Comprehensive Wisconsin Coverage
CARA serves both Public Health and Emergency Management audiences and provides coverage across both regional structures used in Wisconsin preparedness work.
Public Health perspective
- 95 unique public health departments (101 total jurisdiction entries including multi-county secondary mappings)
- 84 local health departments
- 11 federally-recognized tribal nations
- 7 Healthcare Emergency Readiness Coalition (HERC) regions
Emergency Management perspective
- 72 Wisconsin counties (county emergency management agencies)
- 7 Wisconsin Emergency Management (WEM) regions
Each jurisdiction is assessed across 7 primary risk domains using the PHRAT quadratic mean formula to produce an overall risk score, with 2 supplementary domains (Utilities and Cybersecurity) presented for comprehensive planning context.
HERC Region Boundaries: The geographic boundaries used for HERC region mapping and choropleth visualization are approximations based on the official DHS jurisdiction-to-HERC region assignments. HERC region assignments follow the Wisconsin DHS designation for each local health department jurisdiction. Boundary polygon files are not currently published by DHS; the mapped boundaries are derived from the county-level polygons of the member jurisdictions and should be interpreted as planning-level approximations rather than precise administrative boundaries.
Data Sources
All risk calculations are grounded in publicly available, authoritative data sources. The table below lists every data source used, its refresh cadence, and what it provides.
| Data Source | Refresh Cadence | Used For |
|---|---|---|
| FEMA National Risk Index (NRI) | Annual (scheduler every 8,760 h; max acceptable cache age 400 days) | Flood, tornado, winter storm exposure scores; health impact factors |
| CDC Social Vulnerability Index (SVI) | Annual (scheduler every 8,760 h; max acceptable cache age 400 days) | 4 vulnerability themes integrated into EVR framework across all domains |
| CDC NSSP Emergency Department Visits (data.cdc.gov/resource/vutn-jzwm) | Weekly (Fridays) | Wisconsin-specific % of ED visits for Influenza, COVID-19, and RSV; activity level classification and week-over-week trend direction; same underlying NSSP/ESSENCE data that powers WI DHS Tableau dashboards |
| CDC NNDSS Weekly Data (data.cdc.gov/resource/x9gk-5huc) | Weekly (Tuesdays) | Wisconsin state-level weekly case counts for reportable communicable diseases: Measles (Indigenous and Imported), Pertussis, Meningococcal disease. Drives the active_measles_outbreak flag and elevated-disease indicators. State-level signal applied uniformly to all 72 counties (WEDSS county-level case data is not publicly accessible) |
| CDC NHSN Hospital Respiratory Data (data.cdc.gov/resource/ua7e-t2fy) | Weekly (Wednesdays) | Wisconsin state-level weekly ICU bed counts and occupancy, plus confirmed COVID-19, influenza, and RSV hospitalizations and ICU patients, with weekly new-admission breakdowns by age group. Replaces HHS Protect (closed May 2024) and substitutes for the WHA Information Center capacity feed which requires a HERC partner agreement. State-level signal applied uniformly to all 72 counties |
| County Health Rankings (CHR) 2025 | Annual (scheduler every 8,760 h; max acceptable cache age 400 days) | All-ages seasonal flu vaccination rate (BRFSS survey) and primary care physician density (per 100,000 population) for all 72 Wisconsin counties; county range: flu 22% (Taylor/Polk) to 69% (Dane), PC access 9.6/100k (Adams) to 173.8/100k (Ashland) |
| CDC PLACES Local Data for Better Health (Socrata API, keyless) | Annual (scheduler every 8,760 h; max acceptable cache age 400 days) | COPD crude prevalence by county for all 72 Wisconsin counties; BRFSS model-based estimates; county range: 4.1% (Waukesha) to 10.2% (Forest); used as respiratory vulnerability modifier in health metrics EVR calculation |
| WI DHS WIR County Immunization CSV | Annual (scheduler every 720 h; max acceptable cache age 60 days) | MMR (1) vaccination rate for 24-month olds by county (most recent year available); county range: Clark 53.3% to Forest 88.7%; sourced from Wisconsin Immunization Registry (WIR) at dhs.wisconsin.gov/immunization |
| NOAA NCEI Storm Events Database | Weekly (scheduler every 168 h; max acceptable cache age 90 days) | County-level tornado, flood, winter storm, thunderstorm event counts, property damage, injuries, fatalities from bulk CSV data |
| OpenFEMA Disaster Declarations Summaries v2 | Weekly (scheduler cache) | Federal disaster declarations for Wisconsin counties; flood, tornado, winter storm, thunderstorm declaration counts |
| OpenFEMA NFIP Redacted Claims v2 | Weekly (scheduler cache) | National Flood Insurance Program claims by county; total payments and claim counts for flood risk |
| OpenFEMA Hazard Mitigation Assistance Projects v4 | Weekly (scheduler cache) | Hazard mitigation project data for Wisconsin counties; project counts and federal obligations |
| Census ACS Demographics | Annual | Population aged 65+ (%), mobile home %, population density for vulnerability calculations |
| Climate Projection Data (NOAA/IPCC AR6/WICCI/EPA) | Static | Climate zone multipliers in JSON data files for natural hazards and heat projections |
