Florida Public Hurricane Loss Model Submitted in Compliance with the 2007 Standards of the Florida Commission on Hurricane Loss Projection Methodology May 15, 2008

Florida Public Hurricane Loss Model Submitted in compliance with the 2007 Standards of the Florida Commission on Hurricane Loss Projection Methodology May 15, 2008

FPHLM V3.0 2008 1 Model Identification

Name of Model and Version: Florida Public Hurricane Loss Model 3.0

Name of Modeling Organization: Florida International University

Street Address: International Hurricane Research Center, MARC 360

City, State, ZIP Code: Miami, Florida 33199

Mailing Address, if different from above: Same as above

Contact Person: Shahid S. Hamid

Phone Number: 305-348-2727 Fax: 305-348-1761

E-mail Address: [email protected]

Date: May 15, 2008

FPHLM V3.0 2008 2

May 15, 2008

Chair, Florida Commission on Hurricane Loss Projection Methodology c/o Donna Sirmons Florida State Board of Administration 1801 Hermitage Boulevard, Suite 100 Tallahassee, FL 32308

Dear Commission Chairman:

I am pleased to inform you that the Florida Public Hurricane Loss Model (FPHLM) is ready for review by the Professional Team and certification by the Commission. The FPHLM model has been reviewed by professionals having credentials and/or experience in the areas of meteorology, engineering, actuarial science and insurance, statistics and computer science; for compliance with the Standards, as documented by the expert certification forms G1-G7.

Enclosed are 20 bound copies of our submission, which includes the summary statement of compliance with the standards, the forms, and the submission checklist. Also enclosed are 20 CDs containing the submission and forms.

Please contact me if you have any questions regarding this submission.

Sincerely,

Shahid Hamid, Ph.D, CFA Professor of Finance, and Director, Laboratory for Insurance, Economic and Financial Research International Hurricane Research Center RB 202B, Department of Finance, College of Business Florida International University Miami, FL 33199 tel: 305 348 2727 fax: 305 348 4245

Cc: Kevin M. McCarty, Insurance Commissioner

FPHLM V3.0 2008 3 Statement of Compliance and Trade Secret Disclosure Items

The Florida Public Hurricane Loss Model v. 3.0 is intended to comply with each Standard of the 2007 Report of Activities released by the Florida Commission on Hurricane Loss Projection Methodology. The required disclosures, forms, and analysis are contained herein.

The source code for the loss model will be available for review by the Professional Team.

FPHLM V3.0 2008 4 Model Submission Checklist

1. Please indicate by checking below that the following has been included in your submission to the Florida Commission on Hurricane Loss Projection Methodology. Yes No Item X 1. Letter to the Commission X a. Refers to the Expert Certification Forms and states that professionals having credentials and/or experience in the areas of meteorology, engineering, actuarial science, statistics, and computer science have reviewed the model for compliance with the Standards X b. States model is ready to be reviewed by the Professional Team X c. Any caveats to the above statements noted with a complete explanation X 2. Summary statement of compliance with each individual Standard and the data and analyses required in the Disclosures and Forms X 3. General description of any trade secrets the modeler intends to present to the Professional Team X 4. Model Identification X 5. 20 Bound Copies (duplexed) X 6. 20 CDs containing: X a. Submission text in PDF format X b. PDF file highlightable and bookmarked by Standard, Form, and section X c. Data file names include abbreviated name of modeler, Standards year, and Form name (when applicable) X d. Forms V-2, A-3, A-4, A-5, A-7, and S-5 (for models submitted by modeling organizations which have not previously provided the Commission with this analysis) in PDF format X e. Forms V-2, A-1, A-3, A-4, A-5, A-6A, A-6B, and A-7 in Excel format X f. Form S-5 (for models submitted by modeling organizations which have not previously provided the Commission with this analysis) in ASCII format X 7. Table of Contents X 8. Materials consecutively numbered from beginning to end starting with the first page (including cover) using a single numbering system X 9. All tables, graphs, and other non-text items specifically listed in Table of Contents X 10. All tables, graphs, and other non-text items clearly labeled with abbreviations defined X 11. Standards, Disclosures, and Forms in italics, modeler responses in non-italics X 12. Graphs accompanied by legends and labels for all elements X 13. All units of measurement clearly identified with appropriate units used X 14. Hard copy of all Forms included except Forms A-1 and S-5

2. Explanation of “No” responses indicated above. (Attach additional pages if needed.)

Form S-5 was submitted last year, and no changes to the wind model were made.

FPHLM V3.0 2008 5 Florida Public Hurricane Loss Model May 15, 2008

Model Name Modeler Signature Date

FPHLM V3.0 2008 6 Table of Contents MODEL IDENTIFICATION...... 2 STATEMENT OF COMPLIANCE AND TRADE SECRET DISCLOSURE ITEMS ...... 4 MODEL SUBMISSION CHECKLIST ...... 5 TABLE OF CONTENTS ...... 7 LIST OF FIGURES...... 9 LIST OF TABLES...... 11 GENERAL STANDARDS ...... 12 G-1 Scope of the Computer Model and Its Implementation...... 12 G-2 Qualifications of Modeler Personnel and Consultants...... 64 G-3 Risk Location...... 71 G-4 Independence of Model Components ...... 73 G-5 Editorial Compliance ...... 74 Forms G1-7 ...... 75 METEOROLOGICAL STANDARDS ...... 90 M-1 Base Hurricane Storm Set* ...... 90 M-2 Hurricane Parameters and Characteristics*...... 91 M-3 Hurricane Probabilities*...... 98 M-4 Hurricane Windfield Structure*...... 100 M-5 Landfall and Over-Land Weakening Methodologies*...... 106 M-6 Logical Relationships of Hurricane Characteristics...... 111 Form M-1: Annual Occurrence Rates...... 112 Form M-2: Maps of Maximum Winds...... 120 Form M-3: Radius of Maximum Winds and Radii of Standard Wind Thresholds...... 125 VULNERABILITY STANDARDS ...... 128 V-1 Derivation of Vulnerability Functions ...... 128 V-2 Mitigation Measures* ...... 143 Form V-1: One Hypothetical Event...... 147 Form V-2: Mitigation Measures – Range of Changes in Damage...... 150 Form V-3: Mitigation Measures – Mean Damage Ratio ...... 153 ACTUARIAL STANDARDS...... 158 A-1 Modeled Loss Costs...... 158 A-2 Underwriting Assumptions ...... 159 A-3 Loss Cost Projections...... 163 A-4 Demand Surge...... 165 A-5 User Inputs...... 168 A-6 Logical Relationship to Risk...... 175 A-7 Deductibles and Policy Limits...... 181 A-8 Contents ...... 184 A-9 Additional Living Expense (ALE)...... 186 A-10 Output Ranges...... 188 Form A-1: Loss Costs ...... 191

FPHLM V3.0 2008 7 Form A-2: Zero Deductible Loss Costs by ZIP Code...... 193 Form A-3: Base Hurricane Storm Set Average Annual Zero Deductible Statewide Loss Costs ...... 196 Form A-4: Hurricane Andrew Percent of Losses ...... 199 Form A-5: Items A-C: Distribution of Hurricanes by Size of Loss...... 213 Form A-6A: Output Ranges (See Appendix B)...... 219 Form A-6B: Output Ranges (See Appendix B)...... 220 Form A-7: Percentage Change in Output Ranges...... 221 Form A-8: Percentage Change in Output Ranges by County...... 224 STATISTICAL STANDARDS...... 232 S-1 Modeled Results and Goodness-of-Fit...... 232 S-2 Sensitivity Analysis for Model Output ...... 244 S-3 Uncertainty Analysis for Model Output ...... 248 S-4 County Level Aggregation...... 252 S-5 Replication of Known Hurricane Losses ...... 253 S-6 Comparison of Projected Hurricane Loss Costs ...... 256 Form S-1: Probability of Florida Landfalling Hurricanes per Year ...... 257 Form S-2: An Example of a Probable Maximum Loss (PML) Based on a Limited Hypothetical Data Set...... 258 Form S-3: Five Validation and Comparisons ...... 259 Form S-4: Average Annual Zero Deductible Statewide Loss Costs – Historical versus Modeled ...... 262 Form S-5: Hypothetical Events for Sensitivity and Uncertainty Analysis (requirement for models submitted by modeling organizations which have not previously provided the Commission with this analysis) ...... 265 COMPUTER STANDARDS...... 266 C-1 Documentation...... 266 C-2 Requirements ...... 267 C-3 Model Architecture and Component Design ...... 268 C-4 Implementation* ...... 269 C-5 Verification ...... 271 C-6 Model Maintenance and Revision ...... 274 C-7 Security ...... 276 APPENDIX A - EXPERT REVIEW LETTERS ...... 278 APPENDIX B - OUTPUT RANGES BY COUNTY ...... 286

FPHLM V3.0 2008 8 List of Figures FIGURE 1. FLORIDA PUBLIC HURRICANE LOSS MODEL DOMAIN. CIRCLES REPRESENT THE THREAT ZONE. BLUE COLOR INDICATES WATER DEPTH EXCEEDING 656 FT (200 M)...... 13 FIGURE 2. EXAMPLES OF SIMULATED HURRICANE TRACKS. NUMBERS REFER TO THE STOCHASTIC TRACK NUMBER, AND COLORS REPRESENT STORM INTENSITY BASED ON CENTRAL PRESSURE. DASHED LINES REPRESENT TROPICAL STORM STRENGTH WINDS, AND CAT 1-5 WINDS ARE REPRESENTED BY BLACK, BLUE, ORANGE, RED AND TURQUOISE, RESPECTIVELY...... 16 FIGURE 3. COMPARISON BETWEEN THE MODELED AND OBSERVED WILLOUGHBY AND RAHN (2004) B DATA SET...... 17 FIGURE 4. OBSERVED AND EXPECTED DISTRIBUTION FOR RMAX. THE X-AXIS IS THE RADIUS IN STATUTE MILES, AND THE Y-AXIS IS THE FREQUENCY OF OCCURRENCE...... 18 FIGURE 5. COMPARISON OF 100,000 RMAX VALUES SAMPLED FROM THE GAMMA DISTRIBUTION FOR CAT 1-4 STORMS TO THE EXPECTED VALUES...... 19 FIGURE 6. MONTE CARLO SIMULATION PROCEDURE TO PREDICT DAMAGES...... 23 FIGURE 7. PROCEDURE TO CREATE VULNERABILITY MATRIX...... 26 FIGURE 8. WEIGHTED MASONRY STRUCTURE VULNERABILITIES IN THE CENTRAL WIND BONE DEBRIS ZONE...... 33 FIGURE 9. FLOW DIAGRAM OF THE COMPUTER MODEL ...... 37 FIGURE 10. ORGANIZATIONAL STRUCTURE...... 65 FIGURE 11. FLORIDA PUBLIC HURRICANE LOSS MODEL WORKFLOW...... 68 FIGURE 12. COMPARISON OF OBSERVED LANDFALL RMAX (SM) DISTRIBUTION TO A GAMMA DISTRIBUTION FIT OF THE DATA...... 94 FIGURE 13. COMPARISON OF 100,000 RMAX VALUES SAMPLED FROM THE GAMMA DISTRIBUTION FOR CAT 1-4 STORMS TO THE EXPECTED VALUES...... 95 FIGURE 14. AXISYMMETRIC ROTATIONAL WIND SPEED (MPH) VS. SCALED RADIUS FOR B = 1.38, DELP = 49.1 MB....101 FIGURE 15. UPSTREAM FETCH WIND EXPOSURE PHOTOGRAPH FOR CHATHAM MS (LEFT, LOOKING NORTH), AND PANAMA CITY, FL (RIGHT, LOOKING NORTHEAST). AFTER POWELL ET AL., (2004)...... 102 FIGURE 16. COMPARISON OF OBSERVED (RIGHT) AND MODELED (LEFT) LANDFALL WIND FIELDS OF HURRICANES CHARLEY (2004, TOP), AND 2005 HURRICANE KATRINA IN SOUTH FLORIDA (BOTTOM). LINE SEGMENT INDICATES STORM HEADING. HORIZONTAL COORDINATES ARE IN UNITS OF R/RMAX AND WINDS UNITS OF MILES PER HOUR...... 104 FIGURE 17. AS IN FIGURE 16 EXCEPT FOR HURRICANE WILMA OF 2005...... 105 FIGURE 18. OBSERVED (GREEN) AND MODELED (BLACK) MAXIMUM SUSTAINED SURFACE WINDS AS A FUNCTION OF TIME FOR 2004 HURRICANES FRANCESCHARLEY (LEFT) AND CHARLEY (RIGHT). LANDFALL IS REPRESENTED BY THE VERTICAL DASH-DOT RED LINE AT THE LEFT AND TIME OF EXIT AS THE RED LINE ON THE RIGHT...... 108 FIGURE 19. OBSERVED (GREEN) AND MODELED (BLACK) MAXIMUM SUSTAINED SURFACE WINDS AS A FUNCTION OF TIME FOR HURRICANES JEANNE (2004, TOP LEFT), KATRINA (2005 IN SOUTH FLORIDA, TOP RIGHT), AND 2005 WILMA (LOWER LEFT). LANDFALL IS REPRESENTED BY THE VERTICAL DASH-DOT RED LINE AT THE LEFT AND TIME OF EXIT AS THE RED LINE ON THE RIGHT...... 109 FIGURE 20. FORM M-1 COMPARISON OF MODELED AND HISTORICAL LANDFALLING HURRICANE FREQUENCY STATEWIDE...... 116 FIGURE 21. FORM M-1 COMPARISON OF MODELED AND HISTORICAL LANDFALLING HURRICANE FREQUENCY IN REGION A...... 116 FIGURE 22. FORM M-1 COMPARISON OF MODELED AND HISTORICAL LANDFALLING HURRICANE FREQUENCY IN REGION B...... 117 FIGURE 23. FORM M-1 COMPARISON OF MODELED AND HISTORICAL LANDFALLING HURRICANE FREQUENCY IN REGION C...... 117 FIGURE 24. FORM M-1 COMPARISON OF MODELED AND HISTORICAL LANDFALLING HURRICANE FREQUENCY IN REGION D...... 118 FIGURE 25. FORM M-1 COMPARISON OF MODELED AND HISTORICAL LANDFALLING HURRICANE FREQUENCY IN REGION E...... 118 FIGURE 26. FORM M-1 COMPARISON OF MODELED AND HISTORICAL LANDFALLING HURRICANE FREQUENCY IN REGION F...... 119 FIGURE 27. FOR FORM M-1 COMPARISON OF MODELED AND HISTORICAL BY-PASSING HURRICANE FREQUENCY...... 119 FIGURE 28. MAXIMUM ZIP CODE WIND SPEED FOR OPEN TERRAIN WIND EXPOSURE BASED ON SIMULATIONS OF THE HISTORICAL STORM SET ...... 121

FPHLM V3.0 2008 9 FIGURE 29. MAXIMUM ZIP CODE WIND SPEED FOR ACTUAL TERRAIN WIND EXPOSURE BASED ON SIMULATIONS OF THE HISTORICAL STORM SET...... 122 FIGURE 30. 100 YEAR RETURN PERIOD WIND SPEEDS AT FLORIDA ZIP CODES FOR OPEN TERRAIN WIND EXPOSURE....123 FIGURE 31. 100 YEAR RETURN PERIOD WIND SPEEDS AT FLORIDA ZIP CODES FOR ACTUAL TERRAIN WIND EXPOSURE...... 124 FIGURE 32. REPRESENTATIVE SCATTER PLOT OF RADIUS OF MAXIMUM WIND (Y AXIS) VERSUS MINIMUM SEA-LEVEL AIR PRESSURE AT LANDFALL (MB). RELATIVE HISTOGRAMS FOR EACH QUANTITY ARE ALSO SHOWN. EXTRACTED FROM SIMULATION OF 50,000 YEARS CONDUCTED ON 12-10-2007...... 126 FIGURE 33. ONEWAY BOX PLOT (LEFT) OF RMAX (CONTINUOUS) RESPONSE ACROSS 10 MB PMIN GROUPS. BOXES (AND WHISKERS) ARE IN RED, MEANS AND STANDARD DEVIATIONS IN GREEN. HISTOGRAMS (RIGHT) FOR EACH PMIN GROUP. REPRESENTATIVE SAMPLE FROM 50,000 YEAR SIMULATION OF DECEMBER 2007...... 127 FIGURE 34. MONTE CARLO SIMULATION PROCEDURE TO PREDICT DAMAGES...... 131 FIGURE 35. PROCEDURE TO CREATE VULNERABILITY MATRIX ...... 132 FIGURE 36. COMPANY A, HURRICANE ANDREW STRUCTURE VS. CONTENTS LOSSES ...... 134 FIGURE 37. BUILDING DAMAGE VS. 1 MINUTE SUSTAINED WIND SPEED...... 146 FIGURE 38. MITIGATION MEASURES FOR MASONRY HOMES ...... 156 FIGURE 39. MITIGATION MEASURES FOR FRAME HOMES ...... 157 FIGURE 40. MODEL VS. ACTUAL—STRUCTURE LOSS ...... 176 FIGURE 41. MODEL VS. ACTUAL—CONTENT LOSS ...... 177 FIGURE 42. MODEL VS. ACTUAL—ALE LOSS ...... 177 FIGURE 43. MODEL VS. ACTUAL—APP LOSS ...... 178 FIGURE 44. MODELED VS. ACTUAL RELATIONSHIP BETWEEN STRUCTURE AND CONTENT DAMAGE RATIOS...... 185 FIGURE 45. ZERO DEDUCTIBLE LOSS COST BY ZIP CODE FOR OWNERS ...... 193 FIGURE 46. A ZERO DEDUCTIBLE COSTS BY ZIP CODE FOR OWNERS MASONRY ...... 194 FIGURE 47. ZERO DEDUCTIBLE LOSS COSTS BY ZIP CODE FOR MOBILE HOME...... 195 FIGURE 48. HURRICANE ANDREW PERCENT OF LOSSES...... 212 FIGURE 49. COMPARISON OF RETURN TIMES V2.6 AND V3.0 – DIFFERENT EXPOSURE BASES ...... 216 FIGURE 50. COMPARISON OF RETURN TIMES V2.6 AND V3.0 – SAME EXPOSURE BASE...... 217 FIGURE 51. PERCENTAGE CHANGE IN OUTPUT RANGES BY COUNTY – OWNERS FRAME 2% DEDUCTIBLE ...... 225 FIGURE 52. PERCENTAGE CHANGE IN OUTPUT RANGES BY COUNTY – OWNERS MASONRY 2% DEDUCTIBLE...... 226 FIGURE 53. PERCENTAGE CHANGE IN OUTPUT RANGES BY COUNTY – MOBILE HOMES 2% DEDUCTIBLE...... 227 FIGURE 54. PERCENTAGE CHANGE IN OUTPUT RANGES BY COUNTY – RENTERS FRAME 2% DEDUCTIBLE ...... 228 FIGURE 55. PERCENTAGE CHANGE IN OUTPUT RANGES BY COUNTY – RENTERS MASONRY 2% DEDUCTIBLE ...... 229 FIGURE 56. PERCENTAGE CHANGE IN OUTPUT RANGES BY COUNTY – CONDO FRAME 2% DEDUCTIBLE ...... 230 FIGURE 57. PERCENTAGE CHANGE IN OUTPUT RANGES BY COUNTY – CONDO MASONRY 2% DEDUCTIBLE...... 231 FIGURE 58. COMPARISON OF SIMULATED VS HISTORICAL OCCURRENCES...... 233 FIGURE 59. COMPARISON BETWEEN THE MODELED AND OBSERVED WILLOUGHBY AND RAHN (2004) B DATA SET...233 FIGURE 60. OBSERVED AND EXPECTED DISTRIBUTION USING A GAMMA DISTRIBUTION ...... 234 FIGURE 61. COMPARISON OF MODELED (LEFT) AND OBSERVED (RIGHT) SWATHS OF MAXIMUM SUSTAINED OPEN TERRAIN SURFACE WINDS FOR HURRICANE ANDREW OF 1992 IN SOUTH FLORIDA. HURRICANE ANDREW OBSERVED SWATH IS BASED ON ADJUSTING FLIGHT-LEVEL WINDS WITH THE SFMR-BASED WIND REDUCTION METHOD...... 237 FIGURE 62. HISTOGRAM OF THE CVS FOR ALL COUNTIES COMBINED ...... 242 FIGURE 63. STANDARDIZED REGRESSION COEFFICIENTS VS. TIME AT GRID COORDINATES (30,0) FOR CATEGORY 1245 FIGURE 64. STANDARDIZED REGRESSION COEFFICIENTS VS. TIME AT GRID COORDINATES (30,0) FOR CATEGORY 3245 FIGURE 65. STANDARDIZED REGRESSION COEFFICIENTS VS. TIME AT GRID COORDINATES (30,0) FOR CATEGORY 5246 FIGURE 66. EXPECTED PERCENTAGE REDUCTIONS IN THE VAR(MSSWS) FOR A CATEGORY 1 HURRICANE VERSUS TIME AT COORDINATE (30,0)...... 249 FIGURE 67. EXPECTED PERCENTAGE REDUCTIONS IN THE VAR(MSSWS) FOR A CATEGORY 3 HURRICANE VERSUS TIME AT COORDINATE (30,0)...... 249 FIGURE 68. EXPECTED PERCENTAGE REDUCTIONS IN THE VAR(MSSWS) FOR A CATEGORY 5 HURRICANE VERSUS TIME AT COORDINATE (30,0)...... 250 FIGURE 69. SCATTER PLOT BETWEEN TOTAL ACTUAL LOSSES VS. TOTAL MODELED LOSSES2. PROVIDE A COMPLETED FORM S-3, FIVE VALIDATION COMPARISONS...... 254 FIGURE 70. SCATTER PLOTS FOR ACTUAL/EXPOSURE AND MODELED LOSS/EXPOSURE...... 261

FPHLM V3.0 2008 10 List of Tables TABLE 1. DESCRIPTION OF VALUES GIVEN IN THE DAMAGE MATRICES FOR SITE BUILT HOMES………………………. 27 TABLE 2. DESCRIPTION OF VALUES GIVEN IN THE DAMAGE MATRICES FOR MANUFACTURED HOMES………………... 27 TABLE 3. PARTIAL EXAMPLE OF VULNERABILITY MATRIX…………………………………………………………... 30 TABLE 4. AGE CLASSIFICATION OF THE MODELS PER REGION……………………………………………………….. 32 TABLE 5. PROFESSIONAL CREDENTIALS……………………………………………………………………………… 66 TABLE 6. CHI SQUARE GOODNESS OF FIT FOR FLORIDA BY REGION………………………………………………... 115 TABLE 7. SUMMARY OF PROCESSED CLAIMS DATA (NUMBER OF CLAIMS PROVIDED)……………………………….. 133 TABLE 8. COMPANY 1: CLAIM NUMBER FOR EACH YEAR BUILT CATEGORY…………………………………………135 TABLE 9. COMPANY 2: CLAIM NUMBER FOR EACH YEAR BUILT CATEGORY…………………………………………135 TABLE 10. COMPANY 1 AND COMPANY 2: CLAIM NUMBERS COMBINED…………………………………………… 137 TABLE 11. DISTRIBUTION OF COVERAGE FOR COMPANY 1………………………………………………………….. 138 TABLE 12. DISTRIBUTION OF COVERAGE FOR COMPANY 2………………………………………………………….. 138 TABLE 13. OUTPUT REPORT FOR OIR DATA PROCESSING…………………………………………………………… 169 TABLE 14. INPUT FORM FOR FLORIDA PUBLIC HURRICANE LOSS MODEL…………………………………………... 172 TABLE 15. CHECK LIST FOR THE PRE-PROCESSING………………………………………………………………….. 173 TABLE 16. MODELED VS. HISTORICAL LOSS BY CONSTRUCTION TYPE……………………………………………... 179 TABLE 17. VALIDATION TABLE BASED ON ZIP CODE WIND SWATH COMPARISON OF THE PUBLIC WIND FIELD MODEL TO H*WIND. MEAN ERRORS (BIAS) OF MODEL FOR THE SET OF VALIDATION WIND SWATHS. ERRORS (UPPER NUMBER IN EACH CELL) ARE COMPUTED AS MODELED – OBSERVED (OBS) AT ZIP CODES WHERE MODELED WINDS WERE WITHIN WIND THRESHOLDS (MODEL THRESHOLD) OR WHERE OBSERVED WINDS WERE WITHIN RESPECTIVE WIND SPEED THRESHOLD (H*WIND THRESHOLD). NUMBER OF ZIP CODES FOR THE COMPARISONS IS INDICATED AS THE LOWER NUMBER IN EACH CELL…………………………………………………………. 238 TABLE 18. VALIDATION TABLE BASED ON ZIP CODE WIND SWATH COMPARISON OF THE PUBLIC WIND FIELD MODEL TO H*WIND. ROOT MEAN SQUARE (RMS) WIND SPEED ERRORS (MPH) OF MODEL FOR THE SET OF VALIDATION WIND SWATHS. ERRORS ARE BASED ON MODELED – OBSERVED (OBS) AT ZIP CODES WHERE MODELED WINDS WERE WITHIN WIND THRESHOLDS (MODEL THRESHOLD) OR WHERE OBSERVED WINDS WERE WITHIN RESPECTIVE WIND SPEED THRESHOLD (H*WIND THRESHOLD). NUMBER OF ZIP CODES FOR THE COMPARISONS IS INDICATED AS THE LOWER NUMBER IN EACH CELL………………………………………………………………………... 239 TABLE 19. ACTUAL VS. MODEL LOSS……………………………………………………………………………….. 253

FPHLM V3.0 2008 11 GENERAL STANDARDS

G-1 Scope of the Computer Model and Its Implementation

The computer model shall project loss costs for personal lines residential property from hurricane events.

The Florida Public Hurricane Loss Model estimates loss costs from hurricane events for personal lines residential property. The losses are estimated for building, appurtenant structure, contents and ALE.

Disclosures

1. Specify the model and program version number.

The model name is Florida Public Hurricane Loss Model (FPHLM). The current version is 3.0.

2. Provide a comprehensive summary of the model. This summary shall include a technical description of the model including each major component of the model used to produce personal lines residential loss costs in the State of Florida. Describe the theoretical basis of the model and include a description of the methodology, particularly the wind components, the damage components, and the insured loss components used in the model. The description shall be complete and not reference unpublished work.

The model is a very complex set of computer programs. The programs simulate probable future hurricane activity, including where and when hurricanes form, their tracks and intensities, their wind fields, sizes, how they decay and how they are affected by the terrain along the tracks after landfall, how the winds interact with different types of residential structures, how much they can damage house roofs, windows, doors, interior, and contents etc., how much it will cost to rebuild the damaged parts, and how much of the loss will be paid by insurers. The model consists of three major components: wind hazard (meteorology), vulnerability (engineering), and insured loss cost (actuarial). It has over a dozen sub-components. The major components are developed independently before being integrated. The computer platform is designed to accommodate future hookups of additional sub-components or enhancements. Following is the description of each of the major components and their computer platforms.

Atmospheric Science Component

HURRICANE TRACK AND INTENSITY The storm track model generates storm tracks and intensities based on historical storm conditions and motions. The initial seeds for the storms are derived from the HURDAT database. For historical landfalling storms in Florida and neighboring states, the initial positions, intensities and motions are taken from the track fix 36 hours prior to first landfall. For historical storms that do not make landfall but come within 62 sm (100 km) of the coast, the initial conditions are taken from the track fix 36 hours prior to the point which the storm first comes within 62 sm of

FPHLM V3.0 2008 12 the coast (threat zone) and has a central pressure below 1005 mb. Small, uniform random error terms are added to the initial position, storm motion change, and to the storm intensity change. The initial conditions derived from HURDAT are recycled as necessary to generate thousands of years of stochastic tracks. After the storm is initiated, the subsequent motion and intensity changes are sampled from empirically derived probability distribution functions over the model domain (Figure 1).

Figure 1. Florida Public Hurricane Loss Model domain. Circles represent the threat zone. Blue color indicates water depth exceeding 656 ft (200 m).

The time evolution of the stochastic storm tracks and intensity are governed by the following equations

Δ = θ Δ ytcx )cos(/)cos( θ )sin( Δ=Δ tcy Δ=Δ twp where (x,y) are the longitude and latitude of the storm, (c,θ) are the storm speed and heading (in conventional mathematical sense), p is central pressure, w is the rate of change in p, and Δt is the time step. The time step of the model is currently one hour. The storm speed, direction (δc, δθ) are sampled at every 24 hour interval from a probability distribution function (PDF). The intensity change after the initial 24 hours of track evolution is sampled every 6 hours to capture

FPHLM V3.0 2008 13 the more detailed evolution over the continental shelf (shallow water). From the 24 hour change in speed and heading angle, we determine the speed and heading angle at each 1 hour time step by assuming the storm undergoes a constant acceleration that gives the 24 hour sampled change in velocity. For changes in pressure, we first sample from a PDF of relative intensity changes, δr, for the 6 hour period and then determine the corresponding rate of pressure change, w. The relative intensity is a function of the climatological sea surface temperatures and the upper tropospheric 100 mb temperatures. The PDFs of the changes (δc, δθ, δr) depend on spatial location, as well as the current storm motion and intensity. These PDFs are of the form

δ = δ yxaaAaPDF ),,,()( where a is either c, θ, or r and are implemented as discrete bins that are represented by multi- dimensional matrices (arrays), A(l,m,i,j). The indices (i,j) are the storm location bins. The model domain is (100W to 70W, 15N to 40N) and is divided into 0.5 degree boxes. The index m represents the bin interval that a falls into. That is, the range of all possible values of a are divided into discrete bins, the number of which depends on the variable, and the index m represents the particular bin a is in at the current time step. As with a, the range of all possible values of the change in a are also discretely binned. Given a set of indices (m,i,j), which represent the current storm location and state, the quantity A(l,m,i,j) represents the probability that the change in a, δa, will fall into the l'th bin. When A is randomly sampled, one of the bins represented by the l index, say l', is chosen. The change of a is then assigned the midpoint value of the bin associated with l'. A uniform random error term equal to the width of bin l' is added to δa, so that δa may assume any value within the bin l'.

The PDFs described above were generated by parsing the HURDAT database and computing for each track the storm motion and relative intensity changes at every 24 and 6 hour intervals, respectively, and then binning them. Once the counts are tallied, they are then normalized to obtain the distribution function. For intensity reports for which pressure is not available, a wind pressure relation developed by Landsea et al. (2004) is used. In cases where there is no pressure report for a track fix in the historical data but there are two pressure reports within a 24 hour period that includes the track fix, the pressures are derived by linear interpolation. Otherwise the pressure is derived by using the wind-pressure relation. Extra-tropical systems, lows, waves and depressions are excluded. Intensity changes over land are also excluded from the PDFs. To insure a sufficient density of counts to represent the PDFs for each grid box, counts from nearest neighbor boxes, ranging up to 2 to 5 grid units away (both north-south and east-west direction), are aggregated. Thus the effective size of the boxes may range from 1.5 to 5.5 degrees, but are generally a fixed size for a particular variable. The sizes of the bins were determined by finding a compromise between large bin sizes, which ensure a robust number of counts in each bin to define the PDF, and small bin sizes which can better represent the detail of the distribution of storm motion characteristics. Detailed examinations of the distributions as well as sensitivity tests were done. Bin sizes need not be of equal width, and a nonlinear mapping function is used to provide unequal-sized bins. For example, most storm motion tends to be persistent, with small changes in direction and speed. Thus, to capture this detail, the bins are more fine-grained at lower speed and direction changes.

For intensity change PDFs, boxes which are centered over shallow water (defined to be less than 656 ft deep, see Figure 1) are not aggregated with boxes over deeper waters. Deeper waters may

FPHLM V3.0 2008 14 have significantly higher ocean heat content which can lead to more rapid intensification (see, for example, Shay et al. (2000), DeMaria et al. (2005), Wada and Usui (2007)). The depth which defines deep and shallow waters is not too critical, as the continental shelf drops rather sharply. The 200 m (656 ft) bathymetric contour line appears to distinguish well estimates of regions with high and low tropical cyclone heat potential (see http://www.aoml.noaa.gov/phod/cyclone/data/). When gridded long-term analyses of tropical cyclone heat potential, or similar characterization of oceanic heat content, become available, we intend to use that data in lieu of bathymetry.

In Figure 2 we show a sample of tracks generated by the stochastic track and intensity model.

FPHLM V3.0 2008 15

Figure 2. Examples of simulated hurricane tracks. Numbers refer to the stochastic track number, and colors represent storm intensity based on central pressure. Dashed lines represent tropical storm strength winds, and Cat 1-5 winds are represented by black, blue, orange, red and turquoise, respectively.

When a storm is started, the parameters for radius of maximum winds and Holland B are computed and appropriate error terms are added as described below. The Holland B term is modeled as follows:

B −= 007915.074425.1 Lat + 0000084.0 Delp 2 − 005024.0 R max where Lat is the current latitude of the storm center, DelP is the central pressure difference and Rmax is the radius of maximum winds. The random error term for the Holland B is modeled using a Gaussian distribution with a standard deviation of 0.286. Figure 3 shows a comparison between the Willoughby and Rahn (2004) B data set (see Standard M-2.1) and the modeled

FPHLM V3.0 2008 16 results (scaled to equal the 116 measured occurrences in the observed data set). The modeled results with the error term have a mean of about 1.38 and are consistent with the observed results. The figure indicates excellent agreement between model and observations.

Distribution of the B parameter 16 15 14 13 12 11 10 9 Observed 8 Model Scaled 7 6 Occurrence 5 4 3 2 1 0 0.5 1 1.5 2 B param eter

Figure 3. Comparison between the modeled and observed Willoughby and Rahn (2004) B data set

We develop an Rmax model using the revised landfall Rmax database which includes 108 measurements for storms up to 2005. We have opted to model the Rmax at landfall rather than the entire basin for a variety of reasons. One is that the distribution of landfall Rmax may be different than that over open water. An analysis of the landfall Rmax database and the 1988-2007 DeMaria Extended Best Track data shows that there appears to be a difference in the dependence of Rmax on central pressure (Pmin) between the two data sets. The landfall data set provides a larger set of independent measurements, more than 100 storms compared to about 31 storms affecting the Florida threat area region in the Best Track Data. Since landfall Rmax is most relevant for loss cost estimation, and has a larger independent sample size, we have chosen to model the landfall data set. Future studies will examine how the Extended Best Track Data can be used to supplement the landfall data set.

We model the distribution of Rmax using a gamma distribution. Using the maximum likelihood estimation method, we found the estimated parameters for the gamma distribution, kˆ = 53547.5 and θˆ = 67749.4 . With these estimated values, we show a plot of the observed and expected distribution in Figure 4. The Rmax values are binned in 5 sm intervals, with the x-axis showing the end value of the interval.

FPHLM V3.0 2008 17 Gam m a v s Observ ed 22.5

17 5 Observed 15 Gam m a

12 5

7 5

2.5

0 5 10 15 20 25 30 35 40 45 50 55

Figure 4. Observed and expected distribution for Rmax. The x-axis is the radius in statute miles, and the y-axis is the frequency of occurrence.

An examination of the Rmax database shows that intense storms, essentially Category 5 storms, have rather small radii. Thermodynamic considerations (Willoughby, 1998) also suggest that smaller radii are more likely for these storms. Thus, we model Category 5 (DelP>90 mb, where DelP=1013-Pmin and Pmin is the central pressure of the storm) storms using a gamma distribution, but with a smaller value of the θ parameter, which yields a smaller mean Rmax as well as smaller variance. We have found that for Category 1-4 (DelP<80) storms there is essentially no discernable dependence of Rmax on central pressure. This is further verified by looking at the mean and variance of Rmax in each 10 mb interval. Thus we model Category 1-4 storms with a single set of parameters. For a gamma distribution, the mean is given by kθ, and variance is kθ2. For Category 5 storms, we adjust θ such that the mean is equal to the mean of the three Category 5 storms in the database: 1935 No Name, 1969 Camille and 1992 Andrew. An intermediate zone between DelP=80 mb and DelP=90 mb is established where the mean of the distribution is linearly interpolated between the Category 1-4 value and the Category 5 value. As the θ value is reduced, the variance is likewise reduced. Since there are insufficient observations to determine what the variance should be for Category 5 storms, we rely on the assumption that variance is appropriately described by the re-scaled θ, via kθ2.

A simple method is used to generate the gamma-distributed values. A uniformly distributed variable, a product of the random number generator that is intrinsic to the Fortran compiler, is mapped onto the range of Rmax values via the inverse cumulative gamma distribution function. For computational efficiency, a lookup table is used for the inverse cumulative gamma distribution function, with interpolation between table values. Figure 5 shows a test using 100,000 samples of Rmax for Category 1-4 storms, binned in 1 sm intervals, and compared with the expected values.

FPHLM V3.0 2008 18

Figure 5. Comparison of 100,000 Rmax values sampled from the Gamma distribution for Cat 1-4 storms to the expected values.

For Category 5 and intermediate Category 4-5 storms, we utilize the property that the gamma cumulative distribution function is a function of (k,x/θ). Thus, by re-scaling θ, we can use the same function (lookup table), but just rescale x (Rmax). The rescaled Rmax will then still have a gamma distribution, but with different mean and variance.

The storms in the stochastic model will undergo central pressure changes during the storm lifecycle. When a storm is generated, an appropriate Rmax is sampled for the storm. In order to assure the appropriate mean values of Rmax as pressure changes, the Rmax is rescaled every time step as necessary. As long as the storm has DelP < 80 mb, there is in effect no rescaling. In the stochastic storm generator, we limit the range of Rmax from 4 sm to 60 sm.

Storm landfall and decay over land are determined by comparing the storm location (x,y) with a 0.6 sm resolution land-sea mask. This land mask is obtained from the U.S. Geological Survey (USGS) land use cover data, and inland bodies of water have been reclassified as land in order to avoid spurious landfalls. Landfall occurs every time the storm moves from an ocean point to a land point as determined by this land mask. During landfall, the central pressure is modeled by a filling model described in Vickery (2005), and is no longer sampled from the intensity change PDFs. The Vickery (2005) model basically uses an exponentially decaying, in time, function of the central pressure difference with the decay coefficients varying by region based on historical data. The pressure filling model also takes into account the speed and size of the storm. When the storm exits to sea, the land filling model is turned off and sampling of the intensity change PDFs begins again. A storm is dissipated when its central pressure exceeds 1011 mb.

FPHLM V3.0 2008 19 WIND FIELD MODEL Once a simulated hurricane moves to within a threshold distance of a Florida zip code, the wind field model is turned on. The model is based on the slab boundary layer concept originally conceived by Ooyama (1969) and implemented by Shapiro (1983). Similar models based on this concept have been developed by Thompson and Cardone (1996) and Vickery et al. (1995, 2000). The model is initialized by a boundary layer vortex in gradient balance. Gradient balance represents a circular flow caused by balance of forces on the flow whereby the inward directed pressure gradient force is balanced by an outward directed Coriolis and centripetal accelerations. The coordinate system translates with the hurricane vortex moving at velocity c. The vortex translation is assumed to equal the geostrophic flow associated with the large scale pressure gradient. In cylindrical coordinates that translate with the moving vortex, equations for a slab hurricane boundary layer under a prescribed pressure gradient are:

∂u v 2 v ∂ ∂pu u 2 ∂u ∂u u fv +−− + ( 2uK −−∇− ucF 0),() ==+ ∂r r r ∂φ ∂r rr 22 ∂φ ∂t

⎛ ∂v v ⎞ v ∂v v 2 ∂u ∂v u⎜ + ⎟ fu ++ ( 2vK +−∇− vcF 0),() ==+ ⎝ ∂r r ⎠ r ∂φ rr 22 ∂φ ∂t

where u and v are the respective radial and tangential wind components relative to the moving storm, p is the sea-level pressure which varies with radius (r), f is the Coriolis parameter which varies with latitude, φ is the azimuthal coordinate, K is the eddy diffusion coefficient, and F(c,u), F(c,v) are frictional drag terms. All terms are assumed to be representative of means through the boundary layer. The motion of the vortex is determined by the modeled storm track. The symmetric pressure field p(r) is specified by the Holland (1980) pressure profile with the central pressure specified according to the intensity modeling in concert with the storm track. The model for the Holland B pressure profile and the radius of maximum wind are described above. The wind field is solved on a polar grid with a 0.1 R/Rmax resolution. The input Rmax is adjusted to remove a bias caused by a tendency of the wind field solution to place Rmax one grid point radially outward from the input value.

The slab mean boundary layer wind speed is adjusted to the surface based on reduction factors published in Powell et al. (2003) and is adjusted to maximum sustained and peak 3s gust values according to gust factors as described in Vickery and Skerlj (2005). Flow transition from marine to land or from one land roughness to another is dependent on aerodynamic roughness as modeled by Simiu and Scanlon (1996). The roughness database is derived from the Multi- Resolution Land Cover (MRLC) National Land Classification Database (NLCD) of 2001 (Homer et al., 2004). This data set provides land cover classifications on a very high resolution (30 m or 98 ft) grid over the entire U.S. We convert the land cover classifications to a typical roughness value associated with the surface classification via a lookup table. We then use an effective roughness model (Axe 2004) based on the Source Area Model of Schmidt and Oke (1990) to determine an upstream fetch dependent roughness value at all Florida zip codes. The effective roughness model essentially uses a weighting function that decays exponentially as a function of the distance of each roughness element from the zip code centroid. The effective

FPHLM V3.0 2008 20 roughness is computed for 8 possible incoming wind direction intervals for each zip code centroid. The effective roughness is the weighted average of all roughness elements for each octant. We corrected some anomalies where the population centroids were not near residential properties. To remedy this, we set a lower limit of roughness, for each roughness element that is weighted in the effective roughness model, equivalent to that of a low intensity residential area to all land points within 0.311 miles of the centroid. For special cases where the centroid is over water, the roughness was set to that of a low intensity residential area for all points within 0.311 miles of the centroid. For coastal regions, we corrected the roughness by averaging the effective roughness for coastal fetches. Further details on the atmospheric component of the model are contained in Powell et al. (2005).

FPHLM V3.0 2008 21 The Vulnerability Component

The vulnerability model uses a Monte Carlo simulation based on a component approach to determine the external vulnerability of buildings at various wind speeds. The approach explicitly accounts for the resistance capacity of the various building components, the load effects produced by wind from different directions, and associated uncertainties of capacity and loads to predict exterior damage at various wind speeds. The simulation relates probabilistic strength capacities of exterior and structural building components to a series of deterministic 3 sec peak gust wind speeds through a detailed wind and structural engineering analysis that includes effects of wind-borne missiles.

Damage to the structure occurs when the loads from wind or flying debris are greater than the component’s capacity to resist them, where both pressure loads and the likelihood of debris damage increase with wind speed. Once the strength capacities, load demands, and load path(s) are identified and modeled, the vulnerability of a structure at various wind speeds can be estimated by quantifying the amount of damage to the modeled components (shingles, sheathing, walls, connections, windows, doors). Damage to a given component may influence the resistance of other components. For example, load redistribution due to a damaged roof to wall connection, or a change in roof loading from internal pressurization due to a damaged opening. These influences are accounted for through an iterative process of loading, damage assessment, load redistribution and re-loading until convergence is reached.

The damage estimations are affected by uncertainties regarding: the behavior and strength of the various components, and the load effects produced by hurricane winds. Field and laboratory data that better define these uncertain behaviors can thus be directly included in the model by refining the statistical descriptors of the capacities, load paths, and loads.

Once the external damage has been calculated via Monte Carlo simulation, the internal, utilities, and contents damages to the building are then extrapolated from the external damage. The resulting estimates of total building damage result in the formulation of vulnerability matrices for each building type that represents a representative portion of the Florida building stock. The damage model is complemented with estimates of appurtenant structures damage, contents, and additional living expenses (ALE).

The model accounts for a number of construction factors that influence the vulnerability of single family dwellings, including: classification (site built or manufactured home) home size, roof shape, location within Florida, age, and a variety of construction details and mitigation measures. The effects of mitigation measures such as code revisions and post-construction upgrades to the wind resistance of homes (e.g. new roof cover on an older home, shutter protection against debris impact, braced garage door, re-nailed roof decking, etc.) are accounted for both individually and in combination by modifying the statistical descriptors of the capacities of the various components. Thus the comparative vulnerability of older homes as-built, older homes with combinations of mitigation measures, and homes constructed to the new code requirements can be estimated.

FPHLM V3.0 2008 22 The following flow chart (Figure 6) summarizes the Monte Carlo procedure used to predict the external damage. The random variables include wind speed, pressure coefficients, flying debris impact, and the resistances of the various building components (roof cover, roof sheathing, openings, walls, connections).

Selection of Structural Type, Definition of Geometry

Loop for Angle Define Angle of Incidence

Loop for Wind Speed Define Wind Speed

Loop for Building

Simulate an Individual Home: Randomize Wind Speed, Cp’s Sample Resistances

Calculate Initial Loads Repeat until allangles complete

Initial Failure Check: Sheathing, Openings, Walls

Calculate Internal Pressure from Opening Failures and Recalculate Structural Loads

Final Failure Check: Openings, Sheathing angle next to go then complete, speeds all until Repeat Roof Cover all simulations complete until Repeat Connections, Walls Save Damage File Write 1 Row of Damage Array

Figure 6. Monte Carlo Simulation Procedure to Predict Damages

EXPOSURE STUDY The engineers used several sources of information for getting Florida up-to-date building structural type information. One source is the Florida Hurricane Catastrophe Fund (FHCF) exposure database. Another, more important source is the Florida counties property tax appraisers’ databases.

Property appraisers' databases are the most comprehensive and accurate information of building structural characteristics accessible at the present time. Although the databases’ contents and format vary county to county, most of them contain the critical structural information to define the most common structural types in each county. Nine county databases were processed. They

FPHLM V3.0 2008 23 correspond to Brevard, Hillsborough, Pinellas, Walton, Escambia, Leon, Broward, Palm Beach and Monroe counties.

Tax appraiser databases contain large quantities of building information, and it was necessary to extract those characteristics related to the vulnerability of the buildings to wind. First, all the buildings in each county database were divided into three major categories: single family residential buildings, condominiums, and mobile homes. Under each category, the authors chose to extract information on 6 critical building characteristics for analysis and statistical distribution. They are roof cover, roof type, exterior wall material, number of stories, year built and building area.

The aim of the survey was to generate a manageable number of building models to cover the majority of the Florida building stock. To define building models for each county could be cumbersome and unnecessary. Instead, the engineers divided Florida into four regions: North, Central, South, and the Keys. Geography, and the statistics from the Florida Hurricane Catastrophe Fund (FHCF) guided them in defining regions that would have a similar building mix throughout their counties. For example, the FHCF shows that northern Florida has a preponderance of frame houses.

In each region, there were at least two counties: Escambia, Walton, and Leon in the Northern region; Brevard, Pinellas, and Hillsborough in the Central region; Palm Beach, and Broward in the Southeast region. In addition, Monroe County fully covers the Keys region.

To define the structural types, the modelers chose a combination of 4 characteristics: number of stories (either 1 or 2), roof cover (shingle/tile), roof type (either gable or hip) and structural material (either concrete blocks or timber). Based on the information contained in the databases, the authors computed the statistics for each structural type in every sample county and then used weighted average techniques to extrapolate the results to each region.

SITE BUILT MODELS Commonalities for all site built model homes are 15 windows, a garage, a front entrance door, and a sliding glass back door.

In addition to a classification of building by structural types (wood or masonry walls, hip or gable roof), it was also necessary to classify the buildings by relative strength. Residential construction methods have evolved in Florida as experience with severe winds drives the need to reduce vulnerability. To address this, the vulnerability team has developed a strong model, medium strength model, and a weak strength model for each site-built structural type to represent relative quality of original construction as well as post-construction mitigation.

The strong model was developed first and both the weak and medium models were derived from the strong model by adjusting component capacities within the standard model framework. For example, the standard model for the south, concrete block, gable roof construction is converted to a weak model by lowering the roof-to-wall (r2w) connection capacity to toe-nail strength, lowering the garage pressure capacity, reducing the cracking capacity of masonry walls (removing reinforcing steel), and lowering the sheathing and shingle capacities to reduce wind

FPHLM V3.0 2008 24 uplift. Simulations have been generated for gable roof, 1 and 2-story wood and 1-story concrete block wall, north, central and south regions. This has been repeated with plywood window shutters in place. The medium models differ from the weak models by improving the roof to wall connection capacity from toe-nail to metal clip strength. Any given strong, medium or weak model may be altered by additional mitigation measures individually or in combination.

MANUFACTURED HOMES MODELS Based on the exposure study, it was also decided to model four manufactured home (MH) types. These types include: Pre -1994 - Fully Tied down; Pre-1994 - Not Tied down; Post-1994 - HUD Zone II; Post-1994 - HUD Zone III, where 1994 delineates older, much weaker style of manufactured home construction from the post-1994 homes that meet minimum federal construction standards established by HUD.

The partially tied down homes are assumed to have a vulnerability that is an average of the vulnerabilities of fully tied-down and not tied-down homes. Because little information is available regarding the distribution of manufactured home types by size or geometry, it is assumed that all model types are single-wide manufactured homes. The modeled single-wide manufactured homes are 56 ft x 13 ft, have gable roofs, 8 windows, a front entrance door, and a sliding glass back door.

DAMAGE MATRICES The physical damage to single-family homes is estimated by using a component-based Monte Carlo (MC) simulation engine. The simulation estimates probabilistic strength capacities of building components as functions of 3 sec peak gust wind speeds through a detailed wind and structural engineering analysis that includes effects of wind-borne missiles. The component approach taken in the MC simulation explicitly accounts for both the uncertain resistance capacity of the various building components and the load effects produced by wind to predict damage at various wind speeds and directions. The resistance capacity of a building is broken down into the resistance capacity of its components and of their connections. The components include roof cover, roof sheathing, roof-to-wall connections, walls, windows, doors, and garage doors. Damage to the structure occurs when the load effects from wind or flying debris are greater than the component’s capacity to resist them. The output of the Monte Carlo simulation model is an estimate of physical damage to structural and exterior components of the modeled home. The results are in the form of a damage matrix. Each row of the matrix lists results of one model simulation, the amount of damage to each of the 15 modeled components for a simulation being listed in 15 columns of the row (see Table 1). Each damage matrix gives the results of 5000 Monte Carlo simulations. A separate matrix is created for each peak 3-s gust wind speed between 50 and 250 mph in 5 mph increments (50, 55,…, 250 mph) at angles between 0 and 315 degrees in 45-degree increments (50 mph at 0°, 50 mph at 45°, 50 mph at 90°,...). The way the results are produced and stored for the MH models is very similar for the site-built and manufactured home models. A description of the values in each of the 9 columns of the MH damage matrix is given in Table 2.

The following flow chart (Figure 7) summarizes the procedure used to convert the results of the Monte Carlo simulations of physical external damage into a vulnerability matrix.

FPHLM V3.0 2008 25

Figure 7. Procedure to create vulnerability matrix.

FPHLM V3.0 2008 26

Table 1. Description of values given in the damage matrices for site built homes Col.# Description of Value Min Value Max Value 1 % failed roof sheathing 0 100 2 failed roof cover 0 100 3 failed roof to wall connections 0 100 4 # of failed walls 0 4 5 # of failed windows 0 15 6 # of failed doors 0 2 7 y or n failed garage 0 = no 1 = yes 8 y or n envelope breached 0 = no 1 = yes 9 # of windows broken by debris impact 0 15 10 % of gable end panels broken 0 100 11 internal pressure 0 Not defined 12 % failed wall panels – front 0 100 13 % failed wall panels – back 0 100 14 % failed wall panels – side 0 100 15 % failed wall panels – side 0 100

Table 2. Description of values given in the damage matrices for manufactured homes Col # Description of Value Min Value Max Value 1 # of failed windows (out of 8 for single wide) 0 8 2 # of broken windows that were broken by impact load case 0 8 3 # of failed doors (front and back = 2 total) 0 2 4 % of roof sheathing failed 0 100 5 % of roof cover failed 0 100 6 % of wall sheathing failed 0 100 7 # of failed roof to wall connections (out of 58) 0 58 8 sliding (0 = no sliding, 1 = minor sliding, 2 = major sliding) 0 2 9 overturning (0 = not overturned, 1 = overturned) 0 1

Replacement cost ratios provide a key link between modeled physical damage and the corresponding monetary losses. They can be defined as the cost of replacing a damaged component or assembly of a home divided by the cost of constructing a completely new home of the same type. The sum of these ratios is greater than 100% because the replacement costs include the additional costs of removal, repair, and remodeling. Knowing the components of a home and the typical square footage, the cost of repairing all damaged components is estimated using cost estimation resources (e.g. RSMeans Residential Cost Data and CEIA) and expert advice. These resources provide cost data from actual jobs based on successful estimates and represent an average of typical conditions. Unmodeled non-structural interior, plumbing, mechanical, and electrical utilities make up a significant portion of repair costs for a home.

A very simple and explicit procedure is used to convert physical damage of the modeled components to monetary damage. Since the replacement ratio of each modeled component is known, the monetary damage resulting from damage to a component expressed as a percentage

FPHLM V3.0 2008 27 of the home’s value can be obtained by multiplying the damaged percentage of the component by the component’s replacement ratio. For example, if 30% of the roof cover is damaged, and for this particular home type the replacement ratio of roof cover is 14%, the value of the home lost as a result of the damaged roof cover would be 0.30 x 0.14 = 4.2%. If the value of this home were, say, $150,000, the cost to replace 30% of the roof would be $150,000 x 0.042 = $6,300. In addition, the costs will be adjusted as necessary due to certain requirements of the Florida building code that might result in an increase of the repair costs.

INTERIOR AND UTILITIES DAMAGE For the interior and utilities of a home, there is no explicit means by which to compute damages and resulting damage. Unlike the modeled exterior components for which we know that, for each wind speed, loads in excess of the capacity will cause damage and the cost of replacing these components is fairly certain, damage to the interior and utilities occurs when the building envelope is breached allowing wind and rain to enter, and the cost of repairing this damage could be highly variable. Of all the modeled components for site-built homes, damage to roof sheathing, roof cover, walls, windows, doors, and gable ends present the greatest threat of causing interior damage. For manufactured homes, additional interior damage could be caused by sliding or overturning off the foundation.

For each wind speed, interior damage equations are derived as functions of each of the modeled components mentioned earlier. These equations are developed primarily on the basis of experience and engineering judgment. Observations of homes damaged during the 2004 hurricane season helped to validate the predictions. The interior equations are derived by estimating typical percentages of damage to each interior component given a percentage of damage to a modeled component. The interior damage as a function of each modeled component is the same for both site-built and manufactured homes.

To model the uncertainties inherent in the determination of interior damage, the output of the equations is multiplied by a random factor with mean unity. Based on engineering judgment, the factor is assumed to have a Weibull distribution with tail length parameter 2. For the factor to have mean unity, the scale parameter must be 0.7854, resulting in a variance of 0.2732. This choice of Weibull parameters is assumed to be reasonable, and a sensitivity study was done to confirm that assumption and to show that it has no effect on the mean vulnerability, as expected.

To compute the total interior damage for each model simulation, first of all, all values in the damage matrices are converted to percentages of component damage. The interior equations are applied to each component and the total interior damage for each model simulation is taken to be the maximum interior damage value produced by these equations. The maximum value is used to avoid the possibility of counting the same interior damage more than once.

The simplest and most logical method to estimate utilities damage is based upon the prediction of interior damage. To extrapolate the utilities damage, a coefficient is defined for each utility (electrical, plumbing, and mechanical), which is then multiplied by the interior equation defined for each component, and the total damage is taken to be the maximum value. The utilities coefficients are based on engineering judgment. In both site-built and manufactured homes, it is assumed that electrical damage occurs at about half the rate of interior damage, and each interior

FPHLM V3.0 2008 28 equation is multiplied by a coefficient ke=0.5. Plumbing damage is predicted in the same way as electrical damage. However, plumbing damage is assumed to occur at a slower rate than electrical damage. Therefore, the coefficient kp is set equal to 0.35 for site-built homes and for manufactured homes. It is assumed that mechanical damage will occur at a lower rate than electrical damage but at a slightly higher rate than plumbing damage. The value of km is set to 0.4 for site-built homes and for manufactured homes.

CONTENTS DAMAGE Contents include just about anything in the home that is not attached to the structure itself. Like the interior and utilities, the contents of the home are not modeled by Monte Carlo simulations. Contents damage is assumed to be a function of the interior damage caused by each modeled component failure that causes a breach of the building envelope. The functions are based on engineering judgment and validated using actual claims data.

