Journal of Undergraduate Research
Volume 7, Issue 6 - July/August 2006

Hurricane Wind Damage Assessment:  Post 2004 Hurricane Season Damage Survey Efforts

Kristin Brandt

ABSTRACT

The main goal of this research is to quantify the behavior of hurricane winds over land by collecting wind velocities and the envelope wind forces on homes during land falling hurricanes. A companion aspect of the research is to investigate how well old and new residential construction performs in these high winds. This is done by sending out ground teams to document the wind damage to homes affected by hurricane force winds. The efforts of the Hurricane Research team provide the baseline wind load and structural vulnerability data that is needed to improve methods in design for structural engineers and support severe weather precautionary efforts.

INTRODUCTION

The 2005 hurricane season set significant records in Hurricane history.  These include the first season with 27 named storms, 15 hurricanes, four category-five storms, and four major hurricanes hitting the United States.  1995 was the beginning of heightened Atlantic hurricane activity due to the Tropical Multi-Decadal cycle, which is a naturally occurring and lasts for several decades. The active part of the cycle produces very low wind shear and warmer water and conversely higher wind shear and cooler waters during the inactive part of the cycle [4]. While the hurricane climate in Florida cannot be controlled, we do have the ability to reduce our vulnerability to these events through improved methods of building construction. As a state with 85% of its residents within 12 miles of the coastline, it is critical to research ways to prevent structural damage due to high winds [3].

Because we still have the possibility of another 10-20 years of active hurricane seasons in the current cycle, The Hurricane Research Team at the University of Florida is currently seeking ways to prevent wind damage to properties on the coast of the South Eastern US.  The team works year-round preparing to deploy ground level wind velocity measurements, including envelope wind forces, and surveying the effectiveness of currently used wind damage prevention measures and the structural security of coastal homes. 

This report will discuss two aspects of the research being conducted by the UF Hurricane Research Team. The Hurricane Research Team at the University of Florida is a component of a larger effort that works toward improved knowledge of hurricane wind behavior at the ground level. Along with UF, the program includes Florida International University, Clemson University, the Institute for Business and Home Safety, and the Florida Department of Community Affairs to make up the Florida Coastal Monitoring Program.. The first project is the documentation of damage experienced by residential construction during the 2004 hurricane season. The goal is to provide a statistical quantification of the relative performance of homes built according to the two most recent building codes enforced in Florida. The second project is the direct measurement of ground level hurricane wind velocities and the resultant uplift pressures on the roofs of coastal residential structures.

DAMAGE SURVEYS

The 2004 hurricane season produced four land falling hurricanes that affected the state of Florida (Charley, Frances, Ivan and Jeanne). Over the spring of 2005, residential structures in regions affected by the 2004 season were evaluated to document the damage they incurred and the preventative measures taken by the residents. The homes targeted were mainly homes constructed after revisions were made to the building code post hurricane Andrew.  The purpose of the study is to quantify the effectiveness of changes made to the building code after 2001. This is done through direct performance comparisons of homes built between 1994 and 2001 under the Standard Building Code (SBC) to the homes built according to the Florida Building Code (FBC), effective in early 2002.   A detailed study of several hundred homes provides a statistical comparison of the relative effectiveness of the SBC with the FBC.

In the spring of 2005, teams of researchers visited residents affected by Hurricane Charley, Frances, Jeanne and Ivan in three counties (Charlotte, St. Lucie, and Escambia). Each subject home was randomly selected based on age of construction and predicted peak winds experienced by that home. For example, if there were a total of 500 homes in Charlotte County (Hurricane Charley) that were built after 2001 and located in a predicted peak 3 second wind zone of 140 – 150 m.p.h. (based on contour maps of modeled Charley winds), a sample of these homes (23 in this study) were randomly selected for evaluation. Another 24 homes that were built between 1994 and 2001 in this same wind zone were also randomly selected for evaluation [1]. The same is done for lower wind speed zones (in bands of 10 m.p.h.) and the other two counties. Thus the statistical results of the performance evaluation offer a fair comparison of relative performance of SBC and FBC homes that experienced the same peak winds [1].

DAMAGE EVALUATION DETAILS

In teams of two, researchers evaluated each home based on the physical properties before and after the storm, and the behavioral attributes of each.

The following data was collected and input into a PDA by one of the team members [1]:

The other team member’s role was to interview the homeowner. He/she asked the following questions:

This information was collected along with extensive photographic documentation, hand drawn elevation sketches, and any additional details that could possibly contribute to the study. 

Formal letters along with follow-up calls were made months in advance because homeowners were reluctant to allow survey teams into their home.  Initially, an average of 1 in 20 homeowners agreed to participate.

The hard copies of survey evaluation were compared to the PDA input as well as digital photographs of each property.  These were grouped by storm name as well as the weekend the surveying took place.  The data was stored on a multi-functional Microsoft Access database that was often used for applicable statistical analysis.

DAMAGE SURVEY RESULTS

The final report on the results of this study is currently in production and will be available to the public through the Florida Department of Community Affairs. This is a very detailed statistical assessment of all data collected. The results presented in this section are intended to highlight some of the most significant findings.

The data that created the most distinct relationships in this study dealt with window damage vs. window protection and roof cover. Tile roofing resulted in a significant proportion of roof cover loss and window damage. Most homes with tile roofing were not able to escape damage.  These results also provided a correlation to the age of the home and tile [1].

Another important finding in window damage was the effectiveness of window cover.  Window protection prevented damage up to 65% among all homes [1].  The other factors affecting window damage were wind speeds and primary roof cover in the neighborhood, as illustrated in Figure 1 below.

