User Interpretation Guide
Product ID: SPC-CDI-2026-03 Issued: 0600Z 03 MAR 2026 Supersedes: Legacy SIGN (Significant Severe) hatching conventions
1. PURPOSE
This guide assists users in interpreting the Conditional Intensity forecast product, implemented March 3, 2026, as an enhancement to existing SPC Convective Outlooks. The product separates two forecast parameters previously combined in categorical risk assessment: coverage probability (the likelihood that a hazard will occur within a given area) and conditional intensity (the expected severity of the hazard, given occurrence).
Users have historically conflated these parameters. A low-probability event may produce catastrophic intensity. A high-probability event may remain sub-severe across all affected areas. The product exists because this distinction has operational consequences.
2. CATEGORICAL RISK TIERS
The five-tier categorical system remains unchanged. Users should verify their current tier assignment before proceeding to conditional intensity assessment.
| Tier | Designation | Color | Threshold Criteria |
|---|---|---|---|
| 1-MRGL | Marginal | Dark green | Isolated event of limited duration. A single forcing mechanism enters the user's awareness, e.g., 50.0°C at Port Augusta, South Australia, January 30, the southernmost such reading in Australian history. Probability of organized response: ≥2%. |
| 2-SLGT | Slight | Yellow | Organized events in low coverage. Multiple forcing mechanisms present within a 30-day window, e.g., 18 cold-exposure fatalities in New York City, January 25–February 9. User notes co-occurrence with events at other tiers. Probability of organized response: 5–15%. |
| 3-ENH | Enhanced | Orange | Greater concentration of events with varying intensity. User identifies hazard-type convergence within a single geographic area, e.g., D3 Extreme Drought across 43% of Florida producing the National Fire in Big Cypress National Preserve (25,700+ acres), generating smoke advisories across Collier County, February 26. Probability of organized response: 15–30%. |
| 4-MDT | Moderate | Red | Widespread events, several intense. User identifies structural connection between Federal Register Doc. 2026-03157 (effective April 20, 2026) and the forcing mechanisms producing events at lower tiers. Probability of organized response: 30–45%. Issued fewer than once per year. |
| 5-HIGH | High | Magenta | Rare. Fewer than once per 500 assessment periods. All forcing mechanisms active simultaneously. The user cannot locate a discrete boundary between the system and the environment in which the system is being assessed. Probability of organized response: ≥45%. |
3. CONDITIONAL INTENSITY GROUPS
The categorical tier describes coverage probability. It does not describe what happens inside the event.
Conditional Intensity Groups (CIG1, CIG2, CIG3) communicate expected severity of the hazard given occurrence. A user at 1-MRGL with CIG3 faces a different operational situation than a user at 3-ENH with no CIG designation. These groups were derived from 20 years of mesoanalysis data.
| CIG Level | Descriptor | Operational Indicators |
|---|---|---|
| CIG1 | Lower-end severity | Transient physiological response: elevated heart rate, shallow breathing, mild jaw tension. Resolves within minutes of cessation of exposure to the forcing mechanism. Sleep disruption possible but not persistent. |
| CIG2 | Moderate severity | Sustained response not resolving upon cessation of exposure. Indicators include involuntary pattern recognition during unrelated tasks, altered information-consumption behavior, difficulty distinguishing between forecast anxiety and ambient physiological state. Chest tightness may persist without identifiable proximate cause. |
| CIG3 | Higher-end severity | Reserved for rare, historic-level intensity. Response has become indistinguishable from the environmental condition being assessed. The system is no longer external to the observation platform. |
4. DECISION SUPPORT SCENARIOS
Scenario A. 100% of Florida in drought for the first time in the 26-year monitoring record (February 26, 2026). Coverage probability: HIGH. Conditional intensity for a user located outside the affected area: CIG1. The event is large. The user is not in it. If indicators persist beyond the assessment period, proceed to CIG2 evaluation.
Scenario B. EPA determines that Clean Air Act Section 202(a)(1) "does not authorize the Agency to prescribe emission standards in response to global climate change concerns." Coverage probability for any single user on any single day: MRGL. Conditional intensity: CIG3. The probability of encounter is low. The intensity, should encounter occur, is not bounded by the encounter's duration. If the user has difficulty determining whether the assessment period has ended, CIG3 is confirmed.
Scenario C. Sixty-two Australian stations record highest-ever daily maximums between January 26 and 31. User is located in the Northern Hemisphere. Coverage probability: 0%. Conditional intensity: assess locally. Assess locally is issued when the intensity forecast cannot be determined from the product alone. Users should evaluate proximity using all available parameters, including but not limited to geographic distance.
5. PRODUCT LIMITATIONS
The Conditional Intensity product demonstrates skill in discriminating between intensity levels under historical climate conditions. It does not forecast whether historical climate conditions will persist through the assessment period.
No framework exists for evaluating compounding intensity across multiple consecutive periods. Users experiencing persistent CIG2 or CIG3 conditions should note this limitation. No supplementary product is planned.
The product separates probability from intensity because these are operationally distinct parameters. Guidance for conditions in which both parameters are elevated simultaneously falls outside the product's scope.
Next issuance: 1300Z 03 MAR 2026.
Things to follow up on...
- Lawsuits filed same day: On February 18, a broad coalition of health and environmental groups filed suit in the D.C. Circuit challenging the EPA's rescission of the endangerment finding, with the effective date of April 20 still ahead.
- No supplementary product planned: The EPA has simultaneously stopped requiring U.S. polluters to report their emissions, raising the question of whether the data infrastructure underlying any future intensity assessment will exist.
- Attribution as intensity calibration: A World Weather Attribution analysis found that heatwaves like Australia's January event are approximately five times more likely now than in a pre-industrial climate, and that estimate is likely conservative.
- The observation platform erodes: The Washington Post laid off at least 14 climate journalists on February 4, largely reversing a 2022 expansion to more than 30 climate-focused staff, following similar cuts at CBS News.