ADDITIONAL LIVING EXPENSES Additional Living Expense (ALE) is coverage for the increase in living expenses that arise when an insured individual must live away from the insured damaged home. ALE coverage covers only expenses actually paid by the insured. This coverage does not pay all living expenses, only the increase in living expense that results directly from the covered damage, and having to live away from the insured location. The value of an ALE claim is obviously dependent on the time it takes to repair a damaged home as well as the surrounding utilities and infrastructure.

The equations and methods used for manufactured and residential homes are identical. However, it seems logical to reduce the manufactured home ALE predictions because typically a faster repair or replacement time may be expected for these home types. Therefore, a factor Rf was introduced into the manufactured home model. This Rf factor is now set at 0.75 based on engineering judgment, and it multiplies the ALE predictions to adjust the values.

APPURTENANT STRUCTURES Appurtenant structures, typically, are structures not attached to the dwelling or main residence of the home, but located on the insured property. These types of structures could include: detached garages, guesthouses, pool houses, sheds, gazebos, patio covers, patio decks, swimming pools, spas, etc. From insurance claims data there appears to be no obvious relationship between building damage and appurtenant structure claims. One of the primary reasons for this maybe the variability of the structures that are covered by an appurtenant structure policy.

To model appurtenant structure damage, three separate equations were developed. Each determines the appurtenant structure insured damage ratio as a function of wind speed (vulnerability curve). One equation predicts damage for structures highly susceptible to wind damage, the second for moderately susceptible, and the third for structures which are affected only slightly by wind. Because a typical insurance portfolio file gives no indication of the type of appurtenant structure covered under a particular policy, a distribution of the three types (slightly vulnerable, moderately vulnerable, and highly vulnerable) must be assumed, and is validated against the claim data.

FPHLM V3.0 2008 29 VULNERABILITY MATRICES For each Monte Carlo model, 5000 simulations are performed at 8 different angles and 41 different wind speeds. This is 5000 x 8 x 41 = 1,640,000 simulations per model, which are expanded to cover interior, utilities, contents, ALE, and appurtenant structures, as explained above. The simulation results are then transformed into vulnerability matrices. A total of 168 matrices are created for every combination of structural type (frame or masonry), region (North, Central, South), sub-region (high wind velocity zone, wind borne debris region, other), and roof cover type (gable vs. hip, tile vs. shingle).

A partial example of a vulnerability matrix is shown in Table 3.

Table 3. Partial example of vulnerability matrix Damage\Wind Speed (mph) 48.5 to 52.5 52.5 to 57.5 57.5 to 62.5 62.5 to 67.5 67.5 to 72.5 0% to 2% 1 0.99238 0.91788 0.77312 0.61025 2% to 4% 0 0.00725 0.0805 0.21937 0.36138 4% to 6% 0 0.000375 0.001375 0.007 0.0235 6% to 8% 0 0 0.000125 0.000375 0.0025 8% to 10% 0 0 0 0 0.000375 10% to 12% 0 0 0 0 0.000375 12% to 14% 0 0 0 0 0.000625 14% to 16% 0 0 0 0 0.0005 16% to 18% 0 0 0 0 0.000125 18% to 20% 0 0 0 0 0.000125 20% to 24% 0 0 0 0 0.00025 24% to 28% 0 0 0 0 0

The cells of a vulnerability matrix for a particular structural type represent the probability of a given damage ratio occurring at a given wind speed. The columns of the matrix represent the different wind speeds from 50 mph to 250 mph in 5 mph increments. These are 3-s gust wind speeds at a 10 m height. The rows of the matrix correspond to damage ratios (DR) in 2% increments up to 20%, and then in 4% increments up to 100%. At each wind speed, the number of instances of damage within each damage range are counted. For example, if a damage ratio is DR= 15.3%, it is assigned to the interval 14%

One important plot derived from the vulnerability matrix is the vulnerability curve. The vulnerability curve for any structural type is the plot of the mean or average damage ratio per wind speed vs. wind speed. The model can also generate fragility curves for each vulnerability matrix, although these curves are not used in the model. Fragility curves are curves that represent the probability of exceedance of any given damage level, as a function of the wind speed.

FPHLM V3.0 2008 30 Similar vulnerability matrices, and vulnerability curves, are developed for contents, and ALE, one for each structural type. Since the appurtenant structures damage is not derived from the building damage, only one vulnerability matrix is developed for appurtenant structures. The whole process is also repeated for manufactured homes.

Building vulnerability matrices were created for every combination of region (Keys, South, Central, and North), construction type (masonry, wood, or other), roof type (gable or hip), roof cover (tile or shingle), shutters (with or without), and sub-region (standard, windborne debris region, and high velocity zone). However, in general, there is little information available in an insurance portfolio file regarding the structural characteristics and the wind resistance of the insured property. Instead, insurance companies rely on the so-called ISO classification, which is primarily used to define the fire resistance of a home. In addition to the ISO classification, portfolio files will have information on zip code and year built. The ISO classification is used to determine if the home is constructed of masonry, timber, or other. The zip code is used to define the region and sub-region. The year the home was built is utilized to assist in defining whether a home should be considered weak, medium or strong. It is also used for damage predictions for mobile homes.

So from the insurance files, we can easily determine the region, sub-region, construction type, and year built. However this leaves the roof type, roof cover, and shutter options still undefined. But we know from the exposure study, the distribution of different roof types, and to some extent of roof cover per region. Also, some estimation of the percentage of homes with and without shutters in each sub-region can be made. Based on these statistics and estimates, we can define a general matrix for each construction type in each region and sub-region. The general matrices are simply the sum of the model matrices weighted on the basis of their statistical distribution. For example, if we know that a home is masonry construction and is in the windborne debris region of central FL, we also know that 66% of the masonry homes in central FL have gable roofs and 34% have hip roofs, around 85% have shingle cover and 15% tile, and 20% have shutters while 80% do not. Weight factors can be computed for each model matrix based on these statistics. For example, the Central FL, gable, tile, no shutters, masonry matrix would have a weight factor of 66% (masonry percent gable) x 15% (percent tile) x 80% (percent without shutters) = 7.9%, this is the percentage of that home type that would be expected in this region. Each model matrix is multiplied by its weight factor, and the results are summed. The final result is a weighted matrix that is a combination of all the model matrices and can be applied to an insurance policy if only the zip code, year built, and ISO classification are known. As a result, for each sub-region (standard, windborne debris region, and high velocity zone) of each region (Keys, South, Central, and North), there will be a set of weighted matrices (masonry, wood, and others) for weak, medium, and strong structures. Figure 8 shows the weighted matrices for the masonry structures in a Central sub-region.

MODEL'S DISTRIBUTION IN TIME Over time, engineers and builders learned more about the interaction between wind and structures, more stringent building codes were enacted, and when properly enforced, resulted in stronger structures. The weak model, medium strength model, and standard (strong) strength model, developed by the vulnerability team, represent this evolution in time of relative quality of construction in Florida. Each set of models is representative of the prevalent wind vulnerability

FPHLM V3.0 2008 31 of buildings for a certain historical period in time. It is therefore important to define the cut-off date between the different periods, since the overall aggregate losses in any region are determined as a mixture of homes of various strengths (ages). The cut-off dates do not depend only on the evolution of the building code, but also on the prevailing local builder/community code enforcement standards in each era.

This issue of code enforcement has also evolved over time, and it is relatively recent that the State of Florida took an active role in uniform enforcement. Thus a given county may have built to standards that were worse than or exceeded the code in place at the time. After consulting with the building code development experts, the team concluded that the load provisions had some wind provisions since the 1970’s, and the issue is not the code, but rather enforcement of the code. Southern FL construction practice recognized the importance of truss to wall connection as early as the 1950’s, when it became common to use clips rather than toe nails. The clips were not as strong as modern straps, but an improvement over nails only. Northern FL construction suffered from the lack of impact from severe hurricanes over a long period. This sense of safety was compounded by a more localized approach to decision making. Thus northern construction is expected to be weaker than southern in general. The use of clips became relatively standard state-wide by the mid 1980’s, while they were well used in the south prior to this time. The use of rated shingles and resistant garage doors became common after Andrew. Therefore, the classification shown in Table 4 was adopted for characterizing the regions by age and model.

Table 4. Age classification of the models per region

Prior to 1970 1970 to 1983 1984 to 1993 1994 – present All regions ½ weak, ½ medium Medium Medium Strong

However, the year-built or year of last upgrade of a structure in a portfolio might not be available, when performing a portfolio analysis to estimate hurricane losses, in a certain region. In that case, it becomes necessary to assume a certain distribution of ages in the region, in order to come up with an average vulnerability between weak, medium, and strong, and estimate the resulting overall damage to a given county (or zip code).

Although the engineering team did not have detailed information on the building population of every county in Florida, they did have information on 1.5 million homes from insurance company portfolios. The portfolios include an effective year of construction, and thus provide guidance as to how to weigh the combined weak, medium and strong model results when yearbuilt information is not available in other portfolio files. In each region, the data was analyzed to provide the age statistics. These statistics were used to weigh the average of weak, medium, and strong vulnerabilities in each region. The results are shown in Figure 8, for the wind borne debris zone in the Central region. The different weighted vulnerability curves are shown for the weak, medium, and strong models, superimposed with the age weighted vulnerability curve.

FPHLM V3.0 2008 32 Central WBDR Masonry Building Vulnerabilities

50%

Strong 15, February 07

40%

Medium 15, February 07

s 30% Weak 15, February 07

Damage Ratio 20% Age Weighted 0219071-c

10%

0% 50 70 90 110 130 150 170 3 sec gust wind speeds Figure 8. Weighted masonry structure vulnerabilities in the Central wind bone debris zone

Actuarial Component

The actuarial component consists of a set of algorithms. The process involves a series of steps: rigorous check of the input data; selection and use of the relevant output produced by the meteorology component; selection and use of the appropriate vulnerability matrices for building structure, contents, appurtenant structure and additional living expenses; running the actuarial algorithm to produce expected losses; and aggregating the losses in a variety of manner to produce a set of expected annual hurricane wind losses. These losses can be reported by construction type (e.g. masonry, frame, manufactured homes), by county or zip code, by policy form (e.g., HO-3, HO-4 etc.), by rating territory, and combinations thereof.

Expected annual losses are estimated for individual policies in the portfolio. They are estimated for building structure, appurtenant structure, contents and ALE based on their exposures and by using the respective vulnerability matrices for the construction types. There are two methods available for estimating expected losses that theoretically produce the same results. In the first method, for each policy, losses are estimated for all the hurricanes in the stochastic set by using appropriate damage matrices and policy exposure data. The losses are then summed over all hurricanes and divided by the number of years in the simulation to get the annual expected loss. These are aggregated at the zip code, county, territory, or portfolio level and then divided by the respective level of aggregated exposure to get the loss costs. This is a computationally demanding method. Each portfolio must be run through the entire stochastic set of hurricanes.

The second method derives the probability distribution of winds for each zip code from the simulated set of hurricanes. This is done once for each zip code. These distributions are then applied directly to the damage (vulnerability) matrices, and using the insured value and deductible, the expected losses are estimated for each policy. These are then aggregated as needed.

FPHLM V3.0 2008 33

The distribution of losses is driven by both the distribution of damage ratios generated by the engineering component and by the distribution of wind speeds generated by the meteorology component. The meteorology component provides, for each zip code, the associated probabilities for a common set of wind speeds. Thus, zip codes are essentially differentiated by their probability distribution of wind speeds. The meteorology component uses up to 53,500 year simulations to generate a stochastic set of storms. The storms are hurricane events at landfall or when bypassing close by. Each simulated storm has a track and a set of modeled wind fields at successive time intervals. The wind fields generate the 1 minute maximum sustained wind speeds for the storm at various locations (population weighted zip code centroids) along its track. These 1 minute maximum sustained winds are then converted to 3 second peak gusts winds and corrected for terrain roughness by using the gust wind model and the terrain roughness model. For each zip code population centroid, an accounting is then made of all the simulated storms that pass through it. Based on the number of pass through storms and their peak wind speeds, a distribution of the wind speed is then generated for the zip code. Based on this distribution, probabilities are generated for each 5 mph interval of wind speeds, starting at 20 mph. These 5 mph bins constitute the column headings of the damage matrices generated by the engineering component. The wind speeds are generated for the location of the population centroids of the zip codes.

The engineering group has produced vulnerability matrices. Damage ratios are grouped and intervals (or classes) of various lengths are used. Furthermore, damages probabilities for damage intervals are produced for a whole range of wind speeds.

Vulnerability matrices are provided for building structure, contents, appurtenant structures and additional living expenses for a variety of residential construction type and for different policy types. The construction types are: masonry, frame, mobile home, and unknown. The vulnerability matrices are also developed for weak, medium, and strong construction as proxy by year built.

Within each broad construction category, the vulnerability matrices are specific to the roof types and number of stories etc. Since the policy data do not provide this level of specificity, weighted matrices are used instead, where the weights are the proportion of different roof types in given region as determined by a survey of the building blocks and exposure data. The vulnerability matrices are used as input in the actuarial model.

The starting point for the computations is the vulnerability matrix with its set of damage intervals and associated probabilities. Appropriate vulnerability matrices are applied separately for building structure, content, appurtenant structure and ALE. Once the matrix is selected, for a given a wind speed, for each of the mid point of the damage intervals the ground up loss is computed, the appropriate deductibles and limits are applied, and the loss net of deductible is calculated. More specifically, for each damage outcome the damage ratio is multiplied by insured value to get dollar damages, the deductible is deducted and net of deductible loss is estimated, subject to the constraints that net loss is ≥ 0 and ≤ limit – deductible. Percentage deductibles are converted into dollar amounts. Both the replacement cost and actual cash value are generally assumed to equal the coverage limit. Furthermore, if there are multiple hurricanes

FPHLM V3.0 2008 34 in a year in the stochastic set, the wind deductibles are applied to the first hurricane, and any remaining amount is then applied to the second hurricane. If none remains then the general peril deductible can be applied.

The net of deductible loss is multiplied by the probability in the corresponding cell to get the expected loss for the given damage ratio. The results are then averaged across the possible damages for the given wind speed. Next, the wind probability weighted loss is calculated to produce the expected loss for the property. The expected losses are then adjusted by the appropriate expected demand surge factor.

The demand surge factors are estimated by a separate model and applied appropriately to each hurricane in the stochastic set. The surge factors for structures are a function of the size of statewide storm losses and are produced separately for the different regions in Florida. The surge factors for content and ALE are functionally related to the surge factor for structure. To estimate the impact of demand surge on the settlement cost of structural claims following a hurricane, data from 1992 to 2007 on a quarterly construction cost index produced by Marshall & Swift/Boeckh is used. The approach to estimating structural demand surge was to examine the index for specific regions impacted by one or more hurricanes since 1992. From the history of the index we projected what the index would have been in the period following the storm had no storm occurred. Any gap between the predicted and actual index was assumed to be due to demand surge. In total ten storm/region combinations are examined. From these ten observations of structural demand surge the functional relationship is generalized.

After the losses are adjusted for demand surge, they are summed across all structures of the type in the zip code and also across zip codes to get expected aggregate portfolio loss. The model can process any combination of policy type, construction type, deductibles, coverage limits etc. The model output reports include separate loss estimates for structure, content, appurtenant structure, and ALE. These losses are also reported by construction type (e.g. masonry, frame, manufactured homes), by county or zip code, by policy form (e.g., HO-3, HO-4 etc.), by rating territory, and combinations thereof.

Computer System Architecture

FPHLM is a large-scale system, which is designed to store, retrieve, and process huge amounts of hurricane historical data and the simulated data. In addition, intensive computations are supported for hurricane damage assessment and insured loss projection. In order to achieve system robustness, flexibility, and resistance to potential change, a three-tier architecture is adopted and deployed in our system. It aims to solve a number of recurring design and development problems, and hence makes the application development work easier and more efficient. The computer system architecture consists of three layers, namely the user interface layer, application logic layer, and database layer.

The interface layer offers the user a friendly and convenient user interface to communicate with the system. It manages the input/output data and their display. To offer greater convenience to

FPHLM V3.0 2008 35 the users, the system is prototyped on the Web so that the users can access the system with existing web browser software.

The application logic layer handles the controlling functionalities and manipulates the underlying logic connection of the information flow. This is the middle tier in the computer system architecture. It aims to bridge the gap between the user interface and the underlying database and to hide the technical details from the users.

The database layer is responsible for data modeling to store, index, manage, and model the information for this application. Data needed by the application logic layer are retrieved from the database, and the computational results produced by the application logic layer are stored back to the database.

SOFTWARE, HARDWARE, AND PROGRAM STRUCTURE The system is primarily a web-based application that is hosted in Oracle 9i web application server. The backend server environment is Linux and the server side scripts are written in Java Server Pages (JSP) and Java beans. Backend probabilistic calculations are coded in C++ using the IMSL library and called through Java Native Interface (JNI). The system uses an Oracle database that runs on a Sun workstation. Server side software requirements are the IMSL library CNL 5.0, OC4J v1.0.2.2.1, Oracle 9i AS 9.0.2.0.0A, JNI 1.3.1, and JDK 1.3.1.

The end-user workstation requirements are minimal. Internet Explorer 5.5 or 6 running on Windows 2000 or XP are the recommended web browsers. However, other web browsers such as Mozilla Firefox should also deliver optimal user experience. Typically, the vendor’s minimal feature set for a given web browser and operating system combination is sufficient for an optimal operation of the application.

TRANSLATION FROM MODEL STRUCTURE TO PROGRAM STRUCTURE FPHLM uses a component-based approach in converting from model structure to program structure. The model is divided into distinct components or modules, i.e., Storm Forecast Module, Wind Field Module, Damage Estimation Module, and Loss Estimation Module. Each of these modules fulfills its individual functionality and communicates with other modules via well- defined interfaces. The architecture and program flow of each module are defined in its corresponding use case document following software engineering specifications. Each model element is translated into subroutines, functions, or class methods on a one-to-one basis. Changes to the models are strictly reflected in the software code.

FPHLM V3.0 2008 36

3. Provide a flow diagram that illustrates interactions among major model components. User Input

Storm Forecast Module

Retrieves historical storm data set based on Historical Storm user input Database: Generates probability distr bution functions for HURDAT storm motion and intensity Generates initial conditions for the storms Generates storm tracks for simulated storms

Stochastic Storm Database: Simulated Storms Wind Field Module

Estimates open terrain wind speeds Generates actual terrain wind speeds by using Information from roughness data and gust factors Geo Database: Calculates probability of 3-sec gust wind Ground Elevation speeds and Exposure Classification

Engineering Vulnerability Module Building Stock Data Defines structural type Translates and loads wind speeds Quantifies wind resistance Performs Monte Carlo simulation for external Engineering Data damage Quantifies total damage

Policy Data Actuarial Loss Module

Loads winds and vulnerability matrices Adds demand surge factors Calculates probability based insurance loss costs Insurance Claims Calculates scenario based insurance loss Data costs

Output

Figure 9. Flow diagram of the computer model

FPHLM V3.0 2008 37 4. Provide a comprehensive list of complete references pertinent to the submission by Standard grouping, according to professional citation standards.

References

Meteorology Standards

Anctil, F. and M. Donelan, 1996: Air-water momentum flux observations over shoaling waves. J. Phys. Ocean., 26, 1344-1353.

ASTM 1996: Standard practice for characterizing surface wind using a wind vane and rotating anemometer. D 5741-96, Annual Book of ASTM Standards, Vol. 11.03.

Axe, L. M., 2004: Hurricane surface wind model for risk assessment. M. S. Thesis, Department of Meteorology, Florida State University.

Batts, M. E., M. R. Cordes, L. R. Russell, and E. Simiu, 1980: Hurricane wind speeds in the United States, National Bureau of Standards, Report No. BSS-124, U. S. Dept. of Commerce.

Bosart, L. C. S. Velden, W. E. Bracken, J. Molinari, and P. G. Black, 2000: Environmental influences on the rapid intensification of Hurricane Opal (1995) over the Gulf of Mexico. Mon Wea. Rev., 128, 322-352.

Bove, M. C., J. B. Elsner, C. W. Landsea, X. Niu, and J. J. O’Brien, 1998: Effects of El Nino on U.S. land falling hurricanes, revisited. Bull. Amer. Meteor. Soc., 79, 2477–2482.

Darling, R. W. R., 1991: Estimating probabilities of hurricane wind speeds using a large scale empirical model, J. Climate, 4, 1035-1046.

DeMaria, M. and J. Kaplan, 1995: Sea surface temperature and the maximum intensity of Atlantic tropical cyclones. J. Climate, 7, 1324-1334.

DeMaria, M., J. Pennington, and K. Williams, 2002: Description of the Extended Best track file (EBTRK1.4) version 1.4. Available from NESDIS/CIRA/Regional and Mesoscale Meteorology Team, Colorado State University, Fort Collins, CO.

DeMaria, M. M. Mainelli, L.K. Shay, J.A. Knaff and J. Kaplan, 2005: Further improvements to the statistical hurricane intensity prediction scheme, Wea. Forecasting, 20, 531-543.

Donelan, M. A., and co-authors, 2004: On the limiting aerodynamic roughness of the sea in very strong winds. Geophys. Res. Lett., 31, L18306.

Dunion, J. P., C. W. Landsea, and S. H. Houston, 2003: A re-analysis of the surface winds for Hurricane Donna of 1960. Mon. Wea. Rev., 131, 1992-2011.

FPHLM V3.0 2008 38 Dunion, J. P., and M. D. Powell, 2004: A reconstruction of Hurricane Betsy’s (1965) wind field. Final Report to Army Corps of Engineers, New Orleans District.

Editors, 1974: The hurricane disaster potential scale. Weatherwise, 27, 169-186.

Emanuel, K. A. 1987. The Dependence of Hurricane Intensity on Climate. Nature, 326, 483-485.

Evans, J.L., 1993: Sensitivity of tropical cyclone intensity to sea surface temperature. J. Climate, 6, 1133-1140.

Franklin, J. L., M. L. Black, and K. Valde, 2003: GPS dropwindsonde wind profiles in hurricanes and their operational implications. Wea. Forecasting, 18, 32– 44.

Goldenberg, S.B., C.W. Landsea, A.M. Mestas-Nuñez, W.M. Gray, 2001: The Recent Increase in Atlantic Hurricane Activity: Causes and Implications, Science 293, 474-479.

Ho, F. P., J. C. Su, K. L. Hanevich, R. J. Smith, and F. P. Richards, 1987: Hurricane climatology for the Atlantic and Gulf coasts of the United States. NOAA Tech Memo NWS 38, NWS Silver Spring, MD.

Hock, T. R., and J. L. Franklin, 1999: The NCAR GPS drop windsonde. Bull. Amer. Meteor. Soc., 80, 407–420.

Holland, G. J., 1980: An analytic model of the wind and pressure profiles in hurricanes. Mon. Wea. Rev., 108, 1212-1218.

Homer, C. C. Huang, L. Yang, B. Wylie and M. Coan. 2004. Development of a 2001 National Landcover Database for the United States. Photogrammetric Engineering and Remote Sensing, Vol. 70, No. 7, July 2004, pp. 829-840.

Houston, S. H., and M. D. Powell, 2003: Reconstruction of Significant Hurricanes affecting Florida Bay: The Great 1935 Hurricane and Hurricane Donna (1960). J. Coastal Research, 19, 503-513.

Jarvinen, B. R., C. J. Neumann, and M. A. S. Davis, 1984: A tropical cyclone data tape for the North Atlantic basin, 1886-1963: Contents, Limitations, and Uses. NOAA Tech. Memo NWS NHC 22, National Hurricane Center, 22 pp.

Kanamitsu, M., W. Ebisuzaki, J. Woollen, S-K Yang, J.J. Hnilo, M. Fiorino, and G. L. Potter, 2002: NCEP-DEO AMIP-II Reanalysis (R-2): Bull. Amer. Meteor. Soc., 83, 1631-1643.

Kaplan, J. and M. DeMaria, 1995: A simple empirical model for predicting the decay of tropical cyclone winds after landfall. J. Appl.. Meteor., 34, 2499-2512.

Kurihara, Y. M., M. A. Bender, R. E. Tuleya, and R. J. Ross, 1995: Improvements in the GFDL hurricane prediction system. Mon. Wea. Rev., 123, 2791-2801.

FPHLM V3.0 2008 39

Landsea, C. W., R. A. Pielke, Jr, A. M. Mestas-Nunez, and J. A. Knaff, 1999: Atlantic basin hurricanes: Indices of climatic changes, Climatic Change, 42, 89-129.

Landsea, C. W., 2004: The Atlantic hurricane database re-analysis project- documentation for 1850-1910 alterations and additions to the HURDAT database., Chapter in Hurricanes and Typhoons: Past, present, and future, R. Murnane and K. Liu, Eds., Columbia University Press, 178-221.

Large, W. G. and S. Pond, 1981: Open ocean momentum flux measurements in moderate to strong winds. J. Phys. Ocean., 11, 324-336.

Masters, F. J., 2004: Measurement, modeling and simulation of ground-level tropical cyclone winds. Ph.D. Dissertation, University of Florida.

Merrill, R. T. 1988. Environmental Influences on Hurricane Intensification. J. Atmos. Sci.. 45, 1678-1687.

Miller, B. I., 1964: A study on the filling of Hurricane Donna (1960) over land. Mon. Wea. Rev., 92, 389-406.

Moss, M. S. and S. L. Rosenthal, 1975: On the estimation of planetary boundary layer variables in mature hurricanes. Mon. Wea. Rev., 106, 841-849.

Neumann, C. J., B. R. Jarvinen, C. J. McAdie, and G. R. Hammer, 1999: Tropical Cyclones of the North Atlantic Ocean,1871-1998, National Oceanic and Atmospheric Administration, 206 pp.

Ooyama, K. V., 1969: Numerical simulation of the life cycle of tropical cyclones. J. Atmos. Sci., 26, 3-40.

Paulsen, B, M., J. L. Schroeder, M. R. Conder, and J. R. Howard, 2003: Further examination of hurricane gust factors. Preprints, 11th International Conf. on Wind Engineering,, Lubbock, TX, Texas Tech University, 2005-2012.

Pennington, J., M. DeMaria, and K. Williams, cited 2000: Development of a 10-year Atlantic basin tropical cyclone wind structure climatology. (Available online at www.bbsr.edu/rpi/research/demaria/demaria4.html)

Peterson, E. W., 1969: Modification of mean flow and turbulent energy by a change in surface roughness under conditions of neutral stability. Quart. J. Roy. Meteor. Soc., 95, 561-575.

Powell, M.D., 1980: Evaluations of diagnostic marine boundary layer models applied to hurricanes. Mon. Wea. Rev., 108, 757-766.

FPHLM V3.0 2008 40 Powell, M. D., 1982: The transition of the Hurricane Frederic boundary layer wind field from the open Gulf of Mexico to landfall. Mon. Wea. Rev., 110, 1912-1932.

Powell, M. D., 1987: Changes in the low-level kinematic and thermodynamic structure of Hurricane Alicia (1983) at landfall. Mon. Wea. Rev., 115 (1), 75-99.

Powell, M. D., P. P. Dodge, and M. L. Black, 1991: The landfall of Hurricane Hugo in the Carolinas. Wea. Forecasting, 6, 379-399.

Powell, M. D., S. H. Houston, and Ares, I. 1995: Real-time Damage Assessment in Hurricanes. 21st AMS Conference on Hurricanes and Tropical Meteorology, Miami, FL; April 24-28, 1995; Paper 12A.4, pp. 500-502.

Powell, M. D., and S. H. Houston, 1998: Surface wind fields of 1995 Hurricanes Erin, Opal, Luis, Marilyn, and Roxanne at landfall. Mon Wea. Rev., 126, 1259-1273.

Powell, M. D., S. H. Houston, and T. Reinhold, 1996: Hurricane Andrew's landfall in south Florida. Part I: Standardizing measurements for documentation of surface wind fields. Wea. Forecasting, 11, 304-328.

Powell, M. D., and S. H. Houston, 1996: Hurricane Andrew's Landfall in South Florida. Part II: Surface Wind Fields and Potential Real-time Applications. Wea. Forecasting, 11, 329-349.

Powell, M. D., S. H. Houston, L. R. Amat, and N Morisseau-Leroy, 1998: The HRD real-time hurricane wind analysis system. J.Wind Engineer. and Indust. Aerodyn. 77&78, 53-64.

Powell, M. D.,T. A. Reinhold, and R. D. Marshall, 1999: GPS sonde insights on boundary layer wind structure in hurricanes. Wind Engineering into the 21st Century, Larsen, Larose, and Livesey (eds), Balkema, Rotterdam, ISBN 90 5809 059 0.

Powell, M. D., and S. D. Aberson, 2001: Accuracy of United States tropical cyclone landfall forecasts in the Atlantic basin 1976-2000. Bull. Amer. Meteor. Soc., 82, 2749-2767.

Powell, M. D., P. J. Vickery, and T. Reinhold, 2003: Reduced drag coefficient for high wind speeds in tropical cyclones. Nature, 422, 279-283.

Powell, M.D., D. Bowman, D. Gilhousen, S. Murillo, N. Carrasco, and R. St. Fleur, 2004: Tropical Cyclone Winds at Landfall: The ASOS-CMAN Wind Exposure Documentation Project. Bull. Amer. Meteor. Soc., 85, 845-851.

Powell, M. D., G. Soukup, S. Cocke, S. Gulati, N. Morisseau-Leroy, S. Hamid, N. Dorst, and L. Axe, 2005: State of Florida Hurricane Loss Projection Model: Atmospheric Science Component. J.Wind Engineer. and Indust. Aerodyn., 93, 651-674.

Powell, M. D., and T. A. Reinhold, 2007: Tropical cyclone destructive potential by integrated kinetic energy. Bull. Amer. Meteor. Soc., 88, 513-526. [FIU-SCS1].

FPHLM V3.0 2008 41

Reinhold, T. and K. Gurley, 2003: Florida Coastal Monitoring Program. http://www.ce.ufl.edu/~fcmp.

Reynolds, R. W., N. A. Rayner, T. M. Smith, D. C. Stokes and W. Wang, 2002: An improved in situ and satellite SST analysis for climate. J. Climate, 15, 1609-1625.

Rotunno, R. and K. A.. Emanuel, 1987: An air-sea interaction theory for tropical cyclones, Part II: Evolutionary study using a nonhydrostatic axisymmetric numerical model. J. Atmos. Sci., 44, 542-561.

Russell, L. R., 1971: Probability distributions for hurricane effects. Journal of Waterways, Harbors, and Coastal Engineering Division, ASCE, 97, 139-154.

Schmidt, H. P. and T. R. Oke 1990: A model to estimate the source area contributing to turbulent exchange in the surface layer over patchy terrain. Quart. J. Roy. Meteor. Soc., 116, 965-988.

Shapiro, L. 1983: The asymmetric boundary layer flow under a translating hurricane. J. Atmos. Sci., 40, 1984-1998.

Shay, L. K., G. J. Goni, and P. G. Black, 2000: Effects of a warm oceanic feature on Hurricane Opal. Mon. Wea. Rev., 125(5),1366-1383.

Simiu, E., and R. H. Scanlan, 1996: Wind effects on structures: Fundamentals and applications to design. John Wiley and Sons, NY, NY.

Smith, E., “Atlantic and East Coast Hurricanes 1900–98: A Frequency and Intensity Study for the Twenty-first Century”, Bull. Amer. Meteor. Soc., 1999, 18(12), pp. 2717-2720.

Thompson, E. F., and V. J. Cardone, 1996: Practical modeling of hurricane surface wind fields, Journal of Waterways, Port, Coastal, and Ocean Engineering Division, ASCE, 122, 195-205.

Tuleya, R. E., M. A. Bender, and Y. Kurihara, 1984: A simulation study of the landfall of tropical cyclones using a movable nested-mesh model. Mon. Wea. Rev., 112, 124-136.

Uhlhorn, E. and P. G. Black, 2003: Verification of remotely sensed sea surface winds in hurricanes. J. Atmos. and Ocean Tech., 20, 99-116.

Uhlhorn, E. W., P. G. Black, J. L. Franklin, M. Goodberlet, J. Carswell, and A. S. Goldstein, 2006: Hurricane surface wind measurements from an operational stepped frequency microwave radiometer, Mon. Wea. Rev., 135, 3070-3085.

Vickery, P. J., and L. A. Twisdale, 1995: Wind field and filling models for hurricane wind speed predictions. J.Structural Engineering, 121, 1700-1709.

FPHLM V3.0 2008 42 Vickery, P. J., P. F. Skerjl, A. C. Steckley, and L. A. Twisdale, 2000a: A hurricane wind field model for use in simulations. J Structural Engineering, 126, 1203-1222.

Vickery, P. J., P. F. Skerjl,, and L. A. Twisdale, 2000b: Simulation of hurricane risk in the United States using an empirical storm track modeling technique, J. Structural Engineering, 126, 1222-1237.

Vickery, P. J. and P. F. Skerlj, 2000: Elimination of exposure D along the hurricane coastline in ASCE 7. J. Structural Engineering, 126, 545-549.

Vickery, P. J. and P. F. Skerlj, 2005: Hurricane gust factors revisited. J. Structural Engineering, 131, 825-832.

Vickery, P. J., 2005: Simple empirical models for estimating the increase in the central pressure of tropical cyclones after landfall along the coastline of the United States, J. Appl. Meteor., 44, 1807-1826.

Vogelmann, J. E., S. M. Howard, L. Yang, C. R. Larson, B. K. Wylie, N. Van Driel, 2001. Completion of the 1990s National Land Cover Data Set for the Conterminous United States from Landsat Thematic Mapper Data and Ancillary Data Sources, Photogrammetric Engineering and Remote Sensing, 67, 650-652.

Vukovich, F. M. 2005: Climatology of ocean features in the Gulf of Mexico. MMS 2005-031. U. S. Dept. Interior.

Wada, A. and N. Usui, 2007: Importance of tropical cyclone intensity and intensification in the Western North Pacific, J. Phys. Ocean., 63, 427-447.

Walsh, E. J., C. W. Wright, D. Vandemark, W. B. Krabill, A. W. Garcia, S. H. Houston, S. T. Murillo, M. D. Powell, P. G. Black, and F. D Marks, Jr., 2002: Hurricane directional wave spectrum spatial variation at landfall. J. Phys. Ocean., 32, 1667-1684.

Willoughby, H. E., and M. D. Shoreibah, 1982: Concentric eyewalls, secondary wind maxima, and the evolution of the hurricane vortex. J. Atmos. Sci., 39, 395-411.

Willoughby, H.E., 1998: Tropical cyclone eye thermodynamics, Mon. Wea. Rev., 126, 3053- 3067.

Willoughby, H. E., and M. E. Rahn, 2004: Parametric Representation of the Primary Hurricane Vortex. Part I: Observations and Evaluation of the Holland (1980) Model, Mon. Wea. Rev., 132, 3033-3048.

Xue, M., K. K. Droegemeier, and V. Wong, 2000, “The Advanced Regional Prediction System (ARPS) - A Multiscale Nonhydrostatic Atmospheric Simulation and Prediction Model,” Meteor. Atmos. Phys., 75, 161-193.

FPHLM V3.0 2008 43

Vulnerability Standards ACI 530-99/ASCE 5-99/TMS 402-99. (1999). Building Code Requirements for Masonry Structures, American Concrete Institute.

Allen, E. (1999). Fundamentals of Building Constructions: Materials and Methods. Third Edition, Wiley.

American Wood Council. (1997). “National Design Specification for Wood Construction: Allowable Stress Design (ASD) Manual for Engineered Wood Construction”

Amirkhanian, S., Sparks, P. and Watford, S. (1994). “Statistical analysis of wind damage to single family dwellings due to Hurricane Hugo,” Structures Congress.

Ang, A. and. Tang, W. (1975). Probability Concepts in Engineering Planning and Design, John Wiley & Sons.

Aponte, L., Gurley, K., Prevatt, D. and Reinhold, T. (2007). “Uncertainties in the measurement and analysis of full-scale hurricane wind pressures on low-rise structures,” 12th International Conference on Wind Engineering.

ASCE 7-98 Standard. “Minimum Design Loads for Buildings and Other Structures,” American Society of Civil Engineers..

Baskaran, A. (1999). “Discussion on ‘Wind Uplift Model for Asphalt Shingles’," Journal of Architectural Engineering, 5(2), 67-69.

Baskaran, A., and Dutt, O. (1995). “Evaluation of roof fasteners under dynamic loading,” 9th International Conference on Wind Engineering.

Berke, P., Larsen, T. and Ruch, C. (1984). “Computer system for hurricane hazard assessment,” Computers, Environment and Urban Systems.

Beste, F., and Cermak, J. (1997). “Correlation of internal and area-averaged external wind pressures on low-rise buildings,” Journal of Wind Engineering and Industrial Aerodynamics, v 69-71, 557-566.

Bhinderwala, S. (1995). “Insurance loss analysis of single family dwellings damaged in Hurricane Andrew,” MS Thesis, Department of Civil Engineering, Clemson University.

Boswell, M.R., Deyle, R.E., Smith, R.A., and Baker, E.J. (1999). “Quantitative method for estimating probable public costs of hurricanes,” Environmental Management.

Canfield, L., Niu, S. and Liu, H. (1991). “Uplift resistance of various rafter-wall connections,” Forest Products Journal, 41(7/8), 27-34.

FPHLM V3.0 2008 44 Chandler, A.. Jones, E., Patel, M. (2001). “Property loss estimation for wind and earthquake perils,” Risk Analysis, v 21(2), 235-249.

Conner, H., Gromala D., and Burgess D. (1987). “Roof connections in house; Key to wind resistance,” Journal of Structural Engineering, 113(12), 2459-2474.

Cook R. editor. (1993). “Hurricanes of 1992: Lessons Learned and Implications for the Future,” ASCE.

Cope A. (2004). “Predicting the vulnerability of typical residential buildings to hurricane damage,” Ph.D. Dissertation, Department of Civil Engineering, University of Florida.

Cope, A., and Gurley, K. (2001). “Spatial characteristics of pressure coefficients on low rise gable roof structures,” America’s Conference on Wind Engineering.

Cope, A., Gurley, K., Filliben, J., Simiu, E., Pinelli, J.P., Subramanian, C., Zhang, E.L., and Hamid, S. (2003). "A hurricane damage prediction model for residential structures," 9th International Conference on Applications of Statistics and Probability in Civil Engineering.

Cope, A., Gurley, K., Gioffre, M., and Reinhold, T. (2005). “Low-rise gable roof wind loads: characterization and stochastic simulation,” Journal of Wind Engineering and Industrial Aerodynamics, 93(9), 719-738.

Cope, A., Gurley, K., Pinelli, J.P. and Hamid, S. (2003). “A simulation model for wind damage predictions in Florida,” 11th International Conference on Wind Engineering.

Cope, A., Gurley, K., Pinelli, J.P., Murphree, J., Subramanian, C., Gulati, S. and Hamid, S. (2004). “A Probabilistic Model of Damage to Residential Structures from Hurricane Winds,” ASCE joint specialty conference on probabilistic mechanics and structural reliability.

Cunningham, T. (1993). “Roof sheathing fastening schedules for wind uplift,” APA Report T92- 28, American Plywood Association.

DASMA. (2002). “DASMA Garage Door and Commercial Door Wind Load Guide, Technical Data Sheet #155b,” Door & Access Systems Manufacturer’s Association International.

Datin, P., Liu, Z., Prevatt, D., Masters, F., Gurley, K. and Reinhold, T. (2006) “Wind loads on single-family dwellings in suburban terrain – comparing field data with wind tunnel studies”, ASCE Structures Congress.

Dawe, J. and Aridru, G. (1993). “Prestressed concrete masonry walls subjected to uniform out- of-plane loading,” Canadian Journal of Civil Engineering, v 20, 969-979.

Devlin, P. (1996). “Wind resistance of roof coverings,” Natural Hazard Mitigation Insights, Insurance Institute for Property Loss Reduction, v 4.

FPHLM V3.0 2008 45

Dyrbye, C. and Hansen, S. (1997). Wind Loads on Structures, John Wiley & Sons.

Ellingwood, B., Rosowsky, D., Li, Y., Kim, J. (2004). “Fragility assessment of light-frame wood construction subjected to wind and earthquake hazards,” Journal of Structural Engineering, 130(12), 1921-1930.

Federal Emergency Management Agency. (1992). “Building performance: Hurricane Andrew in Florida observations, recommendations, and technical guidance”, Federal Insurance Administration Coastal Archives call # G 409.52 F4 B8.

Federal Emergency Management Agency. (1993). “Building performance: Hurricane Andrew in Florida,” report FIA-22.

Federal Emergency Management Agency. (2005). “Hurricanes’ impact on Florida’s Building Codes & Standards,” Mitigation Section.

Federal Emergency Management Agency. (2003). “Multi-hazard loss estimation methodology hurricane model,” HAZUS®MH Technical Manual.

Federal Emergency Management Agency. “(2005). Home Builder’s Guide to Coastal Construction,” Technical Fact Sheet Series Nos. 1-31.

Florida Building Code, Tallahassee, FL, 2001.

FM Global Technologies, (2002). “Approval standard for class 1 roof covers”

Foliente G., Kasal, B., Paevere, P., Macindoe L., Banks R., Mike S., Seath C., and Collins, M. (2000). “Whole structure testing and analysis of a light frame wood building, phase 1 – test house details and preliminary results”, NAHB Research Center.

Ginger, J., and Letchford, C. (1995). “Pressure factors for edge regions on low rise building roofs,” Journal of Wind Engineering and Industrial Aerodynamics, v 54/55, 337-344.

Ginger, J., and Letchford, C. (1999). “Net pressures on a low-rise full-scale building,” Journal of Wind Engineering and Industrial Aerodynamics, v 83, 239-250.

Gioffre, M., Gurley, K., and Cope, A.(2002). “Stochastic simulation of correlated wind pressure fields on low-rise gable roof structures,” 15th ASCE Engineering Mechanics Conference.

Gioffre, M., Gusella, V. and Grigoriu, M. (2000). “Simulation of non-Gaussian field applied to wind pressure fluctuations,” Probabilistic Engineering Mechanics, v 15, 339-345.

Gurley, K., Cope, A., Pinelli, J.P., and Hamid, S. (2003). "A simulation model for wind damage predictions in Florida," 11th International Conference in Wind Engineering.

FPHLM V3.0 2008 46 Gurley, K., Davis, R., Ferrera, S-P., Burton, J., Masters, F., Reinhold, T. and Abdullah, M. (2006). “Post 2004 hurricane field survey – an evaluation of the relative performance of the Standard Building Code and the Florida Building Code,” ASCE Structures Congress.

Hajj, M., Jordan, D., and Tieleman, H. (1998). “Analysis of atmospheric wind and pressures on a low-rise building,” Journal of Fluids and Structures, v 12, 537-547.

Harris, R. (1990). “The propagation of internal pressures in buildings,” Journal of Wind Engineering and Industrial Aerodynamics, v 34, 169-184.

Ho, T., Davenport, A., and Surry, D. (1995). “Characteristic pressure distribution shapes and load repetitions for the wind loading of low building roof panels,” Journal of Wind Engineering and Industrial Aerodynamics, v 57, 261-279.

Holmes, J. (1996). “Vulnerability curves for buildings in tropical cyclone regions,” Probabilistic Mechanics and Structural Reliability: Proceedings of the 7th Specialty Conference, 78-81.

Hosoya, N. Cermak, J., and Dodge, S. (1999). “Area-averaged pressure fluctuations on surfaces at roof corners and gable peaks,” Wind Engineering in the 21st Century, Larsen, Larose, and Livesey (eds), Bakkema, Rotterdam.

Huang Z. (1999). “Stochastic models for hurricane hazard analysis,” PhD Dissertation. Department of Civil Engineering, Clemson University.

Huang, Z., Rosowsky D., and Sparks, P. (1999). “Event-based hurricane simulation for the evaluation of wind speeds and expected insurance loss,” Wind Engineering into the 21st Century, Larsen, Larose, and Livesey (eds), Bakkema, Rotterdam...

Huang, Z., Rosowsky, D., ,and Sparks, P. (2001). “Long-term hurricane risk assessment and expected damage to residential structures,” Reliability Engineering and System Safety, v 74, 239- 249.

Huang, Z; Rosowsky, D., and Sparks, P. (2001). “Hurricane simulation techniques for the evaluation of wind-seeds and expected insurance losses,” Journal of Wind Engineering and Industrial Aerodynamics, v 89, 605-617.

Institute for Business and Home Safety, (2000). “Industry Perspective: Impact Resistance Standards,” Natural Hazard Mitigation Insights, v 12.

Insurance Information Institute, (2001). “Catastrophes: Insurance Issues”, Insurance Issues Update.

Jordan, D., Hajj, M., Miksad, R., and Tieleman, H. (1999). “Analysis of the velocity-pressure peak relation for wind loads in structures,” 10th International Conference on Wind Engineering, 443-448.

FPHLM V3.0 2008 47 Kareem, A. (1985). “Structural performance and wind speed-damage correlation in Hurricane Alicia,” Journal of Structural Engineering, 111(12), 2596-2610.

Kareem, A. (1986). “Performance of cladding in Hurricane Alicia”, Journal of Structural Engineering, 112(12), 2679-2693.

Kareem, A. (1987). “Wind effects on structures: a probabilistic viewpoint,” Probabilistic Engineering Mechanics, 2(4), 166-200.

Kasperski, M. (1996). “Design wind loads for low-rise buildings: a critical review of wind load specifications for industrial buildings,” Journal of Wind Engineering and Industrial Aerodynamics, v 61, 169-179.

Khanduri, A., and Morrow, G. (2003). “Vulnerability of buildings to windstorms and insurance loss estimation,” Journal of Wind Engineering and Industrial Aerodynamics, 91, 455-467.

Kleindorfer, P., Kunreuther, H. (1999). “The complementary roles of mitigation and insurance in managing catastrophic risks,” Risk Analysis, 19(4), 727-738.

Kumar, S., and Stathopoulos, T. (1998). “Power spectra of wind pressures on low building roofs,” Journal of Wind Engineering and Industrial Aerodynamics, v 74-46, 665-674.

Landsea, C., Pielke, Jr. R., Mestas-Nunez, A. and Knaff, J. (1999). “Atlantic basin hurricanes: indices of climatic changes,” Climatic Change, 42, 89-129.

Langedyk, R., Ticola, V. (2002). “CEIA Cost 2002,” Construction Estimating Institute, Sarasota, FL

Lavelle, F., Vickery, P. , Schauer, B., Twisdale, L., Laatsch, E. (2003). “The HAZUS-MH hurricane model,” 11th International Conference on Wind Engineering.

Liu, Z., Dearhart, E., Prevatt, D., Reinhold, T. and Gurley, K. (2005). “Wind load on components and cladding systems for houses in coastal suburban areas,” 10th Americas Conference on Wind Engineering.

Liu, Z., Prevatt, D., Gurley, K. and Reinhold, T. (2007). “Validating wind tunnel technique using full scale wind pressure data”, 12th International Conference on Wind Engineering (12 ICWE.

Mahendran, M., (1995). “Wind resistant low-rise buildings in the tropics,” Journal of Performance of Constructed Facilities, ASCE.

Marshall, R. (1977). “The measurement of wind loads on a full-scale mobile home,” National Bureau of Standards.

FPHLM V3.0 2008 48 Marshall, R. (1993). “Wind load provisions of the manufactured home construction and safety standards -- A review and recommendations for improvement, NIST IR 5189, Building and Fire Research Laboratory.

Marshall, R. (1994). “Manufactured homes – probability of failure and the Nneed for better windstorm protection through improved anchoring systems,” NIST IR 5370, Building and Fire Research Laboratory.

Marshall R. and Yokel, F. (1995). “Recommended Performance-Based Criteria for the Design of Manufactured Home Foundation Systems to Resist Wind and Seismic Loads,” NISTIR 5664, NIST

Marwood, R. and Wood, C. (1997). “Conical vortex movement and its effect on roof pressures,” Journal of Wind Engineering and Industrial Aerodynamics, v 69-71, 589-595.

Maruta, E., Kanda, M. and Sato, J. (1998). “Effects on surface roughness for wind pressure on glass and cladding of buildings,” Journal of Wind Engineering and Industrial Aerodynamics, v 74-76, 651-663.

Testimony of Dr. Steven L. McCabe on Behalf of the American Society of Civil Engineers Before the Subcommittee on Environment, Technology and Standards Of The Committee on Science, U.S. House of Representatives , October 11, 2001.

Minor, J., and Schneider, P. (2001). “Hurricane loss estimation – The HAZUS preview model,” 1st America’s Conference on Wind Engineering.

Mitsuta, Y., Fujii, T., and Nagashima, I. (1996), “A predicting method of typhoon wind damages,” Probabilistic Mechanics and Structural Reliability: Proceedings of the 7th Specialty Conference, 970-973.

Mizzell, D.P. (1994). “Wind Resistance of Sheathing for Residential Roofs,” MS Thesis, Department of Civil Engineering, Clemson University.

Munson, B., Young, D., and Okiishi, T. (1990). Fundamentals of Fluid Mechanics, John Wiley & Sons.

Murphree J. (2004). “Florida public hurricane loss projection model: development calibration and validation of vulnerability matrices,” MS Thesis, Department of Civil Engineering, Florida Institute of Technology.

NAHB Research Center, (1993). “Assessment of Damage to Single-Family Homes Caused by Hurricanes Andrew and Iniki,” Prepared for U.S. Department of Housing and Urban Development, Office of Policy Development and Research.

NAHB Research Center, (1996). “Assessment of Damage to Homes caused by Hurricane Opal,” Prepared for Florida State Home Builders Association.

FPHLM V3.0 2008 49

NAHB Research Center, (1998). “Factory and site built housing, a comparison for the 21st century,” Prepared for U.S. Department of Housing and Urban Development.

NAHB Research Center, (1999). “Reliability of conventional residential construction: an assessment of roof component performance in Hurricane Andrew and typical wind regions of the United States,” U.S. Department of Housing and Urban Development and National Association of Home Builders.

Owens Corning, 2001. “Certificate of conformance, owens corning select vinyl siding,” Owens Corning.

Pearson, J., Longinow, A., and Donald, F. (1996). “Wind protection tie- downs for manufactured homes,” Practice Periodical on Structural Design and Construction, 1(4).

Peterka, J., Cermak, J,. Cochran, L., Cochran, B., Hosoya, N., Derickson, R., Harper, C. Jones,N. and Metz, B. (1997). “Wind uplift model for asphalt shingles”, Journal of Architectural Engineering, 3(4), 147-155.

Peterka, J., Hosoya, N., Dodge, S., Cochran, L., and Cermak, J. (1998). “Area average peak pressures in a gable roof vortex region,” Journal of Wind Engineering and Industrial Aerodynamics, v77-78, 205-215.

Pettit, C., Jones, N., and Ghanem, R. (1999). “Detection, analysis and simulation of roof-corner pressure transients,” 10th International Conference on Wind Engineering, 1831-1838.

Phang, M. K., (1999). “Wind damage investigation of low rise buildings,” Proceedings of the 1999 Structures Congress, ASCE.

Pielke, Jr., R., and Landsea, C. (1998). "Normalized Atlantic hurricane damage, 1925-1995," Weather Forecasting, v 13, 621-631.

Pinelli, J.P., Gurley, K., Subramanian, C., Hamid, S., and Pita, G. (2007). “Validation of a probabilistic model for hurricane insurance loss projections in Florida ,” Journal of Reliability Engineering and System Safety, accepted for publication.

Pinelli, J.P., Murphree, J., Subramanian, C., Zhang, L.. Gurley, K., Cope, A., Hamid, S., and Gulati, S. (2004). "Hurricane loss estimation: model development, results and validation," Joint International Conference on Probabilistic Safety Assessment and Management and the European Safety and Reliability Conference.

Pinelli, J.P., and O’Neill, S. (2000). “Effect of tornadoes on residential masonry structures,” Wind and Structures Journal, 3(1).

FPHLM V3.0 2008 50 Pinelli, J.P., Simiu, E., Gurley, K., Subramanian, C.. Zhang, L.. Cope, A. Filliben, J., and Hamid, S. (2004). "Hurricane damage prediction model for residential structures," Journal of Structural Engineering (ASCE), 130(11), 1685-1691.

Pinelli, J.P., Subramanian, C., Garcia, F. and Gurley, K. (2007). “A study of hurricane mitigation cost effectiveness in Florida,” European Safety and Reliability Conference, (ESREL ‘07).

Pinelli, J.P., Subramanian, C., Artiles, A., Gurley, K. and Hamid, S. (2006). “Validation of a probabilistic model for hurricane insurance loss projections in Florida,” European Safety and Reliability Conference (ESREL ’06).

Pinelli, J.P., Subramanian, C., Gurley, K. and Hamid, S. (2007). “Validation of the Florida public hurricane loss model,” 12th International Conference on Wind Engineering (12 ICWE).

Pinelli, J.P., Subramanian, C., Murphree, J., Gurley, K., Cope, A., Gulati, S., Simiu, E., and Hamid, S. (2005), “Hurricane loss prediction: model development, results, and validation,” ICOSSAR.

Pinelli, J.-P., Subramanian, C., Murphee, J., Gurley, K.. Hamid, S., and Gulati, S. (2005). "Florida public hurricane loss projection vulnerability model," 10th American Conference on Wind Engineering.

Pinelli, J-P., Subramanian, C., Zhang, L., Gurley, K., Cope, A., Simiu, E., Filliben, J., Diniz, S., and Hamid, S. (2003). "A model to predict hurricane damage for residential structures," 11th International Conference on Wind Engineering.

Pinelli, J.P., Zhang, L., Subramanian, C., Cope, A., Gurley, K., Gulati, S. and Hamid, S., (2003). “Classification of structural models for wind damage predictions in Florida,” 11th International Conference on Wind Engineering.

Porter, K., Scawthorn, C.; Beck, J. (2006). “Cost-effectiveness of stronger woodframe buildings,” Earthquake Spectra, v 22(1), 239 – 266.

Powell, M., Houston, S., and Reinhold, T. (1996). “Hurricane Andrew’s landfall in south Florida. part I: standardizing measurements for documentation of surface wind fields,” Weather Forecasting, v 11, 304-328.

Powell, M., Soukup, G., Cocke, S., Gulati, S., Morisseau-Leroy, N., Hamid, S., Dorst, N., and Axe, L., (2005). "State of Florida hurricane loss projection model: atmospheric science component," Journal of Wind Engineering and Industrial Aerodynamics, v 93, 651-674.

Reed, T., Rosowsky, D., and Schiff, S. (1996). “Structural analysis of light-framed wood roof construction”, Wind Load Test Facility Report for Blue Sky, PBS-9606-02.

Reed, T., Rosowksy, D., and Schiff, S. (1997). “Uplift capacity of light-frame rafter to top plate connections,” Journal of Architectural Engineering, 3(4), 156-163.

FPHLM V3.0 2008 51

Reinhold, T.A. (2002). “13 Homes destroyed,” Disaster Review, ,9-14.

Reinhold, T., Dearhart, A., Gurley, K., and Prevatt, D. (2005). “Wind loads on low-rise buildings: is one set of pressure coefficients sufficient for all types of terrrain?” The Second International Symposium on Wind Effects on Buildings and Urban Environment, Tokyo Polytechnic University, Japan.

Reinhold, T., Gurley, K., Masters, F., and Burton, J. (2005). “US hurricanes of 2004: A clear demonstration that improvements in building codes, enforcement and construction are reducing structural damage”, 6th Asia Pacific Conference on Wind Engineering.

Robertson, A.P. (1992). “The wind-Induced response of a full-scale portal framed building,” Journal of Wind Engineering and Industrial Aerodynamics, v41, 1677-1688.

Rosowsky, D., and Cheng N. (1998). “Reliability of a light frame roof systems subjected to wind uplift”, NAHB Research Center and the National Association of Home Builders, Clemson University,

Rosowsky, D., and Reinhold, T. (1999). “Rate-of-load and duration-of-load effects for wood fasteners”, ASCE Journal of Structural Engineering, 125(7), 719-724.

Rosowsky, D., and Schiff, S. (1999). “Combined loads on sheathing to framing fasteners in wood construction,” Journal of Architectural Engineering, 5(2), 37-43.

Rosowsky, D., Schiff, S., Reinhold, T., Sparks, P. and Sill, B. (1998). “Performance of light- frame wood structures under high wind loads: Experimental and analytical program”, Wind Performance and Safety of Wood Buildings, FPS Specialty Publication.

Rigato, A., Chang, P., and Simiu, E. (2001). “Database-assisted design, standardization, and wind direction effects,” Journal of Structural Engineering, 127(8), 855-860.

Russell, J. (2004). National Renovation & Insurance Repair Estimator, Craftsman Book Company, Carlsbad, CA.

Sadek, F. and Simiu, E. (2002). “Peak non-gaussian wind effects for database-assisted low rise building design,” Journal of Engineering Mechanics, 128(5), 530-539.

