
In mid-March 2024, a series of very severe spring hailstorms careened across Fort Bend County, Texas, directly southwest of Houston. Because of its proximity to the country’s largest municipal user of renewable energy, Fort Bend County is home to more than 1.2GW of large utility-scale solar project capacity. Houston residents – and solar industry stakeholders everywhere – took notice when catastrophic hail losses in the Fighting Jays solar project made headlines.
Though the ‘if-it-bleeds-it-leads’ tendency in journalism is nothing new, the most interesting story coming out of Fort Bend County is not that one-in-500-year hail events in Texas can wreak havoc on the built environment. Rather, it is the news that hail stow protocols successfully prevented widespread physical damage at several utility-scale solar farms near Fighting Jays that were also exposed to very severe hail.
Unlock unlimited access for 12 whole months of distinctive global analysis
Photovoltaics International is now included.
- Regular insight and analysis of the industry’s biggest developments
- In-depth interviews with the industry’s leading figures
- Unlimited digital access to the PV Tech Power journal catalogue
- Unlimited digital access to the Photovoltaics International journal catalogue
- Access to more than 1,000 technical papers
- Discounts on Solar Media’s portfolio of events, in-person and virtual
Or continue reading this article for free
Our forensic investigations show that operational hail stow protocols are effective for weathering 1-in-500-year severe convective storm events – even in Texas. Unfortunately, these success stories are not getting the attention they deserve.
Fort Bend County case study
As the president of a technical advisory firm that specialises in hail risk assessment, mitigation and consulting services, I wanted to learn as much as possible about the estimated multi-million-dollar Fighting Jays solar project hail loss event. What we learned from our meteorological and forensic analyses – with the help of others in the technical due diligence and project stakeholder communities – exceeded our expectations. Here are brief summaries of some of our key findings.
Three severe hail-producing storm events occurred within a 12-hour period. To better understand event meteorology, our team reviewed NOAA’s next-generation weather radar (NEXRAD) data. Specifically, we analysed the maximum expected size of hail (MESH) in the Fort Bend County area around the time of the hail loss event. Based on this analysis, we know that two major hailstorms occurred after noon on Friday 15 March 2024. Additionally, a rare overnight hailstorm struck the area in the early morning hours of Saturday 16 March 2024.
All three of these hailstorms met >500-year event severity criteria. To understand event severity, we compared event-specific maximum hail size estimates to location-specific hail return intervals. The return interval metric characterises the estimated amount of time between discrete events, such as the number of years between baseball-sized hail at a specific location. Comparing NEXRAD hail size estimates to naturally occurring hail return data in ArcGIS Online, we observed that the MESH values for all three of these severe storm events exceeded 500-year fixed hail return intervals for Fort Bend County. Our meteorological models predict that the 500-year fixed hail return interval for the Fort Bend County area is ≥65-mm (≈2.5-in) hail. Based on this hailstone diameter threshold, all three of these severe convective storms were one-in-500-year events.
Four utility-scale solar projects were exposed to >500-year hail. Mapping the path and intensity of all three severe hailstorms, we observed that the first storm’s hail swath had no solar farm overlap, whereas the hail swaths associated with the two subsequent storms crossed over four large utility solar farms. Looking at the two storms with solar farm overlap, our post-event meteorological analysis indicates that each of these four solar farms was exposed to one-in-500-year hail. In other words, we estimate that tennis-ball-sized hail or larger (≥65 mm) fell across some portion of four Fort Bend County-area solar farms.
Three solar farms withstood one-in-500-year hail exposure. These success stories are already providing valuable information for a variety of solar project stakeholders, including owners, operators, investors, insurers and tracker manufacturers. Though the public record provides few technical details regarding the extent of the damages at the Fighting Jays solar project, I can share some general insights regarding the local weather patterns. Specifically, the hail swath associated with the most severe hailstorm passed most directly over Fighting Jays. MESH data indicates that the damaged solar farm was likely exposed to >100-mm (≈4-in) hail, which the National Weather Service broadly categorises as “giant hail” and “an extreme threat to life and property”.

Lessons from null loss events
At the 2024 PV Module Reliability Workshop, an annual event hosted by the National Renewable Energy Laboratory (NREL), our hail team leader, Jon Previtali, presented a poster introducing a public hail forensics database for PV power plants to be located on the DuraMat Data Hub. As part of this effort, VDE Americas is working in collaboration with NREL, SEIA, FM Global, CAC Specialty and others on a null hail event study, which is a study of zero-loss hail events at utility solar farms.
Why so much ado about nothing? Because null loss events are in many ways the missing link in our evolving understanding of solar project hail resilience.
Though we have strong evidence, shown here in Figure 2, that the number of utility solar farms exposed to very large hail is considerably larger than the number of hail-related insurance claims, a database of confirmed null loss events is challenging to compile. After all, no one ever calls a forensic investigator when things go to plan. The Fort Bend County case study is consequential because it definitively validates the efficacy of solar project hail stow protocols.
With the support of Array Technologies, VDE Americas was able to study pre-event hail stow protocols and post-event damage reports for three utility solar farms located within 15 km (≈9.3 mi) of the hail-damaged Fort Bend County solar farm. All three of the projects studied – specifically, Cutlass I, Cutlass II and Old 300 – were exposed to >500-year hail events. Two of the projects sustained no direct hail-related damages. At the third site, a few dozen modules sustained damage due to hail and wind-blown objects. Importantly, hail-related damages at this 270MWp PV project site were limited to an area where a tracker motor issue prevented complete hail stow.
These results provide conclusive evidence that operationalised stow protocols are effective at preventing solar module glass breakage during a severe hail event. On the one hand, we found that PV modules in a hail-stow position sustained no direct hail-related damage from a one-in-500-year hailstorm. On the other hand, we found a relatively modest amount of hail damage at a portion of the plant that we know did not achieve complete hail stow. This clearly indicates that overall site damages could have been much worse if not for successful solar project hail stow protocols.
These real-world results validate outcomes that our team’s work has long predicted. Our deterministic meteorological and financial hail loss models consistently indicate that common sense hail defences can prevent or mitigate hail-related damage. The Fort Bend case study is proof positive that even though severe hail risk is inevitable in some locations, large project losses are not.
John Sedgwick is the president and co-founder of VDE Americas, a technical advisory services firm that de-risks investments in utility-scale solar projects.