| Gun Violence Archive (GVA) 2023 | Static | Mass shooting incident data for active shooter risk |
| NCES SSOCS 2019-2020 | Static | School safety survey data for active shooter risk |
| Wisconsin Tribal Boundaries | Static | GeoJSON boundary data for 11 federally-recognized tribal nations, including reservation and off-reservation trust land polygons. County weights are pre-computed from a spatial area overlay (GeoPandas, EPSG:3174 Great Lakes Albers equal-area projection). |
| EPA AirNow API | Daily (scheduler every 24 h; max acceptable cache age 2 days) | Air quality index readings; cache-only mode populated by background scheduler |
| NOAA/NWS Excessive Heat Outlook | Weekly (scheduler every 168 h; max acceptable cache age 10 days) | Active heat advisories and forecast heat days per county; surfaces as the acute component of the BSTA extreme-heat temporal model. Refreshed by refresh_all_nws_forecasts() under the nws_heat source key. Per review finding M6 (2026-05-20) the freshness window was relaxed from 2 to 10 days so a single missed weekly refresh no longer marks the source stale. |
| WI DHS Heat Vulnerability Index (HVI) (ArcGIS MapServer, keyless) | Quarterly (scheduler every 2,160 h; max acceptable cache age 120 days) | 4,472 Census block groups covering all of WI with composite HVI z-score plus environmental, health, population, and socioeconomic sub-indices. Aggregated to a 72-county table via unweighted mean of block-group z-scores, normalized 0-1, bucketed to DHS quintile categories. Preferred baseline-vulnerability source for the BSTA acute heat path (CDC SVI fallback) |
| WI DHS EPHT (WEDSS) — lyme-county.csv, west-nile-data-county.csv | Weekly (scheduler every 168 h; max acceptable cache age 14 days) | County-level Lyme disease and West Nile Virus confirmed/probable case counts and crude rates per 100,000 for all 72 Wisconsin counties; refreshed by refresh_all_dhs_vbd_surveillance() under the vbd source key |
| WI DNR Dam Safety Database (ArcGIS FeatureServer, primary) / USACE NID (fallback) | Weekly (scheduler every 168 h; max acceptable cache age 60 days) | Wisconsin dam inventory; hazard classification, downstream population exposure for dam failure risk domain |
Hazards Not Yet Modeled (vs Kaiser Permanente HVA)
The hazards listed below appear in the Kaiser Permanente Hazard Vulnerability Analysis reference framework but are not yet quantified by a dedicated CARA risk domain. They are listed here so users complete their planning with an explicit, documented qualitative workflow rather than an undisclosed gap.
| Hazard | Current CARA coverage | Interim qualitative scoring workflow |
|---|---|---|
| HazMat release (fixed-facility, pipeline, rail) | Not modeled. Indirect proxy only via Active Shooter critical-infrastructure exposure and air quality acute events. | Use the local LEPC Tier II inventory and the Wisconsin DOT rail commodity flow study to score Exposure (proximity to pipeline / Class I rail / RMP facility), Vulnerability (downwind population within 1 mile), and Resilience (HazMat team mutual-aid response time) on the same 0.0 to 1.0 scale, then file the worksheet alongside the CARA PDF export. |
| Agricultural disaster (crop loss, livestock disease, ag-water contamination) | Not modeled. Rural land-use factors appear inside the heat and vector-borne vulnerability components but no dedicated ag-economy domain exists. | Score qualitatively using USDA NASS county crop-value and livestock-inventory percentile, recent FSA disaster designations, and known irrigation-aquifer stress. Treat as a population-displacement risk for counties where farm employment exceeds 10% of the workforce. |
| Civil unrest / mass demonstration | Not modeled. Active Shooter covers targeted violence but not protest-related medical surge or police-mutual-aid demand. | Score Exposure from county-seat / state-capital proximity and history of declared events in the last five years; score Vulnerability from hospital bed depth and ED diversion frequency; document any standing mutual-aid arrangements in the Resilience line. |
| Power-grid failure / long-duration outage | Partial proxy only. The Utilities supplementary domain uses SVI and county characteristics; it does not ingest live ATC / MISO reliability data. | Use the PSC of Wisconsin annual reliability report (SAIDI, SAIFI by utility territory) and the local Continuity of Operations Plan to score generator coverage at hospitals, water plants, and dialysis centers. Combine with the extreme-heat and winter-storm composite scores for cascading-failure planning. |
| Mass-gathering medical (festivals, sporting events, large worship services) | Not modeled. Event-driven medical-surge demand is not in the PHRAT composite. | Score qualitatively per recurring event: peak attendance, transport time to nearest receiving hospital, on-site EMS staffing ratio. Use as a planning overlay rather than a contribution to the composite. |
When any of these hazards is added to CARA as a quantified domain, the row will move into the data-sources table above and the PHRAT weight set will be re-normalized in accordance with the renormalize-on-missing-domain logic already documented for the existing seven domains.