Sciaudone, J., Freuerborn, D., Rao, G., and Daneshvaran, S. (1997). “Development of objective wind damage functions to predict wind damage to low-rise structures,” Eighth U.S. National Conference on Wind Engineeringg.

Sharma, R.N. and Richards, P.J. (1997). “The effect of roof flexibility on internal pressure fluctuations,” Journal of Wind Engineering and Industrial Aerodynamics, vol 72, 175-186.

FPHLM V3.0 2008 52 Sill, B.L. and R.T. Kozlowski (1997), “Analysis of storm damage factors for low-rise structures”, Journal of Performance of Constructed Facilities, 11(4), 168-176.

Sill, B. L. and Sparks, P. R. editors. (1990). “Hurricane Hugo one year later,” Proceedings of the Symposium and Public Forum, ASCE.

Simiu, E. and Cordes, M.E. (1980). “Probabilistic assessment of tornado-borne missile speeds,” NBS Interagency Report 80-2117, NBS; Washington, DC .

Simiu, E. and Cordes, M. R. (1983). "Tornado-borne missile probabilities," ASCE Journal of Structural Engineering; v.1, 157-168.

Simiu, E. and Scanlan, R. (1996). Wind Effects on Structures, Fundamentals and Applications to Design, Third Edition, John Wiley & Sons, New York, NY.

Simmons, K., Kruse, J. (2002). “Does a market of mitigation exist?” Disaster Safety Review, 3, 7-8.

Simmons, K., Willner, J. (2001). “Hurricane mitigation: rational choice or market failure,” Atlantic Economic Journal, 12, Page 470.

Simpson Strongtie. (2003), “Connectors for factory built homes, Technical Bulletin T-FBS02,” http://www.strongtie.com/ftp/bulletins/T-FBS02.pdf

Sparks, P. (1991). “Damages and lessons learned from hurricane Hugo,” Proceedings of the 23rd Joint Meeting of the US-Japan Cooperative Program in Natural Resources Panel on Wind and Seismic Effects.

Sparks, p., and Schiff, P. (1994). “Wind damage to the envelopes of houses and consequent insurance losses,” J. of Wind Eng. And Industrial Aerodynamics, 53,145–155.

Stewart, M.G. (2003). Cyclone damage and temporal changes to building vulnerability and economic risks for residential construction,” Journal of Wind Engineering and Industrial Aerodynamics, v 91, n 5, 671-691.

Stewart, M.G., Rosowsky, D., and Huang, Z. (2003). “Hurricane risks and economic viability of strengthened construction subjected to wind and earthquake hazards,” Natural Hazard Review, February, 2003: 12-19.

Stricklin, D.L. Jr. (1996). “Investigation of light-framed wood wall systems under wind uplift loads,” MS Thesis, Department of Civil Engineering, Clemson University.

Topics – Annual Review: Natural Catastrophes 2001 (2002). Munich Re Group, Muenchener Rueckversicherungs-Gesellschaft, D-80791 Munich.

FPHLM V3.0 2008 53 Uematsu, Y., and Isyumov, N. (1999). “Wind pressures acting on low-rise buildings,” Journal of Wind Engineering and Industrial Aerodynamics, vol 82, 1-25.

Vickery, B.J. (1986). “Gust-factors for internal-pressures in low rise buildings,” Journal of Wind Engineering and Industrial Aerodynamics, vol 23, 259-271.

Vickery, B.J. (1994). “Internal pressures and interactions with the building envelope,” Journal of Wind Engineering and Industrial Aerodynamics, vol 53, 125-144.

Vickery, B.J., and Georgiou, P.N. (1991). “A simplified approach to the determination of the influence of internal pressures on the dynamics of large span roofs,” Journal of Wind Engineering and Industrial Aerodynamics, vol 38, 357-369.

Vickery, P.J., Lavelle, F.M., Drury, C., and Schauer, B.A. (2003). “FEMA’s HAZUS hurricane model,” Proceedings of the 11th International Conference on Wind Engineering.

Walpole, R., Myers, R., and Myers, S. (1997). Probability and Statistics for Engineers and Scientists, Sixth edition, Prentice Hall.

Wills, J.A.B., Lee, B. E., and Wyatt, T.A. (2002). “A model of wind-borne debris damage,” Journal of Wind Engineering and Industrial Aerodynamics, vol. 90, 555-565.

Yancey, C.W., Cheok, G.S., Sadek, F., and Mohraz, B. (1988). "A summary of the structural performance of single-family, wood-frame housing," NISTIR 6224, National Institute of Standards and Technology.

Yokel, F., Chung, R., Rankin, F., and Yancey, C. (1982). "Load-displacement characteristics of shallow soil anchors," NBS Building Science Series 142, National Bureau of Standards, Washington D.C.

Young, M. A. (1997). “Effect of open fields on low building wind loads in a suburban environment,” MS Thesis, Department of Civil Engineering, University of Western Ontario.

Zhang, L., (2003), “Public hurricane loss projection model: exposure and vulnerability components,” MS Thesis, Department of Civil Engineering, Florida Institute of Technology,

Actuarial Standards

Hogg, R. and S. A. Klugman (1984), Loss Distribution, (particularly Ch. 4 and 5 and the appendix), Wiley, New York.

Klugman, S., H. Panjer and G. Willmot (1998), Loss Models, Wiley, New York.

Computer Science Standards

FPHLM V3.0 2008 54

Published CS Journal Papers:

S-C. Chen, S. Gulati, S. Hamid, X. Huang, L. Luo, N. Morisseau-Leroy, M.D. Powell, C. Zhan, and C. Zhang, “A Web-based Distributed System for Hurricane Occurrence Projection,” Software: Practice and Experience, May 2004, 34(6), pp. 549-571.

Published CS Conference Proceedings:

K. Chatterjee, K. Saleem, N. Zhao, M. Chen, S-C. Chen, and S. Hamid, “Modeling Methodology for Component Reuse and System Integration for Hurricane Loss Projection Application,” in Proceedings of The 2006 IEEE International Conference on Information Reuse and Integration (IEEE IRI-2006), September 16-18, 2006, Hawaii, USA, pp. 57-62.

S-C. Chen, S. Gulati, S. Hamid, X. Huang, L. Luo, N. Morisseau-Leroy, M. Powell, C Zhan, and C. Zhang, “A Three-Tier System Architecture Design and Development for Hurricane Occurrence Simulation,” in Proceedings of the IEEE International Conference on Information Technology: Research and Education (ITRE 2003), August 10-13, 2003, Newark, New Jersey, USA, pp. 113-117.

S-C. Chen, S. Hamid, S. Gulati, N. Zhao, M. Chen, C. Zhang, and P. Gupta, "A Reliable Web- based System for Hurricane Analysis and Simulation," in Proceedings of IEEE International Conference on Systems, Man and Cybernetics 2004, October 10-13, 2004, Hague, The Netherlands, pp. 5215-5220.

S.-C. Chen, S. Hamid, S. Gulati, G. Chen, X. Huang, L. Luo, C. Zhan, and C. Zhang, “Information Reuse and System Integration in the Development of a Hurricane Simulation System,” in Proceedings of the 2003 IEEE International Conference on Information Reuse and Integration (IRI'2003), October 27-29, 2003, Las Vegas, Nevada, USA, pp. 535-542.

S.-C. Chen, M. Chen, N. Zhao, S. Hamid, K. Saleem, and K. Chatterjee, "Florida Public Hurricane Loss Model (FPHLM): Research Experience in System Integration," accepted for publication, The 9th Annual International Conference on Digital Government Research, May 18- 21, 2008, Montreal, Canada.

Refereed Books and Book Chapters:

B. Bruegge and A.H. Dutoit, Object-oriented Software Engineering Using UML, Patterns, and Java, Second Edition, 2004.

N. Morisseau-leroy, M.K. Solomon, and J. Basu, Oracle8i: Java Component Programming with EJB, CORBA, and JSP, Oracle Press (McGraw-Hill/Osborne), 2000, pp. 286-307.

FPHLM V3.0 2008 55 D. Needham, R. Caballero, S. Demurjian, F, Eickhoff, J. Mehta, and Y. Zhang, “A Reuse Definition, Assessment, and Analysis Framework for UML”, Advances in UML and XML – Based Software Evolution, 2005, 292-307.

Refereed Journals and Magazines:

B. Boehm and C. Abts, “COTS Integration: Plug and Pray?” IEEE Computer, 2000, 32(1), pp. 135-138.

P. Brereton and D. Budgen, “Component-Based Systems: A Classification of Issues”, IEEE Software, Nov. 2000, 33(11), pp. 54-62.

P. Fraternali, “Tools and Approaches for Developing Data-intensive Web Applications: A Survey”, ACM Computing Survey, Sep. 1999, 31(3), pp. 227-263.

Soil Survey Staff, “State Soil Geographic (STATSGO) Data Users Guide.” , USDA National Resources Conservation Service Miscellaneous Publication 1992, pp. 88-1036.

LR. Russell, “Probability distributions for hurricane effects”, Journal of Waterways, Harbors, and Coastal Engineering Division, ASCE 1971, pp. 139-154.

Refereed Conference Proceedings:

All Industry Research Advisory Council (AIRAC), “Catastrophic Losses: How the Insurance Industry Would Handle Two $7 Billion Hurricanes,” The All-Industry Research Advisory Council (AIRAC), Oak Brook, Illinois, 1986.

X. Cai, M. R. Lyu, and K. Wong, “Component-based Software Engineering: Technologies, Development Frameworks, and Quality Assurance Schemes,” in Proceedings of 7th Asia-Pacific Software Engineering Conference (APSEC 2000), Dec. 2000, Singapore, pp. 372-379.

D. Gornik, “UML Data Modeling Profile,” White Paper, Rational Software. May 2002.

M. W. Price, S. A. Demurjian, Sr., and D. Needham, “Reusability Measurement Framework and tool for Ada95,” in Proceedings of TRI-Ada'97, Nov. 11-14, 1997, St. Louis, Missouri, pp. 125- 132.

M. W. Price and S. A. Demurjian, Sr., “Analyzing and Measuring Reusability in Object-Oriented Design,” in Proceedings of the 12th ACM SIGPLAN Conference on Object-oriented programming, systems, languages, and applications, October 05-09, 1997, Atlanta, Georgia, United States, pp. 22-33.

F. T. Sheldon, K. Jerath, Y.-J. Kwon, and Y.-W. Baik, “Case Study: Implementing a Web Based Auction System Using UML and Component-Based Programming,” in Proceedings of 26th

FPHLM V3.0 2008 56 International Computer Software and Applications Conference (COMPSAC 2002), Aug. 2002, Oxford, England, pp. 211-216.

Y. Zhou, Y. Chen, and H. Lu, “UML-based Systems Integration Modeling Technique for the Design and Development of Intelligent Transportation Management System,” in Proceedings of IEEE International Conference on Systems, Man and Cybernetics, October 10- 13, 2004, The Hague, The Netherlands, pp. 6061-6066.

FPHLM V3.0 2008 57

Statistics Standards

Conover, W. J. (1999). Practical Nonparametric Statistics, John Wiley, NY.

Draper, N.R. and H. Smith (1998). Applied Regression Analysis, John Wiley, New York.

Kibria, B. M. G. (2006). Applications of some discrete regression models for count data. Pakistan Journal of Statistics and Operation Research, 2 (1), 1-16.

Greene, W. H. (2003). Econometric Analysis. Fifth Edition, Prentice Hall, New Jersey.

Iman, R. L., Johnson, M. E. and Schroeder, T. (2000a): Assessing Hurricane Effects. Part 1. Sensitivity Analysis, Reliability Engineering & System Safety, 78, 131-145.

Iman, R. L., Johnson, M. E. and Schroeder, T. (2000b): Assessing Hurricane Effects. Part 2. Uncertainty Analysis, Reliability Engineering & System Safety, 78, 147-155.

Tamhane, A. C. and Dunlop, D. (2000). Statistics and Data Analysis, Prentice Hall, NJ.

Lin, L. I. (1989). A concordance correlation coefficient to evaluate reproducibility. Biometrics, 45, 255-268.

Relevant Web Sites

Applied Insurance Research, Inc. (AIR) page. http://www.airboston.com_public/html/rmansoft.asp

Applied Research Associates, Inc. (ARA) page. http://www.ara.com/risk_and_reliability_analysis.htm

ARIS Reference. http://www.idsscheer.com/international/english/products/aris design platform/50324

CIMOSA Reference. http://cimosa.cnt.pl

EQECAT home page. http://www.eqecat.com/

FEMA hurricanes page. http://www.fema.gov/hazards/hurricanes

Global Ecosystems Database (GED). http://www.ngdc.noaa.gov/seg/fliers/se- 2006.shtml

HAZUS Home. http://www.hazus.org/

FPHLM V3.0 2008 58 HAZUS Overview. http://www.nibs.org/hazusweb/verview/overview.php

Hazus manuals page, http://www.fema.gov/hazus/li_manuals.shtm

HURDAT data. http://www.aoml.noaa.gov/hrd/hurdat/Data Storm.html

IMSL Mathematical & Statistical Libraries. http://www.vni.com/products/imsl

Java Native Interface. http://java.sun.com/docs/books/tutorial/native1.1/

Java Server Pages (TM) Technology. http://java.sun.com/products/jsp/

National Hurricane Center. http://www.nhc.noaa.gov/

NOAA Coastal Services Center http://www.csc.noaa,gov/

NOAA EL Nino Page. http://www.elnino.noaa.gov/

NOAA LA Nina Page. http://www.elnino.noaa.gov/lanina.html

Oracle Reference. http://www.oracle.com/ip/deploy/database/oracle9i/

Oracle9iAS Container for J2EE. http://technet.oracle.com/tech/java/oc4j/content.html

Panda D. Oracle Container for J2EE (OC4J). http://www.onjava.com/pub/a/onjava/2002/01/16/oracle.html

PHRLM Manual. http://www.cis.fiu.edu/hurricaneloss

RAMS: Regional Atmospheric Modeling System. http://rams.atmos.colostate.edu/

R.L. Walko, C.J. Tremback, “RAMS: regional atmospheric modeling system, version 4.3/4.4 - Introdcution to RAMS 4.3/4.4.” http://www.atmet.com/html/docs/rams/ug44-rams-intro.pdf

RMS home page. http://www.rms.com

The JDBC API Universal Data Access for the Enterprise. http://java.sun.com/products/jdbc/overview.html

The Interactive Data Language. http://www.rsinc.com/idl/

Track of hurricane Andrew (1992) (Source from NOVA). http://www.pbs.org/newshour/science/hurricane/facts.html

FPHLM V3.0 2008 59 Tropical cyclone heat potential: http://www.aoml.noaa.gov/phod/cyclone/data/

The Ptolemy Java Applet package. http://ptolemy.eecs.berkeley.edu/papers/99/HMAD/html/plotb.html

FPHLM V3.0 2008 60 5. Provide an itemized description of all changes in the model from the prior year’s submission

Changes to the model include;

1. Updating the probability distribution function for storm track and intensity changes using the most recent HURDAT database.

2. The demand surge factors are slightly different due to small changes in weight factors as a result of the new 50,000 year simulation.

6. Provide an itemized description of all substantive changes (identified by the modeler or the Professional Team) since the current year's initial submission, including all interim changes.

1. Revised description of the model to provide more detail and discuss revised Storm Track Model. Additional references are added or cited: Shay et al. (2000), Demaria et al. (2005) and Wada and Uisi (2007).

2. Modified Storm Track Model:

a. We now take indirectly into account oceanic heat content (as measured, for example, by the tropical cyclone heat potential) via water depth in intensity changes. We do this by not aggregating intensity PDFs in shallow waters with those in deeper waters. Deeper waters can have higher oceanic heat content, and may lead to more rapid intensification than in shallow water. A result of this change is that there is somewhat less probability of intensification over the continental shelf, and thus reduced winds affecting the North and Central Florida region. The reduction of intense hurricanes in North and Central Florida is more consistent with observations based on HURDAT.

b. We revised the threat area definition. We now initiate both historically landfalling and bypassing storms at their positions 36 hours prior to landfall or coming within 62 sm of the coast (in addition to random perturbations). This new approach essentially treats the storm genesis locations for stochastic storms based on historical landfalling and bypassing storms more similarly than in the prior method.This change had virtually no significant impact on hurricane frequencies, but we nevertheless opted to include it as it seemed more sound theoretically and more computationally efficient.

c. A slightly more significant change is that we consider all storms with lowest central pressure below 1005 mb as a threat (for the purpose of storm initiation) , as opposed to 990 mb in the previous version. The result of this change is that there are a somewhat higher number of Category 1 storms. Changes in loss costs due to the change in this lowest pressure cutoff were about 1% greater for the state of Florida.

FPHLM V3.0 2008 61

3. All forms affected by the model revision have been redone (M-1, M-2, S-1, S-2, S-3, S-4, A-1, A-2, A-5, A-6, A-7).

4. Form V-1 revised to use 1 minute sustained wind. Forms V-2 and V-3 also revised to correct formatting.

5. The following errors in the vulnerability matrices were corrected:

a. Wrong proportions used in the creation of aged-weighted matrices. These are used only if year built is not available. These proportions were corrected so all the columns in the aged-weighed vulnerability matrices add up to one. b. Round off error created by MATLAB when converting the vulnerability matrices from MAT to CSV format. The MATLAB function that created the round off error was replaced by another function that converts the vulnerability matrices from MAT to CSV format without rounding errors. c. Incorrect mix of North structure mobile-home matrices to create aged-weighted matrices. The North structure mobile-home matrices were updated with the correct matrices.

To avoid these errors from happening again, MATLAB and Java programs were developed to ensure that the columns of all matrices add up to one and that the proper matrices are used. The latter is accomplished by checking that the matrices sent by the engineering team in MAT format are the same as the ones used by the Insurance Loss Model in CSV format. 6. We processed and added 16 new company/hurricane portfolio data points and consolidated 4 company/hurricanes into two, resulting in a total of 39 observations. In addition, we checked the validation of some prior company/hurricane portfolios and made some corrections to these portfolios. As a result, 19/39 company/hurricane portfolios have modeled losses greater than actual losses.

7. References edited to conform to a consistent style and additional references added.

8. Revised method for counting landfalls in Keys region, as they were considerably under- counted using the previous method. Originally, landfall was defined when the storm center was over land at the end of each 1 hour time step. However, due to the small size of islands in the Keys, it was unlikely that landfall would be detected in this manner considering a one hour time step. We now consider landfall to occur if the island is inside the radius of maximum winds of the storm. This counting method is only used near the Keys. Using the original counting method, we only had about 2.8 storms (in 107 years) making landfall in the Keys. The new counting method adds about 5 storms total for Regions B and C, and is more consistent with HURDAT, which has at least 7 storms that make landfall in the Keys only. Note that this change only affects counts reported in Form M-1. It has no impact on loss costs, as the winds associated with uncounted landfalls were already included in the wind probabilities.

9. Minor updates in the code. These updates are adding exceptions to handle abnormal conditions, using a third party library to use functionality not provided by the computer language

FPHLM V3.0 2008 62 C++ such as checking the existence of files, and changing the type of some variables to more- precision types.

10. The demand surge factors are slightly different due to small changes in weight factors as a result of the new 53,500 year simulation.

FPHLM V3.0 2008 63 G-2 Qualifications of Modeler Personnel and Consultants

A. Model construction, testing, and evaluation shall be performed by modeler personnel or consultants who possess the necessary skills, formal education, or experience to develop the relevant components for hurricane loss projection methodologies.

The model was developed, tested, and evaluated by a multi-disciplinary team of professors and experts in the fields of meteorology, wind and structural engineering, computer science, statistics, finance, economics, and actuarial science. The experts work primarily at Florida International University, Florida Institute of Technology, Florida State University, University of Florida, Hurricane Research Division of NOAA, and University of Miami.

B. The model or any modifications to an accepted model shall be reviewed by either modeler personnel or consultants in the following professional disciplines: structural/wind engineering (licensed Professional Engineer), statistics (advanced degree), actuarial science (Associate or Fellow of Casualty Actuarial Society), meteorology (advanced degree), and computer/information science (advanced degree). These individuals shall be signatories on Forms G-1 through G-6 as applicable and shall abide by the standards of professional conduct if adopted by their profession.

The model has been reviewed by modeler personnel and consultants in the required professional disciplines. These individuals abide by the standards of professional conduct if adopted by their profession.

Disclosures

1. Organization Background

A. Describe the ownership structure of the modeling organization. Describe affiliations with other companies and the nature of the relationship, if any. Indicate if your organization has changed its name and explain the circumstances.

The model was developed independently by a multi-disciplinary team of professors and experts. The lead university is the Florida International University. The model was commissioned by the FL Office of Insurance Regulation.

B. If the model is developed by an entity other than a modeling company, describe its organizational structure and indicate how proprietary rights and control over the model and its critical components is exercised. If more than one entity is involved in the development of the model, describe all involved.

FPHLM V3.0 2008 64 University of Miami Florida State (UM) University (FSU)

National Oceanic and Florida International University Office of Insurance Regulation Atmospheric Administration (FIU) (OIR) Hurricane Research Division Lead University Funding Agency (NOAA/HRD) Clients

University of Florida (UF) Florida Institute of Technology (FIT)

Figure 10. Organizational Structure

The Florida Office of Insurance Regulation (OIR) contracted and funded Florida International University to develop the Florida Public Hurricane Loss Model. The model is based at the Laboratory for Insurance, Financial and Economic Research, which is part of the International Hurricane Research Center at Florida International University. The OIR did not influence the development of the model. The model was developed independently by a team of professors, experts, and graduate students working primarily at Florida International University, Florida Institute of Technology, Florida State University, University of Florida, Hurricane Research Division of NOAA, and University of Miami. The copyright for the model belongs to OIR.

C. If the model is developed by an entity other than a modeling company, describe the funding source for the model.

The model was funded by the state legislature at the request of the Florida Office of Insurance Regulation.

D. Describe the modeler’s services.

Currently the modeler provides services to one major client, the FL-OIR.

E. Indicate how long the model has been used for analyzing insurance company exposures or other such uses. Describe these uses.

The first version of the model was developed and completed in May 2005, and was based on the knowledge, and the limited data available prior to the 2004, 2005 hurricane seasons. It was not

FPHLM V3.0 2008 65 used for purposes of estimating loss costs for insurance company exposures. Essentially, it was an internal model that was never implemented.

The next version of the model was developed upon acquiring a limited amount of meteorological, engineering and insurance claim data from the 2004-05 hurricane events. It was implemented in March 2006. This version has been used to process the insurance company data on behalf of the Florida Office of Insurance Regulation.

Last year's version 2.6 of the model has been used for analyzing insurance company exposures since August, 2007.

F. Indicate if the modeling organization has ever been involved in litigation or challenged by a statutory authority where the credibility of one of its U.S. hurricane model versions was disputed. Describe the nature of the case and the conclusion.

None.

2. Professional Credentials

A. Provide in a chart format (a) the highest degree obtained (discipline and University), (b) employment or consultant status and tenure in years, and (c) relevant experience and responsibilities of individuals involved in the primary development of or revisions to the following aspects of the model:

1. Meteorology 2. Vulnerability 3. Actuarial Science 4. Statistics 5. Computer Science

See below.

Table 5. Professional credentials

Degree/ Key Personnel University Employment Status Tenure Experience Discipline Meteorology: Florida State Senior Atmospheric Scientist Meteorology wind field Dr. Mark Powell Ph.D. Meteorology 30 University HRD/NOAA model Univ. Texas Scholar/Scientist Meteorology track, Dr. Steve Cocke Ph.D. Physics 13 Austin FSU, Dept of Meteorology intensity, roughness models Distinguish Professor, FSU, Dr. TN Krishnamurti Ph.D. Meteorology Univ. of Chicago 48 Meteorology Dept of Meteorology MSc Meteorology, Florida State Bachir Annane Meteorologist 15 Meteorology Msc Mathematics University University of Atmospheric Scientist Meteorology. Coding of the Dr. George Soukup Ph.D. Physics 27 Chicago HRD/NOAA wind field model Florida State Neal Durst BSc Meteorology Meteorologist 25 Meteorology University Engineering:

FPHLM V3.0 2008 66 Ph.D. Civil Assoc professor, CE Florida Wind engineering, Dr. Jean-Paul Pinelli Georgia Tech 13 Engineering Institute of Technology vulnerability functions Ph.D. Civil Univ of Notre Assoc professor, CE Wind engineering, Dr. Kurt Gurley 10 Engineering Dame Univ of Florida simulations Ph.D. Mech University of New Professor, Florida Institute of Structural engineering Dr. C. Subramanian 25 Engineering Castle Technology analysis Ph.D. Civil Princeton Distinguish Professor, FIU Dr. Emil Simiu 36 Engineering analysis Engineering University and NIST Fellow

Degree/ Key Personnel University Employment Status Tenure Experience Discipline Actuarial/Finance: Dr. Shahid Hamid Ph.D. Economics Professor of Finance Florida Univ of Maryland 20 Insurance and finance Project manager, PI (financial) International University Assoc Professor of Environ Ph.D Agricultural Resource and agriculture Dr. Mahadev Bhat Univ of Tennessee Studies & Econ, Florida Int’l 16 Economics economics, demand surge University Assistant Professor of Financial and Econometric Dr. Duong Ngyue Ph.D Finance Florida Int’l Univ 2 Finance, U-Mass. Dartmouth Analysis Aguedo Ingco FCAS, Actuary CAS President, AMI Risk Con. 36 Reviewer, Demand Surge

Gail Flannery FCAS, Actuary CAS VP, AMI Risk Consultants 26 Reviewer, Demand Surge Computer Science Ph.D. Electrical and Associate Professor of Software and database Dr. Shu-Ching Chen computer Purdue University 9 Computer Science at FIU development engineering Ph.D. Electrical and Associate Professor of Dr. Mei-ling Shyu computer Purdue University Electrical and Computer 9 Software Quality Assurance engineering Engineering at Univ of Miami Ph.D. Computer Assistant Professor of Com. Software and database Min Chen Florida Int’l Univ 4 Science Sci. at U. Montana development Ph.D. Computer Assistant Vice President Software and database Na Zhao Florida Int’l Univ 4 Science State Street Corp. development Fausto Fleites B.S. Candidate Florida Int’l Univ B.S. Candidate FIU 7 Software development Msc Electrical and University of Guy Ravitz Computer Ph.D. Candidate UM 2 Software Quality Assurance Miami Engineering Florida Msc Computer Database Manager at HRD- Programmer and Database Nirva Morisseau- Leroy International 7 Science NOAA Manager University Statistics Univ of Western Assoc professor, Statistics, Statistical testing and Dr. Golam Kibria Ph.D Statistics 11 Ontario FIU sensitivity analysis Univ of South Dr. S. Gulati Ph.D Statistics Professor, Statistics, FIU 15 Statistical tests Carolina

B. Identify any new employees or consultants (since the previous submission) working on the model.

None.

C. Provide visual business workflow documentation connecting all personnel related to model design, testing, execution, maintenance, and decision-making.

FPHLM V3.0 2008 67 Research and Modeling System Development

Meteorology Team Database Management Hurricane Simulation and Wind Schema Design, Database Field Calculation Development and Maintenance Dr. Mark Powell Min Chen Dr. Steven Cocke Dr. George Soukup Bachir Annane Software Engineering Module Implementation and System Integration Dr. Shu-Ching Chen Structural Engineering Team Min Chen, Na Zhao Statistics Team Vulnerability Modeling and Fausto Fleites Statistical Testing, Validation Sensitivity Analysis, Dr. Jean-Paul Pinelli and Support Dr. Kurtis Gurley Quality Assurance Dr. Golam Kibria Dr. Chelakara Subramanian System Verification and Testing Dr. Sneh Gulati Dr. Mei-Ling Shyu Dr. Duong Nguyen Guy Ravitz

Insured Loss Team Insurance Loss Cost Estimation Documentation Dr. Shahid Hamid Documentation Preparation and Dr. Duong Nguyen Maintenance Dr. Mahadev Bhat Dr. Shu-Ching Chen Gail Flannery Kasturi Chatterjee Aguedo Ingco Fausto Fleites

Services Data Verification Technical Support Result Checking and Data Processing and Technical Verification Services Dr. Shahid Hamid Dr. Shu-Ching Chen Dr. Shu-Ching Chen Na Zhao, Fausto Fleites

Clients

Figure 11. Florida Public Hurricane Loss Model Workflow

D. Indicate specifically whether individuals listed in A. and B. are associated with the insurance industry, consumer advocacy group, or a government entity as well as their involvement with consulting activities.

Dr. Mark Powell, Dr. George Soukup, and Neal Dorst work for the Hurricane Research Division of NOAA. Dr Simiu is a Senior Fellow at the National Institute for Science and Technology.

3. Independent Peer Review

A. Provide dates of external independent peer reviews that have been performed on the following components as currently functioning in the model:

FPHLM V3.0 2008 68

1. Meteorology 2. Vulnerability 3. Actuarial Science 4. Statistics 5. Computer Science

Dr. Gary Barnes, Professor of Meteorology at University of Hawaii performed the external review of the meteorology component in December 2006.

Gail Flannery FCAS and Aguedo Ingco, FCAS, actuaries and vice president and president, respectively, of AMI Risk Consultants in Miami, performed the external review of the actuarial component and submission. Subsequently, they became involved in the development of the demand surge model.

The vulnerability, statistical and computer science components were reviewed by modeler personnel.

B. Provide documentation of independent peer reviews directly relevant to the modeler’s responses to the current Standards, Disclosures, or Forms. Identify any unresolved or outstanding issues as a result of these reviews.

The written independent review of the wind component by Dr. Gary Barnes is presented in Appendix A. No unresolved outstanding issues remain after the review.

Gail Flannery FCAS and Aguedo Ingco FCAS, performed the independent review of the actuarial component. They attended many on site meetings with the model team. They were provided with the relevant submission documents, all relevant forms, and supporting documents. They conducted independent analysis of the A forms and asked questions and provided feedback and suggestions. Their questions were addressed, and the feedback and suggestions were acted upon so that no unresolved outstanding issues remain. A letter from Gail Flannery can be found in Appendix A. See also Form G-4.

C. Describe the nature of any on-going or functional relationship the organization has with any of the persons performing the independent peer reviews.

Dr. Gary Barnes, Professor of Meteorology at University of Hawaii, performed the external review of version 2.0 meteorology component of the model. He has no on-going or functional relationship to FIU or the modeling organization, other than as an independent reviewer. He did not take part in the development or testing of the model. His role in the model has been confined to being an independent external reviewer.

4. Provide a completed Form G-1, General Standards Expert Certification.

See Form G-1

FPHLM V3.0 2008 69 5. Provide a completed Form G-2, Meteorological Standards Expert Certification.

See Form G-2

6. Provide a completed Form G-3, Vulnerability Standards Expert Certification.

See Form G-3

7. Provide a completed Form G-4, Actuarial Standards Expert Certification.

See Form G-4

8. Provide a completed Form G-5, Statistical Standards Expert Certification.

See Form G-5

9. Provide a completed Form G-6, Computer Standards Expert Certification.

See Form G-6

FPHLM V3.0 2008 70 G-3 Risk Location

A. ZIP Codes used in the model shall be updated at least every 24 months using information originating from the United States Postal Service. The United States Postal Service issue date of the updated information shall be reasonable.

Our model uses ZIP Code data primarily from a third-party developer, which bases its information on the ZIP-Code definitions issued by the United States Postal Service. The version we used has a USPS vintage of February 2006. The Zip Code data has not been changed in the current release of the model from last year's submission.

B. ZIP Code centroids, when used in the model, shall be based on population data.

ZIP Code centroids used in the model are the population centroids, and are updated at least every 24 months.

C. ZIP Code information purchased by the modeler shall be verified by the modeler for accuracy and appropriateness.

The methodology employed by the vendor of our model for computing population centroids is identical to the computational methods promulgated by the U.S. Census Bureau.

ZIP-Code information is also checked by experts developing our model for consistency. Maps showing the zip code boundaries and the associated centroids have previously been provided to the Professional Team in the last submission, and will be available for further review for the Professional Team for the current submission.

Disclosures

1. List the current ZIP Code databases used by the model and the components of the model to which they relate. Provide the effective (official United States Postal Service) date corresponding to the ZIP Code databases.

FPHLM uses Dynamap 5-Digit ZIP Codes distributed by MapInfo. The source of the data is Geographic Data Technology, Inc. (GDT). GDT created the data using a combination of its DYNAMAP/2000 data, the United States Postal Service (USPS) ZIP+4 Data File, the USPS National 5-Digit ZIP Code and Post Office Directory, USPS ZIP+4 State Directories, and the USPS City State File.

The ZIP Code data is updated quarterly. The release we used in this submission has a Tele Atlas (GDT, Inc.) vintage of 2006.2 (April 2006) and a USPS vintage of February 2006. 5-Digit ZIP Codes aligns with StreetPro v9.1, MapMarker Plus v11.3, Routing J Server v2006.2, and Census Boundary Products v8.1.

The ZIP Code data is used in the Wind Field Module of the model.

FPHLM V3.0 2008 71

2. Describe in detail how invalid ZIP Codes are handled.

A ZIP Code is defined to be “invalid” if it does not match the list of currently valid ZIP Codes. Exposure in any invalid ZIP Code is not modeled.

FPHLM V3.0 2008 72 G-4 Independence of Model Components

The meteorological, vulnerability, and actuarial components of the model shall each be theoretically sound without compensation for potential bias from the other two components.

The meteorology, vulnerability, and actuarial components of the model are theoretically sound and were developed and validated independently before being integrated. The model components were tested individually.

FPHLM V3.0 2008 73 G-5 Editorial Compliance

All documents provided to the Commission throughout the review process shall be reviewed and edited by a person or persons with experience in reviewing technical documents who shall verify on Form G-7 that the submission has been personally reviewed.

The current submission document has been reviewed and edited by persons who are well qualified to perform such tasks. Future revisions and related documentation will likewise be reviewed and edited by the qualified individual listed in Form G-7.

Disclosure

1. Provide a completed Form G-7, Editorial Certification.

See Form G-7.

FPHLM V3.0 2008 74 Forms G1-7

Scanned images of Forms G1-G7 follows.

FPHLM V3.0 2008 75

FPHLM V3.0 2008 76

FPHLM V3.0 2008 77

FPHLM V3.0 2008 78

FPHLM V3.0 2008 79

FPHLM V3.0 2008 80

FPHLM V3.0 2008 81

FPHLM V3.0 2008 82

FPHLM V3.0 2008 83

FPHLM V3.0 2008 84 FPHLM V3.0 2008 85

FPHLM V3.0 2008 86

FPHLM V3.0 2008 87

FPHLM V3.0 2008 88

FPHLM V3.0 2008 89 METEOROLOGICAL STANDARDS

M-1 Base Hurricane Storm Set* (*Significant Revision)

A. Model validation shall be based upon the National Hurricane Center HURDAT starting at 1900 as of June 1, 2007 (or later), HURDAT as of June 1, 2005 plus the 2005 and 2006 seasons, or HURDAT as of June 1, 2006 plus the 2006 season. Complete additional season increments based on updates to HURDAT approved by the Tropical Prediction Center/National Hurricane Center are acceptable modifications to these storm sets. Peer reviewed atmospheric science literature can be used to justify modifications to the Base Hurricane Storm Set.

Validation of the FPHLM is based on the 1900-2006 period of historical record as provided in the June 2007 version of HURDAT.

B. Any trends, weighting or partitioning shall be justified and consistent with currently accepted scientific literature and statistical techniques. Validation and comparison shall encompass the complete Base Hurricane Storm Set as well as any partitions.

Validation and comparison of the FPHLM encompasses the complete Base Hurricane Storm Set provided in HURDAT. We conduct no trending, weighting, or partitioning of the Base Hurricane Set. Disclosures

1. Identify the Base Hurricane Storm Set, the release date, and the time period included to develop and implement landfall and by-passing storm frequencies into the model.

The National Hurricane Center HURDAT file from June 2007 for the period 1900-2006 is used to establish the official hurricane base set used by our model. All HURDAT storm tracks that have made landfall in Florida or bypassed Florida but passed close enough to produce damaging winds, are documented in our archives.

2. If the modeler has made any modifications to the Base Hurricane Storm Set related to landfall frequency and characteristics, provide justification for such modifications.

For landfall frequency and intensity, the Base Hurricane storm set is derived solely from HURDAT. Although our Base Set is derived from HURDAT, the landfall frequencies for our Base Set differ slightly from the landfall frequencies displayed in Form M-1 of the November 2007 Report of Activities. We can provide more detail on these differences if needed. For pressures or Rmax, NWS-38 was used to make modifications to the base set where there were gaps in the HURDAT information. Additional Rmax information from recent hurricanes comes from the H*Wind archive.

3. Where the model incorporates short-term or long-term modification of the historical data

FPHLM V3.0 2008 90 leading to differences between modeled climatology and that in the entire Base Hurricane Storm Set, describe how this is incorporated.

The FPHLM incorporates no short-term or long-term modifications of the climate record. Storm frequencies are based on historical occurrences derived from HURDAT, and thus implicitly contain any long or short term variations that are contained in the historical record. No attempt is made to explicitly model long or short term variations.

4. Provide a completed Form M-1, Annual Occurrence Rates.

See Form M-1.

M-2 Hurricane Parameters and Characteristics* (*Significant Revision due to new Disclosures and Audit language)

Methods for depicting all modeled hurricane parameters and characteristics, including but not limited to windspeed, radial distributions of wind and pressure, minimum central pressure, radius of maximum winds, strike probabilities, tracks, the spatial and time variant windfields, and conversion factors, shall be based on information documented by currently accepted scientific literature.

All methods used to depict storm characteristics are based on methods described in the peer- reviewed scientific literature. Data sets were developed by our scientists using data from published reports, the HURDAT database, archives, observations, and analyses at NOAA’s Hurricane Research Division, The Florida State University, Florida International University, and the Florida Coastal Monitoring Program.

Disclosures

1. Identify the hurricane parameters (e.g., central pressure or radius of maximum winds) that are used in the model. Describe the historical data used for each of these parameters identifying all storms used.

Characteristics modeled include the storm track (translation speed and direction of the storm), radius of maximum wind (Rmax), Holland surface pressure profile parameter (B), the minimum central sea-level pressure (Pmin), the damage threshold distance, and the pressure decay as a function of time after landfall.

The storm initial position and motion are modeled using the HURDAT database (June 2007). For pressure decay we use the Vickery (2005) decay model. Vickery developed the model based on pressure observations in HURDAT and NWS -38, together with Rmax and storm motion data as described in the publication. The radius of maximum winds at landfall is modeled by fitting a gamma distribution to a comprehensive set of historical data published in NWS-38 by Ho et al, (1987) but supplemented by the extended best track data of DeMaria, NOAA HRD research flight data, and NOAA-AOML-HRD H*Wind analyses (Powell et al., 1996, 1998).

FPHLM V3.0 2008 91 Additional research was used to construct an historical landfall Rmax-Pmin database using existing literature (Ho et al 1987), extended best track data collected by Dr. Mark DeMaria, HRD Hurricane field program data, and the H*Wind wind analysis archive. We developed an Rmax model using the revised landfall Rmax database which includes 108 measurements for hurricanes up to 2005. We have opted to model the Rmax at landfall rather than the entire basin for a variety of reasons. One is that the distribution of landfall Rmax may be different than that over open water. An analysis of the landfall Rmax database and the 1988-2007 DeMaria Extended Best Track data shows that there appears to be a difference in the dependence of Rmax on central pressure (Pmin) between the two data sets. The landfall data set provides a larger set of independent measurements, more than 100 storms compared to about 31 storms affecting the Florida threat area region in the Best Track Data. Since landfall Rmax is most relevant for loss cost estimation, and has a larger independent sample size, we have chosen to model the landfall data set. Future studies will examine how the Extended Best Track Data can be used to supplement the landfall data set.

Recent research results by Willoughby and Rahn (2004) based on the NOAA-AOML-HRD annual hurricane field program and Air Force reconnaissance flight-level observations are used to create a model for the “Holland B” parameter. Ongoing research on the relationship between horizontal surface wind distributions (based on Stepped Frequency Microwave Radiometer observations) to flight level distributions is used to correct the flight level Rmax to a surface Rmax, when developing a relationship for the Holland B term. We multiply the flight level Rmax (from the Willoughby and Rahn (2004) data set) by 0.815 to estimate the surface Rmax (based on SFMR, flight level maxima pair data). This adjustment keeps the Holland pressure profile parameter consistent with a surface Rmax, and (due to the negative term in the equation) produces a larger value of B than if a flight-level value of Rmax were used. This is consistent with the concept of a stronger radial pressure gradient for the mean boundary layer slab than at flight level (due to the warm core of the storm), which agrees with GPS dropsonde wind profile observations showing boundary layer winds that are stronger than those at the 10,000 ft. flight level (which is the level for the most of the B data in Willoughby and Rahn 2004). The B adjustment for a surface Rmax produces an overall stronger surface wind field than if B were not adjusted. In addition, surface pressures from the “Best track” information on HURDAT are used to associate a particular flight-level pressure profile B with a surface pressure.

The NOAA-AOML- HRD H*Wind analysis archive was used to develop a relationship between Rmax and the extent of damaging winds to make sure that the model would only consider zip codes with potential for damaging winds. HRD wind modeling research initiated by Ooyama (1969), and extended by Shapiro (1983) has been used to develop the HRD wind field model. This model is based on the concept of a slab boundary layer model, a concept pioneered at NOAA-AOML- HRD and now in use by other modelers for risk applications (e.g. Thompson and Cardone 1996, Vickery and Twisdale 1995, 2000). The HURDAT historical database is used to develop the track and intensity model. Historical data used for computing the potential intensity is based on the National Centers for Environmental Prediction (NCEP) sea surface temperature archives and the NCEP reanalysis for determining the upper tropospheric outflow temperatures. Furthermore the ability of the model to simulate possible future climate scenarios of El Nino, La Nina, and warm or cold interdecadal periods is based on research on climate cycles including (Bove et al, 1998, Landsea et al., 1999, Goldenberg et al., 2001). Climate

FPHLM V3.0 2008 92 scenarios are disabled in Version 3.0 of the Florida Public Hurricane Loss Model. Use cases describing the various model functions and their research basis are available with the model documentation.

2. Describe the dependencies among variables in the windfield component and how they are represented in the model, including the mathematical dependence of modeled windfield as a function of distance and direction from the center position.

B depends linearly on Pmin, latitude, and Rmax. The gradient wind for the slab boundary layer depends on Pmin (through DelP) and B, the mean slab planetary boundary layer (PBL) wind depends on the gradient wind, the drag coefficient (which depends on wind speed), the air density, the gradients of the tangential and radial components of the wind, and the Coriolis parameter (which also depends on latitude). The wind field model solves the equations of motion on a polar grid with a 0.1 R/Rmax radial grid resolution. The input Rmax is reduced by 10% to correct a small bias in Rmax caused by a tendency of the wind field solution to place Rmax radially outward by one grid point. The wind field model terms and dependencies are further described in Powell et al. (2005).

3. For hurricane parameters modeled as random variables, describe the probability distributions. Identify any parameters that have fixed values and provide justification.

Initial storm positions and motion changes derived from HURDAT are modified by the addition of small uniform random error terms. Subsequent storm motion change and intensity are obtained by sampling from empirically derived PDFs as described in Section G-1.2. The random error term for the B parameter is a normal distribution with zero mean and a standard deviation derived from observed reconnaissance aircraft pressure profile fits for B (Willoughby and Rahn 2004). The radius of maximum winds is sampled from a gamma distribution based on landfall Rmax data.

Based on the semi-boundedness and skewness of Rmax, we sought to model the distribution using a gamma distribution. Using maximum likelihood estimators, we found the parameters for the gamma distribution, k=5.53547, θ=4.67749. With these parameters, we show a plot of the observed and expected distribution in Figure 12. The Rmax values are binned in 5 sm intervals, with the x-axis showing the end value of the interval. The gamma distribution proved to be a reasonable fit.

FPHLM V3.0 2008 93

Figure 12. Comparison of observed landfall Rmax (sm) distribution to a Gamma distribution fit of the data.

An examination of the Rmax database shows that intense storms, essentially category 5 storms, have rather small radii. Thermodynamic considerations (Willoughby, 1998) also suggest that smaller radii are more likely for these storms. Thus, we model category 5 (DelP>90 mb, where DelP=1013-Pmin and Pmin is the central pressure of the storm) storms using a gamma distribution, but with a smaller value of the θ parameter, which yields a smaller mean Rmax as well as smaller variance. We have found that for Category 1-4 (DelP<80) storms there is essentially no discernable dependence of Rmax on central pressure. This is further verified by looking at the mean and variance of Rmax in each 10 mb interval. Thus we model category 1-4 storms with a single set of parameters. For a gamma distribution, the mean is given by kθ, and variance is kθ2. For category 5 storms, we adjust θ such that the mean is equal to the mean of the three category 5 storms in the database: 1935 No Name, 1969 Camille and 1992 Andrew. An intermediate zone between DelP=80 mb and DelP=90 mb is established where the mean of the distribution is linearly interpolated between the Category 1-4 value and the Category 5 value. As the θ value is reduced, the variance is likewise reduced. Since there are insufficient observations to determine what the variance should be for Category 5 storms, we rely on the assumption that variance is appropriately described by the re-scaled θ, via kθ2.

A simple method is used to generate the gamma-distributed values. A uniformly distributed variable, a product of the random number generator that is intrinsic to the Fortran compiler, is mapped onto the range of Rmax values via the inverse cumulative gamma distribution function. For computational efficiency, a lookup table is used for the inverse cumulative gamma distribution function, with interpolation between table values. Figure 13 shows a test using 100,000 samples of Rmax for Category 1-4 storms, binned in 1 sm intervals, and compared with the expected values.

FPHLM V3.0 2008 94

Figure 13. Comparison of 100,000 Rmax values sampled from the Gamma distribution for Cat 1-4 storms to the expected values.

For category 5 and intermediate category 4-5 storms, we utilize the property that the gamma cumulative distribution function is a function of (k,x/θ). Thus, by re-scaling θ, we can use the same function (lookup table), but just rescale x (Rmax). The rescaled Rmax will then still have a gamma distribution, but with different mean and variance.

4. Describe how any hurricane parameters are treated differently in the historical and stochastic storm sets (e.g., has a fixed value in one set and not the other).

All historical storm sets consist of input files containing information derived from HURDAT or other observation sources as described in Standard M-1. All stochastic input storm tracks are modeled.

5. State whether the model simulates surface winds directly or requires conversion between some other reference level or layer and the surface. Describe the process for converting the modeled vortex winds to surface winds including the treatment of the inherent uncertainties in the conversion factor with respect to location of the site compared to the radius of maximum winds over time. Justify the variation in the surface winds conversion factor as a function of hurricane intensity.

FPHLM V3.0 2008 95

Gradient winds are not converted to surface winds in this model. Gradient winds are used to help estimate the initial slab planetary boundary layer (PBL) winds in a given storm. The PBL winds depart from gradient balance due to the effects of friction and the radial advection of tangential momentum. The PBL winds are adjusted to the surface using recent results from Powell et al., (2003) which estimated a mean reduction factor of 77.5%, based on over 300 GPS sonde wind profile observations in hurricanes. The reduction factor is based on the ratio of the surface wind speed at 10 m to the mean wind speed for the 0-500 m layer (Mean Boundary Layer wind speed or MBL) published in Powell et al., (2003). This ratio is much more relevant to a slab boundary layer model than using data based on higher, reconnaissance aircraft flight levels. The depth of the slab boundary layer model is assigned a value of 450 m, which is the level of the maximum mean wind speed from GPS sonde wind profiles published in Powell et al., (2003). The uncertainty of the reduction factor is ~8% based on the standard deviation of the measurements, but no attempt is made to model this uncertainty. No spatial or intensity dependent variation of reduction factor is used at this time.

6. Describe how the windspeeds generated in the windfield model are converted from sustained to gust and identify the averaging time.

Wind speeds from the HRD slab boundary layer wind field model are assumed to represent 10 min averages. A sustained wind is computed by applying a gust factor to account for the highest 1 min wind speed over the 10 min period. A peak 3s gust is also computed. Gust factors depend on wind speed and the upstream fetch roughness which in turn depends on wind direction at a particular location. Gust factor calculations were developed using research in the Engineering Sciences Data Unit (ESDU) series papers as summarized and applied to tropical cyclones by Vickery and Skerlj (2005).

7. Describe how the asymmetric structure of hurricanes is represented in the model.

The asymmetry of the wind field is determined by the storm translation motion (right-left asymmetry), and the associated asymmetric surface friction. A set of form factors for the wind field also contribute to the asymmetry. The proximity of the storm to land also introduces an additional asymmetry due to the affect of land roughness elements on the flow. Azimuthal variation is introduced through the use of two form factors (see Appendix of Powell et al., 2005 for more detail). The form factors multiply the radial and tangential profiles and provide a “factorized” ansatz for both the radial and tangential storm–relative wind components. Each form factor contains three constant coefficients which are variationally determined in such a way that the ansatz constructed satisfies (as far as its numerical degrees of freedom permit) the scaled momentum equations for the storm-relative polar wind components.

8. Describe the historical data used as the basis for the model’s hurricane tracks. Discuss the appropriateness of the model stochastic hurricane tracks with reference to the historical storm database.

FPHLM V3.0 2008 96 The hurricane tracks are modeled as a Markov process. Initial storm conditions are derived from HURDAT. Small uniform random perturbations are added to the historical initial conditions, including initial storm location, change in motion, and intensity.

Storm motion is determined by sampling empirical distributions, based on HURDAT, of change in speed and change in direction, as well as change in relative intensity. These functions are also spatially dependent, binned in variable box sizes (typically 2.5 degree), and are enlarged as necessary to ensure sufficient density of storms for the distribution.

The model has been validated by examining key hurricane statistics at roughly 30 sm milepost locations along the Gulf and Atlantic coasts. The parameters examined include average central pressure deficit, average heading angle and speed, and total occurrence by Saffir-Simpson category.

9. If the historical data are partitioned or modified, describe how the hurricane parameters are affected.

The FPHLM does not partition or modify the historical data.

10. Describe how the coastline is segmented (or partitioned) in determining the parameters for hurricane frequency used in the model. Provide the hurricane frequency distribution by intensity for each segment.

The model does not use coastline segmentation to determine hurricane frequency.

11. Describe any evolution of the functional representation of hurricane parameters during an individual storm life cycle.

Upon landfall, the evolution of the central pressure changes from sampling a PDF, to a decay model described in Vickery (2005). When the storm exits back over water, the pressure is again modeled via the PDF. After landfall, the slab boundary layer surface drag coefficient changes from a functional marine form to a constant based on a mean aerodynamic roughness length of 0.2 m. The slab boundary layer height increases from 450 m to 1 km after the center makes landfall, and decreases back to 450 m if the center exits land to go back to sea.

FPHLM V3.0 2008 97 M-3 Hurricane Probabilities* (*Significant Revision)

A. Modeled probability distributions of hurricane parameters and characteristics shall be consistent with historical hurricanes in the Atlantic basin.

Hurricane motion (track) is modeled based on historical geographic probability distributions of hurricane genesis locations (locations where hurricanes developed or moved into the threat area), translation velocity and velocity change, initial intensity, intensity change, and potential intensity. Modeled probability distributions for hurricane intensity, forward speed, Rmax, and storm heading are consistent with historical hurricanes in the Atlantic basin.

B. Modeled hurricane probabilities shall reflect the Base Hurricane Storm Set used for category 1 to 5 hurricanes and shall be consistent with those observed for each coastal segment of Florida and neighboring states (Alabama, Georgia, and Mississippi).

As shown in Form M-1 and the accompanying plots, our model reflects reasonably the 1900- 2006 Base Hurricane Set for hurricanes of Saffir-Simpson Categories 1-5 in each coastal region of Florida as well as the neighboring states. In addition, a finer scale coastal mile post study of model parameters (occurrence rate, storm translation speed, storm heading, and Pmin) was conducted during the development of the model.

Disclosures

1. List assumptions used in creating the hurricane characteristic databases.

The Holland B database is based on flight-level pressure profiles corresponding to constant pressure surfaces at 700 mb and below. Due to a lack of surface pressure field data, an assumption is made that the Holland B at the surface is equivalent to a B determined from information collected at flight level. The surface pressure profile uses Pmin, DelP, and Rmax at the surface. It would be ideal to have a B data set also corresponding to the surface but such data are not available. The best available data on B are flight-level data from Willoughby and Rahn (2004). Willoughby and Rahn (2004) discuss: “In major hurricanes... they almost invariably flew at 3km (700 mb) .” Few lower level data are available for mature hurricanes so their plot (Fig. 14) of B vs. flight-level “provide no information about average vertical structure”. In lieu of lower level data, we model B using flight data supplied by Dr. Willoughby, but with Rmax adjusted to a surface Rmax, and with surface DelP added from NHC best track for each flight. Since we are modeling hurricane winds during landfall, our Rmax model applies only to landfall and is not designed to model the lifecycle of Rmax as a function of intensity.

2. List data sources used in developing probability distributions for all hurricane parameters and characteristics.

FPHLM V3.0 2008 98 Monthly geographic distributions of climatological sea surface temperatures (Reynolds 1 degree resolution, Reynolds et al., 2002) and upper tropospheric outflow temperatures (NCEP REANALYSIS II 100 mb, Kanamitsu et al., 2002) are used to determine physically realistic potential intensities which help to bound the modeled intensity. Terrain elevation and bathymetry data was obtained from the United States Geological Survey. The radius of maximum wind at landfall is modeled from a comprehensive set of historical data published in NWS-38 by Ho et al, (1987) but supplemented by the extended best track data of DeMaria, (Penington 2000), NOAA HRD research flight data, and NOAA-HRD H*Wind analyses (Powell et al., 1996, 1998). The development of the Rmax frequency distribution fit and it’s comparison to historical hurricane data is discussed in M-2.1 and M-2.3. Comparisons of the modeled radius of maximum wind to the observed data are shown in Form M-3. H*Wind wind field analyses of historical hurricanes are available from the NOAA-AOML-HRD web site: http://www.aoml.noaa.gov/hrd/data_sub/wind.html

FPHLM V3.0 2008 99 M-4 Hurricane Windfield Structure* (*New Standard)

A. Windfields generated by the model shall be consistent with observed historical storms affecting Florida.

As described in Statistical Standards S-1, Disclosure 3, comparisons of FPHLM to gridded H*Wind fields indicate that the FPHLM wind fields are consistent with observed historical wind fields from Florida landfalling hurricanes.

B. The translation of land use and land cover or other source information to geographic surface roughness distribution shall be consistent with current state- of-the-science.

Land friction is modeled according to the currently accepted, state-of-the-science principles of surface layer similarity theory as described in the disciplines of micrometeorology, atmospheric turbulence, and wind engineering. The geographic distribution of surface roughness is determined by careful studies of aerial photography, site visits, and satellite remote sensing measurements used to create land use - land cover classification systems. We use the MRLC NLCD 2001 land use data set. This data set became available in Spring, 2007, and provides detailed (30 m) land use characteristics circa 2001. All population-weighted zip code centroids are assigned roughness values as a function of upstream fetch for each wind direction octant. After landfall, the surface drag coefficient used in the hurricane PBL slab model changes from a marine value to a fixed value associated with a roughness of 0.2 m.

FPHLM V3.0 2008 100

Figure 14. Axisymmetric rotational wind speed (mph) vs. scaled radius for B = 1.38, DelP = 49.1 mb

Disclosures

1. Provide a rotational windspeed (y-axis) versus radius (x-axis) plot of the average or default symmetric wind profile used in the model and justify the choice of this wind profile.

See Figure 14. The Holland B profile has been compared extensively to historical data (Holland, 1980 and Willoughby and Rahn, 2004) and found to be a reasonable fit.

2. If the model windfield has been modified in any way from the previous submission, provide a rotational windspeed (y-axis) versus radius (x-axis) plot of the average or default symmetric wind profile for both the new and old functions. The choice of average or default shall be consistent for the new and old functions.

The wind field model has not been modified since the previous submission.

3. If the model windfield has been modified in any way from the previous submission, describe variations between the new and old windfield functions with reference to historical storms.

The wind field model has not been modified since the previous submission.

FPHLM V3.0 2008 101

4. Describe the relevance of the formulation of gust factor(s) used in the model.

The gust factors used in the model were developed from hurricane wind speed data and the Engineering Sciences Data Unit methods as described in Vickery and Skerlj (2005).

5. Identify all non-meteorological variables that affect windspeed estimation (e.g., surface roughness, topography, etc.).

Upstream aerodynamic surface roughness within a fixed 45 degree sector extending upstream has an effect on the determination of wind speed for a given zip code centroid and is the primary variable that affects estimation of surface wind speeds. The upstream sectors are defined according to the Tropical Cyclone Winds at Landfall Project (Powell et al., 2004), which characterized upstream wind exposure for each of eight wind direction sectors at over 200 coastal automated weather stations (Figure 15).

