Scientific transparency, validated methods, continuous improvement.
Storm Predict computes its own meteorological indices from raw data.
90%+ validation vs AROME on 540 tested data points.
Convective Available Potential Energy — Measures the potential energy available to lift an air parcel. All variants computed using NWS/ECMWF formulations.
Thresholds: 1000 J/kg = strong storms | 2500 J/kg = violent storms
Lifted Index — Galway (1956) — Temperature difference between a lifted parcel and the environment at 500 hPa.
LI < -6 = severe instability | LI > 0 = stable atmosphere
Convective Inhibition — Energy needed to "break the cap" and trigger convection. Computed as surface → LFC integral.
Equivalent Potential Temperature — Critical for elevated storms. Identifies thermal convergence zones and moisture flux.
Storm structuring — Vector wind difference between levels. Essential for identifying supercells and tornadoes.
>20 m/s = supercell-favourable environment
Rotation and tornado potential — Measures horizontal vorticity in low levels. Critical for tornado prediction.
>150 m²/s² = significant tornado risk
Storm-relative rotation — Projected onto Bunkers storm motion vector. Better discriminates environments truly favourable for mesocyclones.
Relevant shear zone — Thompson et al. (2007) method. Calculates shear only in the layer where CAPE is actually present.
Supercell risk — Combines CAPE, effective shear and helicity. Used operationally by the NWS.
SCP > 5 = highly favourable supercell environment
Strong tornado risk — Integrates instability, shear, helicity and cloud base height.
STP > 1 = significant EF2+ tornado risk
Combined energy/rotation — Product of CAPE × helicity. Identifies zones where both energy AND rotation are present simultaneously.
Convection mode — Energy/shear ratio. Discriminates between isolated cells, multicells and squall lines.
Cloud base — Altitude where condensation begins. Determines storm base height and tornado potential.
Free convection onset — Altitude where a parcel becomes warmer than the environment and rises spontaneously.
Theoretical storm top — Altitude where the parcel temperature equals the environment again. Determines maximum vertical development height.
Atmospheric cap — Energy required to reach the LFC. Strong CIN can delay initiation → explosive storms afterwards.
Trigger zones — Identification of convergence lines where air masses meet and rise.
Upper-level forcing — Stratospheric air intrusions destabilising the troposphere.
Divergence/convergence zones — Upper-level rotational structures promoting lift.
Right entrance / Left exit — Upper-level divergence zones in the jet stream favouring storm development.
Available fuel — Quantification of moisture transport and identification of moisture-loaded flows (warm conveyor belt).
Cold pool interactions — Identification of zones where cold pools collide or interact with the inflow, potential triggers for new cells.
Supercell trajectories — Global reference for supercell motion prediction. Deviation calculated relative to mean wind via hodograph analysis.
Automatic storm detection — Density-based spatial clustering of lightning data. Parameterised for European conditions.
Probabilistic simulations — Meteorological parameter variation within realistic ranges to generate alternative scenarios and uncertainty cones.
Venturi effect and channelling — Valley acceleration, terrain blocking, seeder-feeder. SRTM 30m topographic base.
Motion adaptation — Isolated cells, multicells (retrograde propagation), supercells (right-mover deviation), squall lines (ensemble motion).
Our algorithms and forecasts are built on the latest peer-reviewed scientific research
Feldmann et al. (2025) • Science Advances
+50% more supercells on the northern slopes of the Alps by 2050 under 3°C warming. The study underlines the urgent need for European countries to prepare with advanced predictive tools.
Read the full studyRädler et al. (2019) • npj Climate and Atmospheric Science
Significant increase in the frequency of severe weather events across Europe by the end of the century. Convective instability will rise due to increased near-surface moisture.
Read the full studyBattaglioli et al. (2023) • Journal of Applied Meteorology and Climatology
Analysis of lightning and large hail trends across Europe from 1950 to 2021. Confirms the increase in extreme phenomena and validates high-resolution predictive approaches like AROME.
Read the full studyEuropean Environment Agency (2023) • EEA Report
€55 billion in insured losses in Europe in 2023. The EEA stresses that advanced early warning systems are an economic necessity, not a luxury.
Read the full reportBunkers et al. (2000) • Weather and Forecasting
Global reference for supercell trajectory calculation. Used operationally by the NWS, NOAA, and Météo-France. The foundation of Storm Predict's Diamond Trajectory Engine.
Read the full studyAll these studies are peer-reviewed and published in leading scientific journals. Storm Predict relies on research validated by the international scientific community.
Storm Predict is committed to full transparency about its methods, results, and limitations. Meteorology remains complex — perfection does not exist. We tell you this clearly.
Our algorithms improve continuously with every storm tracked. Future AI will learn from millions of real data points collected 24/7.