Overall Risk Score (PHRAT Quadratic Mean)
The overall risk score for each jurisdiction is calculated using the Public Health Risk Assessment Tool (PHRAT) quadratic mean formula. This approach prevents a high risk score in one domain from being diluted by low scores in others.
Overall Risk = √(w₁ × Risk₁² + w₂ × Risk₂² + ... + w₅ × Risk₅²)
Where wi are the domain weights and Riski are individual domain scores (0.0 to 1.0).
| Risk Domain | Weight | Rationale |
|---|---|---|
| Natural Hazards | 28% | Highest annualized expected losses in WI per FEMA NRI; covers 4 sub-types (flood, tornado, winter storm, thunderstorm) |
| Active Shooter | 18% | CDC/ASPR identifies as high-consequence threat requiring dedicated PHEP planning |
| Health Metrics (Infectious Disease) | 17% | Core CDC PHEP capability; respiratory disease surveillance and vaccination coverage |
| Air Quality | 12% | Growing impact from increasing wildfire smoke episodes (EPA); geographically uneven across WI |
| Extreme Heat | 11% | Leading weather-related cause of death nationally (NOAA); tempered by WI's northern latitude |
| Dam Failure | 7% | Standalone EVR domain using WI DNR/USACE NID dam inventory; downstream population inundation exposure |
| Vector-Borne Disease | 7% | Lyme disease and West Nile Virus; real county-level incidence rates from WI DHS EPHT with climate range expansion projections |
| Total | 100% |
config/risk_weights.yaml based on local hazard profiles and planning priorities.
Why quadratic mean? Unlike a simple weighted average, the quadratic mean (root mean square) gives greater emphasis to higher-risk domains. A jurisdiction with one domain at 0.9 risk will score meaningfully higher than one where all domains sit at 0.4, even if the weighted averages would be similar.
Residual Risk Formula (CARA EVR Transform)
Individual risk domains are calculated using the CARA-specific Exposure-Vulnerability-Resilience (EVR) transform. This formula is applied consistently across natural hazards, utilities, and extreme heat domains.
Residual Risk = (Exposure × Vulnerability) × (2.0 - Resilience) × Health Impact Factor
CARA EVR Transform vs FEMA NRI: Not the Same Formula
The CARA EVR transform above is not the FEMA National Risk Index residual-risk formula. The FEMA NRI form is:
Risk (FEMA NRI) = Exposure × AnnualLossRate × (1 + SVI) / (1 + ResilienceRating)
The two formulas differ in two material ways and will produce different numbers even given identical inputs:
- CARA uses (2.0 - Resilience) as a multiplicative amplifier in the numerator. With Resilience in [0.1, 0.9] this amplifier is always between 1.1x and 1.9x, so resilience can only reduce the amount of amplification; it can never attenuate risk below the E × V baseline. FEMA NRI's (1 + R) denominator can attenuate risk below the baseline because raising R divides the numerator down.
- CARA folds vulnerability (including CDC SVI themes) directly into the E × V product as a multiplier in [0, 1]. FEMA NRI uses (1 + SVI), a multiplier in [1, 2].
FEMA NRI Expected Annual Loss Score (EALS) data is used as the source for the CARA Health Impact Factor surface below, so FEMA NRI inputs do feed CARA scores; the residual-risk transform itself is CARA-specific. An EM planner cross-walking county scores between CARA and FEMA NRI should expect different numbers.
Component Definitions
- Exposure (0–1): Hazard likelihood based on historical data and geographic factors
- Vulnerability (0–1): Population and infrastructure susceptibility; incorporates CDC SVI themes and Census demographics
- Resilience (0–1): Community adaptive capacity; acts as an amplifier reduction in the numerator, not a denominator divisor
- Health Impact Factor (0.8–1.5): Derived from FEMA NRI EALS data via
_calculate_normalized_health_factor; adjusts for health consequences of the specific hazard
Resilience as Amplifier Reduction
The (2.0 - Resilience) term means:
- Low resilience (0.1) → 1.9× amplifier
- Medium resilience (0.5) → 1.5× amplifier
- High resilience (0.9) → 1.1× amplifier
High resilience reduces the amplifier toward 1.1x but never below it, so the CARA transform does not allow resilience to attenuate risk below the E × V baseline. This is deliberate (it preserves the floor of structural hazard exposure) and is one of the differences from FEMA NRI noted above.