Figure 15. Upstream fetch wind exposure photograph for Chatham MS (left, looking north), and Panama City, FL (right, looking Northeast). After Powell et al., (2004).

6. Provide the collection and publication dates of the land use and land cover data used in the model and justify their timeliness for Florida.

We use the 2001 Multi-Resolution Land Characteristics Consortium (MRLC) National Land Cover Database released April 25, 2007. To the best of our knowledge, this is the most recent, high resolution (30 m) land cover data set that covers not only Florida, but the entire U.S, and roughly depicts land characteristics circa 2001 (see Homer et al., 2004 for more details).

7. Describe the methodology used to convert land use and land cover information into a spatial distribution of roughness coefficients in Florida and adjacent states.

The land cover classifications provided by the 2001 MRLC Land Cover Database are first mapped to roughness values using a lookup table that associates a representative roughness for the land use category based on peer reviewed literature. An effective roughness model (Axe,

FPHLM V3.0 2008 102 2004) is then used to incorporate upstream roughness elements to provide a more realistic roughness for each zip code centroid.

8. Demonstrate the consistency of the spatial distribution of model-generated winds with observed windfields for hurricanes affecting Florida.

As shown below in Disclosure 9, and in Statistical Standard 1 Disclosure 3, the spatial distribution of model-generated winds is consistent with observed windfields for hurricanes affecting Florida.

9. Describe how the model’s windfield is consistent with the inherent differences in windfields for such diverse storms as Hurricane Charley, Hurricane Katrina, and Hurricane Wilma.

The model can represent a wide variety of storms through variation of parameters for radius of maximum winds, central pressure deficit and Holland Beta (B). Snapshots of model wind fields at landfall are compared to NOAA-AOML-HRD H*Wind analyses below (for further details see disclosure 3 for Standard S-1). In these cases, rather than tune the model to best fit the observations by varying the Holland B parameter, we derived the input B from the H*Wind analyses. Hurricane Charley, a small, fast moving hurricane, was modeled quite well capturing the motion asymmetry and extent of strong winds in the core of the storm. Hurricane Katrina was developing an eyewall while in the process of making landfall. The FPHLM depicts a mature storm whereas the observed H*Wind field shows a storm with and incomplete eyewall. The FPHLM in a simple model without the complicated physics needed to represent convective processes associated with eyewall development. Wilma made landfall in Florida as a very large hurricane. The FPHLM captures the location of maximum winds in the core of the storm and represents the left-right motion asymmetry but tends to produce too broad a wind field.

FPHLM V3.0 2008 103

Figure 16. Comparison of observed (right) and modeled (left) landfall wind fields of Hurricanes Charley (2004, top), and 2005 Hurricane Katrina in south Florida (bottom). Line segment indicates storm heading. Horizontal coordinates are in units of R/Rmax and winds units of miles per hour.

FPHLM V3.0 2008 104

Figure 17. As in Figure 16 except for Hurricane Wilma of 2005.

10. Describe any variations in the treatment of the model windfield for stochastic versus historical storms and justify this variation.

All historical storm sets consist of input files containing information derived from HURDAT or other observation sources as described in Standard M-1. All stochastic input storm tracks are modeled. The wind field is modeled from the stochastic or historical input files in the same manner.

11. Provide a completed Form M-2, Maps of Maximum Winds.

See Form M-2.

FPHLM V3.0 2008 105 M-5 Landfall and Over-Land Weakening Methodologies* (*Significant Revision)

A. The magnitude of land friction coefficients shall incorporate current geographic surface roughness distributions and shall be implemented with appropriate geographic information system data.

We have now incorporated the MRLC NLCD 2001 land use data set. This data set became available in Spring, 2007, and provides detailed (30 m) land use characteristics circa 2001. All population-weighted zip code centroids are assigned roughness values as a function of upstream fetch for each wind direction octant.

B. The hurricane overland weakening rate methodology used by the model shall be consistent with historical records.

Overland weakening rates are based on a pressure decay model developed from historical data as described by a recent paper published in the peer-reviewed atmospheric science literature (Vickery 2005).

C. Models shall use maximum one-minute sustained 10-meter windspeed when defining hurricane landfall intensity. This applies both to the Base Hurricane Storm Set used to develop landfall strike probabilities as a function of coastal location and to the modeled winds in each hurricane which causes damage. The associated maximum one-minute sustained 10-meter windspeed shall be within the range of windspeeds (in statute miles per hour) categorized by the Saffir- Simpson Scale.

Saffir-Simpson Hurricane Scale:

Category Winds (mph) Damage 1 74 – 95 Minimal 2 96 – 110 Moderate 3 111 – 130 Extensive 4 131 – 155 Extreme 5 Over 155 Catastrophic

The HRD wind field model simulates landfall intensity according to the maximum 1 min sustained wind for the 10 m level for both stochastic simulations and the Official Hurricane Set. The Saffir-Simpson damage potential scale is used to further categorize the intensity at landfall and the range of simulated wind speeds (in miles per hour) is within the range defined in the scale.

FPHLM V3.0 2008 106 Disclosures

1. Describe and justify the functional form of hurricane decay rates used by the model.

The hurricane decay rate function acts to decrease the DelP with time after landfall. The functional form is an exponential in time since landfall and is based on historical data (Vickery 2005).

2. Provide a graphical representation of the modeled degradation rates for Florida hurricanes over time compared to wind observations. Reference to the Kaplan-DeMaria decay rates alone is not acceptable.

The degradation of the wind field of a landfalling hurricane is associated with the filling of the central sea-level pressure and the associated weakening of the surface pressure gradient, as well as the fact that the hurricane is over land, where the flow is subject to friction while flowing across obstacles in the form of roughness elements. Maximum wind degradation is shown according to how the maximum sustained surface wind (at the location containing the maximum winds in the storm) changes with time after landfall. At landfall the marine exposure wind is assumed to be representative of the maximum winds occurring onshore. After landfall the open terrain wind is chosen to represent the maximum envelope of sustained winds over land. The NOAA-HRD H*Wind system is used to analyze the maximum winds at a sequence of times following landfalls of Hurricanes Katrina, Charley, Frances, Jeanne, and Wilma. H*Wind uses all available wind observations. The landfall wind field is used as a background field for times after landfall and compared to the available observations at a sequence of times after landfall. An empirical decay is applied to the background field based on the comparisons to the observations. These data are then objectively analyzed to determine the wind field at each time. The model maximum sustained winds are compared to the maximum winds from the H*Wind analyses for the same times and roughness exposures. In general, points after landfall are given for open terrain exposure. At times, even though the storm center is over land, the maximum wind speed may remain over water. For example, in the Frances plot, the first three pairs of points represent marine exposure, the next three open terrain, and the final three marine exposure again, while all Wilma point pairs represent marine exposure. The plots indicate that the Public wind field model realistically simulates decay of the maximum wind speed during the landfall process, as well as subsequent strengthening after exit.

FPHLM V3.0 2008 107

Figure 18. Observed (green) and modeled (black) maximum sustained surface winds as a function of time for 2004 Hurricanes FrancesCharley (left) and Charley (right). Landfall is represented by the vertical dash-dot red line at the left and time of exit as the red line on the right.

FPHLM V3.0 2008 108

Figure 19. Observed (green) and modeled (black) maximum sustained surface winds as a function of time for Hurricanes Jeanne (2004, top left), Katrina (2005 in South Florida, top right), and 2005 Wilma (lower left). Landfall is represented by the vertical dash-dot red line at the left and time of exit as the red line on the right.

FPHLM V3.0 2008 109 3. Describe the transition from over-water to over-land boundary layer simulated in the model.

After landfall, the slab boundary layer surface drag coefficient changes from a functional marine form to a constant based on a mean aerodynamic roughness length of 0.2 m. The slab boundary layer height increases from 450 m to 1 km after the center makes landfall, and decreases back to 450 m if the center exits land to go back to sea. The primary influence on the surface flow is he surface roughness. Roughness is modeled at each zip code and for eight upstream wind direction octants using a method (Schmidt and Oke, 1990, Axe 2004) which integrates the upstream roughness elements, weighting the elements close to the centroid higher than those far away. Therefore, zip codes in which the upstream fetch extends over water, will have more marine-like wind conditions than those that integrate elements existing primarily over land. Therefore the upstream-integrated roughness values determine the transition of wind speeds from over-water to over-land flow exposure.

4. Document any differences between the treatment of decay rates in the model for stochastic hurricanes compared to historical hurricanes affecting Florida.

In the FPHLM model, decay is defined as the change in minimum sea-level pressure (Pmin) with time after landfall. The input file for the wind field model consists of a hurricane track file that contains storm position, Pmin, Rmax, and Holland B at 1h frequency. The wind field model is exactly the same for scenario (historical) or stochastic events. When running the model in scenario mode for historical hurricanes affecting Florida, we use a set of historical hurricane tracks as input to the model. When running the model in stochastic mode, the input hurricane tracks are provided by the track and intensity model. The track and intensity model uses the Vickery (2005) pressure decay after landfall. When a hurricane exits land, the Pmin over water is determined based on the Markov process as described in Disclosure G-1.2

The historical tracks based on HURDAT are detailed on our web site at: http://www.aoml.noaa.gov/hrd/lossmodel/

For historical hurricane tracks the landfall pressure is determined from HURDAT or from the Ho et al., (1987) report. If post-landfall pressure data are available in HURDAT, we interpolate pressure values over land. If post-landfall pressure data are not available, we apply the Vickery (2005) pressure decay model to the landfall pressure. After the storm exits land, the pressure is based on HURDAT data. Therefore, decay rates for historical hurricanes are based on HURDAT data if available, or the Vickery decay rate model applied to the HURDAT or Ho et al, (1987) landfall Pmin, while decay rates for stochastic hurricanes are based on Vickery (2005).

FPHLM V3.0 2008 110 M-6 Logical Relationships of Hurricane Characteristics

A. The magnitude of asymmetry shall increase as the translation speed increases, all other factors held constant.

The storm translation speed causes a major right-left (looking in the direction the storm is moving) asymmetry in the wind field which in turn causes an asymmetry in surface friction since the surface stress is wind speed dependent. The magnitude of the asymmetry increases as the translation speeds increases; there is no asymmetry for a stationary storm except for possible land friction effects if a storm becomes stationary while a large percentage of its circulation is over both land and water.

B. The mean windspeed shall decrease with increasing surface roughness (friction), all other factors held constant.

All other factors held constant, the mean wind speed decreases with increasing surface roughness. However, the gust factor, which is used to estimate the peak one min wind and the peak 3 s gust over the time period corresponding to the model mean wind increases as a function of turbulence intensity, which increases with surface roughness (Paulsen et al., 2003, Masters 2004, Powell et al., 2004). For roughness values representative of zip codes in Florida with residential roughness values on the order of 0.2 - 0.3 m, the roughness effect on decreasing the mean wind speed overwhelms the enhanced turbulence intensity effect that increases the gust factor.

Disclosures

1. Provide a completed Form M-3, Radius of Maximum Winds and Radii of Standard Wind Thresholds.

See Form M-3.

FPHLM V3.0 2008 111 Form M-1: Annual Occurrence Rates

A. Provide annual occurrence rates for landfall from the data set defined by marine exposure that the model generates by hurricane category (defined by windspeed in the Saffir- Simpson scale) for the entire state of Florida and selected regions as defined in Figure 3 [of the 2007 ROA]. List the annual occurrence rate (probability of an event in a given year) per hurricane category. Annual occurrence rates shall be rounded to two decimal places. The historical frequencies below have been derived from the National Hurricane Center’s HURDAT as of June 1, 2007.

FPHLM V3.0 2008 112

Form M-1. Modeled Annual Occurrence Rates Entire State Region A – NW Florida Region B – SW Florida Category Historical Modeled Historical Modeled Historical Modeled 1 0.24 0.29 0.09 0.14 0.08 0.07 2 0.18 0.14 0.07 0.06 0.02 0.05 3 0.18 0.14 0.06 0.04 0.06 0.05 4 0.05 0.07 0.00 0.02 0.02 0.02 5 0.02 0.02 0.00 0.00 0.01 0.00

Florida By-Passing

Region C – SE Florida Region D – NE Florida Hurricanes Category Historical Modeled Historical Modeled Historical Modeled 1 0.07 0.07 0.00 0.01 0.05 0.06 2 0.07 0.04 0.02 0.01 0.03 0.03 3 0.07 0.04 0.00 0.01 0.04 0.03 4 0.03 0.03 0.00 0.00 0.01 0.01 5 0.01 0.02 0.00 0.00 0.00 0.01

Region E - Georgia Region F- AL-MS Category Historical Modeled Historical Modeled 1 0.03 0.01 0.06 0.04 2 0.00 0.00 0.02 0.02 3 0.00 0.00 0.06 0.02 4 0.00 0.00 0.00 0.01

5 0.00 0.00 0.01 0.00

Note: Except where specified, Number of Hurricanes does not include By-Passing Storms. Each time a hurricane goes from water to land (once per region) it is counted as a landfall in the table above.

FPHLM V3.0 2008 113 B. Describe model variations from the historical frequencies.

Form M-1 landfall frequencies were determined from the impact codes listed in the “trailer” information provided in the HURDAT database. In some cases the HURDAT codes did not agree with the 6h HURDAT information. We are corresponding with the HURDAT archivists to learn more about this issue and hope to discuss our results at a future Commission meeting.

The modeled frequencies are consistent with the historical record, to the extent that we may consider the historical record reliable. Statewide, the model produces 71.1 Florida landfalls in 107 years, compared to 69-71 historically. For major (Cat 3-5) storms, the model produces 24.3 storms, compared to about 26 historically.

On a regional basis, the model is also consistent with the historical record. In region A there is relatively somewhat higher number of Category 1 storms, and in region C there is a somewhat lower total number of storms than the historical record, particularly for weaker storms. However, chi square goodness of fit tests for regions A-C have p-values above 0.35 (see Table 6), indicating that any variations could, to a reasonable degree, be due to chance.

FPHLM V3.0 2008 114 Table 6. Chi Square Goodness of Fit for Florida by Region

Region A - NW Florida Cat Model Historical 1 14.49 10 2 5.93 7 3-5 6.66 6 P-value: 0.44

Region B - SW Florida Cat Model Historical 1-2 12.82 11 3-5 7.46 9 P-value: 0.45

Region C – SE Florida Cat Model Historical 1-2 11.53 14 3-5 9.25 11 P-value: 0.35

Entire State of Florida Cat Model Historical 1 31.29 26 2 15.49 19 3 14.52 19 4-5 9.81 7 P-value: 0.27

FPHLM V3.0 2008 115

C. Provide vertical bar graphs depicting distributions of hurricane frequencies by category by region of Florida (Figure 3 [of the 2007 ROA]) and for the neighboring states of Alabama/Mississippi and Georgia. For the neighboring states, statistics based on the closest milepost to the state boundaries used in the model are adequate.

Vertical bar charts are shown in the Figures below. These charts show the number of hurricanes in a 107 year period.

Figure 20. Form M-1 comparison of modeled and historical landfalling hurricane frequency statewide.

Figure 21. Form M-1 comparison of modeled and historical landfalling hurricane frequency in region A.

FPHLM V3.0 2008 116

Figure 22. Form M-1 comparison of modeled and historical landfalling hurricane frequency in region B.

Figure 23. Form M-1 comparison of modeled and historical landfalling hurricane frequency in region C.

FPHLM V3.0 2008 117

Figure 24. Form M-1 comparison of modeled and historical landfalling hurricane frequency in region D.

Figure 25. Form M-1 comparison of modeled and historical landfalling hurricane frequency in region E.

FPHLM V3.0 2008 118

Figure 26. Form M-1 comparison of modeled and historical landfalling hurricane frequency in region F.

Figure 27. For Form M-1 comparison of modeled and historical by-passing hurricane frequency.

D. If the data are partitioned or modified, the modeler shall provide the historical annual occurrence rates for the applicable partition (and its complement) or modification as well as modeled annual occurrence rates in additional Form M-1s.

Not Applicable.

FPHLM V3.0 2008 119 Form M-2: Maps of Maximum Winds

A. Provide color maps of the maximum winds for the modeled version of the Base Hurricane Storm Set for both open terrain and actual terrain.

B. Provide color maps of the maximum winds for a 100-year return period from the stochastic storm set for both open terrain and actual terrain.

C. Provide the maximum winds plotted on each contour map.

“Actual terrain” is the roughness distribution used in the standard version of the model. “Open terrain” uses the same roughness value of 0.03 meters at all land points.

All maps shall be color coded at the ZIP Code level.

Maximum winds in these maps are defined as the maximum one-minute sustained winds over the terrain as modeled and recorded at each location.

The same color scheme and increments shall be used for all maps.

Use the following seven isotach values and interval color coding:

40 mph Blue 75 mph Medium Blue 95 mph Light Blue 110 mph Light Pink 130 mph Pink 140 mph Light Red 155 mph Red

FPHLM V3.0 2008 120

Figure 28 Maximum zip code wind speed for open terrain wind exposure based on simulations of the historical storm set

FPHLM V3.0 2008 121

Figure 29 Maximum zip code wind speed for actual terrain wind exposure based on simulations of the historical storm set.

FPHLM V3.0 2008 122

Figure 30. 100 year return period wind speeds at Florida zip codes for open terrain wind exposure.

FPHLM V3.0 2008 123

Figure 31. 100 year return period wind speeds at Florida zip codes for actual terrain wind exposure.

FPHLM V3.0 2008 124 Form M-3: Radius of Maximum Winds and Radii of Standard Wind Thresholds

A. For the central pressures in the table below, provide ranges for 1) the radius of maximum winds (Rmax) used by the model to create the stochastic storm set and for the radii (R) of 2) Category 3 winds (>110 mph), 3) Category 1 winds (>73 mph), and 4) gale force winds (>40 mph). This information should be readily calculated from the windfield formula input to the model and does not require running the stochastic storm set.

B. Identify the other variables that influence Rmax.

We sample Rmax from a Gamma distribution which only explicitly depends on central pressure.

Central Range of Rmax Range of Range of R Range of R Pressure (mi) R (>110 (>73 mph) (>40 mph) (mb) mph) (mi) (mi) (mi)

900 4.1-19.6 3.9-34.4 2.9-68.0 1.5-132.4 910 4.6-21.0 4.8-30.5 3.4-62.8 1.9-140.0 920 4.0-21.2 3.5-40.3 2.2-83.7 1.3-137.3 930 10.3-42.6 6.2-55.8 4.9-109.4 2.9-157.2 940 12.4-53.3 9.0-53.0 6.0-115.8 3.0-190.8 950 10.0-46.6 7.2-45.9 4.8-88.3 2.4-167.6 960 6.4-56.6 6.8-21.5 4.8-88.8 4.8-187.3 970 6.4-48.4 NA 4.8-67.9 2.4-169.3 980 6.9-38.2 NA 8.3-62.9 2.6-137.2 990 9.3-49.8 NA 10.1-49.8 4.1-126.0

FPHLM V3.0 2008 125 C. Provide a box plot of Central Pressure (x-axis) versus Rmax (y-axis) to demonstrate relative populations and continuity of sampled hurricanes in the stochastic storm set.

A scatter plot and box plot are shown below.

Figure 32 Representative scatter plot of Radius of maximum wind (y axis) versus Minimum sea-level air pressure at landfall (mb). Relative histograms for each quantity are also shown. Extracted from simulation of 50,000 years conducted on 12-10-2007.

FPHLM V3.0 2008 127

Figure 33. Oneway box plot (left) of Rmax (continuous) response across 10 mb Pmin groups. Boxes (and whiskers) are in red, means and standard deviations in green. Histograms (right) for each Pmin group. Representative sample from 50,000 year simulation of December 2007.

FPHLM V3.0 2008 127 VULNERABILITY STANDARDS

V-1 Derivation of Vulnerability Functions

A. Development of the vulnerability functions is to be based on a combination of the following: (1) historical data, (2) tests, (3) structural calculations, (4) expert opinion, or (5) site inspections. Any development of the vulnerability functions based on structural calculations or expert opinion shall be supported by tests, site inspections, or historical data.

The development of the vulnerabilities is based on a component approach that combines engineering modeling and simulations with engineering judgment and observed (historical) data. The determination of external damage to buildings is based on structural calculations, tests, and Monte Carlo simulations. The wind loads and strength of the building components in the simulations are based on laboratory and in-situ tests, manufacturer’s data, expert opinion based on site inspections of actual damage post-hurricane, and codes and standards. The internal and content damage are extrapolated from the external damage based upon expert opinion, and confirmed using historical claims data and site inspections of areas impacted by recent hurricanes.

B. The method of derivation of the vulnerability functions shall be theoretically sound.

The method used in the derivation is based on extrapolating the results of Monte Carlo simulations of physical exterior damage, through simple equations based on engineering judgment, expert opinion, and claims data. Uncertainties at each stage are accounted for by distributing the damage according to reasonable probability distributions and validated with claims data.

C. Any modification factors/functions to the vulnerability functions or structural characteristics and their corresponding effects shall be clearly defined and be theoretically sound.

The Monte Carlo component models take into account many variations in structural characteristics and the result clearly filters through the cost estimation model. There are also different and clearly defined costing considerations applied to each structural type. These adjustments come directly from resources developed exclusively for defining repair costs to structures and therefore are theoretically sound.

D. Construction type and construction characteristics shall be used in the derivation and application of vulnerability functions.

A detailed exposure study was carried out to define the most significant (prevalent) construction types and characteristics in the Florida residential building stock, for different regions of the State. The corresponding engineering models were built for each of the identified common structural types. The models include differing wall types (wood, masonry) of varying strengths

FPHLM V3.0 2008 128 (e.g., reinforced or not, various sill plate connection types), differing roof shapes (hip and gable end) and their affect on uplift loading, various strengths of roof to wall connections (toe nail up through straps), varying window types and sizes, opening protection systems, varying garage door pressure capacities, and one and two story houses.

Models of varying combinations of the above characteristics (e.g. wood frame, gable end, no window shutters) were created for four different regions in Florida, where the region dictates the square footage footprint of the model. In all cases, the probabilistic capacities of the various components were determined by a variety of sources, including testing, test results in the literature, in-field data collection (post hurricane damage evaluations), manufacturer’s specifications as well as manufacturer’s test data when available, and expert opinion.

E. In the derivation and application of vulnerability functions, assumptions concerning building code revisions and building code enforcement shall be justified.

The structural models include options that allow the representation of building code revisions. Three models were derived for each structural type: weak construction, medium construction, and strong construction (post-SSTD 10, “Standard for Hurricane Resistant Residential Construction”, deemed to comply standard). For example, the model for northern wood frame and gable roof homes has weak, medium and strong versions of that same model. The assignment of a given strength level is based on the assumed age of the home being modeled and the available information on construction practice in that region of the state in that era of construction. Florida Building Code requirements that apply to the repair of existing homes are also taken into consideration when computing the repair costs of a structure. Separate models were also developed for manufactured housing constructed based on pre and post 1994 HUD regulations, and for different wind zones. In addition to the various models that reflect construction type, region of Florida, and era of construction (weak, medium or strong construction), each model has numerous additional strength features that can be adjusted before simulations are conducted in order to represent various combinations of mitigation features. For example, a weak constructed home in central Florida with masonry walls (no reinforcing) may have been recently re-roofed with modern code approved shingles. The simulation model is capable of reflecting this combination of weak original construction with new strong roof cover mitigation.

F. Vulnerability functions shall be separately derived for building structures, mobile homes, appurtenant structures, contents, and additional living expenses.

This requirement is fully met. The building structures, mobile homes and appurtenant structures are independently derived. The contents and additional living expenses are separate vulnerabilities, which are functions of (receiving input from) the results of structure vulnerability simulations.

G. The minimum wind speed that generates damage shall be reasonable.

FPHLM V3.0 2008 129 The minimum one-minute average sustained wind speed at which some damage is observed is 38 mph (3 second gust 50 mph) for appurtenant structures. Site-built and manufactured homes have a very small probability of some very minor damage at 42 mph (3 second gust 55 mph). This probability becomes more significant at 46 mph (3 second gust 60 mph) and increases with higher wind speed. Simulations are run for 3-second gusts ranging from 50 mph to 250 mph.

FPHLM V3.0 2008 130 Disclosures

1. Provide a flow chart documenting the process by which the vulnerability functions are derived and implemented.

The flow chart in Figure 34 summarizes the procedure used in the Monte Carlo simulations to predict the external damage to the different structural types. The random variables include wind speed, pressure coefficients, and the resistances of the various building components (roof cover, roof sheathing, openings, walls, connections).

Selection of Structural Type, Definition of Geometry

Loop for Angle Define Angle of Incidence

Loop for Wind Speed Define Wind Speed

Loop for Building

Simulate an Individual Home: Randomize Wind Speed, Cp’s Sample Resistances

Calculate Initial Loads Repeat until all angles complete angles all until Repeat

Initial Failure Check: Sheathing, Openings, Walls

Calculate Internal Pressure from Opening Failures and Recalculate Structural Loads

Figure 34. Monte Carlo Simulation Procedure to Predict Damages

The flow chart in Figure 35 summarizes the procedure used to convert the results of the Monte Carlo simulations of physical external damage into a vulnerability matrix.

FPHLM V3.0 2008 131

Figure 35. Procedure to create vulnerability matrix

2. Describe the nature and extent of actual insurance claims data used to develop the model’s vulnerability functions. Describe in detail what is included, such as, number of policies, number of insurers, date of loss, and number of units of dollar exposure, separated into personal lines, commercial, and mobile home.

FPHLM V3.0 2008 132

At the request of the Florida Department of Financial Services (FDFS), four insurance companies provided insurance claims data for several hurricanes that impacted Florida prior to 2004, including Andrew. The companies provided two types of files: 1. Sample files with 10% of the exposure selected at random, plus the claims on this 10% exposure, since 1996. 2. Hurricane files with premium files for all hurricane claims since 1996, plus all the corresponding claim data since 1996 Because of a confidentiality agreement these companies will remain anonymous (they will be referred to as company A, B, C, and D). They represent between 75 and 85% of the insured exposure in the State, and approximately 70% of the claims. Most of the data provided comes only from minor hurricanes and tropical storms that impacted Florida between 1994 and 2002.

The only significant data for storms prior to 2004 was provided by company A, in particular for Hurricane Andrew. As shown in Table 7 and Figure 36, this data covers the complete range of structural and contents losses. Wind speed estimates are also available, so validation efforts were primarily concentrated on the use of this data. Attempts were made to make use of additional data from Hurricane Opal and other storms. However, the amount of processed data available was too small to be statistically significant for validation.

Table 7. Summary of processed claims data (number of claims provided) Tropical Tropical Hurricane Hurricane Hurricane Storm Storm Hurricane Andrew Georges Opal Irene Earl Erin Company A Concrete 78636 266 1973 3638 59 11460 Timber 1603 1078 9166 776 89 11878 Manufactured 1775 0 256 184 16 690

Note: Only building, contents, and appurtenant structure claims were provided by company A (ALE was not provided).

FPHLM V3.0 2008 133

Figure 36. Company A, Hurricane Andrew structure vs. contents losses

Claim data for the 2004 hurricane season, from a series of insurance companies, was also used to validate FPHLM. Although 21 companies submitted data with a total of almost 675,000 claims, only two main companies are detailed here. These two companies (they will be referred as Company 1 and Company 2) represent 386,000 claims, mainly for site-built homes. These claims are divided between hurricanes Charley, Frances, and Jeanne for central Florida, and hurricane Ivan for the Panhandle. The validation consists of a series of comparisons between the actual claim data and the FPHLM results. The damage from all these hurricanes was reported to the insurance companies (by the insurers), who provided the claims files. Table 8 to Table 10 give the number of policies provided by the two Companies for the four different hurricanes in 2004. As expected, there are more masonry claims in Central Florida, and more timber claims in the Panhandle.

FPHLM V3.0 2008 134 Table 8. Company 1: Claim Number for each year built category Actual Number of Company Hurricane Construction Year Built Claims Company 1 Charley Masonry yb<1970 5026 Company 1 Charley Masonry 1970<=yb<1984 8216 Company 1 Charley Masonry 1984<=yb<1994 11850 Company 1 Charley Masonry yb>=1994 8110 Company 1 Charley Frame yb<1970 956 Company 1 Charley Frame 1970<=yb<1984 1232 Company 1 Charley Frame 1984<=yb<1994 3044 Company 1 Charley Frame yb>=1994 677 Company 1 Charley Manufactured yb<1970 2966 Company 1 Charley Manufactured yb>=1994 212 Company 1 Frances Masonry yb<1970 5009 Company 1 Frances Masonry 1970<=yb<1984 6989 Company 1 Frances Masonry 1984<=yb<1994 7903 Company 1 Frances Masonry yb>=1994 4384 Company 1 Frances Frame yb<1970 902 Company 1 Frances Frame 1970<=yb<1984 2081 Company 1 Frances Frame 1984<=yb<1994 5648 Company 1 Frances Frame yb>=1994 721 Company 1 Frances Manufactured yb<1970 3186 Company 1 Frances Manufactured yb>=1994 222 Company 1 Ivan Masonry yb<1970 2029 Company 1 Ivan Masonry 1970<=yb<1984 2099 Company 1 Ivan Masonry 1984<=yb<1994 1719 Company 1 Ivan Masonry yb>=1994 1769 Company 1 Ivan Frame yb<1970 3048 Company 1 Ivan Frame 1970<=yb<1984 3956 Company 1 Ivan Frame 1984<=yb<1994 4829 Company 1 Ivan Frame yb>=1994 3890 Company 1 Ivan Manufactured yb<1970 634 Company 1 Ivan Manufactured yb>=1994 79 Company 1 Jeanne Masonry yb<1970 3601 Company 1 Jeanne Masonry 1970<=yb<1984 5274 Company 1 Jeanne Masonry 1984<=yb<1994 5698 Company 1 Jeanne Masonry yb>=1994 4999 Company 1 Jeanne Frame yb<1970 825 Company 1 Jeanne Frame 1970<=yb<1984 1386 Company 1 Jeanne Frame 1984<=yb<1994 3430 Company 1 Jeanne Frame yb>=1994 674 Company 1 Jeanne Manufactured yb<1970 2717 Company 1 Jeanne Manufactured yb>=1994 177

Table 9. Company 2: Claim Number for each year built category

FPHLM V3.0 2008 135 Actual Number of Company Hurricane Construction Year Built Claims Company 2 Charley Masonry yb<1970 8677 Company 2 Charley Masonry 1970<=yb<1984 15085 Company 2 Charley Masonry 1984<=yb<1994 18324 Company 2 Charley Masonry yb>=1994 6376 Company 2 Charley Frame yb<1970 1920 Company 2 Charley Frame 1970<=yb<1984 1782 Company 2 Charley Frame 1984<=yb<1994 3786 Company 2 Charley Frame yb>=1994 443 Company 2 Charley Manufactured yb<1970 1843 Company 2 Charley Manufactured yb>=1994 159 Company 2 Frances Masonry yb<1970 8276 Company 2 Frances Masonry 1970<=yb<1984 11978 Company 2 Frances Masonry 1984<=yb<1994 11394 Company 2 Frances Masonry yb>=1994 3224 Company 2 Frances Frame yb<1970 1453 Company 2 Frances Frame 1970<=yb<1984 3202 Company 2 Frances Frame 1984<=yb<1994 7731 Company 2 Frances Frame yb>=1994 601 Company 2 Frances Manufactured yb<1970 1590 Company 2 Frances Manufactured yb>=1994 131

Actual Number of Company Hurricane Construction Year Built Claims Company 2 Ivan Masonry yb<1970 1399 Company 2 Ivan Masonry 1970<=yb<1984 746 Company 2 Ivan Masonry 1984<=yb<1994 449 Company 2 Ivan Masonry yb>=1994 275 Company 2 Ivan Frame yb<1970 4004 Company 2 Ivan Frame 1970<=yb<1984 5546 Company 2 Ivan Frame 1984<=yb<1994 4637 Company 2 Ivan Frame yb>=1994 2229 Company 2 Ivan Manufactured yb<1970 171 Company 2 Ivan Manufactured yb>=1994 41 Company 2 Jeanne Masonry yb<1970 6907 Company 2 Jeanne Masonry 1970<=yb<1984 10767 Company 2 Jeanne Masonry 1984<=yb<1994 9629 Company 2 Jeanne Masonry yb>=1994 4176 Company 2 Jeanne Frame yb<1970 1555 Company 2 Jeanne Frame 1970<=yb<1984 2087 Company 2 Jeanne Frame 1984<=yb<1994 4561 Company 2 Jeanne Frame yb>=1994 484 Company 2 Jeanne Manufactured yb<1970 1401 Company 2 Jeanne Manufactured yb>=1994 128

FPHLM V3.0 2008 136 Table 10. Company 1 and Company 2: Claim Numbers Combined Actual Number of Company Hurricane Construction Claims Company 1 Charley Masonry 33202 Company 1 Charley Frame 5909 Company 1 Charley Manufactured 3178 Company 1 Charley Other 260 Company 1 Frances Masonry 24285 Company 1 Frances Frame 9352 Company 1 Frances Manufactured 3408 Company 1 Frances Other 566 Company 1 Ivan Masonry 7616 Company 1 Ivan Frame 15723 Company 1 Ivan Manufactured 713 Company 1 Ivan Other 100 Company 1 Jeanne Masonry 19572 Company 1 Jeanne Frame 6315 Company 1 Jeanne Manufactured 2894 Company 1 Jeanne Other 331 Company 2 Charley Masonry 48691 Company 2 Charley Frame 7981 Company 2 Charley Manufactured 2002 Company 2 Charley Other 582 Company 2 Frances Masonry 35036 Company 2 Frances Frame 13015 Company 2 Frances Manufactured 1721 Company 2 Frances Other 1134 Company 2 Ivan Masonry 2875 Company 2 Ivan Frame 16466 Company 2 Ivan Manufactured 212 Company 2 Ivan Other 87 Company 2 Jeanne Masonry 31705 Company 2 Jeanne Frame 8716 Company 2 Jeanne Manufactured 1529 Company 2 Jeanne Other 1167

The claims are divided by the type of coverage for structure and contents. Company 1 has two types of coverage, replacement cost, and actual cash value, without specifying whether both structure and contents have the same coverage for each claim.

For company 2, there are 6 types of coverage as shown below.

ACV S/ACV C Structure Actual-Cash-Value, Contents Actual-Cash-Value ACV S/RC C Structure Actual-Cash-Value, Contents Replacement-Cost RC S/ACV C Structure Replacement-Cost, Contents Actual-Cash-Value RC S/RC C Structure Replacement-Cost, Contents Replacement-Cost

FPHLM V3.0 2008 137 SV S/RC C Structure Stated-Value, Contents Replacement-Cost SV S/SV C Structure Stated-Value, Contents Stated-Value

Table 11 and Table 12 summarize the distribution of claims in both companies.

Table 11. Distribution of coverage for Company 1

Premium Policy Coverage Count Claim Policy Count A 44020 1% 2759 2% R 3706219 99% 163692 98% Total 3750240 166451

Table 12. Distribution of coverage for Company 2

Premium Policy Claim Policy Coverage Count Count ACV S/ACV C 13173 3% 3496 3% ACV S/RC C 44805 10% 12150 9% RC S/ACV C 162122 35% 41484 30% RC S/RC C 232688 51% 77146 57% SV S/RC C 235 0% 69 0% SV S/SV C 6019 1% 1717 1% Total 459042 100% 136062 100%

There are 29,372 claims with $0 losses (i.e., Loss structure+Loss app+Loss contents + Loss ALE = 0) though they are listed in the claim file of company 2. They probably correspond to claims whose losses were lower than the deductible.

3. Summarize site inspections, including the source, and a brief description of the resulting use of these data in development, validation, or verification of vulnerability functions.

Several damage surveys were done in 2004. Damage from Charley was reported all across the state with the most severe being where the eye made landfall near the cities of Punta Gorda and Port Charlotte. The extent of the structural damage to homes and manufactured homes in these cities was surveyed by a team that consisted of approximately 30 members from UF, FIU, Clemson, and FIT and was conducted under the leadership of the Institute for Business and Home Safety (IBHS). For several days following the storm the team conducted a detailed statistical survey of damage in the impacted areas. Results of this survey can be found on the IBHS website http://www.ibhs.org/, and other information regarding the damage of Charley and other storms can be found at the Florida Tech Wind and Hurricane Impact Research Laboratory website, http://www.fit.edu/research/whirl/.

Damage from Frances was surveyed in areas from Cocoa Beach to Stuart in eastern FL. Although damage from Frances was not as severe as from Charley, the same extensive survey

FPHLM V3.0 2008 138 conducted in Punta Gorda and Port Charlotte was also conducted in the impacted areas. Great efforts were made to monitor the strength and resulting damage from the storm, as part of the Florida Coastal Monitoring Program. Towers were set up to record wind speeds along the coast in locations where the storm was forecast to make landfall. Sensors to record the wind induced pressure were deployed on the roofs of several homes. Following the storm, members of the same team that surveyed Charley photographed and recorded damage throughout the area. Areas of Fort Pierce appeared to be hardest hit with severe damage to many homes in some areas.

Similar efforts to monitor the winds and survey the damage were made for Jeanne. Towers and pressure sensors were again deployed at various locations near where landfall was forecast. Following the storm members of the team surveyed areas from Stuart to Cocoa Beach. These surveys consisted primarily of cataloging and photographing various observations of damage in the impacted areas, as was done with Frances. Damage from Hurricane Jeanne in many locations was very similar to what was seen from Hurricane Frances. In many cases damage to structures that was initially caused by Frances was compounded by Jeanne. Fatigue of structures from the winds of two hurricanes within three weeks most likely played a roll in the most severe cases of damage in the areas such as Vero Beach and Fort Pierce. In some areas most of the weak trees and components of homes (shingles, screened porches, fences, etc.) were already damaged by Frances, so when Jeanne hit little or no further damage was seen. It is very difficult to tell what damage was caused by Jeanne and what was caused by Frances.

Additionally, engineers working on the physical damage model performed a detailed residential damage study after the 2004 hurricane season in order to assess the performance of housing built to the Florida Building Code and the Standard Building Code. The data was collected as a part of a study conducted by UF and sponsored by the Florida Building Commission. Site built single family homes constructed after Andrew-related changes to the standard building code were in effect were targeted for a detailed investigation of damage as a result of the 2004 hurricane season. The purpose of this study was to provide a quantitative statistical comparison of the relative performance of homes built between 1994 and 2001 with those built after the 2001 Florida Building Code replaced the Standard Building Code. This evaluation was accomplished through a systematic survey of homes built from 1994 to 2004 in the areas that experienced the highest wind speeds from the 2004 storms (Charlotte, St. Lucie, Escambia and Santa Rosa counties). A statistically significant number of homes were surveyed (close to 200) in these regions in order to define correlations between damage, age and construction type. These relationships are referenced to maximum 3-second gust wind speed via wind swath maps. The data from this study was used to modify the residential component capacities as this model evolved. The final report from this study was submitted in the spring of 2006 to the Florida Building Commission.

Another source of field data is the aerial imagery collected by NOAA after Hurricane Katrina. These images provided a quantification of shingle damage relative to estimated wind speed, and were used to validate the roof cover damage output from the physical damage model.

4. Describe the research used in the development of the model’s vulnerability functions

FPHLM V3.0 2008 139

The engineering team adopted a so-called component approach in the development of the vulnerability functions. Although a number of commercial loss projection models have been developed, only a handful of studies are available in the public domain to predict damage for hurricane prone areas. Boswell et al. (1999) attempted to predict the public costs of emergency management and recovery, without taking into account losses to individual homeowners. In 1985, Berke, et al., presented a computer system simulating economical and social losses caused by hurricane disasters, and a Vulnerability Assessment and Mapping System known as VAMS (Berk, Larsen and Ruch, 1984) enabled the user to consider various types of hurricanes with varying surge, wind pattern and point of landfall. This information is of some interest, but it is not directly applicable to residential construction in Florida.

Most studies for residential losses use post-disaster investigations (FEMA, 1993) or available claim data to fit damage versus wind speed vulnerability curves. For example, a relationship between home damage from insurance data and wind speed was proposed for Typhoons Mireille and Flo (Mitsuta et al. 1996). A study by Holmes (1996) presents the vulnerability curve for a fully engineered building with strength assumed to have lognormal distribution, but clearly indicates the need for more thorough post-disaster investigations to better define damage prediction models. A method for predicting the percentage of damage within an area as a function of wind speed and various other parameters was presented by Sill and Kozlowski (1997). The proposed method was intended to move away from curve fitting schemes, but its practical value is hampered by insufficient clarity and transparency. Huang et al. (2001) presented a risk assessment strategy based on an analytical expression for the vulnerability curve. The expression is obtained by regression techniques from insurance claim data for hurricane Andrew. Khanduri et al. (2003) also presented a similar method of assessment of vulnerability and a methodology to translate a known vulnerability curves from one region to another region. Although such approaches are simple, they are highly dependent on the type of construction and construction practices common to the areas represented in the claim data. Recent changes in building codes or construction practices cannot be adequately reflected by Huang et al.’s vulnerability curve. In addition, damage curves obtained by regression from observed data can be misleading, because very often, as was the case for hurricane Andrew, few reliable wind speed data are available. In addition, damage curves regressed from observed data do not adequately represent the influence of primary storm characteristics such as central pressure, forward velocity, radius of maximum wind, the amount of rain, duration, and other secondary parameters such as demand surge and preparedness.

In contrast, a component approach explicitly accounts for both the resistance capacity of the various building components and the load effects produced by wind events to predict damage at various wind speeds. In the component approach the resistance capacity of a building can be broken down into the resistance capacity of its components and the connections between them. Damage to the structure occurs when the load effects from wind or flying debris are greater than the component’s capacity to resist them. Once the strength capacities, load demands, and load path(s) are identified and modeled, the vulnerability of a structure at various wind speeds can be estimated. Estimations are affected by uncertainties regarding on one hand the behavior and strength of the various components and, on the other, the load effects produced by hurricane

FPHLM V3.0 2008 140 winds. A hurricane wind damage prediction model that incorporates a time-stepping component approach was implemented for the FEMA HAZUS project (Lavelle et al., 2003).

5. Describe the number of categories of the different vulnerability functions. Specifically, include descriptions of the structure types, lines of business, and coverages in which a unique vulnerability function is used.

Vulnerability matrices were derived for both manufactured and site built homes.

Table C.1, in Appendix C of Volume III of the Engineering Team final report list the 216 vulnerability matrices developed for site built homes. They correspond to a combination of region (Keys, South, Central, or North), structural type (concrete masonry or wood frame), roof type (gable or hip), roof cover type (tile or shingle), opening protection (with or without shutters), number of stories (1 or 2) and location (wind borne debris region, high velocity hazard, or else).

These 216 matrices were then combined in 30 weighted matrices listed in table 5-4, of the same document. The detail of the weighted procedure is given in section 5.2, of Volume III.

The entire process for site built homes was repeated for weak, medium strength, and strong homes corresponding to homes built in different eras with different building codes, and different building code enforcement.

Four vulnerability matrices were developed for manufactured homes: pre-1994 tied down, pre- 1994 not tied down, post 1992 zone 2, and post-1994 zone 3. The pre-1994 tied down and pre- 1994 not tied down were then combined in a weighted pre-94 matrix. The detail of the weighted procedure is also given in section 5.2, of Volume III.

A contents matrix and an ALE matrix correspond to each structure vulnerability matrix. Finally, one appurtenant structure vulnerability matrix was derived for all site-built homes, and one for all manufactured homes.

6. Identify the one-minute average sustained wind speed at which the model begins to estimate damage.

The wind speeds used in the damage model are 3-second gusts. The lowest 3-second gust is 50 mph. The minimum one-minute sustained wind is approximately 40 mph.

7. Describe how the duration of wind speeds at a particular location over the life of a hurricane is considered.

Duration of the storm is not explicitly modeled. The damage accumulation procedures assume sufficient duration of peak loads to account for duration dependent failures.

8. Provide a completed Form V-1, One Hypothetical Event.

FPHLM V3.0 2008 141 See attached form, the results of which are plotted below for building damage (Figure 37) The model computes the damage based on 3 sec gusts. The losses are then aggregated once according to the 3 sec gusts. At each zip code, the 3 sec gusts are then converted into 1 min sustained winds, and the losses are re-aggregated among the zip codes with the same 1 min sustained winds. The conversion from 3 sec gust to 1 min sustained depends on the roughness of the zip code. Because all the zip codes do not have the same roughness, identical 3 sec gusts results in different 1 min winds depending on location. The bump in the 1 min sustained winds plot is due to this process of conversion and re-aggregation. The modelers do confirm that the structures used in completing the form are identical to those in the table provided in the Standard.

FPHLM V3.0 2008 142 V-2 Mitigation Measures* (*Significant Revision due to new Audit language)

A. Modeling of mitigation measures to improve a structure’s wind resistance and the corresponding effects on vulnerability shall be theoretically sound. These measures shall include fixtures or construction techniques that enhance:

Roof strength Roof covering performance Roof-to-wall strength Wall-to-floor-to-foundation strength Opening protection Window, door, and skylight strength.

Modeling of mitigation measure to improve a structure’s wind resistance is theoretically sound and includes the fixtures mentioned above. The following structures were modeled:

Base case as defined by Commission Mitigated case as defined by Commission Base plus one mitigation at a time

The mitigations included gable bracing, rated shingles, stronger sheathing capacity, stronger roof to wall connections, stronger wall to sill connections, masonry reinforced walls, multiple opening protection options, and wind / missile resistant glass.

B. Application of mitigation measures shall be empirically justified both individually and in combination.

The base cases are very weak cases, where the interior damage is governed by the sheathing loss at low to moderate wind speeds. The application of mitigation measures is justified and the results show the following.

Bracing the gable end or using rated shingles alone does not provide any benefit in the context of weak roof sheathing connections. Regardless of the type of roof cover used, if the home loses its sheathing panels, there will be no benefit in mitigating the roof cover or gable end alone. The observed negative values in Form V-2 corresponding to the braced gable end mitigation are from round off of smaller values within the uncertainty scatter of the model, and indicate zero change.

The hip roof has a greater impact in reducing the losses, especially in the case of frame structures. Because the base frame structure is inherently weaker, there is comparatively a higher gain with the hip timber structure than with the hip masonry structure.

Improving the roof sheathing capacity (8d nails) alone reduces the damage at wind speeds up to 100 and 120 mph sustained winds for wood and masonry structures, respectively, but at higher

FPHLM V3.0 2008 143 wind speeds the mitigation becomes counter effective ( Figure 38 and Figure 39). The behavior of the damage curve with mitigated sheathing after 100 (wood) and 130 (masonry) mph sustained winds is due to the still very weak roof to wall connections. Loss of sheathing reduces the uplift on the roof to wall connections. Thus the stronger deck results in higher loads on the connections, which they are not prepared to absorb.

Clips and straps are very effective for frame structures, less so for masonry structures. The model emphasizes interior damage due to loss of sheathing, roof cover or gable end, which are all independent of the roof to wall connection strength. If the strength of the plywood deck and roof cover are not increased, increasing the roof to wall connections alone will do little good at low to moderate wind speeds. At higher wind speeds, the integrity of the box system in the frame structure is improved by the stronger roof to wall connection. Hence, a more pronounced benefit than for masonry.

Clips and straps for wall to sill plate are very effective at high wind speeds for frame structures. They improve the integrity of the box system. Similarly the reinforcing of the walls for masonry structures is more effective at high wind speeds when un-reinforced walls become vulnerable.

Opening protections are effective, and more so at higher wind speeds.

A mitigated structure with a combination of individual mitigations (as per standards definition) shows improved performance over the base structure and each of the individual mitigations.

The non-zero damage between 40 and 60 mph sustained winds, and the convergence of the base and all mitigation cases in this wind speed range, reflects the incorporation of non-exterior damage related losses in the model. Water penetration through windows and doors is possible even without window or door breach. This portion of the model is not dependent upon mitigations, thus the convergence of curves in Figure 38 and Figure 39 in that wind speed range.

Disclosures

1. Provide a completed Form V-2, Mitigation Measures – Range of Changes in Damage.

See Form V-2. Notice that there are no entries for Wall-Foundation Strength rows for timber structures, because the model does not have the capability to model wall to foundation anchors or straps for timber structures. The model does account for wall-to-sill plate connections, but not the sill plate-to-foundation. There is simply no field data to indicate that this is a failure mode that is significant. The connection to the foundation can be weak and is reflected in the wall-to- sill capacity (toe-nails, clips, straps).

2. Provide a description of the mitigation measures used by the model that are not listed in Form V-2.

FPHLM V3.0 2008 144 The model does not use mitigation measures not listed in Form V-2, except for stronger garage doors.

3. Describe how mitigation is implemented in the model. Identify any assumptions.

The various mitigation options delineated in V-2 and V-3 are implemented in the model by varying the capacity statistics (mean and coefficient of variation) to reflect the strength of a given component. For example, weak (base model) roof covering is represented by a random value for each shingle, with the specific capacity values for a given Monte Carlo simulation randomly assigned based on a specified probability density function, mean and coefficient of variation assigned to weak shingles. If the strong roof cover mitigation option is chosen, a different mean and coefficient of variation, reflecting higher capacity and less variability, is used to randomly assign capacities to the shingles. This same approach is used for every component for which a mitigation option is modeled. One or any combination of mitigation measures may be selected prior to running the Monte Carlo simulation. The stronger resistances of the mitigated components are directly reflected in the randomly assigned capacities of those components, and the un-mitigated components are assigned capacities based on distributions with parameters reflecting weaker resistance.

Each mitigation option (e.g. sheathing, shingles, roof to wall connections) is modeled and accounted for independently, allowing any combination to be chosen. As reflected in the results in Figure 38 and Figure 39, it is assumed that the effect of mitigating one component can change the vulnerability, but not the capacity, of other components via the influence that mitigation has on loading or load sharing. It is also assumed that any given mitigation does not necessarily produce improved overall performance for all wind speeds. An example already discussed is the influence of the roof sheathing strength on the vulnerability of roof to wall connections, caused by the influence of intact strong roof sheathing on the uplift acting on weak roof to wall connections. Another example is the influence of opening vulnerability to the performance of many other components, including walls, sheathing and roof to wall connections, as the change in internal pressure resulting from opening failure changes the overall loading on these other components.

In summary, mitigation options may be selected individually or in combination, but the effects of a given mitigation on other components, and on overall building vulnerability, should not be and is not isolated in the model.

FPHLM V3.0 2008 145 Form V-1

70.0% 60.0% Building damage only 50.0% 40.0% 30.0% 20.0% Damage ratio Damage 10.0% 0.0% 40 65 90 115 140 165 Wind speed mph

Figure 37. Building damage vs. 1 minute sustained wind speed

FPHLM V3.0 2008 146 Form V-1: One Hypothetical Event

A. Wind-speeds for 336 ZIP Codes are provided in the file named “FormV1Input07.xls.” The wind-speeds and ZIP Codes represent a hypothetical hurricane track. The modeler is instructed to model the sample exposure data provided in the file named “FormA1Input07.xls” against these wind-speeds at the specified ZIP Codes and provide the damage ratios summarized by wind-speed (mph) and construction type.

The wind-speeds provided are one-minute sustained 10-meter wind-speeds. The sample exposure data provided consists of three structures (one of each construction type – wood frame, masonry, and mobile home) individually placed at the population centroid of each of the ZIP Codes provided. Each ZIP Code is subjected to a specific wind speed. For completing Part A, Estimated Damage for each individual wind speed range is the sum of Ground Up Loss to all structures in the ZIP Codes subjected to that individual wind speed range, excluding demand surge and storm surge. Subject Exposure is all exposures in the ZIP Codes subjected to that individual wind speed range. For completing Part B, Estimated Damage is the sum of the loss to all structures of a specific type (wood frame, masonry, or mobile home) in all of the wind speed ranges. Subject Exposure is all exposures of that specific type in all of the ZIP Codes.

One reference structure for each of the construction types shall be placed at the population center of the ZIP Codes. Do not include contents, appurtenant structures, or ALE.

Reference Frame Structure: Reference Masonry Structure: One story One story Unbraced gable end roof Unbraced gable end roof Normal shingles (55mph) Normal shingles (55mph) ½” plywood deck ½” plywood deck 6d nails, deck to roof members 6d nails, deck to roof members Toe nail truss to wall anchor Toe nail truss to wall anchor Wood framed exterior walls Masonry exterior walls 5/8” diameter anchors at 48” centers for No vertical wall reinforcing wall/floor/foundation connections No shutters No shutters Standard glass windows Standard glass windows No door covers No door covers No skylight covers No skylight covers Constructed in 1980 Constructed in 1980 Reference Mobile Home Structure: Tie downs Single unit Manufactured in 1980

B. Confirm that the structures used in completing the Form are identical to those in the above table. If additional non-structural assumptions are necessary to complete this Form (for example, regarding duration or surface roughness), the modeler shall provide the reasons why the assumptions were necessary as well as a detailed description of how they were included.

FPHLM V3.0 2008 147

The modelers do confirm that the structures used in completing the form are identical to those in the table provided in the Standard.

C. Provide a plot of the Form V-1, Part A data.

See Figure 37.

FPHLM V3.0 2008 148 Form V-1: One Hypothetical Event

Part A

Wind Speed (mph ) Estimated Damage/ 1 min sustained Subject Exposure Wind 41-50 0.0% 51-60 1.2% 61-70 4.4% 71-80 6.0% 81-90 9.6% 91-100 13.9% 101-110 20.0% 111-120 35.2% 121-130 33.4% 131-140 49.6% 141-150 54.7% 151-160 58.0% 161-170 64.4%

Part B

Construction Estimated Damage/ Type Subject Exposure Wood Frame 4.8% Masonry 3.9% Mobile Home 10.6%

The structures used in completing the form are identical to those in the table provided.

FPHLM V3.0 2008 149 Form V-2: Mitigation Measures – Range of Changes in Damage

A. Provide the change in the zero deductible personal residential reference structure damage rate (not loss cost) for each individual mitigation measure listed in Form V-2 as well as for the combination of the four mitigation measures provided for the Mitigated Frame Structure and the Mitigated Masonry Structure below.

See Form V-2 below.

B. If additional assumptions are necessary to complete this Form (for example, regarding duration or surface roughness), the modeler shall provide the rationale for the assumptions as well as a detailed description of how they are included.

C. Provide this Form on CD in both Excel and PDF format. The file name shall include the abbreviated name of the modeler, the Standards year, and the Form name. A hard copy of Form V-2 shall be included in the submission.

Mitigated Frame Structure: Mitigated Masonry Structure: Rated shingles (110mph) Rated shingles (110mph) 8d nails, deck to roof members 8d nails, deck to roof members Truss straps at roof Truss straps at roof Plywood Shutters Plywood Shutters

FPHLM V3.0 2008 150 Reference and mitigated structures are $100,000 fully insured structures with a zero deductible policy as indicated under “Owners” Policy Type for Form A-6.

Place the reference structure at the population centroid for ZIP Code 33921 located in Lee County.

Windspeeds used in the Form are one-minute sustained 10-meter windspeeds.

FPHLM V3.0 2008 151 Form V-2: Mitigation Measures – Range of Changes in Damage PERCENTAGE CHANGES IN DAMAGE (REFERENCE DAMAGE RATE - MITIGATED DAMAGE RATE)/(REFERENCE DAMAGE RATE)*100 INDIVIDUAL MITIGATION MEASURES FRAME STRUCTURE MASONRY STRUCTURE WIND SPEED (MPH) WIND SPEED (MPH) 60 85 110 135 160 60 85 110 135 160 REFERENCE STRUCTURE ------

ROOF STRE BRACED GABLE ENDS 0% -1% 0% 1% 1% 0% 0% 0% 0% 1% NGTH HIP ROOF 1% 11% 16% 17% 9% 1% 8% 3% 12% 15%

ROOF RATED SHINGLES (110 MPH) 0% 0% 0% 0% 0% -1% 1% 0% 0% 0% COVE MEMBRANE RING NAILING OF DECK 8d 3% 33% -1% -5% -1% 2% 37% 9% -8% -2%

ROOF- WALL STREN CLIPS 0% 4% 16% 20% 12% 0% 1% 2% 12% 18% GTH STRAPS 0% 5% 17% 30% 28% -1% 0% 2% 13% 22%

WALL- FLOOR TIES OR 0% 0% 8% 5% 2% - - - - - STREN CLIPS GTH STRAPS 0% 0% 11% 10% 4% - - - - - WALL FOUN LARGER ANCHORS DATIO ------N OR CLOSER SPACING STRE STRAPS ------NGTH VERTICAL REINFORCING - - - - - 0% -1% 0% 11% 22%

PLYWOOD 0% 1% 5% 3% 1% 0% 2% 6% 3% 2% WINDOW OPENI STEEL 0% 2% 7% 4% 3% 0% 3% 9% 6% 3% NG SHUTTERS PROTE ENGINEERED 0% 3% 10% 7% 5% 0% 3% 12% 8% 5% CTION DOOR AND SKYLIGHT 0% 0% 1% 0% 0% 0% 1% 1% 0% 0% COVERS

WIND OW DOOR LAMINATED 0% 1% 8% 7% 4% 0% 2% 9% 8% 5% , WINDOWS IMPACT SKYLI 0% 2% 9% 9% 7% 0% 3% 11% 11% 8% GHT GLASS STRE NGTH PERCENTAGE CHANGES IN DAMAGE (REFERENCE DAMAGE RATE - MITIGATED DAMAGE RATE)/(REFERENCE DAMAGE RATE)*100 MITIGATION MEASURES IN FRAME STRUCTURE MASONRY STRUCTURE COMBINATION WIND SPEED (MPH) WIND SPEED (MPH) 60 85 110 135 160 60 85 110 135 160 STRU CTUR MITIGATED STRUCTURE 3% 50% 42% 39% 32% 2% 48% 30% 23% 25% E

FPHLM V3.0 2008 152 Form V-3: Mitigation Measures – Mean Damage Ratio

A. Provide the mean damage ratio to the reference structure for each individual mitigation measure listed in Form V-3 as well as the percent damage for the combination of the four mitigation measures provided for the Mitigated Frame Structure and the Mitigated Masonry Structure below.