Figure 1. Homes with Window Damage by Neighboorhood Roof Cover and Window Protection:  Wind Zones 10 and 11 from Hurricane Charley (98) [1]
Figure 1. Homes with Window Damage by Neighboorhood Roof Cover and Window Protection:  Wind Zones 10 and 11 from Hurricane Charley (98) [1]

The data collected on window damage by the FCMP is being used for a study by the Florida Building Commission to determine whether the Panhandle Windborne Debris Region should be expanded. This would implement higher standards in window protection in the region and would have a large economic and social impact. The Hurricane Research Team worked in conjunction with Applied Research & Associates, inc. in providing the information needed for wind tunnel testing that creates windborne models simulating severe wind impact on homes in the coastal panhandle regions [5].  This information provides quantitative data that contributes to the efforts of the Florida Building Commission to make an appropriate decision regarding the size of the region.

Shingle loss was another element that gave appropriate data for comparison. The homes with the most shingle damage were the oldest and in the highest wind zones, likely due to the construction methods under the old code, natural wear and tear, and the effects of prior hurricanes.

Code compliance among counties was evaluated and is useful for all comparisons in the effectiveness of the old and updated Florida Building Code.

HURRICANE WIND DATA COLLECTION

The Hurricane Research Team works year round to prepare for the event of Hurricane Landfall. Wind velocity and envelope wind force measurements on homes during land falling hurricanes are the main objective of this program. Based out of the Florida Department of Transportation in Gainesville, equipment is constantly maintained and updated due to the demanding nature of this field study.

When a storm is predicted to make landfall in the Southern Eastern US, students and faculty deploy to the predicted landfall site (with the assistance of NOAA) and assemble wind towers in the path of the storm. The towers are 30ft high and are composed of a steel lattice weighing 5500lbs. They were developed to endure 200mph winds.  The velocity measurements are taken by two RM Young anemometry systems composed of 3-axis anemometer arrays [2].  Along with the anemometry system, barometric pressure, temperature, relative humidity and rainfall are also determined from the tower trailer.  The towers take approximately 30 minutes to set up and may be assembled by as few as three experienced team members. 

As soon as the tower is erected, the teams evacuate to pre-arranged shelters to wait out the storm and record the results of their testing by the GPS instrumentation uploaded onto the FCMP website. 

The other aspect of Hurricane deployment is home pressure sensor installation to determine uplift pressure on residential structures in the storm.  The pre-wired homes were retrofitted to assist in wind-resistance as an exchange for their on call accessibility.   The pre-wiring includes stainless steal brackets to hold up the pressure sensors and PVC tubing that connects the sensors to the computer systems.  Approximately 25 sensors are temporarily installed per home and are strategically placed in order to evaluate the highest gusts [3]. 

Together, the wind velocity and pressure measurements give ground level hurricane force wind details that are essential in the evolving engineering strategies to improve structural wind resistance.

SUMMARY OF STORM DEPLOYMENTS IN 2005 AND RESEARCH OUTCOMES

In this study, researchers were deployed for Hurricane Dennis and Hurricane Katrina. Figures 2 and 3 pinpoint the tower locations for Hurricane Dennis and Katrina respectively [3].

Figure 2. FCMP Tower Locations for Hurricane Dennis.

Figure 2. FCMP Tower Locations for Hurricane Dennis.


Figure 3. FCMP Tower Locations for Hurricane Katrina.

Figure 3. FCMP Tower Locations for Hurricane Katrina.

In The Florida Building Commissions efforts to make professional recommendations on the weaknesses observed in new construction, a long list of structural attributes was produced to provide quantitative evidence in economically and socially significant building code decisions. The results of the 2004 FCMP damage survey data served as a key contribution to this list. 

The tower project provides wind speed information that essentially gives the “ground truth.”  Previous efforts by NOAA and other such organizations created a basis for categorizing storms and their characteristics by taking measurements high above or on the outskirts of the storms.  They would then extrapolate approximate wind speeds inside the eye by mathematical models.  The FCMP was the first source of wind speed data inside the actual storm on the ground level. 

The findings of the tower data show current classifications of storm wind velocity to be over-estimated by one category.  Building code and construction methods have been following specifications based on incorrect wind speeds.  This information is significant because engineers have been led to believe structural failure occurs at a certain wind speed and pressure when this is in fact happening at lower speeds and pressures than thought. 

Designers build according to the air pressure of the storm, making the pressure sensor data especially important.  The combination of both tower and pressure sensor data creates a relationship that decides the accuracy of the guidelines to which we currently build. 

The contributions of the FCMP are versatile in their applications.  Much of the wind velocity and pressure data provide the means to improve structural engineering standards.  In coastal regions vulnerable to high force winds, this can prevent serious injury, death, and economic disaster among domestic homes and businesses.  Gathering damage survey data can contribute to governmental efforts to make the safest and most economically sound decisions. 

REFERENCES

  1. K.Gurley, R.H. Davis, S.P Ferrera, J. Burton, F. Masters, T. Reinhold, M. Abdullah, “Post 2004 field survey- an evaluation of the relative performance of the standard building code and the florida building code.” Presented at 2006 structures congress, St.Louis, Mi.
  2. F. Masters, “Measurement, modeling and simulation of ground-level tropical cyclone winds.” Dissertation Presented to the Graduate School of the University of Florida, 2004.
  3. J.J. Farnham, L.D. Aponte, F. Masters, J. Jayaro, “Florida coastal monitoring program.” Available at HTTP://users.ce.ufl.edu/~fcmp/index.htm.
  4. “NOAA attributes recent increase in hurricane activity to naturally occurring multi-decadal climate variability,” NOAA magazine online, story 184, available at http://www.magazine.noaa.gov/stories/mag184.htm
  5. The Florida Senate Interim Work Plan, Section 39, Chapter 2005-147, LOF

--top--

Back to the Journal of Undergraduate Research