Hurricane Helene
Hurricane Helene was a significant and devastating storm that impacted the southeastern United States in September 2024. It rapidly intensified from a Category 1 to a Category 4 hurricane within just 36 hours before making landfall in Florida’s Big Bend region. With maximum sustained winds reaching 140 mph, Helene became the strongest hurricane on record to strike this area since 1851.
Hurricane Helene made landfall on September 26, 2024, as a Category 4 hurricane, bringing catastrophic storm surge and flooding to Florida’s Big Bend region. The storm generated surge heights of up to 20 feet, causing extensive damage to coastal communities such as Keaton Beach, where many homes were either destroyed or heavily damaged. The storm unleashed unprecedented rainfall, with areas in western North Carolina receiving over 29 inches within 48 hours, leading to severe flooding and mudslides.
Skytek mapped In Figure 1, the predicted trajectory of Hurricane Helene which is illustrated alongside the anticipated storm surge bands impacting the Florida coast. The figure highlights the storm’s projected path and indicates the areas at risk of significant flooding due to storm surges, which are expected to reach up to 20 feet in certain locations.
Skytek utilised aerial imagery captured on August 12th, 2024, and September 28th, 2024, shortly after Hurricane Helene made landfall near Perry, Florida. Skytek applied its proprietary machine learning algorithm to detect, highlight, and quantify the damage inflicted by the hurricane. The platform ingested aerial images from relevant areas along the hurricane’s trajectory, analysing the impact of the storm surge and flooding, along with wind damage. Figure 2 showcases several selected cities as samples of the available analytics, illustrating the extent of the damage and providing critical insights into the hurricane’s effects on affected communities. This methodology combines advanced imaging techniques with machine learning to deliver precise assessments of hurricane-related destruction.
Apalachicola is a city with a population of approximately 2,231 people and about 1,097 households, located on the Apalachicola River in Franklin County, Florida, USA.
Carrabelle is a city in Franklin County along Florida’s Panhandle, with a population of around 2,606 people and approximately 503 households.
Crawfordville, the county seat of Wakulla County, has a population of about 3,000 residents and around 1,200 households.
Steinhatchee town, of around 1,050 people and about 350 household, located on the Steinhatchee River and part the Taylor district, Florida, USA
Cedar Key, a small island community in Levy County, has a population of approximately 700 people and about 350 households.
Figure 3 – High resolution imagery acquired for August 12th, 2024, for Keaton Beach, Florida, USA.
Figure 4 – High resolution imagery acquired for September 28th, 2024, for Keaton Beach, Florida, USA, with damage results overlayed.
Skytek analysed 1,981 properties and determined that 33% had sustained damage related to Hurricane Helene. Figure 5 consolidates the results of this analysis, categorizing the various types of damage observed across the affected properties. Figure 5 consolidates the results of this analysis, categorizing the various types of damage observed across the affected properties.
Damage Category | Damage Range | Number of Properties | Property Damage % |
Red | 76% - 100% | 413 | 20.85% |
Orange | 26% - 75% | 231 | 11.66% |
Green | 0% - 25% | 1337 | 67.49% |
Skytek can provide a comprehensive table detailing each identified property in the analysed region, complete with geographic coordinates and the results of the machine learning damage detection analysis. This table will indicate the specific category of damage, and the percentage of damage identified for each property. A sample of this detailed table is presented in Figure 6, showcasing the structured data that highlights the extent of damage across the affected area.
ID | Address | Damage % | Lat. | Long. | Damage Category |
22 | **************** | 25.26% | ******** | ******** | Orange |
23 | **************** | 38.90% | ******** | ******** | Orange |
24 | **************** | 100.00% | ******** | ******** | Red |
25 | **************** | 0.00% | ******** | ******** | Green / No Structural Damage |
26 | **************** | 38.97% | ******** | ******** | Orange |
27 | **************** | 42.37% | ******** | ******** | Orange |
28 | **************** | 68.97% | ******** | ******** | Orange |
29 | **************** | 100.00% | ******** | ******** | Red |
30 | **************** | 54.66% | ******** | ******** | Orange |
31 | **************** | 80.24% | ******** | ******** | Red |
32 | **************** | 79.86% | ******** | ******** | Red |
33 | **************** | 64.75% | ******** | ******** | Orange |
34 | **************** | 74.38% | ******** | ******** | Orange |
35 | **************** | 63.66% | ******** | ******** | Orange |
36 | **************** | 98.23% | ******** | ******** | Red |
37 | **************** | 97.98% | ******** | ******** | Red |
38 | King Fisher Rd, Perry, FL, USA | 100.00% | ******** | ******** | Red |
39 | **************** | 83.99% | ******** | ******** | Red |
40 | **************** | 8.84% | ******** | ******** | Green / No Structural Damage |
41 | **************** | 0.00% | ******** | ******** | Green / No Structural Damage |
Figure 6 – Sample tabular listing of properties identified in Keaton Beach, with results of damage detection analysis, after Hurricane Helene.
King Fisher Rd, Perry, FL, USA
Each property identified by the machine learning algorithm can be visualized and examined in detail. Skytek enhances the calculated damage scores with high-resolution imagery captured before and shortly after the event, accompanied by commentary that explains the findings contributing to the damage assessment.
For example, extracting from the listing at Figure 6, the property located at King Fisher Rd, Perry, was assessed by the automated model and found to have sustained 97,98% damage, indicating a total loss. Figure 7 illustrates the area surrounding this property, while Figure 8 highlights the property in red, showcasing its condition after Hurricane Helene. The imagery clearly depicts the devastation, with the structure completely demolished and reduced to rubble.
Figure 7 – Pre-event imagery of Cedar Island, on August 12th, 2024.
Figure 8 – Post-event imagery of Cedar Island on September 28th, 2024.
Skytek’s advanced automated algorithms are designed to efficiently process vast amounts of surface data, enabling rapid analysis and assessment of properties affected by natural disasters. By leveraging cutting-edge technology, these algorithms can swiftly evaluate various factors, including geographic location, building materials, and structural integrity.
The output generated by Skytek’s system includes a comprehensive listing of identified properties, each meticulously scored based on analysing high resolution imagery before and just after passing of the hurricane. This scoring system allows for a nuanced understanding of the overall impact on each property, facilitating informed decision-making for stakeholders involved.
Moreover, the integration of machine learning techniques enhances the accuracy of damage assessments over time, as the algorithms continuously learn from new data inputs and refine their evaluation processes. This capability not only accelerates the identification of at-risk properties but also aids insurance companies and emergency responders in prioritizing their resources effectively.