See Form V-3 below. Notice that for the 60 mph column all the vulnerabilities coincide at 6%. This is because at this low wind speeds, no significant damage is activated to trigger any significant difference between the different cases.

C. Provide a graphical representation of the vulnerability curves for the reference structure and the fully mitigated structure.

See Figure 38 and Figure 39. Because there are too many vulnerability curves to plot in one figure, for the sake of clarity, the mitigations were divided in 2 sets for both masonry and frame structures. In each figure, there are two horizontal axes: the upper axis represents the actual terrain 3 sec gust winds; the lower axis represents the actual terrain one minute sustained winds. The conversion between 3 sec gust and 1 min sustained depends on the roughness of the terrain. Therefore, on each plot, the value of the roughness parameter for Lee County is indicated. Finally, please, note that, as explained in the previous section, mitigating the roof shingles alone, without mitigating the roof deck or the roof to wall connections, does not improve the overall vulnerability of the structure. Consequently, in Figures 38a and 39b, the curves for the base case and the rated shingle case are superimposed on each other.

FPHLM V3.0 2008 153 Reference Frame Structure: Reference Masonry Structure: One story One story Unbraced gable end roof Unbraced gable end roof Normal shingles (55mph) Normal shingles (55mph) ½” plywood deck ½” plywood deck 6d nails, deck to roof members 6d nails, deck to roof members Toe nail truss to wall anchor Toe nail truss to wall anchor Wood framed exterior walls Masonry exterior walls 5/8” diameter anchors at 48” centers for No vertical wall reinforcing wall/floor/foundation connections No shutters No shutters Standard glass windows Standard glass windows No door covers No door covers No skylight covers No skylight covers Constructed in 1980 Constructed in 1980

Reference and mitigated structures are $100,000 fully insured structures with a zero deductible policy as indicated under “Owners” Policy Type for Form A-6.

Place the reference structure at the population centroid for ZIP Code 33921 located in Lee County.

Windspeeds used in the Form are one-minute sustained 10-meter windspeeds.

FPHLM V3.0 2008 154 Form V-3: Mitigation Measures – Mean Damage Ratio MEAN DAMAGE RATIO

INDIVIDUAL MITIGATION MEASURES FRAME STRUCTURE MASONRY STRUCTURE WIND SPEED (MPH) WIND SPEED (MPH) 60 85 110 135 160 60 85 110 135 160 REFERENCE STRUCTURE 6% 13% 38% 59% 73% 6% 12% 30% 44% 66%

ROOF STRE BRACED GABLE ENDS 6% 14% 38% 59% 72% 6% 12% 30% 44% 65% NGTH HIP ROOF 6% 12% 32% 49% 66% 6% 11% 29% 39% 56%

RATED SHINGLES (110 MPH) 6% 13% 38% 59% 73% 6% 12% 30% 44% 66% ROOF COVERI MEMBRANE NG NAILING OF DECK 8d 6% 9% 38% 62% 73% 6% 8% 27% 48% 67%

ROOF- WALL STREN CLIPS 6% 13% 32% 47% 64% 6% 12% 29% 39% 53% GTH STRAPS 6% 13% 32% 42% 52% 6% 12% 29% 39% 51%

WALL- FLOOR TIES OR 6% 13% 35% 56% 71% - - - - - STREN CLIPS GTH STRAPS 6% 13% 34% 53% 70% - - - - - WALL FOUN LARGER ANCHORS DATIO ------N OR CLOSER SPACING STRE STRAPS ------NGTH VERTICAL REINFORCING - - - - - 6% 12% 30% 40% 51%

PLYWOOD 6% 13% 36% 57% 72% 6% 12% 28% 43% 64% OPENIN WINDOW G STEEL 6% 13% 35% 56% 71% 6% 12% 27% 42% 63% PROTE SHUTTERS CTION ENGINEERED 6% 13% 34% 55% 69% 6% 12% 26% 41% 62% DOOR AND SKYLIGHT COVERS 6% 13% 38% 59% 72% 6% 12% 30% 44% 65%

WIND OW DOOR, WINDOWS LAMINATED 6% 13% 35% 55% 69% 6% 12% 27% 41% 62% SKYLI IMPACT GLASS 6% 13% 35% 54% 67% 6% 12% 27% 40% 60% GHT STRE NGTH MEAN DAMAGE RATIO MITIGATION MEASURES IN FRAME STRUCTURE MASONRY STRUCTURE COMBINATION WIND SPEED (MPH) WIND SPEED (MPH) 60 85 110 135 160 60 85 110 135 160 STRU CTUR MITIGATED STRUCTURE 6% 7% 22% 36% 50% 6% 6% 21% 34% 49% E

FPHLM V3.0 2008 155 Vulnerability Curves for Reference Masonry Structure - Mitigation set 1

actual terrain 3 sec gust wind speeds 50 70 90 110 130 150 170 190 210 230 250 90.0%

80.0% Base Braced Gable Lee County z0 = 0.17125 70.0% Rated shingles 8d sheat nails 60.0% R2w clips R2w straps Reinfor mas 50.0% Mitig Struct

40.0%

Damage Ratio 30.0%

20.0%

10.0%

0.0% 39 59 79 99 119 139 159 179 actual terrain 1 min sustained wind speeds (a)

Vulnerability Curves for Reference Masonry Structures - Mitigation set 2

actual terrain 3 sec gust wind speeds 50 70 90 110 130 150 170 190 210 230 250 90.0% Base 80.0% Hip roof Shutt/ply Lee County z0 = 0.17125 70.0% Shutt/steel Shutt/Egnrd 60.0% Door cover Lam Glass 50.0% Impact Glass Mitig Struct 40.0%

Damage Ratio 30.0%

20.0%

10.0%

0.0% 39 59 79 99 119 139 159 179 actual terrain 1 min sustained wind speeds (b) Figure 38. Mitigation measures for masonry homes

FPHLM V3.0 2008 156 Vulnerability Curves for Reference Frame Structure - Mitigation set 1

actual terrain 3 sec gust wind speeds 50 70 90 110 130 150 170 190 210 230 250 90 0% Base

80 0% Braced gable Rated shingles Lee County z0 = 0.17125 8d sheath nails 70 0% R2w clips R2w straps 60 0% Sill clips Sill straps 50 0% Mitig Struct

40 0% Damage Ratio

30 0%

20 0%

10 0%

0 0% 39 59 79 99 119 139 159 179 actual terrain 1 min sustained wind speeds (a)

Vulnerability Curves for Reference Frame Structure - Mitigation set 2

actual terrain 3 sec gust wind speeds 50 70 90 110 130 150 170 190 210 230 250 90.0% Base 80.0% Hip roof Shutt/Ply Lee County z0 = 0.17125 70.0% Shutt/Steel Shutt/Egnrd 60.0% Door cover Lam Glass 50.0% Impact Glass Mitig Struct 40.0%

Damage Ratio 30.0%

20.0%

10.0%

0.0% 39 59 79 99 119 139 159 179 actual terrain 1 min sustained wind speeds

(b) Figure 39. Mitigation measures for frame homes

FPHLM V3.0 2008 157 ACTUARIAL STANDARDS

A-1 Modeled Loss Costs

Modeled loss costs shall reflect all damages from storms that reach hurricane strength and produce minimum damaging wind-speeds or greater on land in Florida.

Modeled loss costs are computed for all hurricanes that affect the State of Florida. Damages are computed for affected land areas in which wind speeds exceed a minimum level.

Disclosure

1. Describe how damage from model generated storms (land falling and by-passing) is excluded or included in the calculation of loss costs for the state of Florida.

Damages are computed for all Florida land falling and certain bypassing storms in the stochastic set that attain hurricane level wind speeds. The following bypassing hurricanes are included:

-Non-land falling hurricanes in regions A, B, C, D, E or F with open terrain winds greater than 30 mph in at least one Florida zip code.

-Land falling hurricanes in regions E or F with open terrain winds greater than 30 mph in at least one Florida zip code.

The Actuaries checked a sample of bypassing storms from the stochastic set to see if they were correctly included or excluded. From the file for each storm they could see the wind speed by zip code for each day and hour of the storm. The storms they selected from the excluded set either had no Florida zip codes impacted, or had Florida zip codes with very low wind speeds. The included storms selected all had one or more Florida zip codes with winds over 30 mph.

FPHLM V3.0 2008 158 A-2 Underwriting Assumptions

A. When used in the modeling process or for verification purposes, adjustments, edits, inclusions, or deletions to insurance company input data used by the modeler shall be based upon accepted actuarial, underwriting, and statistical procedures.

Input data from insurance companies, used for development or validation, were requested and provided in a standardized format. Any adjustments, edits, inclusions or deletions are based upon accepted actuarial, underwriting, and statistical procedures.

Exposure and claim data used in the validation process were collected via a data call issued by the Florida Office of Insurance Regulation. In the case of the 2004 hurricanes, the data call requested policies in force on any of four specific dates in 2004 (i.e. the landfall dates for Frances, Charley, Ivan and Jeanne), and any associated wind claims occurring on those dates. Since the four storms occurred so closely together, the OIR determined there was no need for companies to supply four separate in force files. Several companies supplied separate in force files for each date anyway, but most companies sent the one in force file requested.

All of the data files received were edited for:

-duplicate records -valid entries in each field.

Deletions:

A few duplicate records were found and deleted. In the review process, the Actuaries located and examined several sets of records where a duplicate had been purged from the original data set. These were cases where the policy number and all policy level characteristics were identical. In the process of reviewing these policies, they noted that there did appear to be some policies with legitimate multiple dwellings related to the same policy number, i.e. dwellings with different construction years, coverage A amount of insurance, etc. Therefore it’s possible that some or all of the “duplicates” dropped were not actually duplicates, but a second dwelling with the same policy number.

Policies (and any accompanying claims) with invalid Florida zip codes were also deleted. The Actuaries examined the invalid zip codes found in the Company D data set. They sampled among the deleted zip codes, especially those with multiple policies reported, and checked against their own list of valid Florida zips to ascertain that they were indeed invalid. Only .05% of Company D records and .8% of Company A records were dropped due to invalid zip code. The dropped record count for other companies was reportedly similar to Companies A and D.

Policies excluding wind were dropped. These were reported by two companies even though the data call requested only policies including wind coverage. Such policies were identified by “Ex Wind” or similar identifiers in the hurricane deductible field.

FPHLM V3.0 2008 159 Adjustments

The construction categories and deductible categories reported by the individual companies were mapped to those of the model. The engineers determined how company construction categories should be mapped. The Actuaries examined the mappings of a sample of policies to verify that they were executed as planned.

Percentage deductibles were converted to dollar amounts by multiplying by the structure amount of insurance. The Actuaries examined a sample of these conversions.

Companies B and D did not provide limits for Additional Living Expense in their in force files. Company B subsequently recommended 30% of structure amount of insurance as a reasonable estimate of the limit. For Company D a limit of 10% of structure amount of insurance was selected. This selection was later verified as reasonable based on exposure data submitted with a 2006 Company D rate filing. In that filing the average ALE limit reported by the company was 10.5%. The Actuaries reviewed the correspondence with Company B, and verified the ratio of ALE limit to structure limit from the 2006 Company D rate filing data file.

The edits, deletions and adjustments described above relate to the validation data. The same approach, though, is used with exposures provided by companies in conjunction with rate filings pending with the OIR.

B. For loss cost estimates derived from or validated with historical insured hurricane losses, the assumptions in the derivations concerning (1) construction characteristics, (2) policy provisions, (3) claim payment practices, and (4) relevant underwriting practices underlying those losses, as well as any actuarial modifications, shall be appropriate.

The damages calculated by the model, that subsequently flow into the loss costs, depend on the following characteristics of each exposure:

-Region/Sub-region and zip code -Construction type (Masonry, Frame, Mobile Home, Other) -Year of Construction -Coverages (e.g. contents only or full package homeowners) -Deductible -Limits by coverage.

The following assumptions are implicit in the design of the model:

-Each structure can be appropriately categorized as either Masonry, Frame, Mobile Home or Other. -Within construction types, the relative strength of an exposure can be approximated by the year of construction.

FPHLM V3.0 2008 160 -The values of structures, contents and appurtenant structures are each equal to their policy limit. -There is no difference in loss under Actual Cash Value or Replacement Cost coverage. (The damage model is calibrated to a mix of ACV and RC, but mostly RC.) -Claim practices are stable and do not vary by company. -A company’s underwriting practices relating to any other risk characteristic not considered in the model (i.e. those listed above) will not impact hurricane damages. -The impact on losses of roof type, shutters and other risk characteristics not yet widely available from insurance companies can be approximated using weighted damage matrices.

In responding to this standard the Actuaries reviewed model flow charts, manual calculations of losses for specific policies, and sample vulnerability matrices.

Disclosures

1. Identify the assumptions used to develop loss costs for unknown residential construction types.

Loss costs for unknown construction types are estimated using vulnerability matrices specifically developed for unknown construction types. These unknown construction matrices (called “other” construction in model documentation) are the weighted average of the various vulnerability matrices developed for a given region. The weights depend on the prevailing proportions of various construction types in the region. The proportions were estimated from survey data provided by various counties, and policy data provided by insurance companies. No weight, however, was given to the Mobile Home vulnerability matrix in determining the weighted average matrix for each region.

2. Describe how the modeled loss costs take into consideration storm surge and flood damage to the infrastructure.

The model does not distinguish between direct and indirect losses, but the vulnerability functions are calibrated against claim data that includes both types of losses.

The modeled loss costs do not contain an intentional provision for direct storm surge losses. However, there is certainly a chance that some storm surge claims were paid or partially paid under wind coverage in the validation data, and therefore influenced the damage matrices to some extent.

There is also no specific provision for “ALE only” claims that are due to storm surge damage to the infrastructure with no insured damage to the insured property. Again, though, to the extent “ALE only” claims were present in the validation data, the calibration of the damage model will have allowed for such claims.

3. Describe the assumptions included in model development and validation concerning insurance company claim payment practices.

FPHLM V3.0 2008 161

The implicit assumptions are that such practices are stable over time and do not vary by company.

An option is available that converts any damages over 50% to 100% damage under the assumption that claim adjusters may declare a dwelling uninhabitable. This option was not used in the development or validation of the model. Analysis indicates that activating this option would lead to only a slight increase in loss costs.

Computer code showing that this option is turned off in the production of Commission loss costs can be provided by the Computer Science team.

4. Identify depreciation assumptions and describe the methods and assumptions used to reduce insured losses on account of depreciation. Provide a sample calculation for determining the amount of depreciation and the actual cash value (ACV) losses.

For both replacement cost and ACV policies the value of structures and contents are generally assumed to equal the insured limit. In the rare case where data on property value is available from the insurance company, and that value exceeds the limit, the value provided is used to estimate the ground-up damages.

Depreciation is considered in the model, but not explicitly. The damage ratios were calibrated to insured losses that contained a mix of replacement cost and ACV policies, but primarily replacement cost. Consequently there is an implicit allowance for depreciation (of an unknown degree) built into the modeled losses.

5. Identify property value assumptions and describe the methods and assumptions used to determine the true property value and associated losses. Provide a sample calculation for determining the property value and guaranteed replacement cost losses.

The model assumes that the insured value is the true value of the property except in rare cases when the insurance company provides a separate property value that is higher than the insured value.

6. Describe how loss adjustment expenses are considered within the loss cost estimates.

Loss adjustment expenses are not included in estimates of loss costs. The loss data used for validation do not include loss adjustment expenses. The OIR data call used to collect the validation data required losses excluding LAE.

FPHLM V3.0 2008 162 A-3 Loss Cost Projections

A. Loss cost projections produced by hurricane loss projection models shall not include expenses, risk load, investment income, premium reserves, taxes, assessments, or profit margin.

Loss cost estimates do not include expenses, risk loads, investment income, premium reserves, taxes, assessments, or profit margins. The model produces pure loss costs.

B. Loss cost projections shall not make a prospective provision for economic inflation.

Loss cost estimates do not consider economic inflation.

Disclosures

1. Describe the method or methods used to estimate annual loss costs needed for ratemaking. Identify any source documents used and research performed.

Expected annual losses are estimated for individual policies in the portfolio. Losses are estimated separately for structure, appurtenant structure, contents and ALE.

The meteorological component of the model derives a probability distribution of wind speed for each zip code. This distribution of wind speed is drawn from the stochastic set of hurricanes generated by the model.

The engineering component of the model derives a set of vulnerability matrices. There is a matrix for the various region/coverage/construction combinations. For a given wind speed, the matrix specifies the distribution of damage ratios (i.e. damage as a % of value).

The model first calculates expected loss, net of deductible, for each wind speed. This is done by multiplying the property value by the midpoint of each damage ratio interval, subtracting the deductible, and weighting with the probability of damages in the interval.

Next, the wind probabilities for the property’s zip code are applied to the expected loss by wind speed to produce the expected loss for the property. Finally, the expected loss is adjusted by the expected demand surge factor for the region.

The expected losses can be summed across all properties in a zip code or across all zip codes in a county to obtain expected aggregate loss. The losses can also be aggregated by policy form, construction type, rating territories etc.

The following sources were used in the research:

Hogg and Klugman, Loss Distribution, 1984, particularly Ch. 4 and 5 and the appendix,

FPHLM V3.0 2008 163

Klugman, Panjer and Willmot, Loss Models, 1998

Foundations of Casualty Actuarial Sciences, 4th edition, 2001, Casualty Actuarial Society.

2. Identify the highest level of resolution for which loss costs can be provided. Identify the resolution used for the reported output ranges.

Loss costs can be provided at the individual policy level by coverage. Policy level loss costs can be aggregated by zip code, county, region, rating territory, construction and deductible.

The output ranges are estimated at zip code level.

FPHLM V3.0 2008 164 A-4 Demand Surge

A. Demand surge shall be included in the model’s calculation of loss costs using relevant data.

Demand surge is included in the calculation of loss costs.

B. The methods, data, and assumptions used in the estimation of demand surge shall be actuarially sound.

The method, data, and assumptions used in the estimation of demand surge are actuarially sound.

Disclosures

1. Describe how the model incorporates demand surge in the calculation of loss costs.

How Demand Surge is Incorporated in Loss Cost Calculation

Weighted average demand surge factors across the stochastic set of storms are applied to the modeled losses. There are factors by coverage for each of five regions. The regions are:

Northeast / North Central Northwest Central South (except Monroe County) Monroe County

For each storm in the stochastic set demand surge is assumed to be a function of coverage, region and the storm’s estimated statewide losses before consideration of demand surge.

General Form of the Demand Surge Functions

The functions applied to determine the demand surge for each storm are of the form:

Structure: Surge Factor = c + p1 x ln (statewide storm losses) + p2,

where c is a constant,

p1 is a constant for all regions except Monroe County,

p2 varies by region, and

“statewide storm losses” are the estimated losses, before demand surge, for the storm under consideration.

Appurtenant Structures: Surge Factor = Structural Factor.

FPHLM V3.0 2008 165

Contents: Surge Factor = [ (Structural Factor – 1) x 30% ] + 1.

Additional Living Expenses: Surge Factor = 1.5 x Structural Factor - .5.

Development of the Structural Demand Surge Function

To estimate the impact of demand surge on the settlement cost of structural claims following a hurricane we used a quarterly construction cost index produced by Marshall & Swift/Boeckh. We considered the history of the index from first quarter 1992 through second quarter 2007. There is an index for each of 52 zip codes in Florida with forty-two counties represented. We grouped the indices to produce a set of regional indices, weighting each zip code index with population.

The approach to estimating structural demand surge was to examine the index for specific regions impacted by one or more hurricanes since 1992. From the history of the index we projected what the index would have been in the period following the storm had no storm occurred. Any gap between the predicted and actual index was assumed to be due to demand surge. In total we examined ten storm/region combinations. From these ten observations of structural demand surge we generalized to the functional relationship shown above.

Monroe County was treated as an exception. There were no storms of any severity striking Monroe during the time period of our observations. We believe, though, that the location of and limited access to the Keys will result in an unusually high surge in reconstruction costs after a storm, particularly since the Overseas Highway could be damaged by storm surge or seriously blocked by debris. We have therefore judgmentally selected demand surge parameters for Monroe in excess of those indicated for the remainder of South Florida.

Development of the Contents Demand Surge Function

The approach to determining the contents demand surge function was to relate any surge in consumer prices in Southeast Florida following hurricanes Katrina and Wilma to the estimated structural demand surge following those storms. We used a sub-index of the Miami-Ft. Lauderdale Consumer Price Index for this purpose, and compared the projected and actual index after the storms. Since the surge in consumer prices was roughly 30% of the surge in construction costs, we selected that percentage as the relationship between structural and contents demand surge.

Development of Additional Living Expense (ALE) Demand Surge Function

To estimate ALE demand surge we first examined the relationship between structural losses and ALE losses in the validation data set. This data set includes losses from three storms (Andrew, Charley and Frances) and eleven insurance companies. We then compared the predicted

FPHLM V3.0 2008 166 increase in ALE losses associated with various increases in structural losses. That generalized relationship is the ALE demand surge function shown above.

ALE demand surge is related to structural demand surge in following sense: Structural surge is caused by an inability of the local construction industry to meet the sudden demand for materials and labor following a storm. A high surge in construction costs suggests a more serious mismatch between the demand for repairs and the supply of materials and labor. This mismatch translates into longer delays in the completion of repairs and rebuilding, which in turn implies a higher surge in ALE costs.

Because the model’s ALE surge is determined as a function of structural surge, Monroe County ALE surge factors are higher than those for the remainder of South Florida. We believe this is reasonable because of the unusual delays in repair/rebuilding that will occur following a major storm in the Keys, especially if there is damage to US 1 or to bridges connecting the islands.

Treatment of Demand Surge for Storms Impacting both the Florida Panhandle and Alabama

The Northwest region is segregated from the remainder of the North to allow for demand surge that is a function of combined Florida/Alabama losses from storms impacting both states. The Northwest region consists of all Panhandle counties west of Leon and Wakulla. The definition of this region was selected by considering which counties experienced losses from Ivan, Frederic and Elena, i.e. from storms that impacted both states. Not all counties in the Northwest region experienced losses from these three specific storms, but losses in neighboring counties suggest that that they are nevertheless at risk for inclusion in a combined Florida/Alabama event.

Demand surge factors for the Northwest region are determined as an upward adjustment to the factors for the Northeast/North Central region. The purpose of this adjustment is to correct for an understatement of the model’s demand surge that occurs when only the Florida losses from a combined Florida/Alabama event are used to determine the level of demand surge from a storm.

2. Provide citations to published papers, if any, that were used to develop how the model estimates demand surge.

No published papers were used in the demand surge development.

FPHLM V3.0 2008 167 A-5 User Inputs

All modifications, adjustments, assumptions, and defaults necessary to use the inputs in the model shall be actuarially sound and included with the model output. Treatment of missing values for user inputs required to run the model shall be actuarially sound and described with the model output.

The insurance companies provide policy data in a standardized format. The input format description is available for audit. If observations on the input variables are missing, the provider is often solicited for the information and a determination is made if the data has zero value or is missing. If the data on many key variables are missing the record is dropped from the analysis, otherwise appropriate assumptions are made to retain the record. If, for example, the year built is missing, then weighted average damage matrices are used, with the weights determined by the policy location and construction type. The insured limit is assumed to be the value of the property, and therefore no adjustments are made to the exposure data for building structure, appurtenant structure, contents or additional living expense. In the rare case, when property value data is available and it exceeds the limit, the value is used to calculate the ground-up damage. If limit on ALE is time based and no exposure is provided for ALE, then depending on the policy type, ALE is assumed to be a percentage of either the structure or content coverage for one year. No loss costs are reported for zip codes that are not in the geo-coded set. The number of records deleted and adjustments to the data set are documented.

Disclosures

1. Describe the methods used to distinguish among policy form types (e.g., homeowners, dwelling property, mobile home, tenants, condo unit owners).

The client provides the data on exposure by coverage type, and identifies construction type, policy form, rating territory etc. The model can process any combination of policy type, construction type, deductibles, coverage limits etc. The client is assumed to provide the correct data, though outliers may be investigated. The model output reports include separate loss estimates for structure, content, appurtenant structure, and ALE. These losses are also reported by construction type (e.g. masonry, frame, manufactured homes), by county or zip code, by policy form (e.g., HO-3, HO-4 etc.), by rating territory, and combinations thereof.

FPHLM V3.0 2008 168

2. Disclose, in a model output report, the specific type of input that is required to use the model or model output in a personal residential property insurance rate filing. Such input includes, but is not limited to, optional features of the model, type of data to be supplied by the model user and needed to derive loss projections from the model, and any variables that a model user is authorized to set in implementing the model. Include the model name and version number on the model output report. All items included in the output form submitted to the Commission shall be clearly labeled and defined.

A model output report follows.

Table 13. Output report for OIR data processing

FPHLM V3.0 2008 169 Output Report for OIR Data Processing

Florida Public Hurricane Loss Model: Release 3.0

OIR Data Processing Results:

Report Content: - Original Number of the policies in data set - Process steps to formalize the data set - Numbers of policies which are excluded due to certain reason, e.g. invalid zip codes, invalid format, etc. - Numbers of: Construction Types, Territory Codes, Policy Forms, Program Codes, etc. - Number of policies to generate the estimated losses - Number of files in the final results

The results are aggregated by different combinations upon counties, zip codes, policy forms, program codes, and territory codes.

In case if there are: - more than 1 construction type - more than 1 policy forms - more than 1 program codes - more than 1 territory codes

There will be 47 files in the final results with names as below:

_PILM_Loss_ConstType.xls _PILM_Loss_County.xls _PILM_Loss_PolicyForm.xls _PILM_Loss_ProgramCode.xls _PILM_Loss_TerritoryCode.xls _PILM_Loss_Zipcode.xls

_PILM_Loss_ConstType_PolicyForm.xls _PILM_Loss_ConstType_ProgramCode.xls _PILM_Loss_ConstType_TerritoryCode.xls _PILM_Loss_County_ConstType.xls _PILM_Loss_County_PolicyForm.xls _PILM_Loss_Zipcode_ConstType.xls _PILM_Loss_County_ProgramCode.xls _PILM_Loss_County_TerritoryCode.xls _PILM_Loss_Zipcode_PolicyForm.xls _PILM_Loss_PolicyForm_ProgramCode.xls _PILM_Loss_PolicyForm_TerritoryCode.xls _PILM_Loss_TerritoryCode_ProgramCode.xls _PILM_Loss_Zipcode_ProgramCode.xls _PILM_Loss_Zipcode_TerritoryCode.xls _PILM_Loss_ConstType_PolicyForm_ProgramCode.xls _PILM_Loss_ConstType_PolicyForm_TerritoryCode.xls PILM Loss ConstType TerritoryCode ProgramCode.xls

FPHLM V3.0 2008 170 _PILM_Loss_County_ConstType_PolicyForm.xls _PILM_Loss_County_ConstType_ProgramCode.xls _PILM_Loss_County_ConstType_TerritoryCode.xls _PILM_Loss_County_PolicyForm_ProgramCode.xls _PILM_Loss_County_PolicyForm_TerritoryCode.xls _PILM_Loss_County_TerritoryCode_ProgramCode.xls _PILM_Loss_Zipcode_ConstType_PolicyForm.xls _PILM_Loss_Zipcode_ConstType_ProgramCode.xls _PILM_Loss_Zipcode_ConstType_TerritoryCode.xls _PILM_Loss_Zipcode_PolicyForm_ProgramCode.xls _PILM_Loss_Zipcode_PolicyForm_TerritoryCode.xls _PILM_Loss_Zipcode_TerritoryCode_ProgramCode.xls _PILM_Loss_ConstType_PolicyForm_TerritoryCode_ProgramCode.xls _PILM_Loss_County_ConstType_PolicyForm_ProgramCode.xls _PILM_Loss_County_ConstType_PolicyForm_TerritoryCode.xls _PILM_Loss_County_ConstType_TerritoryCode_ProgramCode.xls _PILM_Loss_County_PolicyForm_TerritoryCode_ProgramCode.xls _PILM_Loss_Zipcode_ConstType_PolicyForm_ProgramCode.xls _PILM_Loss_Zipcode_ConstType_PolicyForm_TerritoryCode.xls _PILM_Loss_Zipcode_ConstType_TerritoryCode_ProgramCode.xls _PILM_Loss_PolicyForm_TerritoryCode_ProgramCode.xls _PILM_Loss_Zipcode_PolicyForm_TerritoryCode_ProgramCode.xls _PILM_Loss_County_ConstType_PolicyForm_TerritoryCode_ProgramCode.xls _PILM_Loss_Zipcode_ConstType_PolicyForm_TerritoryCode_ProgramCode.xls

The final results are zipped and protected by using password

Note: PILM is Probabilistic Insured Loss Model

FPHLM V3.0 2008 171 3. Provide a copy of the input form used by a model user to provide input criteria to be used in the model. The modeler shall demonstrate that the input form relates directly to the model output. Include the model name and version number on the input form. All items included in the input form submitted to the Commission shall be clearly labeled and defined.

A copy of the input form follows.

Table 14. Input form for Florida Public Hurricane Loss Model

Florida Public Hurricane Loss Model Version 3.0

The portfolios should be saved in .txt files with the following format:

PolicyID,Zipcode,YearBuilt,ConstructionType,PropertyValue,StructureCoverage,AppCoverage, ContentCoverage,ALECoverage,Deductible,HurricaneDeductible,NatureOfCoverage,County

1. Attribute Explanation:

PolicyID: the unique ID for this certain portfolio Zipcode: 5-digit zip code where this certain property belongs YearBuilt: 4-digit year number when this property was built ConstructionType: the construction type for this certain property, which is with one of the following four types: Frame, Masonry, Manufactured, or Other PropertyValue: the dollar amount value for this certain property StructureCoverage: the structure coverage amount in dollars AppCoverage: the appurtenant coverage amount in dollars ContentCoverage: the content coverage amount in dollars ALECoverage: the ALE coverage amount in dollars Deductible: deductible amount in dollars for other types of losses HurricaneDeductible: hurricane deductible amount in dollars NatureOfCoverage: using one letter R or A to represent Replacement Cost or Actual Cash Value, respectively County: the name of the county where the property belongs

Note the attributes should be separated by comma only

2. Examples 1,33143,1977,Masonry,162000,162000,16200,124000,0,0,250,R,Miami-Dade

Note: The company may provide more columns, e.g. Policy Form, Program Code, and Territory Code.

4. Describe actions performed to ensure the validity of insurer data used for model inputs or validation/verification.

We developed a set of programs to check and validate the data processing. These programs include the Validation Automation Program and Matlab Plotting Program. Sometimes the computer test results are compared with manually processed results. The following check list is also implemented:

FPHLM V3.0 2008 172 Table 15. Check List for the Pre-processing

Field Name Check that… Checked * There are no null values. PolicyID * All duplicates (if any) have valid policy information. * There are no null values. Zipcode * All values belong to the set of 5-digit zipcodes in Florida. * There are no null values (Note: policies with no YearBuilt should have for value 0). * All values are 4-digit numbers. YearBuilt * There are no values exceeding the current year. * There are no non-zero values less than 1700. * There are no null values. ConstType * All values are either masonry, frame, manufactured, or other. * There are no null values. * There are no negative values. PropValue * If all values are equal to 0, then they are updated to equal LMs. * The actural Property Values will be updated to the larger numeric value between Property Value and Structure Limit * There are no null or non-numeric values. LMs * There are no negative values. * There are no null or non-numeric values.

LMapp * There are no negative values. * If all values are equal to 0 (because it's missing but the company covers Lmapp), then they are updated to 10% of LMs. (double check with Dr. Hamid) * There are no null or non-numeric values. LMc * There are no negative values. * There are no null or non-numeric values. * There are no negative values. LMale * If all values are equal to 0, then he ALE limits will be updated in the program as follows: (1) 20% of LMs or (2) 40% of LMc if LMs is zero but LMc > 0 or (3) 40% of LMapp if both LMs and LMc are zero * There are no null or non-numeric values. * There are no negative values. Deduc * All percentages are converted to numeric values. (Sometimes he percentages are represented as 2, 5, 10, 02, 05, 000002, 000005, 000010 instead of 2%, 5%, 10%) * There are no null or non-numeric values. * There are no negative values. HurrDeduc * All percentages are converted to numeric values. (Sometimes he percentages are represented as 2, 5, 10, 02, 05, 000002, 000005, 000010 instead of 2%, 5%, 10%) * Normally Hurricane Deductible should be no less than 500. * There are no null values. Coverage * The format is correct (i.e. value is equal to A or R). * There are no null values. * All county names are spelled only one way (i.e. all caps & no spelling errors, etc.). County * All names are counties in Florida. * For counties as Miami-Dade (Miami Dade, Dade), St. Johns (Saint Johns, St Johns), St. Lucie (Saint Lucie, St Lucie), make sure only one type of spelling is used. * If the field is present, values cannot be null. PolicyForm * The format is correct (i.e. value is equal to DP-3, HO-6, etc ). * If the field is present, values cannot be null. ProgramCode * The format is correct (i.e. value is equal to A, B, etc ). * If the field is present, values cannot be null. TerritoryCode * The format is correct (i.e. value is equal to 36, 11, etc.).

FPHLM V3.0 2008 173 Note: LMs is coverage limit for building structure; LMapp is coverage limit for appurtenant structure, LMc is coverage limit for contents; and LMale is coverage limit for ALE.

FPHLM V3.0 2008 174 A-6 Logical Relationship to Risk

A. Loss costs shall not exhibit an illogical relation to risk, nor shall loss costs exhibit a significant change when the underlying risk does not change significantly.

The lost costs produced by the FPHLM model do not show illogical relations to risk nor do they change significantly when the underlying risk does not change.

B. Loss costs produced by the model shall be positive and non-zero for all valid Florida ZIP Codes.

The model produces positive and non-zero loss costs for all valid zip codes in the geo-coded set.

C. Loss costs cannot increase as the quality of construction type, materials and workmanship increases, all other factors held constant.

Loss cost decrease as the quality of construction increases.

D. Loss costs cannot increase as the presence of fixtures or construction techniques designed for hazard mitigation increases, all other factors held constant.

Loss cost decreases if loss mitigation measures are considered. See Form V-2.

E. Loss costs cannot increase as the quality of building codes and enforcement increases, all other factors held constant.

Loss cost decreases as the quality of building codes and enforcement increases.

F. Loss costs shall decrease as deductibles increase, all other factors held constant.

Loss cost decrease as deductibles increase, all other factors held constant. See Form A-6.

G. The relationship of loss costs for individual coverages, (e.g., structures and appurtenant structures, contents, and loss of use/additional living expense) shall be consistent with the coverages provided.

Relationship of loss costs for structure, appurtenants, contents, and ALE are consistent with coverages provided. Disclosures

1. Demonstrate that loss cost relationships by type of coverage (structures, appurtenant structures, contents, additional living expenses) are consistent with actual insurance data.

FPHLM V3.0 2008 175 The structure loss consists of external and internal losses. Contents and additional living expense losses are a function of the interior structure loss. Appurtenant structure losses are derived independently. All the losses are based on a combination of engineering principles, empirical equations, and engineering judgment. They were validated against claim data from Andrew, Charley, and Frances. The results are shown in the graphs below for hurricanes Charley and Frances. Each dot represents an insurance portfolio. The square symbols correspond to Charley, while the diamonds corresponds to Frances.

Exposure vs Actual - Structure Loss $500 Millions

$250 Model Structure Losses Structure Model

$0 $0 $250 $500 Millions

Actual Structure Losses

Figure 40. Model vs. Actual—Structure Loss

FPHLM V3.0 2008 176 Model vs Actual - Content Loss

$10 Millions Modeled Content Losses

$0 $0 $10 Millions

Actual Content Losses

Figure 41. Model vs. Actual—Content Loss

Model vs Actual - ALE Loss

$5 Millions Modeled ALE Modeled ALE

$0 $0 $5 Millions

Actual ALE

Figure 42. Model vs. Actual—ALE Loss

FPHLM V3.0 2008 177 Model vs Actual - APP Loss Ratios

$8 Millions Modeled APP Losses

$0 $0 $8 Millions

Actual APP Losses

Figure 43. Model vs. Actual—APP Loss

FPHLM V3.0 2008 178 2. Demonstrate that loss cost relationships by construction type or vulnerability function (frame, masonry, and mobile home) are consistent with actual insurance data.

The validation studies described above were performed for a mix of masonry and frame structures for each portfolio. In addition, portfolios of manufactured homes were validated separately. In general loss costs for masonry are lower than for frame, which are lower than for mobile homes. A comparison of modeled versus historical loss follows.

Table 16. Modeled vs. Historical Loss by Construction Type

Hurricane Charley Actual Modeled Con- Loss/ Loss/ struction Exposure Loss Exposure Exposure Loss Exposure Difference Frame $2,661,997,678 $42,856,539 0.01610 $2,661,997,678 $44,679,205 0.01678 -0.00068 Masonry $15,934,973,183 $230,319,802 0.01445 $15,934,973,183 $193,436,635 0.01214 0.00231 Other $112,688,270 $1,525,991 0.01354 $112,688,270 $1,451,584 0.01288 0.00066

Actual Modeled Con- Loss/Expo Loss/ struction Exposure Loss sure Exposure Loss Exposure Difference Frame $724,372,315 $9,529,621 0.01316 $724,372,315 $12,037,726 0.01662 -0.00346 Masonry $2,968,361,968 $44,677,899 0.01505 $2,968,361,968 $40,936,105 0.01379 0.00126

Actual Modeled Loss/Expo Loss/ County Exposure Loss sure Exposure Loss Exposure Difference Lake $887,803,185 $373,818 0.00042 $887,803,185 $258,173 0.00029 0.00013 Osceola $1,485,169,027 $20,178,018 0.01359 $1,485,169,027 $21,204,446 0.01428 -0.00069 Sarasota $890,840,373 $268,015 0.00030 $890,840,373 $740,082 0.00083 -0.00053 Collier $898,574,742 $656,197 0.00073 $898,574,742 $215,858 0.00024 0.00049

Also see Standard S-5 and Form S-3.

3. Demonstrate that loss cost relationships among coverages, territories, and regions are consistent and reasonable.

Loss costs in regions that have relatively high historical frequency of hurricanes are usually higher. Similarly, the loss costs for inland counties on the average are lower than coastal counties. Also loss costs for northern region are lower than the central and southern region. This is shown in Form A-2 for structural coverage for three types of construction.

4. Explain any anomalies or special circumstances that might preclude any of the above

FPHLM V3.0 2008 179 conditions from occurring.

For some inland zip codes the loss costs may be higher than neighboring zip codes that are closer to the coast because of lower terrain roughness. Similarly a frame structure may have lower lost cost than masonry if the frame is newer and built under a stronger building code.

5. Provide a completed Form A-1, Loss Costs.

See Form A-1.

6. Provide a completed Form A-2, Zero Deductible Loss Costs by ZIP Code.

See Form A-2.

7. Provide a completed Form A-3, Base Hurricane Storm Set Average Annual Zero Deductible Statewide Loss Costs.

See Form A-3.

8. Provide a completed Form A-4, Hurricane Andrew Percent of Losses.

See Form A-4.

9. Provide a completed Form A-5, Distribution of Hurricanes by Size of Loss.

See Form A-5.

FPHLM V3.0 2008 180 A-7 Deductibles and Policy Limits

A. The methods used in the development of mathematical distributions to reflect the effects of deductibles and policy limits shall be actuarially sound.

In practice the insurance companies often allocate deductibles to structure, content, AP, and ALE on a pro-rata loss basis. Thus, if for example, structure and content damages before deductible are $20,000 and $6,000 respectively, and the deductible is $3,000, then (20,000/26,000)(3,000) = $2,308 is allocated to structure and (6,000/26,000)(3,000) = $692 is allocated to contents. This means that the various damages have to be considered and deductibles applied simultaneously. The deductibles must be allocated among the different losses and the truncation applied to each loss separately on a pro-rata basis.

For pro-rata deductible method to work optimally, the functional relationships between structure damage and others should be estimated, and for each interval or class of structural damage, the corresponding mean and variance of the C, AP, and ALE damages should be specified. The conditional probabilities for C, AP, and ALE will then be the same as those for structural damage. An independent content matrix is somewhat problematic and may create biases in estimates of net of deductible losses. For structures we are likely to have damage ratio ranges or intervals of 0 to 2%, 2% to 4%, 4% to 6% etc. For each of these intervals (and its mid points), ideally we may want to use the mean and variance of the corresponding damage ratios for contents, AP and ALE. In practice, since the damage matrix for different types of losses are not directly related, we need to use the mean of the content, or AP, or ALE damage vector conditional on wind speeds, since the wind speed is the only common frame of reference to the various types of damages. L+DS

Expected Structure Loss = E(Ls) = ∑ (DMi - Ds ) pS (xiw) + ∑ LMS pS (xiw)

L+CS

Expected Content Loss = E(LC) = ∑ (f(Xi) - Dc) pC (xiw) + ∑ LMC pC (xiw)

Expected Appurtenant Loss = E(LAP) = ∑ (g(Xi) - DAP) pS (xiw) + ∑ LMAP pS (xiw)

Expected ALE Loss = E(LALE) = ∑ (h(Xi) - DALE) pS (xiw) + ∑ LMALE pS (xiw)

Expected Loss = E (L) = E(LS) + E(LC) + E(LAP) + E(LALE)

Where, each of the losses net of deductible is ≥ 0. And where the deductibles DS, DC, DAP, DALE are applied on a pro-rata basis to the respective damages as follows:

DS = [DMS /(DMS + C + AP + ALE)] * D

DC = [C /(DMS + C + AP + ALE)] * D

DAP = [AP /(DMS + C + AP + ALE)] * D

FPHLM V3.0 2008 181 DALE = [ALE /(DMS + C + AP + ALE)] * D

For this method to work, ideally, the joint probabilities of the losses must be estimated and used. In practice such joint probabilities are hard to estimate and validate. Thus, the engineering component should ideally provide for each structural damage interval, and given a wind speed, the mean and variance of damage ratio for content, AP, ALE. The model uses the mean C, AP, and ALE for the given wind speed to determine the allocation of deductible to the various coverages.

B. The relationship among the modeled deductible loss costs shall be reasonable.

The relationship among the modeled deductible loss costs is reasonable.

C. Deductible loss costs shall be calculated in accordance with s. 627.701(5)(a), F.S.

The deductible loss costs are calculated in accordance with s. 627.701(5)(a), F.S.

Disclosures

1. Describe the methods used in the model to treat deductibles (both flat and percentage), policy limits, replacement costs, and insurance-to-value when projecting loss costs.

In the probabilistic damage matrices, for each possible damage ratio there is a set of probabilities for different wind speeds. For each damage outcome the damage ratio is multiplied by insured value to get dollar damages, the deductible is deducted and net of deductible loss is estimated, subject to the constraints that net loss is ≥ 0 and ≤ limit – deductible. Percentage deductibles are converted into dollar amounts. Both the replacement cost and property value are assumed to equal the coverage limit.

2. Provide an example of how insurer loss (loss net of deductibles) is calculated. Discuss data or documentation used to confirm or validate the method used by the model.

Example: (A) (B) (C) (D)=(A)*(C) (E)=(D)-(B) Structure Policy Damage Zero Deductible Loss Net of Value Limit Deductible Ratio Loss Deductible 100,000 90,000 500 2% 2,000 1,500

Once the damage ratios are generated, then:

FPHLM V3.0 2008 182 Loss net of deductible = (Damage Ratio x Bldg Value) – Deductible If net loss is < 0 then replace it with zero. If net loss > limit – deductible, replace with limit – deductible.

Example Bldg value = $200,000. Limit = $180,000. Deductible = $3,000. Jth Damage ratio = 5%. Loss net of deductible = .05 x 200,000 - 3,000 = $7,000. If the Jth Damage ratio = 1%, then loss net of deductible = 0. If the damage ratio is 95% then the loss net of deductible = $180,000 - $3,000 = $177,000.

The deductible method used by model is based on Hogg and Klugman, Loss Distributions, 1984, Chapters 4 and 5 and the Appendix. Modeled losses net of deductible were validated against insurance company losses for Andrew, Charley and Frances.

3. Describe how the model calculates annual deductibles.

If there are multiple hurricanes in a year in the stochastic set, the wind deductibles are applied to the first hurricane, and any remaining amount is then applied to the second hurricane. If none remains then the general peril deductible can be applied.

FPHLM V3.0 2008 183 A-8 Contents

A. The methods used in the development of contents loss costs shall be actuarially sound.

B. The relationship between the modeled structure and contents loss costs shall be reasonable, based on the relationship between historical structure and contents losses.

A. The methods used in the development of contents loss costs is actuarially sound

B. The relationship between the modeled structure and contents loss costs is reasonable, based on the relationship between historical structure and contents losses.

Disclosure

1. Describe the methods used in the model to calculate loss costs for contents coverage associated with personal residential structures (including mobile homes), tenants, and condo unit owners.

In all cases, contents losses are a function of the internal damage. These empirical functions are based on engineering judgment, and were validated against claim data for hurricane Andrew, Charley, and Frances. A graph showing the vulnerability curves for content loss follows.

FPHLM V3.0 2008 184 .

Structure Damage Ratio Figure 44. Modeled vs. Actual Relationship between Structure and Content Damage Ratios

FPHLM V3.0 2008 185 A-9 Additional Living Expense (ALE)

A. The methods used in the development of ALE loss costs shall be actuarially sound.

B. ALE loss cost derivations shall consider the estimated time required to repair or replace the property.

C. The relationship between the modeled structure and ALE loss costs shall be reasonable, based on the relationship between historical structure and ALE losses.

D. ALE loss costs produced by the model shall appropriately consider ALE claims arising from damage to the infrastructure.

A. The methods used in the development of Additional Living Expense (ALE) loss costs are actuarially sound.

B. ALE loss cost derivations consider the estimated time required to repair or replace the property.

C. The model uses ALE vulnerability function derived from the relationship between structural damage and ALE. The ALE vulnerability functions have been calibrated using historical claim data on structure and ALE.

D. ALE loss costs produced by the model appropriately consider ALE claims arising from damage to the infrastructure. The model does not distinguish explicitly between direct and indirect loss to the structure, but the function is calibrated against claim data that includes both types of losses.

Disclosures

1. Describe the methods used to develop loss cost for ALE coverage. State whether the model considers both direct and indirect loss to the structure. For example, direct loss is for expenses paid to house policyholders in an apartment while their home is being repaired. Indirect loss is for expenses incurred for loss of power (e.g., food spoilage).

The additional living expenses are based on an empirical functional relationship of the interior damage to the structure. The model does not distinguish explicitly between direct and indirect loss to the structure, but the function is calibrated against claim data that includes both types of losses.

2. State the minimum threshold at which ALE loss is calculated (e.g., loss is estimated for structure damage greater than 20% or only for category 3, 4, 5 events). Provide

FPHLM V3.0 2008 186 documentation of validation test results to verify the approach used.

The ALE loss is calculated as a function of interior damage. There is no minimum threshold at which ALE loss is calculated, since it is believed that even with minimum interior damage, some ALE losses might exist when residents are subject to a mandatory evacuation.

FPHLM V3.0 2008 187 A-10 Output Ranges

A. Output ranges shall be logical and any deviations supported.

Output ranges generated by the model are logical. Deviations are explained.

B. All other factors held constant, output ranges produced by the model shall reflect lower loss costs for:

1. masonry construction versus frame construction,

Output ranges produced by the model reflect lower loss costs for masonry versus frame construction. Deviations are explained.

2. residential risk exposure versus mobile home risk exposure,

Output ranges produced by the model reflect lower loss costs for residential versus mobile home risk exposure.

3. in general, inland counties versus coastal counties, and

In general output ranges produced by the model reflect lower loss costs for inland counties versus coastal counties.

4. in general, northern counties versus southern counties.

In general output ranges produced by the model reflect lower loss costs for northern counties versus southern counties.

Disclosures

1. Provide an explanation for all anomalies in the loss costs that are not consistent with the requirements of this Standard.

Loss costs for masonry are lower than frame for every zip code, but the county weighted average loss cost for masonry sometimes exceeds frame because the masonry weights are greater in zip codes with high loss costs.

In a few cases in Form A-1 loss costs are higher for zip codes that are more inland than their neighbors. The reason for this anomaly is that terrain roughness coefficients are lower in these more inland zip codes.

2. Provide an explanation of the differences in the output ranges using the 2002 Florida Hurricane Catastrophe Fund aggregate exposure data between the prior year and the current year submission.

FPHLM V3.0 2008 188

There were changes to the meteorological component of the model that had the impact of increasing the landfall intensity in the south and decreasing intensity in the north.

3. Provide justification for changes using the 2002 Florida Hurricane Catastrophe Fund aggregate exposure data from the prior submission of greater than ten percent in weighted average loss costs for any county, specifically by county.

Owners:

Because of the changes to the meteorological component, many northern and central counties experienced Owners loss cost decreases greater than 10%.

At $0 deductible the counties with decreases in excess of 10% are:

Alachua, Baker, Bradford, Citrus, Clay, Columbia, Dixie, Duval, Escambia, Flagler, Franklin, Gilchrist, Lafayette, Lake, Levy, Marion, Nassau, Putnam, St. Johns, Santa Rosa, Sumter, Suwannee, Union and Volusia.

At the 5% deductible, most counties north of the Lake Okeechobee decreased by more than 10%.

The changes to the meteorological component generally increased loss costs in the south. The only counties with Owners loss cost increases in excess of 10% were Miami-Dade (at 5% deductible) and Broward (at $2,500 and higher deductibles).

The most extreme Owners changes, both at the 5% deductible level, are: -Broward (Owners Masonry) +8.9% ($4.36 to $4.73) -St. Johns (Owners Frame) -30.7%. ($.38 to $.26)

Renters & Condo:

As with Owners, the changes to the meteorological component reduced many northern and central county loss costs for Renters and Condo by more than 10%, and increased some southern county loss costs by more than 10%. The largest changes, again at the 5% deductible, are:

-Broward (Renters Masonry) +16,9% ($.93 to $1.09) -St. Johns (Renters Frame) -32.8% ($.21 to $.14).

Mobile Home:

Mobile Home loss costs were also impacted by the changes in the meteorological component of the model. In addition an error was found in the vulnerability matrix applied last year to northern counties. The combined effect changed loss costs by more than 10% in the following counties at $0 deductible:

FPHLM V3.0 2008 189 Increases of 12% to 16%: Calhoun, Gadsden, Hamilton, Jackson, Jefferson, Leon, Liberty, Madison

Decreases of -11% to -18%: Citrus, Flagler, Hernando, Hillsborough, Lake, Marion, Orange, Osceola, Pasco, Polk, Seminole, Sumter, Volusia.

4. Provide justification for changes using the 2002 Florida Hurricane Catastrophe Fund aggregate exposure data from the prior submission of ten percent or less in the weighted average loss costs for any county, in the aggregate.

All counties were impacted by the changes to the meteorological portion of the model.

5. Provide a completed Form A-6A, Output Ranges using the 2002 Florida Hurricane Catastrophe Fund aggregate exposure data.

See Form A-6A.

6. Provide a completed Form A-6B, Output Ranges using the 2007 Florida Hurricane Catastrophe Fund aggregate exposure data.

See Form A-6B.

7. Provide a completed Form A-7, Percentage Change in Output Ranges using the 2002 Florida Hurricane Catastrophe Fund aggregate exposure data.

See Form A-7.

8. Provide a completed Form A-8, Percentage Change in Output Ranges by County using the 2002 Florida Hurricane Catastrophe Fund aggregate exposure data.

See Form A-8.

FPHLM V3.0 2008 190 Form A-1: Loss Costs

A. Provide the expected annual loss costs by construction type and coverage for each ZIP Code in the sample data set named “FormA1Input07.xls.” Loss costs shall be rounded to six decimal places. There are 1,479 ZIP Codes and three construction types; therefore, the completed file should have 4,437 records in total. The following is a description of the requested file layout. Follow the instructions on Form A-1 below and in the Submission Data description. Note that fields 2-9 are the exposure fields from the sample data set. Fields 10-13 are for the loss costs (net of deductibles).

B. If there are ZIP Codes in the sample data set that the model does not recognize as “valid,” provide a list in the submission document of such ZIP Codes and provide either a) the new ZIP Code to which the original one was mapped, or b) an indication that the insured values from this ZIP Code were not modeled.

Loss cost data shall be provided for all ZIP Codes given in the sample data set. That is, if no losses were modeled, the record should still be included in the completed file with loss cost of zero, and if a ZIP Code was mapped to a new one, the resulting loss costs should be reported with the original ZIP Code.

C. Provide the results on CD in Excel format using the following file layout. The file name shall include the abbreviated name of the modeler, the Standards year, and the Form name. The file shall be provided in PDF format prior to the Commission’s vote on acceptability.

No. Field Name Description 1 Analysis Date Date of Analysis – YYYY/MM/DD Exposure Fields from Sample Data Set 2 County Code FIPS County Code 3 ZIP Code 5-digit ZIP Code 4 Construction Type 1 = Wood Frame, 2 = Masonry, 3 = Mobile Home 5 Annual Deductible 2% (of the Structure Value) policy deductible for each record (i.e., 0.02*$100,000) 6 Structure Value $100,000 for each record 7 Appurtenant Structures Value $10,000 for each record 8 Contents Value $50,000 for each record 9 Additional Living Expense Value $20,000 for each record Loss Costs (net of deductibles) 10 Structure Loss Cost Projected expected annual loss cost for structure divided by the structure value modeled for each record ($100,000) 11 Appurtenant Structures Loss Cost Projected expected annual loss cost for appurtenant structures divided by the appurtenant structures value modeled for each record ($10,000) 12 Contents Loss Cost Projected expected annual loss cost for contents divided by the contents value modeled for each record ($50,000)

FPHLM V3.0 2008 191 13 Additional Living Expense Loss Cost Projected expected annual loss cost for additional living expense divided by the additional living expense value modeled for each record ($20,000) All deductibles are a percentage of the Structure Value and are policy-level deductibles; however, for reporting purposes, the policy deductible shall be pro-rated to the individual coverage losses in proportion to the loss. The default all-other perils deductible is $500.

FPHLM V3.0 2008 192 Form A-2: Zero Deductible Loss Costs by ZIP Code

Provide a map color-coded by ZIP Code (with a minimum of 6 value ranges) displaying zero deductible loss costs for frame, masonry, and mobile home.

Figure 45. Zero Deductible Loss Cost by Zip Code for Owners Frame

FPHLM V3.0 2008 193

Figure 46. A Zero Deductible Costs by Zip Code for Owners Masonry

FPHLM V3.0 2008 194

Figure 47. Zero Deductible Loss Costs by Zip Code for Mobile Home

FPHLM V3.0 2008 195 Form A-3: Base Hurricane Storm Set Average Annual Zero Deductible Statewide Loss Costs

A. Provide the total insured loss assuming all personal residential zero deductible policies from each specific hurricane in the Base Hurricane Storm Set for the 2007 Florida Hurricane Catastrophe Fund’s aggregate exposure data found in the file named “hlpm2007.exe”. Additional storms that are included in the model’s Base Hurricane Storm Set should be included.

B. Provide this Form on CD in both an Excel and a PDF format. The file name shall include the abbreviated name of the modeler, the Standards year, and the Form name. A hard copy of Form A-3 shall be included in the submission.

