Effect of roof age on asphalt shingle performance during high wind events

Kovar, Rachel Nicole

Permalink

https://hdl.handle.net/2142/115453 Copy

Description

Title

Effect of roof age on asphalt shingle performance during high wind events

Author(s)

Kovar, Rachel Nicole

Issue Date

2022-04-14

Director of Research (if dissertation) or Advisor (if thesis)

Lombardo, Franklin T

Department of Study

Civil & Environmental Eng

Discipline

Civil Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• hurricane
• damage
• damage assessment
• roof
• aging
• remote-sensing

Abstract

There are many components to consider when assessing the roof cover damage during a high wind event. A key component to consider is the age of the roof at the time of the event, but the determination of roof age has proven difficult. With the occurrences of Hurricane Rita (2005) and Hurricane Laura (2020) in the same coastal region of Louisiana, there was an opportunity to estimate the roof age of homes during the second event with high confidence and assess the associated level of roof cover damage on asphalt shingle roofs. The homes with roof covers damaged in Hurricane Rita were evaluated and those that sustained substantial roof damage (greater than 25%) were assumed to have endured a full roof replacement, thus the assumed roof age at the time of Hurricane Laura was 15 years. To evaluate a range of roof ages, homes with year-built data available between 2005 and 2020 were collected with the utilization of Zillow. All damage assessments and estimations were determined with the utilization of NOAA aerial imagery. The analyses performed during this study utilized a total of 636 homes collected within the Lake Charles area, as well as the coastal areas of Holly Beach and Cameron. The data was primarily evaluated with respect to roof age and associated roof cover damage levels. Secondary analyses were performed to compare the relationships between roof age, level of damage, and other identifying home and roof characteristics. These characteristics include roof type, the number of stories, and the relative orientation of the home with respect to the estimated wind direction. Each home had a standardized peak 3-s gust wind speed and direction estimation obtained from H*Wind. The range of the peak wind speeds was limited to 120-150 MPH within the data collected, with a distinct separation of the highest wind speeds (140-150 MPH) occurring at the two coastal areas, and further inland, in the Lake Charles area, were the wind speeds at the lower end of the range (120-135 MPH). Thus, in subsequent analyses the homes further inland were evaluated separately to keep a more consistent wind speed and exposure. The single-factor analysis used 2-year roof age bins and estimated roof cover damage states of 0%, less than 25%, 25-50%, 50-75%, and 75-100%. This analysis showed a strong relationship in roof cover damage amounts with an increase in roof age. In review of the individual damage states, significant increases in the percentages of homes damaged were noticed at specific age bins. The 25-50% damage state experiences a significant increase in the percentages of homes damaged in the 8-9-year roof age bin, the 50-75% damage state saw an increase in the 10-11-year roof age bin, and the 75-100% damage state experienced an increase in the 12-13-year roof age bin. Looking at the damage cumulatively with respect to age bin revealed a number of interesting findings. For example, at the 10-11-year roof age bin, the percentage of undamaged roof covers dropped to below 10% of homes, which was a significant decrease from the 30-60% of homes that were classified as undamaged in the roof ages under 10 years. It was also observed that the percentage of homes with greater than 25% damage experienced an increase in the 10-11-year age bin. For the highest two thresholds, 50% and 75%, the percentage of homes experienced an increase in the 12-13-year and 14-15-year age bins, respectively. The results of the cumulative analysis illustrated that if the roof age is not in consideration, the percentages of roofs damaged overall would provide an overestimation for the younger roofs, nine years and under, and an under estimation, in most cases, for the older roofs, ten years and greater. To account for the various characteristics of the home and roof itself, the data was parsed into further subcategories for analysis, while maintaining the same separation by roof age bins. The roof type, number of home stories, and the relative orientation of the home were analyzed with respect to the cumulative damage thresholds. While each characteristic had varying percentages of homes damaged at each age bin, it is observed that regardless of the secondary characteristic analyzed, there is an increase in the percentage of damage to a roof when there is an increase in roof age. The percentage of damage was more severe for certain characteristics than others. To further evaluate this observation, the data was evaluated at specific age bins as well as larger age ranges. The results of these subsequent analyses showed that the hip type roofs were more sensitive to an increase in roof age compared to that of the gable type roofs; the hip roofs showed substantial increases in the percentages of damage as the roof age increased. Another finding was with respect to the number of stories; the multi-story roofs in the 2-3-year age bin sustained higher percentages of damage than the single-story roofs, but the opposite relation was observed in both the 12-13-year and 14-15-year age bins. Damage patterns were also assessed on specific roof types with respect to windward and leeward facing areas of each roof. To keep a more consistent exposure and wind speed, the homes only included those further inland with 120-135 MPH peak wind speeds. In review of each damaged roof cover, the physical area of the damage was separated into two zones, windward and non-windward facing. The zones were determined based upon the approximated wind direction (obtained from H*Wind) at each home and the orientation of the home itself. The average percentage of windward facing and leeward facing damage was then calculated at each respective roof age. From this analysis, it was observed that the hip type roofs sustained higher percentages of windward facing damage with an increase in the roof age. On average, 67% of the roof area damaged was located on a windward facing zone, thus only 33% was located on the leeward facing areas. Within this study, roof age has been determined to be a significant factor contributing to roof cover damage observed. The level of impact that roof age has on roof cover damage is likely dependent on the physical characteristics of the home and roof itself. The applicable codes and standards in place at the time of construction contribute to the damage levels observed, as was observed with the higher damage states seen with lower peak 3-s gust wind speeds in Hurricane Rita compared to Hurricane Laura. To further this study with consideration of the other noted home characteristics, additional data in multiple roof ages would be required. The comparisons of most interest that were limited due to insufficient data were the roof type comparisons at each roof age, as well as the number of home stories at each roof age. Additionally, this study reflects only results for homes analyzed within the limited peak wind speed range of 120-150 MPH; for consideration of additional wind speeds, further analysis would be required.

Graduation Semester

2022-05

Type of Resource

Thesis

Copyright and License Information

Copyright 2022 Rachel Nicole Kovar

Owning Collections