Risk Level Categories
Natural Hazards Domain (28% of Overall)
The natural hazards score is the equal-weighted quadratic mean (RMS, p=2) of four hazard types, each calculated using the EVR framework described above. This is consistent with the outer PHRAT formula and ensures that a single high-severity hazard type elevates the domain score more than a simple average would.
| Flood | Tornado | Winter Storm | Thunderstorm |
|---|---|---|---|
| 25% | 25% | 25% | 25% |
EVR Components by Hazard
- Exposure: FEMA NRI baseline scores supplemented by NOAA Storm Events historical counts and OpenFEMA NFIP claims. Climate change projection multipliers from
data/climate/natural_hazard_climate_projections.jsonare applied per hazard type. Each exposure component is held on its native 0-1 scale with a single layer of documented weights (no hidden pre-scaling). NOAA event counts and NFIP claim counts are normalized to events-per-year and percentile-ranked across all 72 Wisconsin counties so large urban counties do not dominate by area alone. For flood: NRI 30%, NOAA storm-events percentile 20%, NFIP claims percentile 10%, water body proximity 15%, flat terrain 5%, precipitation 5%, climate trend 5%. Tornado, winter storm, and thunderstorm exposures use the same single-weight-layer pattern, each with a NOAA storm-events percentile term (tornado 25%, winter storm 15%, thunderstorm 20%). - Vulnerability: All 4 CDC SVI themes with hazard-specific calibrated weights, plus Census ACS demographics (population aged 65+ (%), mobile home %, population density)
- Resilience: Inverse SVI socioeconomic and housing scores as proxies for community adaptive capacity, with no hard-coded county adjustments. The EVR formula uses a
(2.0 - Resilience)amplifier rather than a simple divisor, so low resilience magnifies risk and high resilience attenuates it without eliminating it entirely. Earlier versions of CARA applied flat bonuses to short lists of "well-resourced" or "EOC-capable" counties; those lists have been removed because they created abrupt cliffs between adjacent counties and were not backed by a cited capacity dataset. A continuous capacity index can be reintroduced in the future if a defensible data source (e.g. emergency-management staffing per capita, hospital beds per capita) becomes available.
SVI Vulnerability Weights by Hazard Type
| Factor | Flood | Tornado | Winter Storm | Thunderstorm |
|---|---|---|---|---|
| SVI: Housing/Transportation | 30% | 25% | 20% | 25% |
| SVI: Socioeconomic Status | 20% | 15% | 10% | 10% |
| SVI: Household Composition | 15% | 10% | 15% | 10% |
| SVI: Minority Status | 10% | 10% | 5% | 5% |
| Census: Population Aged 65+ (%) | 15% | 10% | 20% | 10% |
| Census: Mobile Home % | 10% | 25% | 5% | 10% |
| Census: Population Density | — | 5% | — | — |
| Power Grid Vulnerability | — | — | 15% | — |
| Rural Isolation | — | — | 10% | — |
| Flood Susceptibility | — | — | — | 15% |
| Tree Coverage | — | — | — | 15% |
Climate Change Projections
Wisconsin is divided into 3 climate zones (Southern, Central, Northern) with RCP4.5/SSP2-4.5 scenario projections applied as exposure multipliers:
| Hazard Type | Projected Change Range | Source |
|---|---|---|
| Flood | +15% to +25% | NOAA/WICCI |
| Tornado | +8% to +15% | NOAA/IPCC AR6 |
| Winter Storm | +5% to +12% | NOAA/WICCI |
| Thunderstorm | +12% to +22% | NOAA/WICCI |
data/climate/natural_hazard_climate_projections.json (derived from 20 years of Thunderstorm Wind, Hail, Lightning, and Funnel Cloud event records for all 72 Wisconsin counties), rather than being derived from tornado data.