Date Year Name Loss 8/10/1901 1901 NoName4 $513,922,939 9/11/1903 1903 NoName3 $10,138,149,180 10/16/1904 1904 NoName3 $3,211,235,513 6/16/1906 1906 NoName2 $1,612,430,903 9/25/1906 1906 NoName6 $590,647,279 10/8/1906 1906 NoName8 $10,458,861,743 10/11/1909 1909 NoName10 $1,171,882,909 10/17/1910 1910 NoName5 $13,138,301,302 8/8/1911 1911 NoName2 $279,709,756 8/23/1911 1911 NoName3 $0 9/11/1912 1912 NoName3 $2,891,677 9/3/1915 1915 NoName4 $506,546,474 7/4/1916 1916 NoName1 $1,807,660 10/17/1916 1916 NoName13 $689,380,621 11/15/1916 1916 NoName14 $267,116,235 9/26/1917 1917 NoName3 $1,073,728,006 9/9/1919 1919 NoName2 $858,663,197 10/24/1921 1921 NoName6 $20,019,382,778 9/13/1924 1924 NoName4 $170,722,671 10/20/1924 1924 NoName7 $8,308,611,291 11/30/1925 1925 NoName2 $3,172,843,862 7/27/1926 1926 NoName1 $8,306,805,476 9/18/1926 1926 NoName6 $29,803,897,270 10/20/1926 1926 NoName10 $385,385,031 8/7/1928 1928 NoName1 $6,798,038,561 9/16/1928 1928 NoName4 $32,683,862,150 9/27/1929 1929 NoName2 $16,209,136,137

FPHLM V3.0 2008 196 Date Year Name Loss 8/29/1932 1932 NoName3 $1,275,943,125 7/29/1933 1933 NoName5 $1,899,703,073 9/3/1933 1933 NoName12 $10,279,592,647 9/2/1935 1935 NoName2 $10,334,789,126 11/4/1935 1935 NoName6 $6,856,207,471 7/27/1936 1936 NoName5 $685,541,371 8/11/1939 1939 NoName2 $6,310,184,642 8/5/1940 1940 NoName3 $0 10/5/1941 1941 NoName5 $20,350,960,198 10/18/1944 1944 NoName11 $22,371,939,242 6/22/1945 1945 NoName1 $11,396,368,852 9/15/1945 1945 NoName9 $16,771,758,798 10/7/1946 1946 NoName5 $11,898,447,236 9/17/1947 1947 NoName4 $19,138,475,981 10/11/1947 1947 NoName8 $6,722,262,400 9/21/1948 1948 NoName7 $4,889,959,858 10/5/1948 1948 NoName8 $1,898,802,265 8/26/1949 1949 NoName2 $19,965,156,381 8/29/1950 1950 BAKER $371,664,580 9/3/1950 1950 EASY $13,846,844,436 10/17/1950 1950 KING $4,528,373,062 9/25/1953 1953 FLORENCE $437,884,666 9/24/1956 1956 FLOSSY $583,489,864 9/9/1960 1960 DONNA $19,747,544,690 9/14/1960 1960 ETHEL $0 8/26/1964 1964 CLEO $10,580,462,138 9/9/1964 1964 DORA $5,180,632,426 10/14/1964 1964 ISBELL $8,101,218,178 9/7/1965 1965 BETSY $5,607,434,889 6/8/1966 1966 ALMA $11,126,203,436 9/21/1966 1966 INEZ $331,655,097 10/16/1968 1968 GLADYS $5,537,368,628 8/16/1969 1969 CAMILLE $0 6/18/1972 1972 AGNES $262,561,043 9/22/1975 1975 ELOISE $875,518,933 9/3/1979 1979 DAVID $7,944,104,311 9/12/1979 1979 FREDERIC $756,730,048 8/29/1985 1985 ELENA $269,809,576 11/20/1985 1985 KATE $360,369,327

FPHLM V3.0 2008 197 Date Year Name Loss 10/12/1987 1987 FLOYD $122,676,019 8/24/1992 1992 ANDREW $19,886,312,089 8/1/1995 1995 ERIN $6,354,560,497 10/3/1995 1995 OPAL $1,837,383,385 7/16/1997 1997 DANNY $75,978,033 9/1/1998 1998 EARL $20,387,357 9/25/1998 1998 GEORGES $519,549,400 10/15/1999 1999 IRENE $4,338,957,681 8/13/2004 2004 CHARLEY $9,676,451,596 9/4/2004 2004 FRANCES $11,446,300,742 9/14/2004 2004 IVAN $603,104,320 9/20/2004 2004 IVAN $0 9/25/2004 2004 JEANNE $12,660,026,029 7/7/2005 2005 DENNIS $653,115,916 8/24/2005 2005 KATRINA $4,114,455,413 9/18/2005 2005 RITA $156,867,393 10/20/2005 2005 WILMA $15,936,075,610

FPHLM V3.0 2008 198 Form A-4: Hurricane Andrew Percent of Losses

A. Provide the percentage of personal residential zero deductible losses, rounded to four decimal places, from Hurricane Andrew for each affected ZIP Code. Include all ZIP Codes where losses are equal to or greater than $500,000.

B. Provide a map color-coded by ZIP Code depicting the percentage of total losses from Hurricane Andrew below latitude 27°N using the following interval coding:

Red Over 5% Light Red 2% to 5% Pink 1% to 2% Light Pink 0.5% to 1% Light Blue 0.2% to 0.5% Medium Blue 0.1% to 0.2% Blue Below 0.1%

C. Provide this Form on CD in both an Excel and a PDF format. The file name shall include the abbreviated name of the modeler, the Standards year, and the Form name. A hard copy of Form A-4 shall be included in the submission.

Rather than using directly a published windfield for Hurricane Andrew, the winds underlying the loss cost calculations must be produced by the model being evaluated and should be the one most emulated by the model. Use the 2007 Florida Hurricane Catastrophe Fund’s aggregate exposure data found in the file named “hlpm2007.exe”.

Zipcode V3mph Total loss Percent loss 41 70 $0 0.0000% 43 131 $0 0.0000% 53 54 $0 0.0000% 97 131 $0 0.0000% 98 68 $0 0.0000% 33001 37 $0 0.0000% 33002 95 $1,340 0.0000% 33004 77 $18,784,001 0.0945% 33008 85 $18,510 0.0001% 33009 86 $39,714,431 0.1997% 33010 104 $48,138,641 0.2421% 33011 105 $25,077 0.0001% 33012 98 $77,007,592 0.3872% 33013 99 $52,502,636 0.2640% 33014 96 $61,290,987 0.3082% 33015 91 $80,147,522 0.4030%

FPHLM V3.0 2008 199 Zipcode V3mph Total loss Percent loss 33016 92 $38,583,835 0.1940% 33017 89 $86,382 0.0004% 33018 93 $56,087,168 0.2820% 33019 91 $39,430,759 0.1983% 33020 80 $47,052,656 0.2366% 33021 79 $105,830,538 0.5322% 33022 82 $80,010 0.0004% 33023 82 $115,702,930 0.5818% 33024 79 $117,718,950 0.5920% 33025 84 $69,273,743 0.3483% 33026 80 $82,798,295 0.4164% 33027 84 $92,119,023 0.4632% 33028 80 $41,910,036 0.2107% 33029 81 $107,319,521 0.5397% 33030 130 $181,955,990 0.9150% 33031 136 $122,587,999 0.6164% 33032 135 $214,378,718 1.0780% 33033 132 $211,039,254 1.0612% 33034 124 $59,644,096 0.2999% 33035 127 $34,278,942 0.1724% 33036 45 $0 0.0000% 33037 70 $70,188,939 0.3530% 33039 130 $147,534 0.0007% 33040 31 $0 0.0000% 33041 29 $0 0.0000% 33042 38 $0 0.0000% 33043 33 $0 0.0000% 33045 29 $0 0.0000% 33050 41 $0 0.0000% 33051 36 $0 0.0000% 33052 36 $0 0.0000% 33054 93 $32,515,008 0.1635% 33055 88 $71,100,999 0.3575% 33056 89 $54,435,045 0.2737% 33060 63 $31,125,336 0.1565% 33061 64 $26,013 0.0001% 33062 71 $43,653,627 0.2195% 33063 62 $43,378,599 0.2181% 33064 60 $48,126,225 0.2420%

FPHLM V3.0 2008 200 Zipcode V3mph Total loss Percent loss 33065 62 $44,027,942 0.2214% 33066 62 $11,885,525 0.0598% 33067 62 $47,595,157 0.2393% 33068 64 $49,890,379 0.2509% 33069 63 $12,916,162 0.0650% 33070 49 $0 0.0000% 33071 63 $74,065,933 0.3724% 33072 63 $30,560 0.0002% 33073 60 $23,393,183 0.1176% 33074 65 $1,613 0.0000% 33075 60 $54,162 0.0003% 33076 59 $44,921,802 0.2259% 33077 62 $3,385 0.0000% 33081 80 $14,470 0.0001% 33082 78 $4,332 0.0000% 33083 79 $18,031 0.0001% 33084 78 $15,836 0.0001% 33090 127 $90,018 0.0005% 33092 137 $131,784 0.0007% 33093 63 $959 0.0000% 33097 61 $4,385 0.0000% 33101 113 $155,665 0.0008% 33102 110 $14,421 0.0001% 33107 110 $0 0.0000% 33109 143 $73,197,941 0.3681% 33110 102 $36,331 0.0002% 33111 137 $67,959 0.0003% 33112 110 $11,185 0.0001% 33114 119 $78,115 0.0004% 33116 147 $207,997 0.0010% 33119 126 $0 0.0000% 33121 139 $11,423 0.0001% 33122 107 $57,921 0.0003% 33124 97 $64,564 0.0003% 33125 115 $78,045,830 0.3925% 33126 117 $55,369,674 0.2784% 33127 108 $34,943,451 0.1757% 33128 116 $3,438,915 0.0173% 33129 138 $135,950,880 0.6836%

FPHLM V3.0 2008 201 Zipcode V3mph Total loss Percent loss 33130 117 $11,268,983 0.0567% 33131 142 $38,475,295 0.1935% 33132 142 $10,591,448 0.0533% 33133 132 $388,189,858 1.9520% 33134 126 $326,013,846 1.6394% 33135 114 $47,831,400 0.2405% 33136 114 $5,741,627 0.0289% 33137 117 $40,097,944 0.2016% 33138 102 $60,571,815 0.3046% 33139 115 $78,033,454 0.3924% 33140 118 $161,027,781 0.8097% 33141 120 $74,330,176 0.3738% 33142 110 $67,479,376 0.3393% 33143 140 $659,398,545 3.3158% 33144 122 $94,229,719 0.4738% 33145 124 $153,404,230 0.7714% 33146 137 $290,239,064 1.4595% 33147 102 $57,038,847 0.2868% 33148 131 $323,054 0.0016% 33149 154 $324,774,765 1.6332% 33150 103 $33,071,346 0.1663% 33151 102 $19,023 0.0001% 33152 110 $132,759 0.0007% 33153 100 $9,599 0.0000% 33154 99 $38,816,341 0.1952% 33155 128 $411,335,059 2.0684% 33156 154 $1,466,604,575 7.3749% 33157 145 $1,071,848,598 5.3899% 33158 146 $229,706,407 1.1551% 33159 128 $303,007 0.0015% 33160 108 $63,244,876 0.3180% 33161 96 $56,284,116 0.2830% 33162 91 $52,511,897 0.2641% 33163 87 $50,631 0.0003% 33164 88 $62,204 0.0003% 33165 132 $489,693,747 2.4625% 33166 111 $58,663,398 0.2950% 33167 97 $26,683,070 0.1342% 33168 93 $38,323,795 0.1927%

FPHLM V3.0 2008 202 Zipcode V3mph Total loss Percent loss 33169 88 $53,307,093 0.2681% 33170 138 $106,065,475 0.5334% 33172 121 $31,756,502 0.1597% 33173 144 $551,126,743 2.7714% 33174 122 $91,737,104 0.4613% 33175 132 $473,905,916 2.3831% 33176 150 $1,218,466,822 6.1272% 33177 146 $610,703,728 3.0710% 33178 107 $50,603,884 0.2545% 33179 87 $54,370,468 0.2734% 33180 92 $37,053,781 0.1863% 33181 98 $28,294,795 0.1423% 33182 120 $54,534,589 0.2742% 33183 140 $360,756,712 1.8141% 33184 126 $124,376,485 0.6254% 33185 132 $172,181,080 0.8658% 33186 145 $1,047,895,280 5.2694% 33187 148 $312,265,152 1.5703% 33189 141 $228,904,476 1.1511% 33190 140 $71,530,890 0.3597% 33193 141 $333,719,115 1.6781% 33194 129 $24,527,105 0.1233% 33195 115 $47,402 0.0002% 33196 145 $504,372,365 2.5363% 33197 138 $388,552 0.0020% 33199 123 $10,332 0.0001% 33231 139 $138,847 0.0007% 33233 134 $100,927 0.0005% 33234 119 $0 0.0000% 33238 103 $41,062 0.0002% 33239 121 $14,578 0.0001% 33242 108 $2,120 0.0000% 33243 134 $56,961 0.0003% 33245 124 $0 0.0000% 33247 110 $0 0.0000% 33255 132 $0 0.0000% 33256 150 $475,405 0.0024% 33257 137 $56,070 0.0003% 33261 94 $2,201 0.0000%

FPHLM V3.0 2008 203 Zipcode V3mph Total loss Percent loss 33265 130 $116,155 0.0006% 33266 105 $16,403 0.0001% 33269 87 $16,083 0.0001% 33280 94 $9,762 0.0000% 33283 141 $201,833 0.0010% 33296 150 $153,928 0.0008% 33299 108 $0 0.0000% 33301 69 $38,807,267 0.1951% 33302 72 $56,018 0.0003% 33303 72 $54,460 0.0003% 33304 71 $25,142,378 0.1264% 33305 68 $27,469,280 0.1381% 33306 70 $9,296,434 0.0467% 33307 65 $14,315 0.0001% 33308 66 $54,918,153 0.2762% 33309 66 $38,703,791 0.1946% 33310 66 $209,801 0.0011% 33311 67 $45,597,778 0.2293% 33312 73 $89,439,672 0.4498% 33313 68 $45,421,561 0.2284% 33314 75 $25,691,640 0.1292% 33315 72 $20,427,691 0.1027% 33316 83 $39,078,060 0.1965% 33317 72 $81,536,267 0.4100% 33318 68 $81,103 0.0004% 33319 68 $58,569,272 0.2945% 33320 66 $49,114 0.0002% 33321 66 $57,588,313 0.2896% 33322 68 $76,153,188 0.3829% 33323 68 $42,994,804 0.2162% 33324 72 $63,659,479 0.3201% 33325 72 $56,736,828 0.2853% 33326 72 $64,561,621 0.3247% 33327 71 $36,187,203 0.1820% 33328 77 $71,951,866 0.3618% 33329 74 $81,526 0.0004% 33330 77 $49,659,547 0.2497% 33331 77 $64,755,730 0.3256% 33332 76 $31,623,808 0.1590%

FPHLM V3.0 2008 204 Zipcode V3mph Total loss Percent loss 33334 63 $34,041,822 0.1712% 33335 65 $17,151 0.0001% 33336 66 $0 0.0000% 33337 71 $1,042 0.0000% 33338 71 $10,269 0.0001% 33339 67 $33,354 0.0002% 33340 66 $0 0.0000% 33345 67 $32,824 0.0002% 33346 77 $3,816 0.0000% 33348 72 $164 0.0000% 33349 72 $10,583 0.0001% 33351 66 $36,578,714 0.1839% 33355 73 $19 0.0000% 33359 70 $7,361 0.0000% 33388 71 $7,093 0.0000% 33394 67 $3,893 0.0000% 33401 45 $0 0.0000% 33402 53 $22,423 0.0001% 33403 44 $0 0.0000% 33404 45 $0 0.0000% 33405 47 $0 0.0000% 33406 48 $0 0.0000% 33407 45 $0 0.0000% 33408 47 $0 0.0000% 33409 47 $0 0.0000% 33410 44 $0 0.0000% 33411 45 $0 0.0000% 33412 44 $0 0.0000% 33413 47 $0 0.0000% 33414 47 $0 0.0000% 33415 47 $0 0.0000% 33416 49 $0 0.0000% 33417 47 $0 0.0000% 33418 46 $0 0.0000% 33419 43 $0 0.0000% 33420 43 $0 0.0000% 33421 45 $0 0.0000% 33422 46 $0 0.0000% 33424 50 $1,327 0.0000%

FPHLM V3.0 2008 205 Zipcode V3mph Total loss Percent loss 33425 50 $898 0.0000% 33426 51 $184,681 0.0009% 33427 57 $317,249 0.0016% 33428 59 $50,203,231 0.2525% 33429 61 $77,056 0.0004% 33430 46 $0 0.0000% 33431 54 $14,855,425 0.0747% 33432 62 $31,155,289 0.1567% 33433 60 $61,908,819 0.3113% 33434 57 $21,557,544 0.1084% 33435 52 $260,085 0.0013% 33436 51 $412,689 0.0021% 33437 52 $848,546 0.0043% 33438 42 $0 0.0000% 33439 44 $0 0.0000% 33440 48 $0 0.0000% 33441 60 $18,232,823 0.0917% 33442 60 $23,629,275 0.1188% 33443 59 $34,637 0.0002% 33444 53 $8,897,021 0.0447% 33445 53 $21,300,800 0.1071% 33446 55 $32,314,821 0.1625% 33447 54 $38,156 0.0002% 33448 54 $76,279 0.0004% 33454 48 $0 0.0000% 33455 0 $0 0.0000% 33458 42 $0 0.0000% 33459 47 $0 0.0000% 33460 49 $0 0.0000% 33461 47 $0 0.0000% 33462 49 $0 0.0000% 33463 50 $412,114 0.0021% 33464 49 $0 0.0000% 33465 52 $213 0.0000% 33466 50 $1,414 0.0000% 33467 48 $0 0.0000% 33468 44 $0 0.0000% 33470 45 $0 0.0000% 33471 44 $0 0.0000%

FPHLM V3.0 2008 206 Zipcode V3mph Total loss Percent loss 33474 50 $562 0.0000% 33476 44 $0 0.0000% 33477 45 $0 0.0000% 33478 41 $0 0.0000% 33480 56 $52,857,829 0.2658% 33481 57 $27,994 0.0001% 33482 54 $50,875 0.0003% 33483 58 $24,414,318 0.1228% 33484 54 $17,953,034 0.0903% 33486 57 $18,140,290 0.0912% 33487 56 $19,237,016 0.0967% 33488 62 $17,189 0.0001% 33493 47 $0 0.0000% 33496 55 $44,011,289 0.2213% 33497 57 $83,961 0.0004% 33498 56 $21,124,552 0.1062% 33499 56 $6,501 0.0000% 33901 52 $182,544 0.0009% 33902 52 $1,189 0.0000% 33903 49 $0 0.0000% 33904 55 $35,191,611 0.1770% 33905 50 $306,973 0.0015% 33906 54 $37,182 0.0002% 33907 54 $9,493,576 0.0477% 33908 57 $37,722,801 0.1897% 33909 53 $17,345,782 0.0872% 33910 51 $647 0.0000% 33911 52 $328 0.0000% 33912 56 $33,625,268 0.1691% 33913 55 $19,854,506 0.0998% 33914 55 $44,084,145 0.2217% 33915 51 $511 0.0000% 33916 51 $78,156 0.0004% 33917 51 $320,839 0.0016% 33918 56 $35,942 0.0002% 33919 55 $25,724,588 0.1294% 33920 50 $0 0.0000% 33921 58 $15,675,002 0.0788% 33922 54 $3,561,886 0.0179%

FPHLM V3.0 2008 207 Zipcode V3mph Total loss Percent loss 33924 60 $8,594,131 0.0432% 33927 46 $0 0.0000% 33928 59 $32,499,254 0.1634% 33930 55 $259,840 0.0013% 33931 62 $19,405,382 0.0976% 33932 62 $8,043 0.0000% 33935 48 $0 0.0000% 33936 53 $21,277,166 0.1070% 33938 43 $0 0.0000% 33944 44 $0 0.0000% 33945 54 $109,970 0.0006% 33946 49 $0 0.0000% 33947 48 $0 0.0000% 33948 46 $0 0.0000% 33949 47 $0 0.0000% 33950 46 $0 0.0000% 33951 45 $0 0.0000% 33952 44 $0 0.0000% 33953 45 $0 0.0000% 33954 45 $0 0.0000% 33955 50 $0 0.0000% 33956 56 $4,081,055 0.0205% 33957 58 $34,800,706 0.1750% 33960 41 $0 0.0000% 33965 59 $293 0.0000% 33970 52 $573 0.0000% 33971 53 $25,378,155 0.1276% 33972 53 $10,670,866 0.0537% 33975 48 $0 0.0000% 33980 46 $0 0.0000% 33981 47 $0 0.0000% 33982 45 $0 0.0000% 33983 46 $0 0.0000% 33990 52 $389,424 0.0020% 33991 54 $17,442,572 0.0877% 33993 53 $19,148,197 0.0963% 33994 50 $265 0.0000% 34101 73 $353,344 0.0018% 34102 75 $83,646,803 0.4206%

FPHLM V3.0 2008 208 Zipcode V3mph Total loss Percent loss 34103 70 $48,519,501 0.2440% 34104 75 $47,650,480 0.2396% 34105 72 $45,889,381 0.2308% 34106 73 $198,430 0.0010% 34107 73 $78,180 0.0004% 34108 68 $83,729,660 0.4210% 34109 70 $51,414,337 0.2585% 34110 63 $53,536,493 0.2692% 34112 77 $59,668,014 0.3000% 34113 82 $46,946,033 0.2361% 34114 87 $40,363,563 0.2030% 34116 71 $42,716,138 0.2148% 34117 70 $27,841,472 0.1400% 34119 68 $77,127,989 0.3878% 34120 67 $39,574,314 0.1990% 34133 62 $35,334 0.0002% 34134 63 $59,795,531 0.3007% 34135 63 $72,733,033 0.3657% 34136 64 $13,806 0.0001% 34137 94 $190,161 0.0010% 34138 125 $2,540,507 0.0128% 34139 109 $4,499,859 0.0226% 34140 110 $2,088,612 0.0105% 34141 102 $380,224 0.0019% 34142 57 $2,479,883 0.0125% 34143 58 $109,233 0.0005% 34145 100 $111,092,692 0.5586% 34146 98 $200,751 0.0010% 34223 45 $0 0.0000% 34224 47 $0 0.0000% 34229 40 $0 0.0000% 34231 40 $0 0.0000% 34232 40 $0 0.0000% 34233 40 $0 0.0000% 34238 41 $0 0.0000% 34239 39 $0 0.0000% 34241 41 $0 0.0000% 34242 41 $0 0.0000% 34269 44 $0 0.0000%

FPHLM V3.0 2008 209 Zipcode V3mph Total loss Percent loss 34272 42 $0 0.0000% 34274 42 $0 0.0000% 34275 42 $0 0.0000% 34277 40 $0 0.0000% 34284 42 $0 0.0000% 34285 43 $0 0.0000% 34286 44 $0 0.0000% 34287 44 $0 0.0000% 34288 44 $0 0.0000% 34289 44 $0 0.0000% 34292 44 $0 0.0000% 34293 44 $0 0.0000%

FPHLM V3.0 2008 210 Form A-4: Hurricane Andrew Percent of Losses (Map)

FPHLM V3.0 2008 211

Figure 48. Hurricane Andrew Percent of Losses.

FPHLM V3.0 2008 212 Form A-5, Items A-C: Distribution of Hurricanes by Size of Loss

A. Provide a detailed explanation of how the Expected Annual Hurricane Losses and Return Times are calculated.

B. Complete Form A-5 showing the Distribution of Hurricanes by Size of Loss. For the Expected Annual Hurricane Losses column, provide personal residential, zero deductible statewide loss costs based on the 2007 Florida Hurricane Catastrophe Fund’s aggregate exposure data found in the file named “hlpm2007.exe.”.

C. In the column, Return Time (Years), provide the return time associated with the average loss within the ranges indicated on a cumulative basis.

For example, if the average loss is $4,705 million for the range $4,501 million to $5,000 million, provide the return time associated with a loss that is $4,705 million or greater.

For each loss range in millions ($1,001-$1,500, $1,501-$2,000, $2,001-$2,500) the average loss within that range should be identified and then the return time associated with that loss calculated. The return time is then the reciprocal of the probability of the loss equaling or exceeding this average loss size.

We estimated the losses, using 2007 exposure data provided by Cat Fund, for each of the 34,098 hurricanes in the simulated stochastic set. The hurricanes were then grouped by ranges of loss size. The return period, for a given loss size, is the reciprocal of the probability of equaling or exceeding the loss size. To smooth out the relationship between return period and loss size, the return period is defined as the mean of a geometric distribution, where the probability is based on Poisson distribution.

FPHLM V3.0 2008 213 Form A-5: Distribution of Hurricanes by Size of Loss, Part A

Losses are in millions of dollars.

Range Range Average Number of Expected Return Start End Total Loss Loss Hurricanes Annual Loss Time (Yrs) 0 500 1538849.67 197.41 7795 28.76 2.32 500 1000 2739434.80 723.00 3789 51.20 2.74 1000 1500 2505032.47 1221.37 2051 46.82 3.03 1500 2000 2184556.00 1737.91 1257 40.83 3.23 2000 2500 2151534.84 2245.86 958 40.22 3.39 2500 3000 2283619.52 2748.04 831 42.68 3.53 3000 3500 2571127.68 3250.48 791 48.06 3.67 3500 4000 2700327.63 3750.46 720 50.47 3.81 4000 4500 2816922.21 4242.35 664 52.65 3.96 4500 5000 2992761.41 4742.89 631 55.94 4.11 5000 6000 6578490.84 5495.82 1197 122.96 4.34 6000 7000 7596257.48 6492.53 1170 141.99 4.69 7000 8000 7763654.26 7501.12 1035 145.12 5.08 8000 9000 8535139.19 8492.68 1005 159.54 5.51 9000 10000 9205211.14 9499.70 969 172.06 6.02 10000 11000 10152332.47 10487.95 968 189.76 6.63 11000 12000 9601744.77 11499.10 835 179.47 7.33 12000 13000 8980968.68 12473.57 720 167.87 8.08 13000 14000 9358527.59 13504.37 693 174.93 8.90 14000 15000 8651013.92 14490.81 597 161.70 9.88 15000 16000 8494770.71 15501.41 548 158.78 10.90 16000 17000 7490474.33 16498.84 454 140.01 12.03 17000 18000 7535141.87 17482.93 431 140.84 13.25 18000 19000 7106406.43 18506.27 384 132.83 14.60 19000 20000 7489168.65 19503.04 384 139.98 16.19 20000 21000 5699294.24 20501.06 278 106.53 17.88 21000 22000 6007959.42 21457.00 280 112.30 19.65 22000 23000 6136381.00 22477.59 273 114.70 21.65 23000 24000 5325396.16 23459.89 227 99.54 24.05 24000 25000 4386772.81 24507.11 179 82.00 26.26 25000 26000 4841273.54 25480.39 190 90.49 28.95 26000 27000 4002057.57 26503.69 151 74.80 31.61 27000 28000 4615849.15 27475.29 168 86.28 34.95 28000 29000 3622541.56 28523.95 127 67.71 38.37

FPHLM V3.0 2008 214 Range Range Average Number of Expected Return Start End Total Loss Loss Hurricanes Annual Loss Time (Yrs) 29000 30000 4388671.26 29454.17 149 82.03 42.73 30000 35000 14759433.72 32367.18 456 275.88 55.94 35000 40000 11210703.07 37369.01 300 209.55 91.03 40000 45000 6671414.40 42224.14 158 124.70 151.64 45000 50000 5002356.19 47192.04 106 93.50 230.12 50000 55000 3660133.41 52287.62 70 68.41 379.94 55000 60000 2278777.46 56969.44 40 42.59 615.46 60000 65000 1438178.28 62529.49 23 26.88 922.94 65000 70000 1218436.19 67690.90 18 22.77 1338.04 70000 75000 431509.97 71918.33 6 8.07 2058.25 75000 80000 697664.37 77518.26 9 13.04 2816.37 80000 90000 856333.24 85633.32 10 16.01 6688.20 90000 100000 185274.39 92637.19 2 3.46 17834.37 100000 Maximum 108645.32 108645.32 1 2.03 26751.30 Total 246568525.28 34098 4608.74

FPHLM V3.0 2008 215 Form A-5, Item D: Comparison of Return Times

D. Provide a graphical comparison of the current submission Return Times to the prior year’s submission Return Times. Return Time (Years) shall be shown on the y-axis on a log 10 scale with Losses in Billions shown on the x-axis. The legend shall indicate the corresponding submission with a solid line representing the current year and a dotted line representing the prior year.

Comparison of the Current Submission Return Times to Prior Year's Submission Return Times

100000.00

10000.00

1000.00 V3.0 V2.6

100.00 Return TimesReturn (Years) 10.00

1.00 0.0 20.0 40.0 60.0 80.0 100.0 120.0

Average Losses (Billions) Figure 49. Comparison of Return Times V2.6 and V3.0 – Different Exposure Bases

FPHLM V3.0 2008 216 Comparison of the Current Submission Return Times to Prior Year's Submission Return Times 100000.00

10000.00

1000.00 V3.0 V2.6 100.00

Return Times (Years) (Years) Times Return 10.00

1.00 0.0 20.0 40.0 60.0 80.0

Average Losses (Billions)

Figure 50. Comparison of Return Times V2.6 and V3.0 – Same Exposure Base

FPHLM V3.0 2008 217 Form A-5, Item E: Estimated Loss by Return Period

Form A-5: Distribution of Hurricanes by Size of Loss, Part B

E. Provide the estimated loss for each of the return periods given in Part B.

F. Provide this Form on CD in both an Excel and a PDF format. The file name shall include the abbreviated name of the modeler, the Standards year, and the Form name. A hard copy of Form A-5 shall be included in the submission.

Part B

Losses are in millions of dollars.

Return Time (years) Estimated Loss 500 $55,268.03 250 $47,841.56 100 $38,247.09 50 $31,151.43 20 $21,664.51 10 $14,623.57 5 $7,292.96

FPHLM V3.0 2008 218 Form A-6A: Output Ranges (See Appendix B)

A. Provide output ranges in the format shown in the file named “2007FormA6A.xls” by using an automated program or script. A hard copy of the output range spreadsheets shall be included in the submission. Provide the output ranges on CD in Excel format as specified. The file name shall include the abbreviated name of the modeler, the Standards year, and the Form name. The file shall be provided in PDF format prior to the Commission’s vote on acceptability.

B. Provide loss costs by county. Within each county, loss costs shall be shown separately per $1,000 of exposure for personal residential, tenants, condo unit owners, and mobile home; for each major deductible option; and by construction type. For each of these categories using ZIP Code centroids, the output range shall show the highest loss cost, the lowest loss cost, and the weighted average loss cost based on the 2002 Florida Hurricane Catastrophe Fund aggregate exposure data provided to each modeler in the file named “hlpm2002.exe.” A file named “02FHCFWts.xls” has also been provided for use in determining the weighted average loss costs. Include the statewide range of loss costs (i.e., low, high, and weighted average). For each of the loss costs provided, identify what that loss cost represents by line of business, deductible option, construction type, and coverages included, i.e., structure, contents, appurtenant structures, or additional living expenses as specified.

C. If a modeler has loss costs for a ZIP Code for which there is no exposure, then the modeler shall give the loss costs zero weight (i.e., assume the exposure in that ZIP Code is zero). Provide a list in the submission document of those ZIP Codes where this occurs.

D. If the modeler does not have loss costs for a ZIP Code for which there is some exposure, the modeler shall not assume such loss costs are zero, but shall use only the exposures for which it has loss costs in calculating the weighted average loss costs. Provide a list in the submission document of the ZIP Codes where this occurs.

E. All anomalies in loss costs that are not consistent with the requirements of Standard A-10 and have been explained in Disclosure A-10.1 shall be shaded.

Output ranges should be computed assuming an average structure.

Modelers shall indicate if per diem is used in producing loss costs for Coverage D (ALE) in the output ranges. If a per diem rate is used in the submission, a rate of $150.00 per day per policy shall be used.

See Appendix B for Form A-6.

FPHLM V3.0 2008 219 Form A-6B: Output Ranges (See Appendix B)

A. Provide output ranges in the format shown in the file named “2007FormA6B.xls” by using an automated program or script. A hard copy of the output range spreadsheets shall be included in the submission. Provide the output ranges on CD in Excel format as specified. The file name shall include the abbreviated name of the modeler, the Standards year, and the Form name. The file shall be provided in PDF format prior to the Commission’s vote on acceptability.

B. Provide loss costs by county. Within each county, loss costs shall be shown separately per $1,000 of exposure for personal residential, tenants, condo unit owners, and mobile home; for each major deductible option; and by construction type. For each of these categories using ZIP Code centroids, the output range shall show the highest loss cost, the lowest loss cost, and the weighted average loss cost based on the 2007 Florida Hurricane Catastrophe Fund aggregate exposure data provided to each modeler in the file named “hlpm2007.exe.” A file named “07FHCFWts.xls” has also been provided for use in determining the weighted average loss costs. Include the statewide range of loss costs (i.e., low, high, and weighted average). For each of the loss costs provided, identify what that loss cost represents by line of business, deductible option, construction type, and coverages included, i.e., structure, contents, appurtenant structures, or additional living expenses as specified.

E. All anomalies in loss costs that are not consistent with the requirements of Standard A-10 and have been explained in Disclosure A-10.1 shall be shaded.

FPHLM V3.0 2008 220 Form A-7: Percentage Change in Output Ranges

A. Provide the percentage change in the weighted average loss costs using the 2002 Florida Hurricane Catastrophe Fund’s aggregate personal residential exposure data found in the file named “hlpm2002.exe”, as of August 1, 2003 only, from the output ranges from the prior year submission for the following:

- statewide (overall percentage change),

- by region, as defined in Figure 4 – North, Central and South,

- by coastal and inland counties, as defined in Figure 5.

FPHLM V3.0 2008 221

Form A7: Percentage Change In Output Ranges

$0 Deductible $500 $1,000 $2,500 1% 2% 5% Struct- Con- App. ALE Deduct- Deduct- Deduct- Deduct- Deduct- Deduct- ure tents Struct. ible Total ible Total ible Total ible Total ible Total ible Total Frame Owners Coastal -3.65% -6.78% -5.37% -7.44% -4.52% -4.86% -5.54% -4.86% -5.27% -6.62% Inland -9.10% -11.38% -9.67% -13.64% -9.99% -10.96% -12.16% -10.96% -11.74% -14.25% North -10.36% -11.82% -10.08% -13.03% -11.42% -12.80% -13.52% -12.80% -13.49% -13.72% Central -7.60% -11.35% -8.13% -13.48% -8.59% -9.37% -10.69% -9.37% -10.19% -13.18% South 2.38% 0.91% 1.15% 0.14% 1.74% 1.40% 0.85% 1.40% 1.05% 0.32% State- wide -4.75% -7.50% -6.33% -8.20% -5.58% -5.97% -6.62% -5.97% -6.37% -7.59% Masonry Owners Coastal 2.38% 3.25% -0.90% 2.67% 2.20% 2.16% 2.32% 2.16% 2.26% 3.17% Inland -8.27% -9.90% -9.13% -12.06% -9.03% -9.92% -10.94% -9.92% -10.56% -12.68% North -10.76% -12.62% -10.95% -14.20% -12.01% -13.81% -14.78% -13.81% -14.76% -15.03% Central -6.94% -9.96% -7.69% -12.37% -7.83% -8.66% -9.86% -8.66% -9.40% -12.22% South 5.76% 8.04% 2.16% 7.46% 5.84% 5.84% 6.20% 5.84% 6.06% 7.52% State- wide 0.83% 1.74% -2.24% 1.29% 0.63% 0.56% 0.75% 0.56% 0.68% 1.69% Mobile Homes Coastal -4.92% -9.59% -4.59% -9.55% -6.46% -6.91% -8.20% -6.46% -6.91% -8.20% Inland -8.83% -16.79% -8.46% -17.45% -11.26% -12.09% -14.59% -11.26% -12.09% -14.59% North 7.19% -15.05% -9.79% -15.95% 0.41% -1.71% -8.25% 0.41% -1.71% -8.25% Central -9.96% -15.57% -7.88% -16.05% -11.71% -12.31% -14.06% -11.71% -12.31% -14.06% South -1.18% -4.41% 0.97% -4.44% -2.25% -2.54% -3.40% -2.25% -2.54% -3.40% State- wide -6.12% -11.39% -6.03% -11.58% -7.85% -8.39% -9.94% -7.85% -8.39% -9.94% Frame Renters Coastal 0.00% -8.39% 0.00% -10.89% -10.32% -10.20% -10.01% -10.38% -10.32% -10.15% Inland 0.00% -11.08% 0.00% -13.29% -15.95% -15.84% -15.43% -15.97% -15.95% -15.75% North 0.00% -12.18% 0.00% -13.51% -14.39% -14.31% -14.14% -14.42% -14.39% -14.27% Central 0.00% -11.95% 0.00% -14.29% -17.42% -17.47% -17.20% -17.35% -17.42% -17.43% South 0.00% 2.46% 0.00% 0.17% 2.10% 2.34% 2.62% 1.95% 2.10% 2.43% State- wide 0.00% -9.02% 0.00% -11.39% -11.15% -11.02% -10.81% -11.22% -11.15% -10.97%

Form A7: Percentage Change In Output Ranges Masonry Renters Coastal 0.00% 1.96% 0.00% 1.40% 3.00% 3.31% 3.52% 2.80% 3.00% 3.40% Inland 0.00% -9.96% 0.00% -12.06% -14.52% -14.36% -13.92% -14.57% -14.52% -14.25% North 0.00% -11.71% 0.00% -14.31% -14.12% -14.06% -13.91% -14.14% -14.12% -14.02%

FPHLM V3.0 2008 222 Form A7: Percentage Change In Output Ranges Central 0.00% -10.19% 0.00% -12.44% -15.86% -15.88% -15.52% -15.79% -15.86% -15.82% South 0.00% 7.38% 0.00% 6.80% 9.11% 9.48% 9.71% 8.87% 9.11% 9.58% State- wide 0.00% 0.50% 0.00% -0.11% 1.82% 2.13% 2.31% 1.60% 1.82% 2.22% Frame Condos Coastal -1.64% -3.71% 0.00% -4.34% -4.02% -4.42% -4.09% -4.02% -4.42% -4.09% Inland -9.32% -11.57% 0.00% -13.84% -13.54% -16.66% -16.41% -13.54% -16.66% -16.41% North -10.24% -10.75% 0.00% -12.60% -12.11% -12.63% -12.52% -12.11% -12.63% -12.52% Central -7.78% -12.36% 0.00% -14.30% -14.58% -17.51% -17.65% -14.58% -17.51% -17.65% South 2.32% 1.70% 0.00% 0.92% 1.41% 1.37% 1.84% 1.41% 1.37% 1.84% State- wide -2.64% -4.22% 0.00% -4.94% -4.54% -4.88% -4.53% -4.54% -4.88% -4.53% Masonry Condos Coastal 3.38% 5.06% 0.00% 3.90% 4.96% 5.63% 6.26% 4.96% 5.63% 6.26% Inland -8.31% -9.98% 0.00% -12.31% -11.94% -15.18% -14.87% -11.94% -15.18% -14.87% North -9.06% -10.36% 0.00% -11.75% -11.48% -12.02% -11.93% -11.48% -12.02% -11.93% Central -5.95% -10.56% 0.00% -12.47% -12.88% -16.10% -16.26% -12.88% -16.10% -16.26% South 5.27% 7.74% 0.00% 6.64% 7.79% 8.72% 9.42% 7.79% 8.72% 9.42% State- wide 3.07% 4.86% 0.00% 3.66% 4.77% 5.48% 6.11% 4.77% 5.48% 6.11%

FPHLM V3.0 2008 223 Form A-8: Percentage Change in Output Ranges by County

Provide color-coded maps by county reflecting the percentage changes in the weighted average 2% deductible loss costs for frame owners, masonry owners, mobile homes, frame renters, masonry renters, frame condos, and masonry condos from the output ranges using the 2002 Florida Hurricane Catastrophe Fund’s aggregate personal residential exposure data found in the file named “hlpm2002.exe”, as of August 1, 2003 only.

Counties with a negative percentage change (reduction in loss costs) shall be indicated with shades of blue; counties with a positive percentage change (increase in loss costs) shall be indicated with shades of red, and counties with no percentage change shall be white. The larger the percentage change in the county, the more intense the color-shade.

FPHLM V3.0 2008 224

Figure 51. Percentage Change in Output Ranges by County – Owners Frame 2% Deductible

FPHLM V3.0 2008 225

Figure 52. Percentage Change in Output Ranges by County – Owners Masonry 2% Deductible

FPHLM V3.0 2008 226

Figure 53. Percentage Change in Output Ranges by County – Mobile Homes 2% Deductible

FPHLM V3.0 2008 227

Figure 54. Percentage Change in Output Ranges by County – Renters Frame 2% Deductible

FPHLM V3.0 2008 228

Figure 55. Percentage Change in Output Ranges by County – Renters Masonry 2% Deductible

FPHLM V3.0 2008 229

Figure 56. Percentage Change in Output Ranges by County – Condo Frame 2% Deductible

FPHLM V3.0 2008 230

Figure 57. Percentage Change in Output Ranges by County – Condo Masonry 2% Deductible

FPHLM V3.0 2008 231 STATISTICAL STANDARDS

S-1 Modeled Results and Goodness-of-Fit

A. The use of historical data in developing the model shall be supported by rigorous methods published in currently accepted scientific literature.

The historical data for the period 1900-2006 was modeled using scientifically accepted methods that have been published in accepted scientific literature.

B. Modeled and historical results shall reflect agreement using currently accepted scientific and statistical methods in the appropriate disciplines.

Modeled and historical results are in agreement as indicated by appropriate statistical and scientific tests. Some of these tests will be discussed below.

Disclosures

1. Identify the form of the probability distributions used for each function or variable, if applicable. Identify statistical techniques used for the estimates and the specific goodness-of-fit tests applied. Describe whether the p-values associated with the fitted distributions provide a reasonable agreement with the historical data.

Historical initial conditions are used to provide the seed for storm genesis in the model. Small uniform random error terms are added to the historical starting positions, intensities and changes in storm motion. Subsequent storm motion and intensity is determined by randomly sampling empirical probability distribution functions derived from the HURDAT historical record.

Figure 58 shows the occurrence rate of both modeled and historical landfalling hurricanes in Florida. The figure shows a good agreement between historical and modeled occurrences. A Chi square goodness of fit test, for the number of years with 0, 1, and 2 or more hurricanes per year (3 bins with 5 or more occurrences giving 2 degrees of freedom), gives a p-value of approximately 0.82. An analysis of landfalls by each region and intensity in Florida is given in Form M-1. The landfall occurrences by region are reasonable as indicated by the Chi square tests shown in Form M-1, especially given the large uncertainties in the historical record, particularly with regard to intensity.

FPHLM V3.0 2008 232 0.7

0.6

0.5

0.4

0.3 Historical Modeled 0.2 Occurrence rate

0.1

0 0123456

Events per Year

Figure 58. Comparison of simulated vs historical occurrences

Distribution of the B parameter 16 15 14 13 12 11 10 9 Observed 8 Model Scaled 7 6 Occurrence 5 4 3 2 1 0 0.5 1 1.5 2 B param eter

Figure 59. Comparison between the modeled and observed Willoughby and Rahn (2004) B data set

The random error term for the Holland B is modeled using a Gaussian distribution with a standard deviation of 0.286. Figure 59 shows a comparison between the Willoughby and Rahn (2004) B data set (see Standard M-2.1) and the modeled results (scaled to equal the 116 measured occurrences in the observed data set). The modeled results with the error term have a mean of about 1.38 and are consistent with the observed results. The figure indicates excellent agreement, and the Chi square goodness of fit gives a p-value about 0.57 using 8 degrees of freedom (re-binning to 11 bins and two estimated parameters). A KS goodness of fit yields a p- value of 0.845 (ks=0.057).

FPHLM V3.0 2008 233 We develop an Rmax model using the revised landfall Rmax database which includes 108 measurements for storms up to 2005. We have opted to model the Rmax at landfall rather than the entire basin for a variety of reasons. One is that the distribution of landfall Rmax may be different than that over open water. An analysis of the landfall Rmax database and the 1988-2007 DeMaria Extended Best Track data shows that there appears to be a difference in the dependence of Rmax on central pressure (Pmin) between the two data sets. The landfall data set provides a larger set of independent measurements, more than 100 storms compared to about 31 storms affecting the Florida threat area region in the Best Track Data. Since landfall Rmax is most relevant for loss cost estimation, and has a larger independent sample size, we have chosen to model the landfall data set. Future studies will examine how the Extended Best Track Data can be used to supplement the landfall data set.

Based on the semi-boundedness and skewness of Rmax, we sought to model the distribution using a gamma distribution. Using the maximum likelihood estimation method, we found the estimated parameters for the gamma distribution, kˆ = 53547.5 and θˆ = 67749.4 . With these estimated values, we show a plot of the observed and expected distribution in Figure 60. The Rmax values are binned in 5 sm intervals, with the x-axis showing the end value of the interval.

Gam m a v s Observ ed 22.5

17 5 Observed 15 Gam m a

12 5

7 5

2.5

0 5 10 15 20 25 30 35 40 45 50 55

Figure 60. Observed and expected distribution using a Gamma distribution

The gamma distribution showed a reasonable fit. A Chi square goodness of fit test yields a p- value of 0.26 with 5 degrees of freedom (re-binning to 8 bins to ensure more than 5 occurrences per bin and 2 estimated parameters). A KS goodness of fit yields a p-value of 0.80 (ks= 0.063).

2. Provide the source and the number of years of the historical data set used to develop probability distributions for specific hurricane characteristics. If any modifications have been made to the data set, describe them in detail and their appropriateness.

FPHLM V3.0 2008 234 The model uses the National Hurricane Center HURDAT file from June 2007 for the period 1900-2006. This information is provided in detail in Disclosures M-1.1, M-1.2 and M-2.1. (Met Standards).

3. Describe the nature and results of the tests performed to validate the wind speeds generated.

We compare the cumulative effect of a series of modeled and observed wind fields by comparing the peak winds observed at a particular zip code during the entire storm life-cycle. We also compare our modeled wind fields to those that have been constructed from all available observations which are freely available on the NOAA AOML-HRD web site. A subsequent section describes the process for recording the peak modeled and observed wind speeds (wind swaths) from which the validation statistics are generated. Our validation is based on nine hurricanes that by-passed or made landfall in Florida. These hurricanes were well observed and we will have the ability to add new storms and quickly conduct new validation studies as our validation set grows and we make enhancements to the model. In order to run the Loss Model in “scenario” mode for doing validation studies, we had to construct detailed storm track histories for recent storms affecting Florida using the HURDAT, Rmax and Holland B databases. The validation suite included 1992 Hurricane Andrew and the following 2004 and 2005 storms: Charley, Frances, Jeanne, Ivan, Dennis, Katrina, Rita, and Wilma. The validations make use of the Hurricane Research Division’s Surface Wind Analysis System (H*Wind).

H*WIND The HRD approach to hurricane wind analysis employed in H*Wind evolved from a series of peer-reviewed, scientific publications analyzing landfalls of major hurricanes including Frederic of 1979, Alicia of 1983, Hugo of 1989, and Andrew of 1992 (Powell et al., 1991, Powell and Houston, 1996, 1998, Powell et al., 1998). In Powell et al., (1991) which described Hurricane Hugo's landfall, a concept was developed for conducting a real-time analysis of hurricane wind fields.. The system was first used in real-time during Hurricane Emily in 1993 (Burpee et al., 1994). Since 1994, HRD wind analyses have been conducted on an experimental basis to create real time hurricane wind field guidance for forecasters at the National Hurricane Center. During Hurricane landfall episodes from 1995-2005, HRD scientists have conducted research side by side with hurricane specialists at NHC analyzing wind observations on a regular 3 or 6 hour schedule consistent with NHC's warning and forecast cycle.

An HRD wind analysis requires the input of all available surface weather observations (e.g., ships, buoys, coastal platforms, surface aviation reports, reconnaissance aircraft data adjusted to the surface, etc.). Observational data are downloaded on a regular schedule and then processed to fit the analysis framework. This includes the data sent by NOAA P3 and G4 research aircraft during the HRD hurricane field program, including the Step Frequency Microwave Radiometer measurements of surface winds and U.S. Air Force Reserves (AFRES) C-130 reconnaissance aircraft, remotely sensed winds from the polar orbiting SSM/I and ERS, the QuikScat platform and TRMM microwave imager satellites, and GOES cloud drift winds derived from tracking low level near-infrared cloud imagery from these geostationary satellites. These data are composited relative to the storm over a 4-6 hour period. All data are quality controlled and processed to conform to a common framework for height (10 m or 33 feet), exposure (marine or open terrain

FPHLM V3.0 2008 235 over land), and averaging period (maximum sustained 1 minute wind speed) using accepted methods from micrometeorology and wind engineering (Powell et al., 1996, Powell and Houston, 1996). This framework is consistent with that used by the National Hurricane Center (NHC), and is readily converted to wind load frameworks used in building codes. Based on a qualitative examination of various observing platforms and methods used to standardize observations, Powell et al., 2005 suggest that the uncertainty of the maximum wind from a given analysis ranges from 10-20% depending on the observing platform. In general the uncertainty of a given H*Wind analysis is of the order of 10% for analysis of Hurricanes Ivan, Frances, Jeanne, and Katrina, all of which incorporated more accurate surface wind measurements from the Stepped Frequency Microwave Radiometer (SFMR) aboard the NOAA research aircraft. The SFMR data used for those analyses was post-processed during the fall of 2005 using the latest geophysical model function relating wind speed to sea surface foam emissivity. Hurricanes Charley, Dennis, Rita, Wilma, and Andrew did not have the benefit of SFMR measurements but relied on adjusting Air Force reconnaissance observations at the 3 km altitude to the surface with empirical reduction methods. The method used was based on how SFMR measurements compared to flight level winds and depended on storm relative azimuth. Preliminary results suggest that this method has an uncertainty of 15%. We created wind swaths for both the modeled and observed winds. We also computed the maximum winds at zip codes for both the observed and modeled winds, and from that we derived the mean and root-mean-square error (see Table 17 and Table 18).

WIND SWATHS For each storm in the validation set, the peak sustained surface wind speed is recorded at each zip code in Florida for the duration of the storm event. Observed wind fields from H*Wind and modeled wind fields from the public model are moved along the exact same tracks, which are the observed high-resolution storm tracks assembled from reconnaissance aircraft and radar data. For each storm, the recorded peak of the observed and modeled wind speed is saved at each grid point as well as of each zip code, and the resulting zip code comparison pairs provide the basis for the model validation statistics. The peak grid point values are color contoured and mapped as graphics showing the “swath” of maximum winds swept out by the storm passage. Wind swaths are sometimes confused with a wind field. The winds depicted in a wind swath do not have time continuity, cannot depict a circulation, and therefore cannot be described as a wind field. A wind field represents a vector field that represents a representative instance of the surface wind circulation.

Wind swaths were constructed for both the modeled and observed winds. Maximum marine exposure winds were compared at all zip codes for both the observed and modeled winds (Figure 61) from which we derived the mean and root-mean-square error statistics shown in Table 18. This type of comparison provides an unvarnished assessment of model performance.

FPHLM V3.0 2008 236

Figure 61. Comparison of modeled (left) and observed (right) swaths of maximum sustained open terrain surface winds for Hurricane Andrew of 1992 in South Florida. Hurricane Andrew observed swath is based on adjusting flight-level winds with the SFMR- based wind reduction method.

FPHLM V3.0 2008 237 Table 17. Validation Table based on zip code wind swath comparison of the Public wind field model to H*Wind. Mean errors (bias) of model for the set of validation wind swaths. Errors (upper number in each cell) are computed as Modeled – Observed (Obs) at zip codes where modeled winds were within wind thresholds (model threshold) or where observed winds were within respective wind speed threshold (H*Wind threshold). Number of zip codes for the comparisons is indicated as the lower number in each cell.

56-74 75-112 >112mph >56mph 56-74 75-112 >112mph >56mph Storms Year Model Model Model Model H*Wind H*Wind H*Wind H*Wind Threshold Thresh. Thresh. Thresh. Thresh. Thresh. Thresh. Thresh.

5.25 13.86 2.73 7.49 10.26 12.47 0.66 7.68 Andrew 1992 92 107 100 299 139 54 88 281

12.96 21.36 -7.36 17.80 8.58 -3.09 -8.91 3.47 Charley 2004 112 244 13 369 122 63 17 202

3.99 -0.99 3.38 -0.59 -4.48 -1.38 Frances 2004 None None 693 96 789 372 96 468

-6.95 -3.35 -4.59 -5.76 -3.73 -4.44 Ivan 2004 None None 20 38 58 22 41 63

6.78 3.95 5.56 2.67 -3.87 0.38 Jeanne 2004 None None 250 190 440 225 121 346

2.45 6.98 5.87 5.22 7.57 -4.37 5.87 Dennis 2005 None 15 46 61 29 29 3 61

Dennis -12.65 -12.65 2005 None None None None None None Keys 5 5

-11.43 -2.42 -6.34 -8.93 -11.57 -10.55 Katrina 2005 None None 77 100 177 93 149 242

6.28 14.54 9.38 12.01 12.01 Rita 2005 None None None 5 3 8 5 5

0.44 -9.99 -7.35 6.54 -13.35 -9.77 Wilma 2005 None None 133 394 527 87 396 483

FPHLM V3.0 2008 238 Table 18. Validation Table based on zip code wind swath comparison of the Public wind field model to H*Wind. Root mean square (RMS) wind speed errors (mph) of model for the set of validation wind swaths. Errors are based on Modeled – Observed (Obs) at zip codes where modeled winds were within wind thresholds (model threshold) or where observed winds were within respective wind speed threshold (H*Wind threshold). Number of zip codes for the comparisons is indicated as the lower number in each cell. 56-74 75-112 >112mph >56mph 56-74 75-112 >112mph >56mph Storms Year Model Model Model Model H*Wind H*Wind H*Wind H*Wind Threshold Thresh. Thresh. Thresh. Thresh. Thresh. Thresh. Thresh.

Andrew 1992 6.11 15.75 7.024 10.81 12.19 14.26 5.82 11.10

Charley 2004 19.84 26.59 10.08 24.30 16.65 8.60 11.69 14.21

Frances 2004 8.08 11.20 None 8.52 4.99 10.20 None 6.41

Ivan 2004 7.07 5.20 None 5.91 6.11 5.51 None 5.72

Jeanne 2004 10.14 9.65 None 9.93 10.88 6.16 None 9.50

Dennis 2005 3.06 9.19 None 8.12 6.15 9.93 4.59 8.12

Dennis 2005 None None None None 12.67 None None 12.67 Keys

Katrina 2005 14.66 8.25 None 11.49 12.50 17.97 None 16.09

Rita 2005 6.4992 14.54 None 10.28 12.41 None None 12.41

Wilma 2005 14.73 14.05 None 14.22 12.51 14.83 None 14.44

RMS 10.18 14.87 6.26 12.37 9.75 12.79 6.71 11.19 All N 1397 1218 113 2728 1099 949 108 2156

Comparison of model and H*Wind sustained marine exposure wind speeds at zip codes receiving model wind speeds over the given thresholds (Table 17) indicates a positive bias. For zip codes where model wind speeds exceeded 56 mph, the bias is +3.3 mph and negative bias was apparent in Hurricanes Ivan, Katrina, and Wilma. At other wind speed thresholds, low bias is evident for winds > 112 mph in Charley, and winds of 75-112 mph in Frances, Ivan, Katrina, and Wilma. For winds of 56-74 mph, low bias is noted in Hurricanes Ivan, and Katrina. Errors

FPHLM V3.0 2008 239 for Andrew are relatively high but the lack of observations for Andrew makes it difficult to determine if it was a Cat 4 or Cat 5 hurricane during its landfall in South Florida. Rita in the Keys also shows relatively high bias but observations indicate that there were fluctuations in intensity over a short period of time during its passage past the Keys. Model errors for Charley are also relatively highly likely due to the model producing too broad a wind field. When model winds are compared to H*Wind at zip codes exceeding H*Wind, and sustained wind speed thresholds of 56 mph are considered, the mean bias is -2.2 mph. However, bias at other wind speed thresholds is larger, primarily caused by large model - H*Wind differences in Hurricane Andrew, Charley, and Rita.

When swaths are evaluated at zip codes, a positive wind speed bias of ~3 mph is indicated. However, the model can also under-predict swaths for individual cases. While bias correction is an accepted practice for numerical weather prediction, there is no evidence that the model has a consistent bias. The swath bias is probably associated with limitations in specifying the radial pressure profile after landfall. The tendency for the Holland pressure profile parameter to produce too broad an area of strong winds near the eyewall is the most likely cause of bias and is likely a feature found in many of the current risk models. Therefore we have decided to forego any corrective measures at this point.

