Hail Forecast

Overview

The Central United States experiences frequent hailstorms due to the collision of warm, moist air from the Gulf of Mexico with dry, unstable air from the interior of North America. These conditions fuel powerful thunderstorms where hailstones form and fall once they become too heavy.

For the solar industry, hail poses serious risk to photovoltaic (PV) systems. Reliable hail forecasting provides early warnings and supports preventive actions, helping reduce damage from one of the most severe and fast-developing weather hazards affecting solar power plants.

Solargis approach to forecast the hail events

Hail is challenging to forecast with exact precision, but certain atmospheric conditions—such as strong instability, deep storm development, and powerful updrafts—are clear indicators of hail potential.

Solargis Forecast targets hail stone sizes that exceed standard PV design limits. Standard PV modules are tested under IEC 61215 using a 25 mm (approximately 1 inch) ice ball propelled at high velocity. This test represents resistance to moderate hail, not extreme events. Hail around 1.5 inches (~38 mm) can generate micro-cracks in tempered glass, leading to latent performance loss. Hail of 2 inches (~50 mm) or more often exceeds certification thresholds and can cause glass fracture and immediate module failure.

Input data and processing methods

Our hail forecast for CONUS (Continental United States) region is based on the NOAA High-Resolution Rapid Refresh (HRRR) model. HRRR is a real-time, convection-allowing atmospheric model with 3 km spatial resolution, updated hourly and assimilating radar observations. 

The model generates detailed grid-level signals indicating potential hail occurrence and expected hailstone size. Because these signals are highly localized and uncertain in time and space, they are aggregated across both dimensions to improve reliability. The processed output is a probabilistic hail risk metric that reflects the strength and spatial consistency of the forecast signal. 

Raw hail forecast data are post-processed using temporal and spatial aggregation to improve reliability. The processing includes:

• Time aggregation: A 6-hour evaluation window (±3 hours) is applied, with the highest weight assigned to the central forecast hour and gradually decreasing weights toward the outer hours.

• Spatial aggregation: The hail risk footprint is expanded around detected hail cores to a radius of up to 50 km (~31 miles). The highest probability is assigned to the core, with probability gradually decreasing toward the periphery.

• Signal consistency: Hail risk increases for larger or clustered hail kernels and when the signal is confirmed across multiple hourly forecast cycles.

Additional post-processing harmonizes different HRRR forecast cycles. The main HRRR runs at 00, 06, 12, and 18 UTC provide forecasts up to 48 hours, while intermediate hourly runs extend to 18 hours. In Solargis, these outputs are homogenized to deliver a seamless 43–48 hour forecast horizon.

HRRR data are typically available with a 2–3 hour latency. To account for this, the highest weighting for the nearest upcoming forecast hour is generally taken from the fourth forecast hour of the respective model run, rather than from the earliest lead times.

Figure 2: Visualisation  of Solargis hail forecast data for central USA on 23 Sep 2025 at 22:00UTC.

The Map on the left shows potential for hail events [%] with 25 mm (1 inch) size, on the right satellite image from GOES-E GeoColor (NASA GIBS) documenting the extensive atmospheric cyclone. The chart shows the time series of the hail forecast signal for 25 mm, 38 mm and 50 mm size potential at the location marked on the map

Why global models are difficult to use Worldwide

HRRR is particularly well-suited for hail forecasting compared to global NWP models (e.g., ECMWF IFS). With a high-resolution 3 km grid and hourly updates, HRRR can explicitly simulate individual convective cells, strong updrafts, and storm-scale processes that drive hail formation. It also incorporates near real-time radar observations, improving the accuracy of storm initiation, intensity, and location. In contrast, global models operate at coarser resolution and rely on convective parameterization, which limits their ability to capture localized, short-lived severe storms. This makes HRRR more effective at providing realistic timing, structure, and spatial detail for hail-producing thunderstorms—essential for short-range, impact-based hail forecasts.

Access Solargis Hail Forecast

Currently, Solargis hail forecasts are available for the CONUS region, including southern Canada and northern Mexico, corresponding to the HRRR model domain. Forecasts are delivered via the Solargis Forecast API and push (FTP/SFTP) provided hourly for subscribed locations.

Three hail-risk parameters are available, each corresponding to a specific hail-size threshold:

• HAILRISK25MM

• HAILRISK38MM

• HAILRISK50MM

For more details about FTP/ SFTP , go to Push Delivery for Solargis Monitor & Forecast.

Parameters via SFTP

To retrieve this parameters via SFTP, you can request via support@solargis.com

For an API request the following parameters are mandatory:

• Site coordinates: Defined by geographic latitude and longitude

• Time period: Defined by datetime values

• Summarization: Provided as HOURLY

For more details about the Solargis Forecast API request, go to Monitor & Forecast API endpoint.

Figure 3: Solargis Forecast API request and response example.

Real example

The screenshot depicts data from a real world event occured in the US in 2003 resulting in damage of the PV plant.