Health Metrics / Infectious Disease Domain (17% of Overall)
Health metrics risk uses the EVR (Exposure-Vulnerability-Resilience) framework. Respiratory disease surveillance from Wisconsin DHS drives exposure; county-level COPD burden, MMR vaccination rates, and primary care access drive vulnerability and resilience.
| EVR Component | Inputs | Data Source |
|---|---|---|
| Exposure | Flu activity (40%) + COVID-19 activity (40%) + RSV activity (20%); each normalized via sigmoid function from cases per 100k | CDC NSSP Emergency Department Visits (data.cdc.gov/resource/vutn-jzwm, weekly) |
| Vulnerability | MMR vaccination gap (primary driver, 0.1–0.9) + COPD prevalence modifier (up to +0.15 for high-COPD counties; normalized from 4.1%–10.2% WI range) | WI DHS WIR County Immunization CSV (MMR) + CDC PLACES Socrata API (COPD) |
| Resilience | Inverse of MMR vaccination gap (base) + primary care physician density modifier (up to +0.15 for counties at or above 100 physicians per 100k) | WI DHS WIR County Immunization CSV (MMR) + CHR 2025 (PC access) |
| Health Impact Factor | 1.5 (maximum, applied uniformly — infectious disease carries the highest direct health consequence of all domains) | FEMA NRI / CARA calibration |
Strategic Vaccination Risk Multiplier
After the base EVR score is computed, a strategic vaccination risk multiplier (0.7–2.0) is applied based on county-specific herd immunity gap analysis:
- MMR gap: Each percentage point below the 95% measles herd immunity threshold adds 4% to the base multiplier (using county MMR rates from WI DHS WIR).
- Active measles outbreak flag: +30% multiplier when an active measles outbreak is present and the county is below threshold.
- School vulnerability: +20% multiplier when the school vulnerability index exceeds 0.7.
- Flu gap: Counties more than 10 percentage points below the 70% flu herd immunity threshold receive a SEASONAL PRIORITY policy flag (using county flu rates from CHR 2025).
- Multiple simultaneous gaps: +15% compounding adjustment when 2 or more vaccination gaps exceed 5 percentage points.
Respiratory activity levels (Influenza, COVID-19, RSV) are fetched weekly from the CDC NSSP Emergency Department Visits dataset (data.cdc.gov/resource/vutn-jzwm), which provides Wisconsin-specific percent of ED visits per pathogen updated every Friday. This is the same NSSP/ESSENCE data that underlies the WI DHS Tableau respiratory dashboards. County-specific MMR rates (WI DHS WIR), flu vaccination rates (CHR), COPD prevalence (CDC PLACES), and primary care access (CHR) are refreshed annually with a 30-day cache.
Vector-Borne Disease Domain (7% of Overall)
Vector-borne disease risk covers Lyme disease and West Nile Virus (WNV) using real county-level incidence data from Wisconsin DHS, supplemented by environmental and climate range expansion factors.
| Component | Weight | Data Source |
|---|---|---|
| Lyme Disease Incidence Rate | 40% | WI DHS EPHT — lyme-county.csv; crude rate per 100,000 |
| West Nile Virus Incidence Rate | 20% | WI DHS EPHT — west-nile-data-county.csv; crude rate per 100,000 |
| Environmental Risk Factors | 25% | Habitat suitability, deer density, land cover (forested/rural proportion) |
| Climate Range Expansion | 15% | NOAA/IPCC AR6 projections for tick and mosquito range northward shift by 2050 |
Data Refresh
WI DHS EPHT CSV files (lyme-county.csv, west-nile-data-county.csv) are downloaded automatically each week by the background scheduler. These files cover all 72 Wisconsin counties with confirmed and probable case counts sourced from WEDSS (Wisconsin Electronic Disease Surveillance System) as published on the DHS Environmental Public Health Tracking portal. All 72 counties are represented; counties with zero reported cases receive a rate of 0.0 rather than a fallback estimate.
Active Shooter Domain (18% of Overall)
The active shooter domain evaluates population-level environmental and social indicators correlated with elevated risk conditions. This framework is intended for public health preparedness planning, not individual threat assessment.
| Sub-Domain | Weight |
|---|---|
| Historical Incident Density | 25% |
| School & Youth Vulnerability | 20% |
| Social & Community Fragility | 20% |
| Mental & Behavioral Health Risk | 20% |
| Access to Lethal Means | 15% |
Data Sources: Gun Violence Archive 2023 (historical incident density, annual static download); NCES SSOCS 2019-2020 (school safety microdata, static file); Census ACS (isolation, demographics); CDC SVI 2022 (social fragility); County Health Rankings 2025 (mental health providers per 100k and poor mental health days per month); CDC PLACES MHLTH crude prevalence (Socrata API); RAND 2022 Wisconsin Firearm Law Score (static fixed value, 0.65).
View detailed Active Shooter methodology →
Extreme Heat Domain (11% of Overall)
Extreme heat risk combines the EVR residual risk calculation with wet bulb temperature and climate trend adjustments. As of the 2026-05-20 horizon reconciliation (review finding Q7), the calculator returns TWO horizons in the same payload and they are kept strictly separated downstream.