Our validation set is unique in that the values of storm position, motion, Rmax and Pmin are observed, and B is determined independently from the H*Wind field. In other words, there is no tuning dial that we can turn to try to improve our results. Although additional validation storms are desired, we believe the positive bias for locations with winds > 56 mph is a characteristic of models that use the Holland B pressure profile parameter, which tends to produce model fields that are too broad outside the radius of maximum winds. Our validation method provides an objective means of assessing model performance by evaluating the portion of the wind field that contains damaging winds.

The root mean square (RMS) error (Table18) provides a better estimate of model uncertainty. For zip codes in which model winds were 56-74 mph, the RMS error is +/- 10 mph (~ 15%), for 75-112 mph the error is +/- 15 mph (~16%), and for winds > 112 mph the errors is +/- 6 mph (~ 5%). In general, for winds > 56 mph, the RMS error is +/- 12 mph or ~ 13%. RMS errors are similar for zip codes in which H*Wind wind speeds fell into the respective thresholds.

FPHLM V3.0 2008 240 SUMMARY OF WIND SWATH VALIDATION Validation of the winds from the wind model against the H*WIND analyses was prepared by considering winds that would be strong enough to be associated with damage. Threshold-based comparisons could miss places where the observed winds were greater than the model and the model was below the threshold. Conversely, observed winds over the same thresholds can be compared to the co-located model grid points but would miss places where the observed winds were below the threshold. It is important to evaluate the errors both ways to see if a consistent bias is evident. According to our validation statistics, albeit for a relatively small number of cases, wind swath zip code comparisons show evidence of a 3 mph positive bias but it is not consistent for all storms. The bias is likely related to the limitations of the Holland B pressure profile specification. The model uncertainty, as estimated by the RMS error, is on the order of 15%.

4. Provide the date of loss of the insurance company data available for validation and verification of the model.

The following hurricane data from different insurance companies are used to validate the model:

Andrew 1992 Charley 2004 Frances 2004

FPHLM V3.0 2008 241 5. Provide an assessment of uncertainty in loss costs for output ranges using confidence intervals or other accepted scientific characterizations of uncertainty.

While the model does not automatically produce confidence intervals for the output ranges, the data does allow for the calculation of confidence intervals. The mean and the standard deviation of the losses are calculated for each county and found that the standard errors are within less than 2.5% of the means for all counties. We have calculated the coefficient of variation (CV) for all counties and drew a histogram which is provided in Figure 62. We noticed that the range of the CVs is found to be (2.81 to 5.03).

Histogram of CVs

0 2.53.03.54.04.55.05.5 CV

Figure 62. Histogram of the CVs for all counties combined

6. Justify any differences between the historical and modeled results using current accepted scientific and statistical methods in the appropriate disciplines.

The various statistical tests as well as other validation tests presented here and elsewhere indicate that any differences between modeled results and historical are not statistically significant given the large known uncertainties in the historical record.

7. Provide graphical comparisons of modeled and historical data and goodness-of-fit tests. Examples include hurricane frequencies, tracks, intensities, and physical damage

For hurricane frequencies as a function of intensity by region, see Form M-1 plots and goodness of fit table. The histogram in Figure 58 compares the modeled and historical annual landfall distribution by number of events per year. The agreement between the two distributions is quite close and the histogram shows a good fit. The chi-square goodness of fit test gives a p-value of 0.82 as described in S-1.1. Plots and goodness-of-fit for the radius of maximum wind and the Holland pressure profile parameter are shown in Disclosure 1 of this standard. Plots and statistical comparisons of historical and modeled losses are shown in Standard S-5, Form S-3 and Form S-4.

FPHLM V3.0 2008 242 8. Provide a completed Form S-1, Probability of Florida Landfalling Hurricanes per Year.

See the completed Form S-1 at the end of this section.

9. Provide a completed Form S-2, An example of a Probable Maximum Loss(PML) Based on a Limited Hypothetical Data set. See completed Form S-2 at the end of this section.

FPHLM V3.0 2008 243 S-2 Sensitivity Analysis for Model Output

The modeler shall have assessed the sensitivity of temporal and spatial outputs with respect to the simultaneous variation of input variables using currently accepted scientific and statistical methods in the appropriate disciplines and have taken appropriate action.

We have assessed the sensitivity of temporal and spatial outputs with respect to the simultaneous variation of input variables using currently accepted scientific and statistical methods and submitted to the commission with the original submission in 2007.

Disclosures

1. Provide a detailed explanation of the sensitivity analyses that have been performed on the model above and beyond those completed for the original submission of Form S-5 and provide specific results. (Requirement for modeling organizations that have previously provided the Commission with Form S-5. This disclosure can be satisfied with an updated Form S-5 that incorporates changes to the model since the previous submission of the Form).

We have not done any sensitivity analyses on the model above and beyond those completed for the original submission Form S-5. In Form S-5, the following input variables were used.

CP = central pressure (in millibars) Rmax = radius of maximum winds (in statute miles) VT = translational velocity (forward speed in miles per hour) Holland B pressure profile parameter

2. Provide a description of the statistical methods used to perform the sensitivity analysis.

We have followed the procedures as described in the paper “Assessing Hurricane Effects. Part 1. Sensitivity Analysis,” by Ronald L. Iman, Mark E. Johnson, and Tom Schroeder (2000a).

3. Identify the most sensitive aspect of the model and the basis for making this determination. Provide a full discussion of the degree to which these sensitivities affect output results and illustrate with an example.

For the sensitivity analysis, some selected graphs of the standardized regression coefficients vs time and for Category 1, 3 and 5 hurricanes are provided in Figure 63- Figure 65. From these graphs, we observed that the maximum sustained surface wind speed (MSSWS) is most sensitive to Rmax parameter followed by VT, Holland B and CP. At hour 0, MSSWS is the most sensitive to Rmax, where as at hour 12, MSSWS is the most sensitive to VT. We also noticed that the sensitivity of MSSWS depends on the time, grid points and the category of hurricanes.

FPHLM V3.0 2008 244 Grid 30,0

1 0.8 0.6 0.4 HollandB 0.2 Rmax 0 -0.2 VT

Coefficient -0.4 CP -0.6 -0.8 Standardized Regression Regression Standardized -1 0123456789101112 Time (hr)

Figure 63. Standardized Regression Coefficients vs. Time at Grid Coordinates (30,0) for Category 1

Grid 30,0

1 0.8 0.6 0.4 HollandB 0.2 Rmax 0 -0.2 VT

Coefficient -0.4 CP -0.6 -0.8 Standardized Regression Regression Standardized -1 0123456789101112 Time (hr)

Figure 64. Standardized Regression Coefficients vs. Time at Grid Coordinates (30,0) for Category 3

FPHLM V3.0 2008 245 Grid 30,0

1.2 1 0.8 0.6 HollandB 0.4 0.2 Rmax 0 VT -0.2 Coefficient -0.4 CP -0.6 -0.8

Standardized Regression Regression Standardized -1 0123456789101112 Time (hr)

Figure 65. Standardized Regression Coefficients vs. Time at Grid Coordinates (30,0) for Category 5

FPHLM V3.0 2008 246 4. Describe how other aspects of the model may have a significant impact on the sensitivities in output results and the basis for making this determination.

Validation studies (described in Standard S-1.3) indicated that air density, boundary layer height, fraction of the boundary layer depth over which the turbulent stresses act, the drag coefficient, the averaging time chosen to represent the boundary layer slab winds, and the reduction factor to adjust slab winds to the surface all have a significant effect on the output results. These quantities were evaluated during the validation process, resulting in the selection of physically consistent values. For example, the values chosen for air density, marine boundary layer height , and reduction factor from the mean boundary layer to the surface are representative of near surface GPS dropsonde measurements in hurricanes.

Model wind speeds are very sensitive to zip code roughness, which in turn depend on land use/land cover determined from satellite remote sensing, and the assignment of roughness to mean land use / land cover classifications as well as the upstream filtering or weighting factor applied to integrate the upstream roughness elements within a 45 degree sector to windward of the zip code. When zip codes are updated to reflect annual changes and population centroids are updated, the roughness table is also updated. Zip code location changes will generate different wind speeds. Experiments with different land use land cover filtering factors suggest that extending the filtering further upstream has the effect of a small reduction in roughness at Florida zip codes (probably due to proximity to the coast or smoother Everglades areas) with slightly higher wind speeds. However, loss cost sensitivity was found to be small (~ $0.24B).

5. Describe actions taken in light of the sensitivity analyses performed.

No actions were taken in light of the aforementioned sensitivity experiments.

6. Provide a completed Form S-5, Hypothetical Events for Sensitivity and Uncertainty Analysis (requirement for models submitted by modeling organizations which have not previously provided the Commission with this analysis).

A Completed Form S-5 has been submitted with the original submission in 2007.

FPHLM V3.0 2008 247 S-3 Uncertainty Analysis for Model Output

The modeler shall have performed an uncertainty analysis on the temporal and spatial outputs of the model using currently accepted scientific and statistical methods in the appropriate disciplines and have taken appropriate action. The analysis shall identify and quantify the extent that input variables impact the uncertainty in model output as the input variables are simultaneously varied.

We have performed an uncertainty analysis on the temporal and spatial outputs of the model using currently accepted scientific and statistical methods and submitted to the commission with the original submission in 2007.

Disclosures

1. Provide a detailed explanation of the uncertainty analyses that have been performed on the model above and beyond those completed for the original submission Form S-5 and provide specific results. (Requirement for modeling organizations that have previously provided the Commission with Form S-5. This disclosure can be satisfied with an updated Form S-5 that incorporates changes to the model since the previous submission of the Form).

We have not done any uncertainty analyses on the model above and beyond those completed for the original submission Form S-5. In Form S-5, the following input variables were used.

CP = central pressure (in millibars) Rmax = radius of maximum winds (in statute miles) VT = translational velocity (forward speed in miles per hour) Holland B pressure profile parameter

2. Provide a description of the statistical methods used to perform the uncertainty analysis.

We have followed the procedures as described in the paper “Assessing Hurricane Effects. Part 2. Uncertainty Analysis,” by Ronald L. Iman, Mark E. Johnson, and Tom Schroeder (2000b).

3. Identify the major contributors to the uncertainty in model outputs and the basis for making this determination. Provide a full discussion of the degree to which these uncertainties affect output results and illustrate with an example.

Expected Percentage Reductions in the variance of Maximum Sustained Surface Wind Speed (MSSWS) for Category 1, 3 and 5 Hurricanes versus Time at Coordinate (30,0) are presented in Figure 66 –Figure 68. The major contribution to the uncertainty in the model is R-max followed by VT, Holland B and CP. At hour 0, Rmax produces the most uncertainty and at hour 12 VT contributed the highest uncertainty in the model. It is also noted that at hour 2 there is no uncertainty among the four parameters except VT.

FPHLM V3.0 2008 248 Grid 30,0

125

100 HollandB 75 Rmax 50 VT

Reduction CP 25 Expected Percentage Percentage Expected 0 0123456789101112 Time (hr)

Figure 66. Expected Percentage Reductions in the Var(MSSWS) for a Category 1 Hurricane versus Time at Coordinate (30,0)

Grid 30,0

125

100 HollandB 75 Rmax 50 VT

Reduction CP 25 Expected Percentage Percentage Expected 0 0123456789101112 Time (hr)

Figure 67. Expected Percentage Reductions in the Var(MSSWS) for a Category 3 Hurricane versus Time at Coordinate (30,0)

FPHLM V3.0 2008 249 Grid 30,0

125

100 HollandB 75 Rmax VT 50 CP Reduction

25 Expected Percentage Expected Percentage

0 0123456789101112 Time (hr)

Figure 68. Expected Percentage Reductions in the Var(MSSWS) for a Category 5 Hurricane versus Time at Coordinate (30,0)

4. Describe how other aspects of the model may have a significant impact on the uncertainties in output results and the basis for making this determination.

Limitations in the HURDAT record contribute to the uncertainty of modeled tracks and pressures. Surface pressure measurements are not always available in HURDAT and estimating surface pressures by pressure-wind relationships are also fraught with uncertainty since well- observed hurricanes can demonstrate a large variation in maximum wind speeds for a given minimum surface pressure. The HURDAT record prior to the advent of satellites in the mid 1960s probably missed many hurricanes that affected Florida in the early 20th century. There is still considerable uncertainty in the assessment of hurricane intensity, even today. Recent preliminary research results based on SFMR measurements indicate that some Saffir-Simpson 1- 3 Category hurricanes may be rated too high while the Category 4 and 5 storms are probably rated accurately.

Uncertainty in zip code roughness has a significant impact on wind uncertainty. For a given zip code, changing the zip code roughness tables from 2004 to 2006 demonstrated some instances of large changes in roughness for the same zip code for specific upstream flow directions. This is due to a shift in the zip code population-weighted centroid location, and a resulting incorporation of different upstream land use / land cover elements.

5. Describe actions taken in light of the uncertainty analyses performed.

No actions were taken in light of the aforementioned uncertainty analysis.

6. For models submitted by modeling organizations, which have not previously provided this analysis to the Commission, Form S-5 was disclosed under Standard S-2 and will be used

FPHLM V3.0 2008 250 in the verification of Standard S-3.

A Completed Form S-5 has been submitted with the original submission in 2007.

FPHLM V3.0 2008 251 S-4 County Level Aggregation

At the county level of aggregation, the contribution to the error in loss cost estimates attributable to the sampling process shall be negligible.

The error in the county level loss costs induced by the sampling process can be quantified by computing standard errors for the county level loss costs. These loss costs have been computed for all counties in the state of Florida using 53500 years of simulation. The results indicate that the standard errors are less than 2.5% of the average loss cost estimates for all counties.

Disclosures

1. Describe the sampling plan used to obtain the average annual loss costs and output ranges. For a direct Monte Carlo simulation, indicate steps taken to determine sample size. For an importance sampling design, describe the underpinnings of the design.

The number of simulation runs was determined through the following process:

The average loss cost, X Y , and standard deviation sY, was determined for each county Y using an initial run of a 10,700 year simulation. Then the maximum error of the estimate will be 2.5% of the estimated mean loss cost, if the number of simulation runs for county Y is: 2 ⎛ s ⎞ ⎜ Y ⎟ NY = ⎜ ⎟ ⎝ 025.0 XY ⎠ Based on the initial 10,700 year simulation runs, the minimum number of years required is NY = 35080 for Lafayette County, which had the highest number of years required of all the counties. Therefore, we have decided to use 53,500 (500x107) years of simulation for our final results. From the 53,500 simulation year runs we find that the standard errors are less than 2.5% for each county.

FPHLM V3.0 2008 252 S-5 Replication of Known Hurricane Losses

The model shall estimate incurred losses in an unbiased manner on a sufficient body of past hurricane events from more than one company, including the most current data available to the modeler. This Standard applies separately to personal residential and, to the extent data are available, to mobile homes. Personal residential experience may be used to replicate structure-only and contents-only losses. The replications shall be produced on an objective body of loss data by county or an appropriate level of geographic detail.

The following Table 19 compares the modeled and actual total losses by hurricane and company for residential coverage. Moreover, Figure 69 indicates reasonable agreement between the observed and modeled losses (r=0.99, which indicates a strong positive correlation).

Disclosures

1. Describe the nature and results of the analyses performed to validate the loss projections generated by the model.

For model validation purposes, the actual and modeled losses for some selected companies and hurricanes are provided in the Table 19.

Table 19 Actual vs. Model Loss Name Event Total Actual Loss Total Modeled Loss A Charley 110471361 135347764 A Frances 20201407 78044136 B Andrew 2984373067 2558464266 B Charley 1037108745 735800992 B Charley_Mob 23395988 25990713 B Frances 614006549 428145647 B Frances_Mob 18467176 8466678 B Erin 50519119 60458572 C Charley 63889029 32714722 C Frances 122776727 88435747 D Charley 274702333 239567424 D Frances 224656954 142289724 E Charley 62086256 53273107 E Frances 43799401 18568918 F Charley 111013524 269149904 F Frances 94272660 380701388 G Charley 952353 900161 G Frances 10007410 4176704 H Charley 13157215 8547382 H Frances 15499060 7563073 H Jeanne 8403121 9178144 I Charley 54207520 52973831

FPHLM V3.0 2008 253 I Frances 121893725 46891133 J Charley 2015902 2434734 J Frances 2659551 3716741 K Charley 113313510 51133868 K Frances 78377163 62858848 K Jeanne 40245030 67467706 L Charley 32316645 28614453 L Jeanne 3125588 10242976 M Jeanne 31066792 36722644 N Charley_Mob 79751698 82168190 N Jeanne_Mob 81552694 110169688 J Jeanne_Mob 29144703 34822744 J Jeanne 2059383 3696970 O Jeanne 84545829 83045444 P Charley 15135021 27026195 P Frances 9399468 19818798 P Jeanne 9048905 27597126

The following Figure 69 provides a comparison of total actual losses vs. total modeled losses by different hurricanes. The comparison indicates a reasonable agreement between the actual and modeled losses. The correlation (measure of precision) between actual and modeled losses is found to be 0.988, which indicates a very strong positive correlation between actual and modeled losses. When we test the difference in paired mean values equals zero, the paired t-test ( t = 0.9882, df = 38, p-value = 0.3293) indicates that we fail to reject the null hypothesis based on this data, and conclude that there is insufficient evidence to suggest a difference between actual and modeled losses. We also observed from Table 19 that about 51% of the actual losses are more than the corresponding model losses and 49% of the model losses are more than the corresponding actual losses. Following Lin (1989), the bias correction factor (measure of accuracy) is obtained as 0.985 and the sample concordance correlation coefficient is found to be 0.973 which showed a very good agreement between actual and model losses.

Scatter plot between Total Actual Losses and Modeled Losses Model Losses 0 5*10^8 10^9 1.5*10^9 2.5*10^9

0 5*10^8 10^9 1.5*10^9 2*10^9 2.5*10^9 3*10^9

Actual Losses

Figure 69. Scatter plot between Total Actual Losses vs. Total Modeled losses2. Provide a completed Form S-3, Five Validation Comparisons.

FPHLM V3.0 2008 254

See Form S-3. Reference:

Lin, L. I. (1989). A concordance correlation coefficient to evaluate reproducibility. Biometrics, 45, 255-268.

FPHLM V3.0 2008 255 S-6 Comparison of Projected Hurricane Loss Costs

The difference, due to uncertainty, between historical and modeled annual average statewide loss costs shall be reasonable, given the body of data, by established statistical expectations and norms.

The difference, due to uncertainty, between historical and modeled annual average statewide loss costs is reasonable as shown in the following description.

Disclosures

1. Describe the nature and results of the tests performed to validate the expected loss projections generated. If a set of simulated hurricanes or simulation trials was used to determine these loss projections, specify the convergence tests that were used and the results. Specify the number of hurricanes or trials that were used.

Loss costs are generated using a simulated number of hurricanes. The number of years used in the simulations was calculated using Standard S-4, which is found to be 53500. The standard errors are within less than 2.5% of the means for all counties. Extensive validation tests for generated loss costs are not possible owing to a lack of a sufficient amount of claims data. From Form S-4 we found that, the 95% confidence interval on the difference between the mean of the historical and modeled losses contains 0 indicating that the modeled losses do not differ significantly from historical losses. These forms show that FPHLM (V.3.0) losses are in good agreement with the historical losses.

2. Identify differences, if any, in how the model produces loss costs for specific historical events versus loss costs for events in the stochastic hurricane set.

A specific historical event is treated in the scenario mode, in which the peak 3s gust wind speed at a zip code is associated with a particular level of damage to a specific types of structures that are the characteristic of that zip code. In stochastic mode, the damage is computed according to the probability of the 3s gust wind speed being within wind speed consecutive 5 mph wind speed bands.

3. Provide a completed Form S-4, Average Annual Zero Deductible Statewide Loss Costs – Historical versus Modeled. See Form S-4.

FPHLM V3.0 2008 256 Form S-1: Probability of Florida Landfalling Hurricanes per Year

Complete the table below showing the probability of landfalling Florida hurricanes per year. Modeled probability should be rounded to four decimal places. The historical probabilities below have been derived from the Commission’s Official Hurricane Set. If the National Hurricane Center’s HURDAT or other hurricanes in addition to the Official Hurricane Set as specified in Standard M-1 are used by the modeler, then the historical probabilities should be modified accordingly. If the National Hurricane Center’s HURDAT is used, provide the HURDAT revision date. Model Results Probability of Florida Landfalling Hurricanes per Year

Number Historical Of Hurricanes Probability Modeled Per Year (Hurdat June 2007) Probability 0.5794 0.6026 0 0.2710 0.2668 1 0.1215 0.0939 2 0.0280 0.0283 3 0.0000 0,0083 4 0.0000 0.0001 5 0.0000 0.0000 6 0.0000 0.0000 7 0.0000 0.0000 8 0.0000 0.0000 9 0.0000 0.0000 10 or more

FPHLM V3.0 2008 257 Form S-2: An Example of a Probable Maximum Loss (PML) Based on a Limited Hypothetical Data Set

Provide projections of the insured loss for various probability levels using the hypothetical data set provided in the file named “FormA1Input07.xls.” Provide the total average annual loss for the PML distribution. If the methodology of your model does not allow you to produce a viable answer, please state so and why.

The PML is provided in the table below. Losses are in thousands.

Part A

Return Time Probability of (Years) Exceedance Estimated Loss 53500 $103,728,414 10000 0.01% $70,655,088 5000 0.02% $68,977,344 2000 0.05% $58,506,876 1000 0.10% $53,610,128 500 0.20% $47,154,509 250 0.40% $41,384,473 100 1.00% $33,235,114 50 2.00% $26,297,815 20 5.00% $16,867,502 10 10.00% $9,482,954 5 20.00% $3,245,185

Part B

Mean $2,772,311 Median $0 Standard Deviation $6,709,689 Interquartile Range $1,690,706 Sample size 53500 years

FPHLM V3.0 2008 258 Form S-3: Five Validation and Comparisons

A. Provide five validation comparisons of actual exposures and loss to modeled exposures and loss. These comparisons must be provided by line of insurance, construction type, policy coverage, county or other level of similar detail in addition to total losses. Include loss as a percent of total exposure. Total exposure represents the total amount of insured values (all coverages combined) in the area affected by the hurricane. This would include exposures for policies that did not have a loss. If this is not available, use exposures for only those policies that had a loss. Specify which was used. Also, specify the name of the hurricane event compared.

B. Provide scatter plot(s) of modeled vs. historical losses for each of the five validation comparisons. (Plot the historical losses on the x-axis and the modeled losses on the y-axis.) Rather than using directly a specific published hurricane wind field, the winds underlying the modeled loss cost calculations must be produced by the model being evaluated and should be the wind field most emulated by the model.

Comparison #1: Hurricane Charley and Company A by Coverage Company Actual Modeled Difference Coverage Loss/Exposure Loss/Exposure Building 0.01788 0.02297 -0.00508 Contents 0.00349 0.00310 0.00039 Appurtenants 0.02764 0.01191 0.01573 ALE 0.00227 0.00214 0.00013 Total 0.01298 0.01590 -0.00292

Comparison #2: Different Companies by Different Hurricanes Company Actual Modeled Difference Company Event Loss/ExposureLoss/Exposure A Charley 0.01298 0.01590 -0.00292 B Andrew 0.10426 0.08938 0.01488 C Frances 0.01769 0.01275 0.00495 H Jeanne 0.01916 0.02093 -0.00177 M Jeanne 0.00916 0.01082 -0.00167

Comparison #3: Company B by Hurricane Andrew, Charley, and Frances

FPHLM V3.0 2008 259 Company Actual Modeled Difference Company Event Loss/Exposure Loss/Exposure B Andrew 0.10426 0.08938 0.01488 B Charley 0.01573 0.01116 0.00457 B Frances 0.01288 0.00898 0.00390

Comparison #4: Construction Type for Hurricane Charley Company Actual Modeled Difference Construction Loss/Exposure Loss/Exposure Frame 0.01610 0.01678 -0.00068 Masonry 0.01445 0.01214 0.00231 Manufactured 0.04466 0.04962 -0.00495

Comparison #5: County-wide for Company A and Hurricane Charley Company Actual Modeled Difference County Loss/Exposure Loss/Exposure Lake 0.00042 0.00029 0.00013 Osceola 0.01359 0.01428 -0.00069 Sarasota 0.00030 0.00083 -0.00053 Collier 0.00073 0.00024 0.00049 St. Johns 0.00013 0.00033 -0.00020

FPHLM V3.0 2008 260

Scatter plot for comparison #1 Scatter plot for comparison #2 Modeled Loss/Exposure Modeled Modeled Loss/Exposure 0.0 0.02 0.04 0.06 0.08 0.10 0.0 0.005 0.010 0.015 0.020 0.025 0.030 0.0 0.02 0.04 0.06 0.08 0.10 0.0 0.005 0.010 0.015 0.020 0.025 0.030 Actual Loss/Exposure Actual Loss/Exposure

Scatter plot for comparison #3 Scatter plot for comparison #4 Modeled Loss/Exposure Modeled Loss/Exposure Modeled 0.0 0.02 0.04 0.06 0.08 0.10 0.0 0.01 0.02 0.03 0.04 0.05

0.0 0.02 0.04 0.06 0.08 0.10 0.0 0.01 0.02 0.03 0.04 0.05 Actual Loss/Exposure Actual Loss/Exposure

Scatter plot for comparison #5 0.015 0.010 Modeled Loss/ExposureModeled 0.005 0.0

0.0 0.005 0.010 0.015 Actual Loss/Exposure

Figure 70. Scatter plots for Actual/Exposure and Modeled Loss/Exposure

FPHLM V3.0 2008 261 Form S-4: Average Annual Zero Deductible Statewide Loss Costs – Historical versus Modeled

Part A

A. Provide the average annual zero deductible statewide loss costs produced using the list of hurricanes in the Base Hurricane Storm Set based on the 2002 Florida Hurricane Catastrophe Fund’s aggregate personal residential exposure data, as of August 1, 2003 (hlpm2002.exe).

Average Annual Zero Deductible Statewide Loss Costs (million)

Time Period - 2002 FHCF Historical Hurricanes Produced by Model Exposure Data Current Year $2,813.27 $2,646.17 Previous Year $2,804.47 $2,673.36 Second Prior - - Percentage Change Current Year/Previous Year 0.31% -1.02% Percentage Change Current Year/Second Prior - -

Total: $301,019,382,590 $141,569,986,397,759 Mean (Ann. Hurricane Losses): $2,813,265,258 $2,646,167,970 Standard Deviation: $5,182,849,942 $5,675,183,217

B. Provide a comparison with the statewide loss costs produced by the model on an average industry basis.

The loss cost produced by the model on an average industry basis is 2.6 billion and the corresponding historical average loss is 2.8 billion.

C. Provide the 95% confidence interval on the differences between the mean of the historical and modeled loss.

The 95% confidence interval on the difference between the mean of the historical and the mean of the modeled losses is between -1.24 and 0.91 billion dollars. Since the interval contains 0, we are 95% confident that there is no significant difference between the historical and the modeled losses. Using Splus, we have also done statistical test of equality means using a parametric test (t-test) (t=-0.3043, df=53605, p-value=0.7609) and equality of CDFs using a nonparametric (KS) test (ks=0.0629, p-value=0.7887) . In both parametric and non-parametric cases, we have high p-values. Therefore, we may conclude that the average historical and modeled losses are not statistically different.

FPHLM V3.0 2008 262 D. If the data are partitioned or modified, the modeler shall provide the average annual zero deductible statewide loss costs for the applicable partition (and its complement) or modification as well as the modeled average annual zero deductible statewide loss costs in additional tables.

Not applicable

Part B

A. Provide the average annual zero deductible statewide loss costs produced using the list of hurricanes in the Base Hurricane Storm Set based on the 2007 Florida Hurricane Catastrophe Fund’s aggregate personal residential exposure data (hlpm2007.exe).

Average Annual Zero Deductible Statewide Loss Costs (millions)

Time Period - 2007 FHCF Historical Hurricanes Produced by Model Exposure Data Current Year $4,937.12 $4,608.76

Total: $528,272,124,095 $246,568,525,277,870 Mean (Ann. Hurricane Losses): $4,937,122,655 $4,608,757,482 Standard Deviation: $9,090,112,106 $9,773,538,196

B. Provide a comparison with the statewide loss costs produced by the model on an average industry basis.

The loss cost produced by the model on an average industry basis is 4.6 billion and the corresponding historical average loss is 4.9 billion.

C. Provide the 95% confidence interval on the differences between the mean of the historical and modeled loss.

The 95% confidence interval on the difference between the mean of the historical and the mean of the modeled losses is between -2.18 and 1.53 billion dollars. Since the interval contains 0, we are 95% confident that there is no significant difference between the historical and the modeled losses. Using Splus, we have also done statistical test of equality means using a parametric test (t-test) (t=-0.3472, df=53605, p-value=0.7284) and equality of CDFs using a nonparametric test (KS) (ks = 0.066, p-value = 0.738). In both parametric and non-parametric cases, we have high p-values. Therefore, we may conclude that the average historical and modeled losses are not statistically different.

D. If the data are partitioned or modified, the modeler shall provide the average annual zero deductible statewide loss costs for the applicable partition (and its complement) or modification as well as the modeled average annual zero deductible statewide loss costs in

FPHLM V3.0 2008 263 additional tables.

Not applicable

FPHLM V3.0 2008 264

Form S-5: Hypothetical Events for Sensitivity and Uncertainty Analysis (requirement for models submitted by modeling organizations which have not previously provided the Commission with this analysis)

The sensitivity and uncertainty analyses required in Form S-5 was submitted to the commission under 2006 standards.

FPHLM V3.0 2008 265 COMPUTER STANDARDS

C-1 Documentation

A. The modeler shall maintain a primary document binder, containing a complete set of documents specifying the model structure, detailed software description, and functionality. Development of each section shall be indicative of accepted software engineering practices.

Florida Public Hurricane Loss Model (FPHLM) maintains a primary document binder, both in electronic and physical formats, to satisfy the above mentioned requirements. In addition to these, a user manual is maintained with the end user in focus, to give a high level introduction and step by step guideline through the whole system. All the documents are easily available for inspection; electronic copies are also available on-line. Accepted software engineering practices are used to make all the documents more readable, self contained, consistent, and easy to understand. Every component of the system is documented with standard Use Case Diagrams, Class Diagrams, Data Flow Diagrams, Sequence Diagrams, etc. The diagrams describe the structure, logic flow, information exchange among sub-modules, etc. of each component in detail and increase the visibility of the system. These diagrams describing the component functionality and structure also make each component of the system reusable and easily maintainable.

B. All computer software (i.e., user interface, scientific, engineering, actuarial, data preparation, and validation) relevant to the modeler’s submission shall be consistently documented and dated.

The Primary Document Binder consists of all the required documents arranged in different sub folders which are linked to one another based on their mutual relationships. Thus, the entire document can be viewed as a hierarchical referencing scheme where each module is linked to its sub module which ultimately refers to the corresponding codes.

C. Documentation shall be created separately from the source code.

Databases and formats of all the input/output data files are comprehensively documented. All source code is properly documented in terms of both in-line detailed comments and external higher level documentation, and they are maintained under version control systems. Source code documentation has been created separately from the source code.

FPHLM V3.0 2008 266 C-2 Requirements

The modeler shall maintain a complete set of requirements for each software component as well as for each database or data file accessed by a component.

FPHLM is divided into several major modules, where each of them provides one or more inputs to other modules. Requirements of each of the modules, including input/output formats, are precisely documented. Apart from maintaining a detailed documentation of each module of the system using standard software practices, several other documents are maintained as part of a large-scale project management requirement such as quality assurance document, system hardware and software specification document, training document, model maintenance document, testing document, user manual document, etc. Moreover, a detailed documentation is designed for the database consisting of the schemas and information about each table. Additionally, for each data file (in the form of excel or text file) accessed by different programs, the information about its format is also documented.

Disclosures

1. Provide a description of the documentation for interface, human factors, functionality, documentation, data, human and material resources, security, and quality assurance.

The user interface, functionality requirements, and material resources of each of the modules are described in the relevant module documentation. Database schemata and table formats are separately documented for the whole system and attached to the primary document binder. A separate software testing and quality assurance document describes the system quality, performance, and stability concerns. Additionally, a user manual and a human resource management document are maintained. Apart from these, security, software and hardware specifications for the system as well as training plans are documented.

FPHLM V3.0 2008 267 C-3 Model Architecture and Component Design

The modeler shall maintain and document (1) detailed control and data flow diagrams and interface specifications for each software component, and (2) schema definitions for each database and data file. Documentation shall be to the level of components that make significant contributions to the model output.

Interface specifications for each of the modules are included in the module documentation. In addition, the User Manual provides further information about the user interface specification. Control and data flow diagrams are presented at various levels of the model documentation. High level flow diagrams are used to illustrate the flow of the whole system and interactions between modules, while more technical and detailed diagrams are used in module level descriptions.

The database schema is documented and attached as part of the document binder. A detailed schema representation of the active database is documented with additional information like database maintenance, tuning, data loading methodologies, etc. to provide with a complete picture of the database maintained for the project.

These documents will be made available to the Professional Team during its site visit.

FPHLM V3.0 2008 268 C-4 Implementation* (*Significant Revision)

A. The modeler shall maintain a complete procedure of coding guidelines consistent with accepted software engineering practices.

FPHLM has developed and followed a set of coding guidelines that is consistent with accepted software practices. These documents include guidelines for version control, code revision history maintenance, etc. All the developers involved in the system development adhere to the instructions in these documents.

B. The modeler shall maintain a complete procedure used in creating, deriving, or procuring and verifying databases or data files accessed by components.

FPHLM uses an Oracle database to store the related data necessary for the model. The database documentation includes the procedures for creating and deriving the database. Data files are generated by different modules and used as interfaces between modules. Several data verification techniques are undertaken to assure the correctness. Details about these are included in the module documentation.

C. All components shall be traceable, through explicit component identification in the flow diagrams, down to the code level.

Traceability, from requirements to the code level and vice versa, is maintained throughout the system documentation.

D. The modeler shall maintain a table of all software components affecting loss costs, with the following table columns: (1) Component name, (2) Number of lines of code, minus blank and comment lines; and (3) Number of explanatory comment lines.

The FPHLM primary document binder includes a table that gives the above requested information. The table is available for review by the professional team.

E. Each component shall be sufficiently and consistently commented so that a software engineer unfamiliar with the code shall be able to comprehend the component logic at a reasonable level of abstraction.

All the software codes are properly provided with code-level comments and a consistent format is maintained throughout the software modules. These code level comments include a summary of important changes, names of developers involved in each modification, function headers, and in-line comments to explain potentially ambiguous software code.

F. The modeler shall maintain the following documentation for all components or data modified by items identified in Standard G-1, Disclosures 5 and 6:

FPHLM V3.0 2008 269

1. A list of all equations and formulas used in documentation of the model with definitions of all terms and variables.

2. A cross-referenced list of implementation source code terms and variable names to items within F.1.

Tables that map the equations and formulas used in documentation of the model implementation source code terms and variable names were added as glossaries to the model’s documentation, thus combining F.1 and F.2 into the same table. These tables enhance the model’s documentation and include the equations and formulas for each module (not just the modified ones from the prior year’s submission).

Disclosures

1. Specify the hardware, operating system, other software, and all computer languages required to use the model.

The system is mainly a web-based application that is hosted over an Oracle 9i web application server. The backend server environment is Linux and the server side scripts are written in Java Server Pages (JSP) and Java beans. Many backend calculations are coded in C++ using the IMSL library and called through Java Native Interface (JNI). The system uses an Oracle database running on a Sun workstation. Server side software requirements are IMSL library CNL 5.0, +v1.0.2.2.1, Oracle 9i AS 9.0.2.0.0A, JNI 1.3.1, and JDK 1.3.1.[mc129987e2]

The end-user workstation requirements are minimal. Internet Explorer 5.5 or 6 running on Windows 2000 or XP[mc129987e3] is the recommended web browser. However, other web browsers such as Mozilla Firefox should also deliver the optimal user experience. Typically, the manufacturer’s minimal feature for a given web browser and operating system combination is sufficient for an optimal operation of the application.

FPHLM V3.0 2008 270 C-5 Verification

A. General

For each component, the modeler shall maintain procedures for verification, such as code inspections, reviews, calculation crosschecks, and walkthroughs, sufficient to demonstrate code correctness. Verification procedures shall include tests performed by modeler personnel other than the original component developers.

FPHLM software verification is done in three stages.

1. Code inspection and verification by the code developer. 2. Inspection of the input and validation of the output by the system modeler. 3. Review and extensive testing of the code by modeler personnel who are not part of the original component development.

The first level of verification includes code-level debugging, walking through the code to ensure a proper flow, inspection of internal variables through intermediate output printing and error logging, use of exception handling mechanisms, calculation cross checks, and verification of the output against sample calculations provided by the system modeler.

In the second level of the verification, the modeler is provided with sample inputs and corresponding outputs . The modeler then conducts black box testing to verify the results against his/her model. Finally, each component is rigorously tested by modeler personnel not responsible for original component development.

B. Component Testing

1. The modeler shall use testing software to assist in documenting and analyzing all components.

Component testing (C-5.B) and Data testing (C-5.C) are done in the third level of verification. The system is rigorously checked for the correctness, precision, robustness and stability of the whole system. Calculations are performed outside the system and compared against the system generated results to ensure the system correctness. Extreme and unexpected inputs are given to the system to check the robustness. Wide series of test cases are developed to check the stability and the consistency of the system.

These verification procedures are properly documented and are available for inspection.

2. Unit tests shall be performed and documented for each component.

FPHLM V3.0 2008 271 Unit testing is done at the first and third levels of verification. The developer tests all the units as the unit is developed and modified. Then all the units are tested again by the external testing team. Both “black-box” and “white-box” tests are performed and documented in a separate Testing Document.

3. Regression tests shall be performed and documented on incremental builds.

Regression testing is performed for each module. In this kind of testing methodology, the modules which have undergone some changes and revisions are retested to ensure that the changes have not affected the entire system in any undesired manner.

4. Aggregation tests shall be performed and documented to ensure the correctness of all model components. Sufficient testing shall be performed to ensure that all components have been executed at least once.

Aggregation testing is performed at all three levels of verification. Aggregation testing is performed by running each major module as a complete package. It is ensured that all components have been executed at least once during the testing procedure. All the test cases executed are described in Software Testing and Verification documentation.

C. Data Testing

1. The modeler shall use testing software to assist in documenting and analyzing all databases and data files accessed by components.

FPHLM uses an Oracle database to store the required data. Data integrity and consistency are maintained by the database itself. Moreover, different queries are issued and PL/SQL is implemented to check the database. Oracle 9i has a very robust loader. It is used to load the data into the database. The loader maintains a log which depicts if the loading procedure has taken place properly and completely without any discrepancy. Data files are manually tested using commercial data manipulation software such as Excel and Access.

2. The modeler shall perform and document integrity, consistency, and correctness checks on all databases and data files accessed by the components.

All the tests are well documented in a separate Testing Document.

Disclosures

1. State whether the model produces the same loss costs if it runs the same information more than once without changing the seed of the random number generator.

FPHLM V3.0 2008 272 The model produces the same loss costs if it runs the same information more than once without changing the seed of the random number generator.

2. Provide an overview of the component testing procedure.

FPHLM software testing and verification is done in three stages.

[A] Code inspection and the verification by the code developer

The code developer performs a sufficient amount of testing on the code, and wouldn’t deliver the code until he/she is convinced of proper functionality and robustness of the code.

The first level of verification includes code-level debugging, walking through the code to ensure proper flow, inspection of internal variables through intermediate output printing and error logging, use of exception handling mechanisms, calculation cross checks, and verification of the output against sample calculations provided by the system modeler.

[B] Verification of results by the person who developed the system model

Once the first level of testing is done, the developer should send the sample inputs and the generated results back to the modeler. Then the system modeler would double-check the results against his/her model. The code is not used in the production environment unless approved by the modeler.

[C] Review and extensive testing of the code by modeler personnel other than the original component developers.

The system is rigorously checked by modeler personnel (testers) other than the original component developers for the correctness, precision, robustness and stability of the whole system. Calculations are performed outside the system and compared against the system generated results to ensure the system correctness. Extreme and unexpected inputs are given to the system to check the robustness. Wide series of test cases are developed to check the stability and the consistency of the system.

Unit testing, Regression testing, and Aggregation testing (both white-box and black-box) are performed and documented.

Any flaw in the code is reported to the developer, and the bug-corrected code is again sent to the tester. The tester then performs unit testing again on the modified units. Additionally, Regression testing is performed to determine if the modification affects any other parts of the code.

Different Testing Tools and Software packages are used to test different components of the system. The detailed list of the various testing tools and/or techniques used for different components of the system is provided in the main document for audit.

FPHLM V3.0 2008 273 C-6 Model Maintenance and Revision

A. The modeler shall maintain a clearly written policy for model revision, including verification and validation of revised components, databases, and data files.

FPHLM model will be periodically enhanced to reflect the new knowledge acquired on hurricanes and Florida ZIP code information. FPHLM maintains a clearly written policy for model revision.

B. A revision to any portion of the model that results in a change in any Florida residential hurricane loss cost shall result in a new model version number.

Whenever a revision results in a change in any Florida residential hurricane loss cost, a new model version number will be assigned to the revision. Verification and validation of the revised units is repeated according to the above mentioned “software verification procedures” document.

C. The modeler shall use tracking software to identify all errors, as well as modifications to code, data, and documentation.

FPHLM uses CVS (Concurrent Versions System) for version control. CVS is a production quality system used widely in commercial and educational research-oriented projects around the world. We can record the history of source files and documents by utilizing CVS.

Disclosures

1. Identify procedures used to maintain code, data, and documentation.

FPHLM’s software development team employs source revision and control software for all software development. In particular, FPHLM employs Concurrent Versions System (CVS), an accepted and effective system for managing simultaneous development of files. It is in common use in large programming projects to track the modifications to the source code and documentation files. CVS maintains a record of the changes to each file, and allows the user to revert to a previous version, merge versions, and track changes. This software is able to record the information for each file, the date of each change, the author of each change, the file version, and the comparison of the file before and after the changes. The detailed information will be made available to the Professional Team during its site visit.

2. Describe the rules underlying the model and code revision numbering systems.

The model numbering system consists of the scheme: “V[major].[minor]”. The terms "[major]" and "[minor]" are positive numeric numbers that correspond to major and minor changes in the model respectively; a minor change causes the minor number to be incremented by one, and

FPHLM V3.0 2008 274 similarly a major change causes the major number to be incremented by one with the minor number reset to zero. The rules that prompt major or minor changes in the model are the following:

Rules that trigger a change in the major number:

- Updates in any of the main modules of FPHLM: any change resulting in the partial or total modification of the algorithm/model of the Storm Forecast, Wind Field, Damage Estimation, and/or Insurance Loss models.

Rules that trigger a change in the minor number:

- Slight changes to the Storm Forecast, Wind Field, and/or Damage Estimation modules: small updates such as a change in the Holland B parameter or any change to correct deficiencies that do not result in a new algorithm for the component.

- Updates to correct errors in the computer code: modifications in the code to correct deficiencies or errors such as a code bug in the computer program.

- Changes in the probability distribution functions using updated or corrected historical data, such as the updates of the HURDAT database: each year the model updates its HURDAT database with the latest HURDAT data released by the National Hurricane Center, which is used as the input in the Storm Forecast Model.

- Updates of the ZIP Code list: every two years the ZIP Codes used in the model must be updated according to information originating from the United States Postal Service.

- Updates in the validation of the vulnerability matrices: the incorporation of new data, such as updated winds and insurance data may trigger a tune up of the vulnerability matrices used in the Insurance Loss Model.

If any change results in a change in lost costs estimates there will be at least a change in the minor revision number.

Consequently, for the re-submission of May 15, 2008, the Florida Public Hurricane Loss Model changed its version number from 2.7 to 3.0 due to modest change in the Storm Track Generator.

FPHLM V3.0 2008 275 C-7 Security

The modeler shall have implemented and fully documented security procedures for: (1) secure access to individual computers where the software components or data can be created or modified, (2) secure operation of the model by clients, if relevant, to ensure that the correct software operation cannot be compromised, (3) anti-virus software installation for all machines where all components and data are being accessed, and (4) secure access to documentation, software, and data in the event of a catastrophe.

FPHLM maintains a set of security procedures to protect data and documents from deliberate and inadvertent changes. These procedures include both physical and electronic measures. There are a set of policies identifying different security issues and addressing each of them. All the security measures are properly documented and attached to the primary document binder.

Disclosures

1. Describe methods used to ensure the security and integrity of the code, data, and documentation.

Electronic measures include the use of different authorization levels, special network security enforcements, and regular backups. Each developer is given a separate username and password, and assigned with a level of authorization so that even a developer cannot change another developer’s code. The users of the system are given usernames and passwords so that unauthorized users cannot use the system. External users are not allowed direct access to any of the data sources of the system. The network is extensively monitored for any unauthorized actions using standard industry practices. Since the system runs on a Linux sever environment, minimal virus attacks are expected.

Any sensitive or confidential data (insurance data, for example) are kept on an unshared disk on a system which has user access control and requires a login. Screen locks are enforced whenever the machine is left unattended. In addition, for system security and reliability purposes, we also deploy a development environment besides the production environment. Modifications to the code and data are done in the development environment and tested by in-house developers. The final production code and data can only be checked into the production environment by the authorized personnel. The models resulting from FPHLM project can only be used by the authorized users. Authorized user accounts are created by the project manager. Regular backups of the server are taken and stored in two ways: physically and electronically. Backups are performed on a daily basis and are kept for six weeks. Nightly backups of all UNIX data disks and selected Windows data disks (at user requests) are performed over the network onto LT02 and LT03 tapes. The tape drives have built in diagnostics and verification to ensure that the data is written correctly to the tapes. This ensures that if the tape is written successfully, it will be readable, provided no physical damage occurred to the tape. A copy of each backup is placed in

FPHLM V3.0 2008 276 a secure and hurricane protected building. Additionally, the application server and the database server are physically secured in a secure server room with alarm systems. In case of disasters, we have implemented a set of preparation procedures and recovery plans as outlined in “FIU SCS Hurricane Preparation Procedures”.

FPHLM V3.0 2008 277 Appendix A - Expert Review Letters

FPHLM V3.0 2008 278 Assessment of the meteorological portion of the State of Florida Public Hurricane Model

February 15, 2007

Gary M. Barnes Professor, Department of Meteorology School of Ocean and Earth Science and Technology University of Hawaii at Manoa

Introduction

My review of the State of Florida Public Hurricane Model is based on a three day visit to Florida International University in December, and an examination of the submission draft provided to me in February. I have had full access to the meteorological portion of the model, access to the draft for the Florida commission, and access to prior submittals to the commission from several other groups in order to establish a sense of what is desired by the commission. I am pleased to report that the issues that I have raised have received their attention and I believe that the model meets all the standards set forth by the commission. Ultimately this model, when linked to engineering and actuarial components, will provide objective guidance for the estimation of wind losses from hurricanes for the state of Florida. It does not address losses from other aspects of a tropical cyclone such as storm surge, or fresh water flooding. I now offer specific comments on each of the six meteorological standards established by the commission to ascertain this model’s suitability.

M-1 Official Hurricane Set

The consortium of scientists working on the Public model have adopted HURDAT (1900- 2006) to determine landfall frequency and intensity at landfall. The NWS report by Ho et al. (1987), DeMaria’s extension of the best track, H*Wind analyses (Powell et al. 1996a, 1996b, 1998) and NOAA Hurricane Research Division aircraft data are used to estimate the radius of maximum winds (RMW) at landfall. The strength of HURDAT is that it is the most complete and accessible historical record for hurricanes making landfall or passing closely by Florida. HURDAT weaknesses include the abbreviated record and questionable intensity estimates for those hurricanes early in the record, especially those that remain offshore. Evidence for the shortness of record is the impact of the last few hurricane seasons on landfall return frequency. The meteorological team has scrutinized the base set developed by the commission and made a number of adjustments to the dataset based on refereed literature and the HURDAT record. I have looked at several of these adjustments in detail and find the corrections to be an improvement over the initial base set.

FPHLM V3.0 2008 279 M-2 Hurricane Characteristics

The model has two main components. The track portion of the model produces a storm with either an initial location or genesis point and an intensity that is derived from an empirical distribution derived from HURDAT (2006). Storm motion and intensity is then initialized by using a Monte Carlo approach, drawing from probability density functions (PDFs) based on the historical dataset to create a life for a bogus hurricane. Examination of the PDFs reveals that they are faithful to the observed patterns for storms nearing Florida, and the evolution of any particular hurricane appears realistic.

The second component of the meteorological model is the wind field generated for a given hurricane, which only comes into play when the hurricane comes close enough to place high winds over any given zip code of Florida. To generate a wind field the minimum sea-level pressure (MSLP) found in the eye, the RMW at landfall, and a distant environmental pressure (1013 mb) are entered into the Holland (1980) B model for the axisymmetric pressure distribution around the hurricane. The behavior of the RMW is based on a variety of sources that include Ho et al. (1987), DeMaria’s extension of the best track data, H*wind analyses, and aircraft reconnaissance radial wind profiles. The B coefficient is based on the extensive aircraft dataset acquired in reconnaissance and research flights over the last few decades. RMW and B use a random or error term to introduce variety into the model. The Holland pressure field is used to produce a gradient wind at the top of the boundary layer. The winds in the boundary layer are estimated following the work proposed by Ooyama (1969) and later utilized by Shapiro (1983) which includes friction and advection effects. These boundary layer winds are reduced to surface winds (10 m) using reduction factors based on the work of Powell et al. (2003). Maximum sustained winds and 3 second gusts are estimated using the guidance of Vickery and Skerlj (2005). Once the hurricane winds come ashore there are further adjustments to the wind to account for local roughness as well as the roughness of the terrain found upstream of the location under scrutiny. The pressure decay of the hurricane is modeled to fit the observations presented by Vickery (2005).

Gradient balance has been demonstrated to be an accurate representation for vortex scale winds above the boundary layer by Willoughby (1990) and is a fine initial condition. The slab boundary layer concept of Ooyama and Shapiro has been shown to produce wind fields much like observed once storm translation and surface friction come into play. The reduction to 10 m altitude is based on Powell et al. (2003); they use the state of the art Global Positioning System sondes to compare surface and boundary layer winds.

Perhaps the most questionable part of the wind portion of the model is the reliance on the estimates of the RMW at landfall. The scatter in RMW for a given MSLP is large; larger RMWs coupled with the B parameter control the size of the annulus of the damaging winds. The typical length of an aircraft leg from the eye is about 150 km so the choice of the B parameter is based on a small radial distance in the majority of hurricanes. The collection of quality wind observations over land in hurricanes remains a daunting task; therefore the actual response of the hurricane winds to variations in roughness is less certain. Applying roughness as a function of zip code is a coarse approximation to reality. However, this is the approach chosen by the commission, and given the data limitations, a reasonable course to take.

FPHLM V3.0 2008 280 M-3 Landfall Intensity

The model uses one minute winds at 10 m elevation to determine intensity at landfall and categorizes each hurricane according to the Saffir-Simpson classification. The model considers any hurricane that makes landfall or comes close enough to place high winds over Florida. Multiple landfalls are accounted for, and decay over land between these landfalls is also estimated. Maximum wind speeds for each category of the Saffir-Simpson scheme are reasonable as is the worst possible hurricane the model generates. Simulations are conducted for a hypothetical 60,000 years. Any real climate change would alter results, but maybe not as much as have an actual record of order of 1,000 years to base the PDFs on.

M-4 Hurricane Probabilities

Form M-1 demonstrates that the model is simulating the landfalls very well for the entire state, region A (NW Florida) and region B (SW Florida). There are subsections of the state where the historical and the simulated landfalls have a discrepancy. In region C (SE Florida) the observations show an unrealistic bias toward category 3 storms. This is likely due to an overestimate of intensity for the hurricanes prior to the advent of aircraft sampling or advanced satellite techniques. The historical distribution for region C also does not fit any accepted distributions that we typically see for atmospheric phenomena. This discrepancy is probably due to the shortness of the historical record. I note that other models also have difficulty with this portion of the coast. I believe the modeled distribution, based on tens of thousands of years, is more defensible than the purported standard. Regions D (NE Florida) and E (Georgia) have virtually no distribution to simulate, again pointing to a very short historical record. There is no documented physical reason why these two regions have escaped landfall events. Perhaps a preferred shape of the Bermuda High may bias the situation, but this remains speculative.

M-5 Land Friction and Weakening

Land use and land cover are based on high resolution satellite imagery. Roughness for a particular location is then based on HAZUS tables that assign a roughness to a particular land use. There are newer assessments from other groups but the techniques were not consistently applied throughout the state, nor are the updated HAZUS maps for 2000 available yet. Winds at a particular location are a function of the roughness at that point and conditions upwind. A pressure decay model based on the work of Vickery (2005) produces weakening winds that are reasonable approximations of the observed decay rates of several hurricanes that made landfall in Florida in 2004 and 2005.

The maps (Form M-2) of the 100 year return period maximum sustained winds shows the following trends: (1) a reduction in the sustained winds from south to north, (2) a reduction of winds from coastal to inland zip codes, and (3) the highest winds in the Keys and along the SE and SW coasts. The plotting thresholds requested by the commission partially obfuscate the gradients in wind speed, but Form M-2 produced with finer contours highlights the above trends clearly. The open terrain maps look logical; the actual terrain maps are perhaps overly sensitive to the local roughness. Convective scale motions, which cannot be resolved in this type of model, would probably be responsible for making the winds closer to the open terrain results.

FPHLM V3.0 2008 281 M-6 Logical Relationships of Hurricane Characteristics

The RMW is a crucial but poorly measured variable. Making RMW a function of intensity and latitude explains only a small portion of the variance (~20%). Examination of aircraft reconnaissance radial profiles shows that RMW is highly variable. Currently there are no other schemes available to explain more of the variance. Form M-3 reflects the large range of RMW. Note that only the more intense hurricanes (MSLP < 940 mb) show a trend, and only with the upper part of the range. Even open ocean studies of the RMW show such large scatter.

Tests done during my visits show that wind speed decreases as a function of roughness, all other variables being held constant. The evolution of the wind field as a hurricane comes ashore is logical.

Summary

The consortium that has assembled the meteorological portion of the Public Model for Hurricane Wind Losses for the State of Florida is using the HURDAT with corrections based on other refereed literature. These data yield a series of probability density functions that describe frequency, location, and intensity at landfall. Once a hurricane reaches close enough to the coast the gradient winds are estimated using the equations by Holland (1980), then a sophisticated wind model (Ooyama 1969, Shapiro 1983) is applied to calculate the boundary layer winds. Reduction of this wind to a surface value is based on recent boundary layer theory and observations. Here the consortium has exploited other sources of data (e.g., NOAA/AOML/HRD aircraft wind profiles and GPS sondes) to produce a surface wind field. As the wind field transitions from marine to land exposure changes in roughness are taken into account. Form M-1 (frequency and category at landfall as a function of coastal segment) and Form M-2 (100 year return maximum sustained winds for Florida) highlight the good performance of the model.

I suspect that the differences between the historical record and the simulation are largely due to the shortness and uncertainty of the record. If the consortium had the luxury of 1000 years of observations agreement between the record and the simulation would be improved. I believe that the meteorological portion of the model is meeting all the standards established by the commission. Tests of the model against H*Wind analyses and the production of wind speed swaths go beyond the typical quality controls of prior models and demonstrate that this model is worthy of consideration by the commission.

FPHLM V3.0 2008 282

February 22, 2008

Dr. Shahid Hamid Professor of Finance, Department of Finance, CBA and International Hurricane Research Center Florida International University, RB 202 B Miami, FL 33199

Re: Florida Public Hurricane Loss Model Version 2.7 Independent Actuarial Review

Dear Dr. Hamid:

AMI Risk Consultants, Inc. was engaged by the International Hurricane Research Center (“IHRC”) at Florida International University (“FIU”) to review the actuarial components of its hurricane model, Florida Public Hurricane Loss Model, Version 2.7. I am a Fellow of the Casualty Actuarial Society, a Member of the American Academy of Actuaries, and have twenty-five years of actuarial experience in the property/casualty insurance industry. I am an employee of the actuarial consulting firm AMI Risk Consultants, Inc., and assisted Aguedo (Bob) Ingco in his review of Version 2.6 of the Model last year.

It is my understanding that between Versions 2.6 and 2.7 the changes to the Florida Public Hurricane Loss Model (“FPHLM”) are minimal. Although the base set of historical storms that underlie the model has been updated, and a new set of stochastic storms generated, the core meteorological, engineering and actuarial assumptions have not changed. Furthermore, the computer code that executes the entire process has not been altered.