If an API that forecast issues and alerts about hail had been available, the operators could have taken measures to avoid damage to the PV plant, such as tow tracking PV modules/arrays into a more secure position.

Solargis Hail Forecast data interpretation

Hail events are typically highly localized in space and time. Solargis hail forecast is given in percent [%] across three hail-size thresholds:

• 25 mm (~1 inch): Awareness level for PV operators

• 38 mm (~1.5 inches): Readiness-to-action level

• 50 mm (~2.0 inches): Elevated readiness-to-action level

The percentage indicates the likelihood that a hail event will occur in the surrounding area of 3-10 km, in rare cases reaching distances of up to 50 km. The reported hail size represents the expected severity of the event.

Forecast values should be interpreted as the potential for hail reaching the specified size in the vicinity of the storm core. The reported hail size represents the expected severity of the event. In practice, the largest hailstones typically occur over very limited areas, while smaller hail can affect a much broader region. However, the exact location of the largest hail sizes during the storm event cannot be forecast with high spatial accuracy.

Solargis Hail Forecast data accuracy

Hail events are difficult to validate comprehensively, as many significant occurrences go unobserved and unreported. NOAA’s Storm Prediction Center (SPC) collects and archives hail observations across the United States, forming the primary reference dataset for observed hail occurrence and size.

Hail observations are naturally incomplete, with reporting density varying across space and time and biased toward populated areas, cities, and major transportation routes. Some regions provide much more detailed reporting than others. Despite these limitations, SPC records offer reliable confirmation that a hail event occurred and document the reported hailstone size.

Figure 4: SPC records on observed hail events. Density of reported observations correlate with patterns of urban areas and transportation corridors

We compared the Solargis hail forecast with official SPC hail reports from 2021–2023. During this period, the SPC database recorded 20,905 hail events with hail sizes of at least 1 inch.

The HRRR weather model detected approximately 94% of these events in at least one forecast run, within a ±3-hour window and a 50 km radius. In practical terms, this means that nearly all damaging hail events would have triggered an early warning in the Solargis hail forecast. Given the complexity of hail formation, this level of performance is considered very good and provides PV operators valuable time to implement mitigation measures.

Around 3.5% of events were predicted as smaller, non-damaging hail, while about 2.5% were missed entirely.

The spread between forecasted and observed hail sizes is more pronounced. For moderate hail events (1–3 inches), the densest areas in the 2D histogram lie mostly above the line of equality, indicating that the HRRR model tends to predict slightly larger hail sizes than those reported by SPC, providing a conservative estimate. For very large hail (above 3 inches), the pattern reverses: the model generally underestimates hail size compared to observations. These extreme events are relatively rare, which is reflected by lower counts in the histogram.

Figure 5:  Comparison of SPC events with hail size larger than 1''  in the period 2021-2023 to HRRR model outputs, postprocessed by Solargis method, presented in the density plot.

Insights Gained

On 23 June 2023, a violent storm approached Scottsbluff in western Nebraska, where a 5.2 MWp PV power plant with single-axis trackers is located.

Around 9 PM local time (02:00 UTC on 24 June), hailstones up to 4 inches in diameter struck the site, causing extensive damage to the majority of PV modules.

Figure 6: PV power plant in Scottsbluff, western Nebraska after damaging hail event.

Source: renewableenergyworld.com

The maps and chart below illustrate how the Solargis hail forecast would have represented the Scottsbluff event several hours in advance. The forecast signal began rising about three hours before the peak impact, reaching a 95% probability for 25 mm (~1 inch) hail and a 35% probability for 50 mm (~2 inch) hail at the time of maximum impact. Maximum model probabilities reached 100% for 25 mm and 58% for 50 mm hail in an area approximately 20 km northeast of the plant.

This case demonstrates that access to short-term hail risk information could provide actionable lead time for PV operators to adjust tracker positions and reduce potential damage. According to news reports, the affected PV modules were later replaced, and the plant returned to service in early 2024.

Figure 7: Reconstructed Solargis hail forecast for 23–24 June 2023. The Scottsbluff PV power plant location is marked on the maps. Forecast potential for hail exceeding 25 mm (left) and 50 mm (right) at 9 PM local time (02:00 UTC). The chart shows the time series of the hail forecast signal at the plant location.

PV Plants hail damage prevention

In regions with frequent hail activity, tracker-based PV systems are preferred because they can actively reduce impact risk. Effective mitigation relies on reliable hail forecasting to provide timely warnings before storm arrival. During hail alerts, trackers can be parked to minimize impact energy, with modules aligned as parallel as possible to the expected hail trajectory to reduce direct impacts.

For fixed-tilt PV systems, where active mitigation is not possible, the use of anti-hail PV modules with reinforced or strengthened glass is strongly recommended. Several manufacturers already offer certified anti-hail modules on the market. Regions with elevated hail risk can be identified using the hail risk map in Solargis Prospect (coming soon).