Five Components
- Exposure: annual heat-day count per county normalized to [0, 0.95]. v28.9 source hierarchy: CDC Environmental Public Health Tracking measure 421 (days at or above 90F per county per year) as the canonical observed source, with the legacy NCEI Climate-at-a-Glance heuristic (cache-only) and the statewide constant as cascading fallbacks. The displayed dashboard caption ("Source: ...") reflects which level of the hierarchy was actually used. Threshold reconciliation: EPHT measure 421 counts days at or above 90F, while the legacy NCEI heuristic was anchored on a ~100F monthly-maximum proxy. The same [0, 20] normalization band is retained because the [0, 0.95] mapping was calibrated against typical EPHT 90F observed counts in Wisconsin (roughly 5 in the far north to 20 in southern urban counties); the 100F-anchored heuristic values fall inside the same band by construction, so cascading between the two sources does not distort the EVR exposure scale. The threshold mismatch is the explicit reason EPHT is now canonical and the 100F heuristic is a fallback. For the present-day 12-month score the frequency multiplier is 1.0; the 2050 trajectory applies the 1.4 multiplier.
- Vulnerability: Weighted CDC SVI 2022 themes (socioeconomic 30%, housing-transportation 20%, household-composition 15%, minority-status 10%) plus Census ACS population aged 65+ factor (25%). Same value for both horizons.
- Resilience: Inverse CDC SVI socioeconomic and housing-transportation themes from a 0.5 baseline, clamped to [0.1, 0.9]. Same value for both horizons.
- Wet Bulb Temperature: Statewide NOAA Great Lakes baseline of 0.60. Present-day horizon applies humidity factor 1.0; 2050 trajectory applies 1.2.
- Climate Trend: Present-day horizon uses 1.0 (no mid-century warming applied). 2050 trajectory uses the NOAA WICCI mid-century warming midpoint of 1.35.
Present-day (PHRAT-composited) Heat Risk = (0.7 x Exposure + 0.3 x WetBulb) x Vulnerability x (2.0 - Resilience) x 1.3
Trajectory 2050 (planning panel only) Heat Risk = (0.7 x Exposure_2050 + 0.3 x WetBulb_2050) x V x (2.0 - R) x min(1.6, 1.3 x 1.35)
The trend multiplier is capped at 1.6 on the trajectory horizon to prevent runaway compounding.
Climate Adjustments (NOAA/IPCC) — applied ONLY to the 2050 trajectory panel
- 40% increase in heat event frequency by 2050 (RCP4.5) applied as the NOAA frequency multiplier in the trajectory exposure component
- 2–4°F warming projected for Wisconsin by 2050 (NOAA WICCI midpoint 1.35 used as the trajectory climate-trend factor)
- 20% wet-bulb humidity rise applied to a statewide NOAA Great Lakes baseline of 0.60 in the trajectory wet-bulb component
Data Sources: CDC Environmental Public Health Tracking measure 421 (annual days at or above 90F per county; v28.9 canonical observed source, 12-24 month publication lag), NCEI Climate-at-a-Glance (legacy heuristic fallback), NOAA climate normals (long-term baseline), CDC SVI 2022, U.S. Census ACS, WICCI 2021 Assessment. Per-county wet-bulb grid and a continuous urban-heat-island signal are not yet integrated; both currently use statewide values.
Utilities Domain (Supplementary Assessment)
- Exposure proxies: County geographic position, population density (Census ACS 2022), and utility service territory maps (WI PSC)
- Vulnerability proxies: CDC SVI housing/transportation and socioeconomic themes; Census ACS poverty rate and percentage of population aged 65+
- Resilience proxies: County urbanization level (Census) as proxy for redundant infrastructure and restoration capacity
Four components are assessed, each using the EVR framework:
| Component | Weight |
|---|---|
| Electrical Outage Risk | 30% |
| Utilities Disruption Risk | 30% |
| Supply Chain Risk | 20% |
| Fuel Shortage Risk | 20% |
Exposure, vulnerability, and resilience for each component are modeled based on county infrastructure characteristics, population density, geographic factors, and SVI vulnerability themes.
Cybersecurity Domain (Supplementary Assessment)
- Threat proxies: Population size (Census ACS 2022) as proxy for target visibility; institutional presence from WI Blue Book (government, healthcare, finance)
- Vulnerability proxies: County per-capita revenue (WI DOR 2022) as proxy for IT modernization budget; urban/rural classification (Census); CDC SVI socioeconomic theme
- Capability proxies: County government staffing levels (WI DOR 2022) and urbanization as proxy for access to IT security specialists
The cybersecurity risk score models the relative exposure of public health infrastructure to cyber threats based on three components:
- Threat Landscape (35%): County population size and institutional presence as proxy for target visibility
- Vulnerability (40%): County fiscal resources as proxy for IT infrastructure age, adjusted by CDC SVI socioeconomic factor (up to 25% increase for more vulnerable jurisdictions)
- Capability (25%): County staffing and urbanization as proxy for organizational security readiness (inverted: higher capability = lower risk)
config/risk_weights.yaml and can be set to 0.0 to disable SVI influence entirely.
