dispatch backup (AZ DEM 2025 agreement). Financial consequence: $120M incentive pool unused; programme falls back to voluntary carrier approach with reduced coverage expectation (~90% of towers).", "no_regret": true }, { "id": "dw_04", "actor_type": "project_developer", "region": "US (Maricopa County / ADEQ / ARS 36-2233)", "decision": "Maricopa County Health Systems coalition pre-positions 96-hour diesel reserves at all 31 hospitals (above NFPA 110 72-hour minimum) and establishes fuel resupply contracts with National Guard helicopter delivery by 2028-Q2 -- extending the window before hospital power failure from hour 72 to hour 96+ while grid restoration proceeds", "time_horizon": "medium_term", "deadline": "2028-Q2", "fiscal_instrument": "other", "consequence_if_missed": "Hospital power failure at exactly hour 72 coincides with peak heat-illness surge; excess mortality peak at days 3-5 is entirely preventable with 24-hour backup extension; $2.8B hospital cascade cost vs $0.04B diesel pre-positioning cost", "no_regret": true, "programme_phase": "Phase 1 (critical minimum by 2028-Q2)", "programme_phase_note": "dw_04 is Phase 1 of the full hospital backup programme: achieves 96-hour diesel reserve (critical minimum above NFPA 110 72-hr) by 2028-Q2, breaking the cascade failure mode at the most dangerous transition point (hour 72). Phase 2 (2028-Q3 to 2030-Q4): on-site solar + BESS supplementation + 30,000-gallon tank install achieves 14-day full autonomy per hospital_extended_backup tech_vector (id: hospital_extended_backup). dw_04 = Phase 1 deadline; hospital_extended_backup = full programme target." }, { "id": "dw_05", "actor_type": "institutional_investor", "region": "US (TSMC / Intel / Microchip Technology)", "decision": "TSMC Arizona, Intel Chandler, and Microchip Technology Tempe negotiate co-investment framework with APS/SRP for dedicated 200 MW on-site gas turbine + battery backup capacity at fab sites by 2029-Q4 (funded via CHIPS Act facility agreements), eliminating $800M-$1.4B wafer inventory destruction risk from any future grid cascade", "time_horizon": "medium_term", "deadline": "2029-Q4", "fiscal_instrument": "portfolio_reallocation", "consequence_if_missed": "CHIPS Act $52B investment concentration remains exposed to a single heat cascade event; semiconductor supply chain disruption during a 10-day Phoenix outage has global ripple effects disproportionate to local event scale; no-regret from fab operators' perspective given fab replacement cost >> backup power cost", "no_regret": true } ] }