This domain is included to support comprehensive preparedness planning. Jurisdictions seeking detailed cyber risk assessments should consult CISA resources and conduct dedicated cybersecurity evaluations.
Air Quality Domain (12% of Overall)
Air quality risk combines current EPA AirNow data with strategic climate projections to assess long-term environmental health planning needs.
Components
- Current AQI Baseline: EPA AirNow API data, retrieved via background scheduler and served from cache
- Climate Projections: 40% ozone increase and 110% wildfire smoke increase projected by 2050
- Strategic Assessment: 5-year historical baseline patterns integrated with forward-looking projections
Dam Failure Domain (7% of Overall)
Dam failure risk is a standalone EVR domain assessing the probability and downstream consequence of dam failure for each Wisconsin county.
EVR Components
- Exposure: Dam count, hazard class (High/Significant/Low per USACE/WI DNR classification), and dam condition ratings for dams within or upstream of the jurisdiction
- Vulnerability: Downstream population exposure within modeled inundation zones; population density (Census ACS); CDC SVI housing/transportation theme
- Resilience: Emergency Action Plan (EAP) availability, dam inspection recency, and watershed management capacity
Data Sources
| Source | Role | Refresh |
|---|---|---|
| WI DNR Dam Safety ArcGIS FeatureServer | Primary dam inventory (name, hazard class, condition, EAP status) | Weekly (scheduler cache) |
| USACE National Inventory of Dams (NID) ArcGIS FeatureServer | Fallback when WI DNR unavailable | Weekly (scheduler cache) |
| Census ACS / CDC SVI | Downstream population and vulnerability inputs | Annual |
| FEMA NRI Flood Data | Supplementary inundation zone context | Annual |
CDC Social Vulnerability Index (SVI) Integration
The CDC/ATSDR Social Vulnerability Index is integrated throughout CARA as a core component of vulnerability assessment.
Four SVI Themes
- Socioeconomic Status — poverty, unemployment, housing cost burden, education, health insurance
- Household Composition & Disability — age 65+, age 17 and under, disability, single-parent households
- Minority Status & Language — racial/ethnic minority, limited English proficiency
- Housing Type & Transportation — multi-unit housing, mobile homes, crowding, no vehicle, group quarters
How SVI Is Used
- Natural Hazards: Integrated directly into the EVR vulnerability component with hazard-specific calibrated weights (see Natural Hazards section)
- Air Quality: SVI housing/transportation and socioeconomic themes applied as a combined multiplier (capped at 1.5x) to the base score before PHRAT
- Dam Failure: SVI housing/transportation theme applied as a multiplier (up to 1.2x) to the base EVR score before PHRAT
- Vector-Borne Disease: SVI socioeconomic theme applied as a multiplier (up to 1.15x) to the base EVR score before PHRAT
- Active Shooter and Extreme Heat: SVI themes integrated as weighted inputs within sub-component calculations
SVI data is sourced from the CDC/ATSDR SVI API with a 365-day cache, providing county-level percentile rankings for each theme.
Temporal Framework (BSTA)
CARA provides a Baseline-Seasonal-Trend-Acute (BSTA) temporal framework as a complementary analytical lens for each risk domain. BSTA decomposes domain risk into time-horizon components to support strategic planning context. It is displayed alongside the primary PHRAT score but does not modify the PHRAT calculation directly. Temporal factors such as long-term storm event trends, climate projections, and seasonal patterns are embedded within each domain's underlying data inputs. The BSTA display shows how those temporal dimensions break down for each hazard type.
As of 2026-05 the Acute component was formally retired from every non-infectious domain. CARA is a strategic planning tool, not an emergency-response dashboard, and the previous acute signals (active weather alerts, current AQI, recent storm event counts) duplicated real-time situational-awareness tools that public health departments already use. Non-infectious domains now use a three-component Baseline-Seasonal-Trend (BST) model. Infectious disease retains the Acute component because emerging-pathogen risk genuinely manifests in waves rather than smooth multi-year trends, and CDC NSSP respiratory surveillance provides a defensible weekly signal.
In strategic planning mode, long-term structural risks receive the highest emphasis.
| Component | Non-infectious domains | Infectious disease | Description |
|---|---|---|---|
| Baseline | 60% | 60% | Long-term structural risks (1–10 year horizon), from historical data trimmed-mean (median 60% of values) |
| Seasonal | 25% | 25% | Predictable cyclical variations; 70% canonical hazard pattern + 30% jurisdiction-specific historical adjustment |
| Trend | 15% | Not used | Medium-term directional changes from NOAA Storm Events (10-20 year event-frequency trends), OpenFEMA NFIP claims, NOAA/WICCI/IPCC climate projections, and Census ACS demographic aging. Infectious disease replaces Trend with Acute (county-level multi-year ID trend signal is too noisy to be useful). |
| Acute | Retired | 15% | Active surveillance: CDC NSSP respiratory ED visit activity levels (flu, COVID-19, RSV), MMR / flu vaccination coverage gaps, outbreak alerts. Retired from non-infectious domains 2026-05 along with the Operational Alert Overlay. |
Seasonal Patterns by Hazard
| Hazard | Peak Season |
|---|---|
| Flood | Spring (March–May) |
| Tornado | Late Spring / Summer (May–August) |
| Winter Storm | Winter (November–March) |
| Thunderstorm | Summer (June–August) |
| Extreme Heat | Summer (June–September) |
| Respiratory Illness | Fall / Winter (October–March) |
| Air Quality (Wildfire Smoke) | Summer / Early Fall (June–October) |
CDC Public Health Emergency Preparedness (PHEP) Capabilities Integration
CARA aligns with the CDC's Public Health Emergency Preparedness (PHEP) Cooperative Agreement 2024-2028. All risk assessment results, action plans, and preparedness recommendations support the following PHEP capabilities:
- Community Preparedness — Building community resilience through risk-informed planning
- Community Recovery — Supporting recovery operations for public health systems
- Emergency Operations Coordination — Establishing effective incident command structures
- Emergency Public Information and Warning — Facilitating timely information exchange
- Information Sharing — Maintaining situational awareness across partners
- Responder Safety and Health — Protecting public health workforce
- Medical Countermeasure Dispensing and Administration
- Medical Materiel Management and Distribution
- Medical Surge — Supporting healthcare systems during crises
- Nonpharmaceutical Interventions — Implementing community mitigation strategies
- Public Health Laboratory Testing — Identifying biological and chemical threats
- Public Health Surveillance and Epidemiological Investigation
PHEP Capability Implementation in Risk Assessment and Action Planning
For each risk type, specific PHEP capabilities are identified and integrated into a three-phase implementation approach:
| Implementation Phase | Timeframe | PHEP Capability Focus |
|---|---|---|
| Immediate | 0–30 days |
|
| Short-term | 1–3 months |
|
| Long-term | 3–12 months |
|
Tribal Nation Risk Assessment Methodology
CARA supports all 11 federally-recognized tribal nations in Wisconsin as distinct health jurisdictions, each with its own dashboard, risk scores, and action plans. Because tribal land bases span multiple counties and include both reservation land and off-reservation trust land, a simple single-county mapping would produce inaccurate risk scores. CARA uses a geometric area-weighted approach to resolve this.
County Weight Computation
County weights for each tribal jurisdiction are derived from a spatial area overlay (using GeoPandas in EPSG:3174, Great Lakes Albers equal-area projection) between two authoritative GeoJSON datasets:
- Wisconsin tribal boundary polygons sourced from the U.S. Census Bureau TIGER/Line files, which include both reservation polygons and off-reservation trust land polygons as separate features.
- Wisconsin county boundary polygons (72 counties) from the same TIGER/Line source.
For each tribe, the intersection area between tribal features and each county polygon is computed. Weights are assigned proportionally to intersection area, normalized so all county weights for a tribe sum to 1.0. Counties that contribute less than 0.5 percent of a tribe's total land area are excluded to avoid incorporating boundary slivers or mapping artifacts.
Trust Land Inclusion
Off-reservation trust lands are included in the area-weighted calculation. This has a material effect on several jurisdictions:
- Ho-Chunk Nation: approximately 78.5 percent of total tribal land area is off-reservation trust land distributed across 12 counties (Adams, Clark, Crawford, Eau Claire, Jackson, Juneau, Marathon, Monroe, Shawano, Trempealeau, Vernon, Wood). Omitting trust lands would have reduced the county count from 13 to 5 and would have systematically overweighted Sauk County (which contains the main reservation) to 74 percent.
- Forest County Potawatomi Community: approximately 15.9 percent trust land, including parcels in Fond du Lac and Milwaukee counties that are absent from reservation-only mappings.
- St. Croix Chippewa Indians of Wisconsin: approximately 16.2 percent trust land, concentrated in Burnett County and adding a small Polk County component.
- Bad River Band: trust land in Forest County (1.1 percent of total area) adds to the reservation's primary Ashland County base.
Tribes with no trust land features in the source GeoJSON (Lac du Flambeau, Sokaogon) are represented by reservation-only weights.
Risk Score Aggregation
For each risk domain, county-level risk scores are multiplied by the corresponding county area weight and summed to produce the tribal composite score. This is equivalent to a weighted mean where the weights reflect the geographic distribution of the tribe's land base.
Known Limitations
Last Updated: April 2026