My review is based the IHRC’s February 28, 2008 model submission to the Commission. I focused my review on Standards A-6 and A-10; however I revisited each of the Actuarial Standards, and have the following comments:

Standard A-1: I was directly involved in the testing of included and excluded bypassing storms for Version 2.6 of the model. Although Version 2.7 incorporates a new set of stochastic storms, the criteria for inclusion/exclusion have not changed, and the computer code categorizing each storm is also unchanged. Therefore, I did not sample bypassing storms again this year.

FPHLM V3.0 2008 283 Standard A-2: There was no additional validation data collected or validation studies performed in the past year. The model’s assumptions as disclosed in this standard have not changed.

FPHLM V3.0 2008 284

Standard A-3: The approach to estimating loss costs has not changed in this version of the model.

Standard A-4: The approach to incorporating demand surge in the loss costs has not changed in this version of the model. Demand surge factors were calculated for each storm in the new stochastic set using the same functions and parameters as were applied last year.

Standard A-5: I visited the computer lab at FIU and examined a sample of insurance company data files that had been recently processed through the model. The editing process for the input, and the process for handling missing values have not changed since last year’s submission.

Standard A-6: I examined Forms A-1, A-2, A-3, A-4 and A-5 for reasonability, and compared them to last year’s submission, taking into consideration the impact of the updated exposure base for A-3, A- 4 and A-5, and the change in deductible for A-1.

Standard A-7: The methods used by the model to reflect the impact of deductibles and policy limits has not changed since last year’s submission.

Standard A-8: The method used by the model to estimate contents losses has not changed since last year’s submission.

Standard A-9: The method used by the model to estimate additional living expense losses has not changed since last year’s submission.

Standard A-10: I tested the loss costs for compliance with this standard. I received assistance from the Computer Science team in testing at the zip code level in instances where compliance could not be verified from the weighted averages in Forms A-6A & B. I also examined the change in loss costs compared to last year, and investigated changes in excess of 10% at the county level.

My conclusion is that the IHRC hurricane model reflects reasonable actuarial assumptions, and meets the Commission’s standards A-1 through A-10.

If you have any questions about my review, I would be happy to discuss them.

Sincerely,

Gail Flannery, FCAS, MAAA Consulting Actuary

FPHLM V3.0 2008 285 Appendix B - Output Ranges by County

FPHLM V3.0 2008 286 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 Personal Residential -- Owners -- FRAME $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Alachua LOW 0 9243 0 0947 0.0628 0 0227 0.7528 0.4034 0.2115 0.4034 0 2434 0.1118 HIGH 1.1533 0.1272 0.0813 0 0326 0.9834 0.5745 0.3211 0 5745 0 3667 0.1783 WGHTD AVE 1 0195 0.1091 0.0707 0 0270 0.8506 0.4772 0.2574 0.4772 0 2954 0.1384

Baker LOW 0 5677 0 0542 0.0367 0 0122 0.4400 0.2112 0.1043 0 2112 0.1197 0.0554 HIGH 0 8755 0 0911 0.0587 0 0220 0.7155 0.3858 0.2052 0 3858 0 2350 0.1118 WGHTD AVE 0 8358 0 0865 0.0560 0 0208 0.6806 0.3634 0.1918 0 3634 0 2199 0.1039

Bay LOW 1 2576 0.1381 0.0875 0 0357 1.0702 0.6235 0.3484 0 6235 0 3977 0.1942 HIGH 3 6548 0.7095 0.2089 0 2585 4.1917 3.5535 2.9320 3 5535 3 0789 2.4173 WGHTD AVE 2 0619 0 2867 0.1394 0 0913 2.0180 1.4644 1.0242 1.4644 1.1172 0.7216

Bradford LOW 0 8953 0 0923 0.0602 0 0219 0.7272 0.3866 0.1998 0 3866 0 2305 0.1038 HIGH 1 0851 0.1174 0.0753 0 0293 0.9116 0.5180 0.2823 0 5180 0 3237 0.1533 WGHTD AVE 0 9191 0 0949 0.0623 0 0226 0.7482 0.3995 0.2075 0 3995 0 2392 0.1086

Brevard LOW 2 9589 0 2072 0.1066 0 0585 2.8071 2.2855 1.5104 2 2855 1.7523 0.7646 HIGH 6 2668 0 8464 0.2286 0 2898 6.8328 6.1104 4.8555 6.1104 5 2520 3.5138 WGHTD AVE 3 8129 0 2957 0.1363 0 0883 3.7140 3.1016 2.1299 3.1016 2.4354 1.1624

Broward LOW 6 8054 0 8291 0.1977 0 2907 7.3574 6.5943 5.1156 6 5943 5 5907 3.5343 HIGH 10.1466 1 6773 0.2787 0 6065 11 8495 10 9922 9.2009 10.9922 9.7784 7.1512 WGHTD AVE 8 0213 1.1048 0.2297 0 3957 8.9448 8.1358 6.5107 8.1358 7 0339 4.7276

Calhoun LOW 1 0433 0.1093 0.0715 0 0268 0.8610 0.4733 0.2507 0.4733 0 2887 0.1327 HIGH 1 3090 0.1519 0.0947 0 0407 1.1416 0.6890 0.3981 0 6890 0.4504 0.2322 WGHTD AVE 1.1608 0.1260 0.0814 0 0318 0.9800 0.5619 0.3060 0 5619 0 3515 0.1647

Charlotte LOW 4 0272 0 2905 0.1498 0 0856 3.9168 3.2827 2.2385 3 2827 2 5681 1.1724 HIGH 6 2648 0.7969 0.2274 0 2755 6.8339 6.1052 4.8003 6.1052 5 2136 3.3840 WGHTD AVE 4.7034 0 3914 0.1710 0.1251 4.6959 4.0194 2.8709 4 0194 3 2340 1.6745

Citrus LOW 2 3820 0.1372 0.0821 0 0344 2.1982 1.7629 1.1126 1.7629 1 3163 0.4921 HIGH 3 2340 0 2165 0.1149 0 0613 3.0989 2.5735 1.7228 2 5735 1 9914 0.8674 WGHTD AVE 2 9064 0.1817 0.1019 0 0490 2.7404 2.2445 1.4669 2 2445 1.7117 0.7003

Clay LOW 0 8589 0 0872 0.0576 0 0206 0.6933 0.3644 0.1882 0 3644 0 2168 0.0989 HIGH 1.1803 0.1307 0.0830 0 0336 1.0077 0.5900 0.3309 0 5900 0 3776 0.1847 WGHTD AVE 0 9468 0.1000 0.0646 0 0246 0.7830 0.4323 0.2330 0.4323 0 2669 0.1271

Collier LOW 4 8882 0 3557 0.1857 0.1064 4.7733 4.0132 2.7511 4 0132 3.1495 1.4568 HIGH 7 2367 0 8529 0.2755 0 2977 7.8072 6.9535 5.3906 6 9535 5 8863 3.6842 WGHTD AVE 5 5278 0.4775 0.2101 0.1536 5.5829 4.8029 3.4501 4 8029 3 8784 2.0255

Columbia LOW 0 6045 0 0578 0.0395 0 0129 0.4691 0.2258 0.1105 0 2258 0.1272 0.0575 HIGH 0 9238 0 0992 0.0627 0 0245 0.7680 0.4277 0.2317 0.4277 0 2651 0.1274 WGHTD AVE 0 8536 0 0894 0.0578 0 0216 0.6991 0.3774 0.1998 0 3774 0 2291 0.1079

De Soto LOW 4 0237 0 2562 0.1439 0 0711 3.8368 3.1812 2.1148 3.1812 2.4517 1.0376 HIGH 4 5496 0 3372 0.1662 0.1021 4.4714 3.7899 2.6342 3.7899 3 0004 1.4322 WGHTD AVE 4.1662 0 2844 0.1500 0 0822 4.0222 3.3637 2.2805 3 3637 2 6229 1.1747

Dixie LOW 0 9509 0.1061 0.0655 0 0274 0.8098 0.4717 0.2677 0.4717 0 3038 0.1546 HIGH 1 6311 0 2247 0.1124 0 0693 1.5693 1.1029 0.7543 1.1029 0 8256 0.5241 WGHTD AVE 1.1402 0.1376 0.0783 0 0372 1.0142 0.6380 0.3925 0 6380 0.4384 0.2469

*Includes contents and A.L.E.

FPHLM V3.0 2008 287 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 Personal Residential -- Owners -- FRAME $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Duval LOW 0 5482 0 0520 0.0360 0 0116 0.4234 0.2013 0.0983 0 2013 0.1130 0.0517 HIGH 1 3334 0 2018 0.0864 0 0656 1.3248 0.9644 0.7115 0 9644 0.7615 0.5427 WGHTD AVE 0 9101 0 0992 0.0621 0 0251 0.7608 0.4272 0.2386 0.4272 0 2702 0.1384

Escambia LOW 1.4819 0.1908 0.1102 0 0580 1.3922 0.9455 0.6171 0 9455 0 6825 0.4046 HIGH 4 6206 1 0023 0.2385 0 3737 5.5827 4.9318 4.2443 4 9318 4.4115 3.6365 WGHTD AVE 2 5539 0.4237 0.1657 0.1481 2.7207 2.1538 1.6601 2.1538 1.7687 1.2900

Flagler LOW 2.7374 0.1707 0.0967 0 0461 2.5761 2.1036 1.3713 2.1036 1 6015 0.6543 HIGH 3 5853 0 3130 0.1317 0 0987 3.5877 3.0485 2.1659 3 0485 2.4439 1.2676 WGHTD AVE 3 0241 0 2254 0.1081 0 0662 2.9212 2.4279 1.6505 2.4279 1 8946 0.8812

Franklin LOW 1.7797 0 2509 0.1260 0 0791 1.7322 1.2306 0.8486 1 2306 0 9261 0.5964 HIGH 3 0451 0 5875 0.1759 0 2125 3.4619 2.9048 2.3923 2 9048 2 5094 1.9805 WGHTD AVE 2 2574 0 3729 0.1440 0.1270 2.3712 1.8427 1.4062 1 8427 1.4997 1.0908

Gadsen LOW 0.7958 0 0794 0.0530 0 0184 0.6356 0.3241 0.1620 0 3241 0.1875 0.0827 HIGH 1.1013 0.1186 0.0764 0 0297 0.9246 0.5250 0.2840 0 5250 0 3265 0.1523 WGHTD AVE 0 8443 0 0861 0.0568 0 0205 0.6825 0.3592 0.1840 0 3592 0 2124 0.0954

Gilchrist LOW 0 8799 0 0930 0.0596 0 0227 0.7251 0.3968 0.2112 0 3968 0 2424 0.1139 HIGH 1.1609 0.1344 0.0826 0 0359 1.0166 0.6214 0.3629 0 6214 0.4110 0.2118 WGHTD AVE 1 0770 0.1220 0.0749 0 0318 0.9286 0.5536 0.3171 0 5536 0 3602 0.1823

Glades LOW 4 5736 0 2915 0.1644 0 0813 4.3644 3.6205 2.4095 3 6205 2.7921 1.1856 HIGH 4.7418 0 3132 0.1708 0 0893 4.5545 3.7965 2.5507 3.7965 2 9446 1.2849 WGHTD AVE 4.7317 0 3119 0.1705 0 0888 4.5434 3.7862 2.5425 3.7862 2 9357 1.2791

Gulf LOW 1 3350 0.1593 0.0960 0 0442 1.1831 0.7338 0.4405 0.7338 0.4941 0.2695 HIGH 1 8586 0 2556 0.1332 0 0795 1.7895 1.2541 0.8507 1 2541 0 9319 0.5867 WGHTD AVE 1.7801 0 2353 0.1219 0 0729 1.6853 1.1646 0.7802 1.1646 0 8566 0.5322

Hamilton LOW 0 5619 0 0539 0.0367 0 0121 0.4363 0.2100 0.1032 0 2100 0.1186 0.0543 HIGH 0.7819 0 0817 0.0529 0 0198 0.6398 0.3452 0.1829 0 3452 0 2097 0.0991 WGHTD AVE 0 6932 0 0703 0.0462 0 0166 0.5567 0.2893 0.1497 0 2893 0.1718 0.0804

Hardee LOW 3 9552 0 2462 0.1417 0 0676 3.7582 3.1081 2.0564 3.1081 2 3885 0.9948 HIGH 4 3828 0 3086 0.1603 0 0912 4.2682 3.5959 2.4655 3 5959 2 8235 1.2993 WGHTD AVE 4 0462 0 2566 0.1457 0 0713 3.8592 3.2014 2.1301 3 2014 2.4686 1.0453

Hendry LOW 4.7473 0 3185 0.1728 0 0920 4.5794 3.8307 2.5940 3 8307 2 9851 1.3274 HIGH 5.4157 0.4135 0.1973 0.1268 5.3453 4.5397 3.1699 4 5397 3 6040 1.7503 WGHTD AVE 5 0873 0 3636 0.1852 0.1083 4.9618 4.1840 2.8794 4.1840 3 2925 1.5335

Hernando LOW 2.7395 0.1576 0.0952 0 0397 2.5436 2.0479 1.3006 2 0479 1 5349 0.5788 HIGH 3 3393 0 2261 0.1195 0 0653 3.2091 2.6704 1.7973 2 6704 2 0730 0.9158 WGHTD AVE 3 0136 0.1954 0.1066 0 0542 2.8637 2.3588 1.5603 2 3588 1 8118 0.7664

Highlands LOW 3.7763 0 2191 0.1328 0 0567 3.5310 2.8800 1.8582 2 8800 2.1800 0.8466 HIGH 4 6097 0 2972 0.1667 0 0837 4.4133 3.6720 2.4547 3 6720 2 8396 1.2188 WGHTD AVE 4 0891 0 2513 0.1462 0 0681 3.8732 3.1924 2.0992 3.1924 2.4442 1.0026

Hillsborough LOW 2 9792 0.1868 0.1062 0 0511 2.8209 2.3206 1.5293 2 3206 1.7786 0.7393 HIGH 4.7113 0.4487 0.1750 0.1490 4.8527 4.2234 3.1255 4 2234 3.4735 1.9529 WGHTD AVE 3 5160 0 2527 0.1279 0 0748 3.4189 2.8671 1.9638 2 8671 2 2491 1.0418

Holmes LOW 1.1945 0.1317 0.0852 0 0343 1.0247 0.6058 0.3406 0 6058 0 3892 0.1886 HIGH 1 5478 0.1909 0.1104 0 0557 1.4169 0.9311 0.5888 0 9311 0 6561 0.3741 WGHTD AVE 1 2837 0.1459 0.0916 0 0394 1.1205 0.6824 0.3984 0 6824 0.4514 0.2320

*Includes contents and A.L.E.

FPHLM V3.0 2008 288 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 Personal Residential -- Owners -- FRAME $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Indian River LOW 4 0612 0 3002 0.1473 0 0883 3.9313 3.2682 2.2086 3 2682 2 5420 1.1593 HIGH 6.4165 0 8883 0.2184 0 3042 7.0856 6.3460 5.0573 6 3460 5.4647 3.6825 WGHTD AVE 4 9648 0.4574 0.1789 0.1474 5.0218 4.3046 3.1077 4 3046 3.4855 1.8801

Jackson LOW 0.7861 0 0770 0.0516 0 0176 0.6192 0.3082 0.1518 0 3082 0.1756 0.0772 HIGH 1 2637 0.1414 0.0897 0 0375 1.0937 0.6570 0.3770 0 6570 0.4289 0.2144 WGHTD AVE 1 0316 0.1083 0.0713 0 0267 0.8536 0.4718 0.2514 0.4718 0 2891 0.1339

Jefferson LOW 0 6811 0 0705 0.0456 0 0170 0.5524 0.2926 0.1551 0 2926 0.1770 0.0858 HIGH 0 8187 0 0894 0.0559 0 0234 0.6816 0.3945 0.2287 0 3945 0 2575 0.1377 WGHTD AVE 0 6968 0 0727 0.0467 0 0177 0.5680 0.3041 0.1628 0 3041 0.1855 0.0908

Lafayette LOW 0.7391 0 0775 0.0498 0 0190 0.6058 0.3281 0.1770 0 3281 0 2018 0.0989 HIGH 0 8787 0 0957 0.0598 0 0242 0.7371 0.4176 0.2325 0.4176 0 2643 0.1326 WGHTD AVE 0 8728 0 0950 0.0595 0 0240 0.7318 0.4140 0.2303 0.4140 0 2618 0.1312

Lake LOW 2.1902 0.1215 0.0755 0 0286 1.9931 1.5725 0.9724 1 5725 1.1593 0.4114 HIGH 3 9921 0 2790 0.1441 0 0819 3.8696 3.2444 2.2114 3 2444 2 5380 1.1550 WGHTD AVE 3.1828 0.1898 0.1128 0 0501 2.9811 2.4298 1.5697 2.4298 1 8404 0.7254

Lee LOW 4.4442 0 3315 0.1662 0 0982 4.3516 3.6643 2.5188 3 6643 2 8806 1.3462 HIGH 6 2974 0.7140 0.2381 0 2448 6.7236 5.9550 4.5703 5 9550 5 0092 3.0669 WGHTD AVE 5 2730 0.4386 0.1918 0.1443 5.2906 4.5500 3.2703 4 5500 3 6755 1.9250

Leon LOW 0 5975 0 0575 0.0389 0 0129 0.4649 0.2249 0.1105 0 2249 0.1272 0.0578 HIGH 0 9848 0.1086 0.0682 0 0276 0.8342 0.4811 0.2677 0.4811 0 3054 0.1499 WGHTD AVE 0 8036 0 0829 0.0542 0 0201 0.6547 0.3531 0.1864 0 3531 0 2139 0.1002

Levy LOW 0 8702 0 0931 0.0603 0 0227 0.7162 0.3881 0.2052 0 3881 0 2347 0.1115 HIGH 1.4524 0.1808 0.1030 0 0519 1.3269 0.8678 0.5470 0 8678 0 6097 0.3479 WGHTD AVE 1.1989 0.1390 0.0838 0 0378 1.0506 0.6459 0.3843 0 6459 0.4326 0.2318

Liberty LOW 1 0071 0.1054 0.0690 0 0255 0.8275 0.4501 0.2343 0.4501 0 2706 0.1215 HIGH 1 0456 0.1122 0.0721 0 0280 0.8728 0.4897 0.2665 0.4897 0 3048 0.1472 WGHTD AVE 1 0107 0.1066 0.0694 0 0261 0.8348 0.4589 0.2419 0.4589 0 2787 0.1274

Madison LOW 0 5980 0 0584 0.0390 0 0134 0.4702 0.2333 0.1171 0 2333 0.1345 0.0619 HIGH 0.7896 0 0836 0.0533 0 0206 0.6509 0.3566 0.1934 0 3566 0 2204 0.1080 WGHTD AVE 0.7411 0 0769 0.0497 0 0185 0.6033 0.3224 0.1708 0 3224 0.1953 0.0933

Manatee LOW 3 5529 0 2649 0.1339 0 0796 3.4766 2.9239 2.0131 2 9239 2 3004 1.0797 HIGH 5.4896 0 6725 0.2002 0 2303 5.9208 5.2511 4.0836 5 2511 4.4525 2.8331 WGHTD AVE 4 0372 0 3392 0.1485 0.1089 4.0309 3.4462 2.4651 3.4462 2.7748 1.4458

Marion LOW 1.7807 0 0936 0.0604 0 0208 1.5846 1.2162 0.7248 1 2162 0 8766 0.2845 HIGH 3.1919 0 2058 0.1130 0 0571 3.0369 2.5079 1.6632 2 5079 1 9296 0.8173 WGHTD AVE 2.7398 0.1599 0.0936 0 0407 2.5410 2.0509 1.3055 2 0509 1 5394 0.5862

Martin LOW 5 5029 0.4668 0.1718 0.1529 5.5300 4.7683 3.3573 4.7683 3 8094 1.9207 HIGH 8 8069 1 2848 0.2545 0.4607 9.9439 9.0835 7.3514 9 0835 7 9084 5.4443 WGHTD AVE 6 3367 0 6388 0.1927 0 2177 6.5804 5.7870 4.2726 5.7870 4.7587 2.6911

Miami-Dade LOW 6 3691 0.7503 0.1869 0 2637 6.8431 6.1184 4.7117 6.1184 5.1637 3.2098 HIGH 12.3543 2.4188 0.3275 0 8765 15 0084 14.1356 12 2773 14.1356 12.8762 10.1047 WGHTD AVE 8 3931 1 2095 0.2397 0.4404 9.4879 8.6855 7.0361 8 6855 7 5679 5.1929

Monroe LOW 7.1985 0.7628 0.2059 0 2879 7.7536 7.0540 5.5083 7 0540 6 0115 3.7381 HIGH 12.1357 2 2055 0.3086 0 8453 14 5620 13.7846 12 0217 13.7846 12.5941 9.8970 WGHTD AVE 9 5750 1 5284 0.2671 0 5489 11.1044 10 3475 8.6476 10.3475 9 2004 6.6451

Nassau LOW 0 5997 0 0617 0.0400 0 0148 0.4799 0.2456 0.1301 0 2456 0.1471 0.0754 HIGH 0 8718 0.1039 0.0586 0 0286 0.7626 0.4645 0.2882 0.4645 0 3193 0.1884 WGHTD AVE 0.7736 0 0862 0.0523 0 0222 0.6507 0.3693 0.2126 0 3693 0 2384 0.1298

*Includes contents and A.L.E.

FPHLM V3.0 2008 289 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 Personal Residential -- Owners -- FRAME $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Okaloosa LOW 1.4958 0.1898 0.1097 0 0567 1.3875 0.9250 0.5953 0 9250 0 6597 0.3870 HIGH 3 0581 0 5280 0.1963 0.1891 3.3402 2.7105 2.1297 2.7105 2 2611 1.6775 WGHTD AVE 2 5776 0.4118 0.1689 0.1433 2.7087 2.1234 1.6137 2.1234 1.7258 1.2337

Okeechobee LOW 4.1892 0 2698 0.1529 0 0740 3.9651 3.2470 2.1249 3 2470 2.4772 1.0210 HIGH 4.4990 0 3091 0.1652 0 0883 4.3178 3.5766 2.3871 3 5766 2.7616 1.2010 WGHTD AVE 4.4478 0 2965 0.1634 0 0832 4.2463 3.5036 2.3195 3 5036 2 6921 1.1436

Orange LOW 2 8607 0.1607 0.0993 0 0395 2.6596 2.1350 1.3431 2.1350 1 5915 0.5808 HIGH 3 9662 0 2783 0.1413 0 0811 3.8348 3.2051 2.1760 3 2051 2 5012 1.1339 WGHTD AVE 3 3835 0 2036 0.1198 0 0540 3.1787 2.5976 1.6861 2 5976 1 9731 0.7859

Osceola LOW 3 2248 0.1823 0.1129 0 0460 2.9954 2.4274 1.5492 2.4274 1 8254 0.6883 HIGH 3 9236 0 2514 0.1386 0 0698 3.7164 3.0706 2.0343 3 0706 2 3612 0.9978 WGHTD AVE 3.4809 0 2064 0.1221 0 0539 3.2628 2.6650 1.7264 2 6650 2 0220 0.7976

Palm Beach LOW 5.7979 0 5339 0.1772 0.1772 5.9093 5.1337 3.6847 5.1337 4.1492 2.2020 HIGH 10.2483 1 6224 0.2896 0 5815 11 8186 10 8977 8.9881 10.8977 9 6036 6.8432 WGHTD AVE 7.1182 0 8339 0.2166 0 3018 7.6517 6.8349 5.2492 6 8349 5.7586 3.5704

Pasco LOW 2 8557 0.1725 0.1024 0 0436 2.7207 2.2343 1.4563 2 2343 1.7160 0.6482 HIGH 3 6290 0 2632 0.1303 0 0800 3.4848 2.9082 1.9709 2 9082 2 2671 1.0520 WGHTD AVE 3 2510 0 2233 0.1178 0 0639 3.1248 2.5937 1.7421 2 5937 2 0105 0.8877

Pinellas LOW 3 0593 0 2173 0.1131 0 0629 2.9494 2.4477 1.6486 2.4477 1 8999 0.8499 HIGH 5 9985 0 8854 0.2147 0 3112 6.7554 6.1095 4.9691 6.1095 5 3293 3.7285 WGHTD AVE 3 6384 0 3135 0.1355 0 0993 3.6370 3.0939 2.2028 3 0939 2.4835 1.2896

Polk LOW 3 3369 0.1916 0.1174 0 0488 3.1180 2.5352 1.6250 2 5352 1 9115 0.7304 HIGH 4 6183 0 3527 0.1686 0.1082 4.5540 3.8624 2.6976 3 8624 3 0664 1.4913 WGHTD AVE 3.7659 0 2358 0.1347 0 0647 3.5782 2.9578 1.9543 2 9578 2 2712 0.9453

Putnam LOW 1 0061 0.1064 0.0691 0 0258 0.8337 0.4582 0.2425 0.4582 0 2793 0.1281 HIGH 1 2420 0.1418 0.0888 0 0377 1.0833 0.6583 0.3824 0 6583 0.4337 0.2210 WGHTD AVE 1.1667 0.1297 0.0824 0 0336 0.9999 0.5896 0.3332 0 5896 0 3797 0.1875

St. Johns LOW 0 8701 0 0908 0.0582 0 0220 0.7110 0.3831 0.2046 0 3831 0 2339 0.1127 HIGH 1 5166 0 2036 0.1045 0 0627 1.4439 1.0024 0.6803 1 0024 0.7454 0.4691 WGHTD AVE 1.1753 0.1422 0.0813 0 0400 1.0461 0.6571 0.4092 0 6571 0.4546 0.2635

St. Lucie LOW 5 0478 0.4061 0.1576 0.1304 5.0205 4.3016 2.9842 4 3016 3.4060 1.6530 HIGH 6.7974 0.7710 0.2046 0 2677 7.2366 6.4350 4.8837 6.4350 5 3820 3.2452 WGHTD AVE 5 8816 0 5561 0.1809 0.1869 6.0315 5.2622 3.8132 5 2622 4 2779 2.3178

Santa Rosa LOW 1.4854 0.1820 0.1075 0 0533 1.3626 0.8990 0.5687 0 8990 0 6342 0.3590 HIGH 3 0990 0 5482 0.1961 0.1978 3.4253 2.8113 2.2319 2 8113 2 3645 1.7743 WGHTD AVE 2 3900 0 3787 0.1562 0.1296 2.4909 1.9341 1.4607 1 9341 1 5637 1.1129

Sarasota LOW 3.4445 0 2571 0.1288 0 0770 3.3643 2.8231 1.9379 2 8231 2 2170 1.0366 HIGH 4 5825 0 3938 0.1728 0.1250 4.6151 3.9581 2.8312 3 9581 3.1878 1.6510 WGHTD AVE 4 0241 0 3261 0.1524 0.1017 4.0058 3.4088 2.4002 3.4088 2.7190 1.3532

Seminole LOW 2 9479 0.1750 0.1049 0 0445 2.7302 2.1906 1.3853 2.1906 1 6371 0.6155 HIGH 3 3624 0 2285 0.1212 0 0641 3.2004 2.6270 1.7350 2 6270 2 0149 0.8614 WGHTD AVE 3.1413 0.1902 0.1126 0 0494 2.9305 2.3696 1.5139 2 3696 1.7821 0.6866

Sumter LOW 2 8209 0.1627 0.0973 0 0411 2.6107 2.1018 1.3326 2.1018 1 5739 0.5926 HIGH 3 3910 0 2080 0.1197 0 0563 3.2008 2.6290 1.7191 2 6290 2 0060 0.8143 WGHTD AVE 3 0223 0.1781 0.1058 0 0464 2.8160 2.2867 1.4689 2 2867 1.7260 0.6720

*Includes contents and A.L.E.

FPHLM V3.0 2008 290 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 Personal Residential -- Owners -- FRAME $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Suwanee LOW 0 6147 0 0597 0.0402 0 0136 0.4818 0.2374 0.1188 0 2374 0.1365 0.0629 HIGH 1 0217 0.1140 0.0711 0 0294 0.8724 0.5105 0.2879 0 5105 0 3277 0.1634 WGHTD AVE 0.7325 0 0749 0.0491 0 0180 0.5927 0.3136 0.1651 0 3136 0.1891 0.0899

Taylor LOW 0.7444 0 0786 0.0502 0 0195 0.6132 0.3357 0.1828 0 3357 0 2081 0.1029 HIGH 1.1551 0.1444 0.0805 0 0409 1.0391 0.6593 0.4141 0 6593 0.4588 0.2703 WGHTD AVE 0 9346 0.1083 0.0641 0 0294 0.8136 0.4942 0.2976 0.4942 0 3331 0.1850

Union LOW 0 6883 0 0672 0.0452 0 0152 0.5397 0.2658 0.1315 0 2658 0.1515 0.0683 HIGH 0 9837 0.1038 0.0668 0 0254 0.8123 0.4470 0.2393 0.4470 0 2745 0.1293 WGHTD AVE 0 9598 0.1012 0.0652 0 0246 0.7912 0.4328 0.2308 0.4328 0 2648 0.1245

Volusia LOW 2 2937 0.1285 0.0792 0 0314 2.1008 1.6711 1.0431 1 6711 1 2393 0.4499 HIGH 4 0644 0 3703 0.1453 0.1198 4.1207 3.5445 2.5709 3 5445 2 8790 1.5577 WGHTD AVE 3.1585 0 2105 0.1109 0 0590 3.0056 2.4784 1.6477 2.4784 1 9091 0.8244

Wakulla LOW 0 8054 0 0839 0.0535 0 0204 0.6569 0.3525 0.1883 0 3525 0 2150 0.1043 HIGH 1 5252 0 2048 0.1072 0 0620 1.4500 1.0029 0.6695 1 0029 0.7373 0.4523 WGHTD AVE 0 9609 0.1027 0.0598 0 0274 0.8069 0.4733 0.2762 0.4733 0 3108 0.1668

Walton LOW 1 3398 0.1586 0.0968 0 0442 1.1868 0.7364 0.4436 0.7364 0.4973 0.2717 HIGH 3.1032 0 5257 0.1975 0.1877 3.3741 2.7359 2.1394 2.7359 2 2770 1.6703 WGHTD AVE 2 0204 0 2961 0.1362 0 0962 2.0097 1.4682 1.0442 1.4682 1.1326 0.7524

Washington LOW 1.1730 0.1281 0.0816 0 0331 0.9955 0.5774 0.3232 0 5774 0 3685 0.1814 HIGH 1 9642 0 2771 0.1396 0 0884 1.9206 1.3738 0.9515 1 3738 1 0376 0.6691 WGHTD AVE 1.4897 0.1804 0.1065 0 0516 1.3486 0.8698 0.5378 0 8698 0 6024 0.3332

STATEWIDE LOW 0 5482 0 0520 0.0360 0 0116 0.4234 0.2013 0.0983 0 2013 0.1130 0.0517 HIGH 12.3543 2.4188 0.3275 0 8765 15 0084 14.1356 12 2773 14.1356 12.8762 10.1047 WGHTD AVE 3.1427 0 2939 0.1270 0 0973 3.0794 2.5434 1.8180 2 5434 2 0320 1.1226

*Includes contents and A.L.E.

FPHLM V3.0 2008 291 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL -Owners -- MASONRY 0% 0% $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Alachua LOW 0 8981 0 0934 0.0628 0 0222 0.7258 0 3773 0.1935 0 3773 0.2216 0.1037 HIGH 1.1107 0.1244 0.0813 0 0316 0.9383 0 5307 0 2889 0 5307 0.3290 0.1604 WGHTD AVE 0 9780 0.1061 0.0700 0 0259 0.8076 0.4374 0 2295 0.4374 0.2625 0.1241

Baker LOW 0 5565 0 0538 0.0367 0 0121 0.4287 0 2004 0 0974 0 2004 0.1109 0.0530 HIGH 0 8502 0 0896 0.0587 0 0215 0.6890 0 3601 0.1870 0 3601 0.2133 0.1028 WGHTD AVE 0 8148 0 0856 0.0561 0 0203 0.6584 0 3411 0.1761 0 3411 0.2008 0.0964

Bay LOW 1 2116 0.1351 0.0875 0 0345 1.0214 0 5762 0 3135 0 5762 0.3570 0.1749 HIGH 3 2644 0 5978 0.2089 0 2201 3.6547 3 0201 2.4271 3 0201 2.5612 1.9541 WGHTD AVE 1 8710 0 2551 0.1380 0 0790 1.7907 1 2443 0 8339 1 2443 0.9146 0.5685

Bradford LOW 0 8707 0 0912 0.0602 0 0215 0.7019 0 3622 0.1835 0 3622 0.2105 0.0970 HIGH 1 0484 0.1151 0.0753 0 0285 0.8729 0.4805 0 2554 0.4805 0.2918 0.1393 WGHTD AVE 0 8964 0 0939 0.0625 0 0223 0.7244 0 3758 0.1912 0 3758 0.2192 0.1016

Brevard LOW 2 6438 0.1804 0.1066 0 0484 2.4577 1 9386 1 2519 1 9386 1.4616 0.6084 HIGH 5 3507 0 6313 0.2286 0 2148 5.5796 4 8622 3.7422 4 8622 4.0899 2.5791 WGHTD AVE 3 3890 0 2545 0.1373 0 0731 3.2391 2 6295 1.7655 2 6295 2.0317 0.9275

Broward LOW 5 2412 0 5070 0.1977 0.1716 5.3590 4 6029 3 3705 4 6029 3.7579 2.1116 HIGH 7.4921 0 9744 0.2787 0 3525 8.2480 7.4005 5 8986 7.4005 6.3736 4.2648 WGHTD AVE 6 0346 0 6474 0.2272 0 2253 6.3336 5 5332 4.1788 5 5332 4.6060 2.7630

Calhoun LOW 1 0114 0.1077 0.0715 0 0262 0.8280 0.4414 0 2286 0.4414 0.2619 0.1224 HIGH 1 2550 0.1475 0.0947 0 0390 1.0832 0 6321 0 3546 0 6321 0.4005 0.2061 WGHTD AVE 1.1221 0.1239 0.0815 0 0310 0.9394 0 5222 0 2775 0 5222 0.3175 0.1501

Charlotte LOW 3 5647 0 2492 0.1498 0 0696 3.4004 2.7697 1 8427 2.7697 2.1296 0.9233 HIGH 5 2802 0 5990 0.2274 0 2048 5.5861 4 8628 3 6994 4 8628 4.0613 2.4757 WGHTD AVE 3 9370 0 3005 0.1686 0 0886 3.8275 3.1619 2.1603 3.1619 2.4708 1.1535

Citrus LOW 2.1466 0.1267 0.0821 0 0303 1.9502 1 5168 0 9425 1 5168 1.1186 0.4091 HIGH 2 8715 0.1895 0.1149 0 0509 2.7018 2.1791 1.4260 2.1791 1.6592 0.6909 WGHTD AVE 2 6255 0.1652 0.1032 0 0425 2.4366 1 9391 1 2446 1 9391 1.4589 0.5783

Clay LOW 0 8356 0 0862 0.0576 0 0203 0.6698 0 3417 0.1729 0 3417 0.1980 0.0923 HIGH 1.1366 0.1276 0.0830 0 0325 0.9612 0 5448 0 2975 0 5448 0.3387 0.1661 WGHTD AVE 0 9303 0 0999 0.0657 0 0245 0.7645 0.4109 0 2169 0.4109 0.2474 0.1193

Collier LOW 4 3130 0 3036 0.1857 0 0859 4.1298 3 3739 2 2539 3 3739 2.6005 1.1391 HIGH 6 0929 0 6475 0.2755 0 2220 6.3891 5 5422 4.1521 5 5422 4.5847 2.6836 WGHTD AVE 4.7926 0 3859 0.2087 0.1189 4.7247 3 9511 2.7525 3 9511 3.1247 1.5293

Columbia LOW 0 5925 0 0576 0.0395 0 0129 0.4573 0 2144 0.1033 0 2144 0.1181 0.0553 HIGH 0 8944 0 0974 0.0627 0 0239 0.7372 0 3978 0 2104 0 3978 0.2397 0.1164 WGHTD AVE 0 8297 0 0881 0.0578 0 0212 0.6742 0 3532 0.1829 0 3532 0.2088 0.0999

De Soto LOW 3 5794 0 2273 0.1439 0 0596 3.3556 2.7034 1.7590 2.7034 2.0516 0.8332 HIGH 3 9941 0 2851 0.1662 0 0820 3.8476 3.1700 2.1453 3.1700 2.4639 1.1128 WGHTD AVE 3.7223 0 2519 0.1521 0 0694 3.5374 2 8797 1 9121 2 8797 2.2122 0.9519

Dixie LOW 0 9163 0.1033 0.0655 0 0264 0.7724 0.4353 0 2405 0.4353 0.2722 0.1386 HIGH 1 5289 0 2075 0.1124 0 0632 1.4458 0 9814 0 6497 0 9814 0.7131 0.4416 WGHTD AVE 1 0249 0.1199 0.0739 0 0316 0.8845 0 5213 0 3002 0 5213 0.3377 0.1793

*Includes contents and A.L.E.

FPHLM V3.0 2008 292 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL -Owners -- MASONRY 0% 0% $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Duval LOW 0 5378 0 0517 0.0360 0 0115 0.4131 0.1914 0 0921 0.1914 0.1050 0.0498 HIGH 1 2377 0.1805 0.0864 0 0581 1.2017 0 8427 0 6009 0 8427 0.6457 0.4476 WGHTD AVE 0 8769 0 0962 0.0618 0 0240 0.7251 0 3933 0 2133 0 3933 0.2410 0.1233

Escambia LOW 1 3955 0.1806 0.1102 0 0540 1.2936 0 8488 0 5362 0 8488 0.5941 0.3456 HIGH 4 0399 0 8185 0.2385 0 3106 4.7592 4.1124 3.4544 4.1124 3.6084 2.8929 WGHTD AVE 2 3773 0 3885 0.1679 0.1343 2.4974 1 9301 1.4533 1 9301 1.5529 1.1100

Flagler LOW 2.4460 0.1530 0.0967 0 0391 2.2624 1.7923 1.1450 1.7923 1.3443 0.5290 HIGH 3.1334 0 2555 0.1317 0 0773 3.0601 2 5241 1.7407 2 5241 1.9826 0.9660 WGHTD AVE 2 6082 0.1843 0.1069 0 0508 2.4605 1 9759 1 3000 1 9759 1.5078 0.6523

Franklin LOW 1 6620 0 2314 0.1260 0 0719 1.5901 1 0908 0.7272 1 0908 0.7960 0.5003 HIGH 2.7181 0.4952 0.1759 0.1805 3.0135 2.4594 1 9699 2.4594 2.0766 1.5924 WGHTD AVE 2.1263 0 3490 0.1498 0.1184 2.2167 1 6907 1 2687 1 6907 1.3548 0.9733

Gadsen LOW 0.7762 0 0789 0.0530 0 0182 0.6165 0 3058 0.1502 0 3058 0.1726 0.0786 HIGH 1 0643 0.1165 0.0764 0 0290 0.8859 0.4876 0 2574 0.4876 0.2947 0.1389 WGHTD AVE 0 8191 0 0849 0.0566 0 0201 0.6570 0 3354 0.1682 0 3354 0.1930 0.0889

Gilchrist LOW 0 8535 0 0916 0.0596 0 0222 0.6976 0 3702 0.1926 0 3702 0.2200 0.1049 HIGH 1.1122 0.1302 0.0826 0 0344 0.9635 0 5696 0 3233 0 5696 0.3658 0.1875 WGHTD AVE 1 0386 0.1191 0.0751 0 0308 0.8868 0 5122 0 2856 0 5122 0.3238 0.1637

Glades LOW 4 0641 0 2583 0.1644 0 0681 3.8123 3 0724 2 0002 3 0724 2.3323 0.9489 HIGH 4 2021 0 2744 0.1708 0 0739 3.9648 3 2110 2.1078 3 2110 2.4498 1.0205 WGHTD AVE 4.1984 0 2739 0.1706 0 0737 3.9605 3 2071 2.1048 3 2071 2.4465 1.0185

Gulf LOW 1 2753 0.1535 0.0960 0 0419 1.1169 0 6692 0 3894 0 6692 0.4366 0.2362 HIGH 1.7403 0 2373 0.1332 0 0727 1.6485 1.1154 0.7316 1.1154 0.8036 0.4944 WGHTD AVE 1 6240 0 2141 0.1225 0 0636 1.5087 0 9981 0 6416 0 9981 0.7070 0.4265

Hamilton LOW 0 5508 0 0536 0.0367 0 0120 0.4251 0.1993 0 0964 0.1993 0.1099 0.0520 HIGH 0.7594 0 0805 0.0529 0 0194 0.6164 0 3226 0.1671 0 3226 0.1907 0.0915 WGHTD AVE 0 6868 0 0707 0.0471 0 0168 0.5485 0 2780 0.1413 0 2780 0.1612 0.0770

Hardee LOW 3 5226 0 2199 0.1417 0 0570 3.2922 2 6455 1.7141 2 6455 2.0025 0.8019 HIGH 3 8618 0 2648 0.1603 0 0741 3.6901 3 0216 2 0199 3 0216 2.3311 1.0184 WGHTD AVE 3 5875 0 2260 0.1449 0 0593 3.3604 2.7066 1.7600 2.7066 2.0532 0.8308

Hendry LOW 4.1988 0 2774 0.1728 0 0757 3.9775 3 2329 2.1372 3 2329 2.4770 1.0494 HIGH 4.7430 0 3471 0.1973 0.1012 4.5853 3.7845 2 5707 3.7845 2.9481 1.3516 WGHTD AVE 4 5533 0 3207 0.1878 0 0911 4.3664 3 5854 2.4133 3 5854 2.7774 1.2398

Hernando LOW 2.4603 0.1455 0.0952 0 0350 2.2542 1.7619 1.1013 1.7619 1.3042 0.4808 HIGH 2 9585 0.1969 0.1195 0 0536 2.7906 2 2548 1.4819 2 2548 1.7211 0.7247 WGHTD AVE 2.7533 0.1816 0.1101 0 0488 2.5845 2 0771 1 3567 2 0771 1.5792 0.6566

Highlands LOW 3 3871 0 2008 0.1328 0 0493 3.1195 2.4717 1 5673 2.4717 1.8465 0.6976 HIGH 4 0899 0 2622 0.1667 0 0697 3.8486 3.1112 2 0335 3.1112 2.3676 0.9721 WGHTD AVE 3 6416 0 2244 0.1456 0 0574 3.3902 2.7131 1.7466 2.7131 2.0456 0.8061

Hillsborough LOW 2 6563 0.1667 0.1062 0 0432 2.4726 1 9748 1 2745 1 9748 1.4907 0.5957 HIGH 4 0479 0 3557 0.1750 0.1142 4.0656 3.4405 2.4658 3.4405 2.7689 1.4588 WGHTD AVE 3 0998 0 2141 0.1278 0 0598 2.9497 2 3994 1 5974 2 3994 1.8457 0.8045

Holmes LOW 1.1490 0.1287 0.0852 0 0331 0.9763 0 5588 0 3059 0 5588 0.3487 0.1695 HIGH 1.4676 0.1820 0.1104 0 0524 1.3263 0 8423 0 5166 0 8423 0.5762 0.3232 WGHTD AVE 1 2237 0.1405 0.0906 0 0374 1.0547 0 6191 0 3499 0 6191 0.3962 0.2016

*Includes contents and A.L.E.

FPHLM V3.0 2008 293 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL -Owners -- MASONRY 0% 0% $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Indian River LOW 3 5991 0 2567 0.1473 0 0718 3.4128 2.7533 1 8145 2.7533 2.1039 0.9095 HIGH 5.4121 0 6607 0.2184 0 2252 5.7796 5 0452 3 8948 5 0452 4.2522 2.7018 WGHTD AVE 4.4060 0 3797 0.1777 0.1177 4.3487 3 6307 2 5553 3 6307 2.8884 1.4800

Jackson LOW 0.7688 0 0766 0.0516 0 0175 0.6020 0 2918 0.1414 0 2918 0.1623 0.0740 HIGH 1 2131 0.1376 0.0897 0 0361 1.0392 0 6041 0 3371 0 6041 0.3828 0.1910 WGHTD AVE 0 9949 0.1054 0.0707 0 0257 0.8133 0.4337 0 2244 0.4337 0.2572 0.1200

Jefferson LOW 0 6625 0 0694 0.0456 0 0166 0.5330 0 2738 0.1418 0 2738 0.1611 0.0791 HIGH 0.7924 0 0865 0.0559 0 0224 0.6537 0 3632 0 2046 0 3632 0.2300 0.1224 WGHTD AVE 0 6722 0 0707 0.0462 0 0170 0.5423 0 2805 0.1461 0 2805 0.1659 0.0819

Lafayette LOW 0.7174 0 0761 0.0498 0 0185 0.5830 0 3060 0.1612 0 3060 0.1830 0.0906 HIGH 0 8493 0 0936 0.0598 0 0235 0.7057 0 3871 0 2100 0 3871 0.2381 0.1199 WGHTD AVE 0 8464 0 0933 0.0596 0 0234 0.7032 0 3855 0 2090 0 3855 0.2370 0.1193

Lake LOW 1 9834 0.1139 0.0755 0 0262 1.7776 1 3588 0 8292 1 3588 0.9906 0.3479 HIGH 3 5237 0 2403 0.1441 0 0668 3.3508 2.7291 1 8141 2.7291 2.0978 0.9068 WGHTD AVE 3 0545 0.1894 0.1209 0 0485 2.8354 2 2604 1.4495 2 2604 1.7000 0.6679

Lee LOW 3 9209 0 2815 0.1662 0 0800 3.7627 3 0792 2 0627 3 0792 2.3775 1.0521 HIGH 5 3424 0 5469 0.2381 0.1836 5.5459 4.7828 3 5505 4.7828 3.9339 2.2551 WGHTD AVE 4 3749 0 3333 0.1879 0 0991 4.2629 3 5341 2.4215 3 5341 2.7669 1.2967

Leon LOW 0 5855 0 0572 0.0389 0 0128 0.4529 0 2135 0.1033 0 2135 0.1179 0.0554 HIGH 0 9499 0.1061 0.0682 0 0268 0.7971 0.4451 0 2413 0.4451 0.2745 0.1353 WGHTD AVE 0.7822 0 0820 0.0544 0 0198 0.6326 0 3313 0.1713 0 3313 0.1956 0.0933

Levy LOW 0 8445 0 0917 0.0603 0 0221 0.6894 0 3622 0.1870 0 3622 0.2129 0.1029 HIGH 1 3781 0.1719 0.1030 0 0486 1.2422 0.7849 0.4795 0.7849 0.5350 0.3001 WGHTD AVE 1.1314 0.1318 0.0838 0 0345 0.9756 0 5715 0 3240 0 5715 0.3660 0.1891

Liberty LOW 0 9775 0.1041 0.0690 0 0251 0.7972 0.4209 0 2145 0.4209 0.2463 0.1129 HIGH 1 0118 0.1105 0.0721 0 0273 0.8377 0.4558 0 2412 0.4558 0.2748 0.1337 WGHTD AVE 0 9796 0.1052 0.0693 0 0256 0.8028 0.4280 0 2206 0.4280 0.2529 0.1175

Madison LOW 0 5847 0 0579 0.0390 0 0132 0.4567 0 2204 0.1086 0 2204 0.1239 0.0586 HIGH 0.7656 0 0821 0.0533 0 0201 0.6257 0 3321 0.1757 0 3321 0.1995 0.0987 WGHTD AVE 0.7186 0 0757 0.0495 0 0180 0.5804 0 3008 0.1558 0 3008 0.1773 0.0859

Manatee LOW 3.1339 0 2248 0.1339 0 0641 3.0051 2.4554 1 6469 2.4554 1.8969 0.8425 HIGH 4 6508 0 5096 0.2002 0.1719 4.8653 4 2004 3.1605 4 2004 3.4830 2.0805 WGHTD AVE 3.4479 0 2694 0.1474 0 0809 3.3606 2.7829 1 9184 2.7829 2.1860 1.0487

Marion LOW 1 6256 0 0898 0.0604 0 0194 1.4258 1 0589 0 6258 1 0589 0.7564 0.2473 HIGH 2 8414 0.1822 0.1130 0 0479 2.6561 2.1299 1 3823 2.1299 1.6134 0.6554 WGHTD AVE 2.4514 0.1456 0.0941 0 0352 2.2370 1.7511 1 0954 1.7511 1.2970 0.4804

Martin LOW 4.4397 0 3188 0.1718 0 0951 4.2669 3 5110 2 3368 3 5110 2.7041 1.1912 HIGH 6 6145 0.7610 0.2545 0 2687 7.0449 6.1935 4.7441 6.1935 5.2006 3.2257 WGHTD AVE 4 9689 0.4076 0.1925 0.1298 4.9069 4.1214 2 8636 4.1214 3.2581 1.6092

Miami-Dade LOW 4 8946 0.4577 0.1869 0.1544 4.9735 4 2558 3 0851 4 2558 3.4531 1.8923 HIGH 8 9491 1.4042 0.3275 0 5162 10 3236 9.4621 7 8946 9.4621 8.3910 6.1510 WGHTD AVE 6 2246 0 6925 0.2320 0 2445 6.5974 5 8069 4.4385 5 8069 4.8707 2.9873

Monroe LOW 6 9457 0 6925 0.2059 0 2640 7.4068 6.7073 5.1270 6.7073 5.6425 3.3324 HIGH 11.1642 1.7971 0.3086 0.7054 13 0137 12.1735 10.3662 12.1735 10.9509 8.2199 WGHTD AVE 9 9021 1.4814 0.2833 0 5623 11.4240 10.6336 8 8120 10.6336 9.4054 6.6700

Nassau LOW 0 5847 0 0605 0.0400 0 0143 0.4637 0 2299 0.1188 0 2299 0.1336 0.0693 HIGH 0 8347 0 0984 0.0586 0 0267 0.7191 0.4219 0 2531 0.4219 0.2807 0.1627 WGHTD AVE 0.7301 0 0800 0.0508 0 0200 0.6013 0 3246 0.1789 0 3246 0.2007 0.1073

*Includes contents and A.L.E.

FPHLM V3.0 2008 294 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL -Owners -- MASONRY 0% 0% $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Okaloosa LOW 1.4135 0.1803 0.1097 0 0529 1.2938 0 8331 0 5192 0 8331 0.5762 0.3322 HIGH 2.7682 0.4661 0.1963 0.1668 2.9696 2 3435 1.7902 2 3435 1.9090 1.3787 WGHTD AVE 2 3210 0 3656 0.1686 0.1251 2.3977 1 8179 1 3368 1 8179 1.4365 0.9963

Okeechobee LOW 3.7485 0 2403 0.1529 0 0623 3.4867 2.7721 1.7777 2.7721 2.0836 0.8255 HIGH 4 0022 0 2695 0.1652 0 0729 3.7698 3 0325 1 9781 3 0325 2.3034 0.9554 WGHTD AVE 3 9639 0 2603 0.1628 0 0691 3.7144 2 9754 1 9262 2 9754 2.2497 0.9128

Orange LOW 2 5640 0.1500 0.0993 0 0352 2.3530 1 8422 1.1424 1 8422 1.3574 0.4870 HIGH 3 5071 0 2399 0.1413 0 0663 3.3260 2 6997 1.7884 2 6997 2.0708 0.8928 WGHTD AVE 3 0034 0.1822 0.1180 0 0455 2.7724 2.1971 1 3970 2.1971 1.6438 0.6304

Osceola LOW 2 9001 0.1688 0.1129 0 0405 2.6545 2 0892 1 3124 2 0892 1.5518 0.5722 HIGH 3 5011 0 2225 0.1386 0 0585 3.2566 2 6070 1 6900 2 6070 1.9736 0.7995 WGHTD AVE 3.1607 0.1918 0.1248 0 0478 2.9238 2 3222 1.4799 2 3222 1.7399 0.6690

Palm Beach LOW 4 6504 0 3571 0.1772 0.1097 4.5233 3.7536 2 5469 3.7536 2.9247 1.3634 HIGH 7 6807 0 9644 0.2896 0 3431 8.3645 7.4535 5 8512 7.4535 6.3574 4.1389 WGHTD AVE 5 5471 0 5347 0.2095 0.1801 5.6561 4 8499 3 5313 4 8499 3.9457 2.1962

Pasco LOW 2 5480 0.1594 0.1024 0 0384 2.3820 1 8980 1 2296 1 8980 1.4350 0.5384 HIGH 3 2170 0 2209 0.1303 0 0640 3.0337 2.4537 1 6219 2.4537 1.8791 0.8147 WGHTD AVE 2 8359 0.1940 0.1165 0 0531 2.6798 2.1598 1.4199 2.1598 1.6482 0.6999

Pinellas LOW 2.7123 0.1877 0.1131 0 0515 2.5691 2 0700 1 3613 2 0700 1.5797 0.6725 HIGH 4 9907 0 6496 0.2147 0 2286 5.4343 4.7935 3.7745 4.7935 4.0907 2.6991 WGHTD AVE 3 0934 0 2424 0.1314 0 0718 3.0003 2.4675 1 6908 2.4675 1.9305 0.9206

Polk LOW 2 9858 0.1766 0.1174 0 0429 2.7582 2.1783 1 3732 2.1783 1.6215 0.6042 HIGH 4 0456 0 2959 0.1686 0 0863 3.9067 3 2192 2.1870 3 2192 2.5075 1.1511 WGHTD AVE 3 3534 0 2101 0.1347 0 0545 3.1329 2 5153 1 6272 2 5153 1.9021 0.7606

Putnam LOW 0 9722 0.1048 0.0691 0 0253 0.8019 0.4275 0 2212 0.4275 0.2535 0.1182 HIGH 1.1906 0.1376 0.0888 0 0362 1.0275 0 6041 0 3410 0 6041 0.3863 0.1959 WGHTD AVE 1.1260 0.1270 0.0827 0 0326 0.9567 0 5473 0 3017 0 5473 0.3429 0.1698

St. Johns LOW 0 8450 0 0892 0.0582 0 0215 0.6848 0 3577 0.1866 0 3577 0.2124 0.1036 HIGH 1.4253 0.1904 0.1045 0 0579 1.3364 0 8967 0 5899 0 8967 0.6478 0.3990 WGHTD AVE 1.1690 0.1446 0.0849 0 0408 1.0400 0 6438 0 3961 0 6438 0.4389 0.2558

St. Lucie LOW 4.1105 0 2844 0.1576 0 0828 3.9190 3 2053 2.1085 3 2053 2.4512 1.0449 HIGH 5 2937 0.4826 0.2046 0.1596 5.3435 4 5490 3 2550 4 5490 3.6614 1.9493 WGHTD AVE 4 6219 0 3586 0.1786 0.1110 4.5100 3.7526 2 5583 3.7526 2.9324 1.3806

Santa Rosa LOW 1.4073 0.1735 0.1075 0 0501 1.2746 0 8128 0.4985 0 8128 0.5566 0.3098 HIGH 2.7953 0.4805 0.1961 0.1736 3.0333 2.4228 1 8709 2.4228 1.9909 1.4539 WGHTD AVE 2 2751 0 3596 0.1599 0.1189 2.3457 1.7839 1 3202 1.7839 1.4162 0.9918

Sarasota LOW 3 0411 0 2184 0.1288 0 0621 2.9103 2 3721 1 5865 2 3721 1.8293 0.8096 HIGH 3 9941 0 3215 0.1728 0 0977 3.9310 3 2779 2 2765 3 2779 2.5874 1.2594 WGHTD AVE 3 5624 0 2748 0.1540 0 0817 3.4738 2 8756 1 9710 2 8756 2.2515 1.0587

Seminole LOW 2 6629 0.1608 0.1049 0 0390 2.4279 1 8907 1.1774 1 8907 1.3957 0.5127 HIGH 3 0028 0 2006 0.1212 0 0534 2.8050 2 2344 1.4438 2 2344 1.6869 0.6900 WGHTD AVE 2 8358 0.1738 0.1128 0 0430 2.6040 2 0447 1 2851 2 0447 1.5182 0.5694

Sumter LOW 2 5392 0.1500 0.0973 0 0361 2.3138 1 8072 1.1277 1 8072 1.3364 0.4916 HIGH 3 0277 0.1871 0.1197 0 0478 2.8112 2 2424 1.4375 2 2424 1.6862 0.6601 WGHTD AVE 2.7878 0.1691 0.1092 0 0423 2.5677 2 0315 1 2885 2 0315 1.5175 0.5806

*Includes contents and A.L.E.

FPHLM V3.0 2008 295 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL -Owners -- MASONRY 0% 0% $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Suwanee LOW 0 6014 0 0592 0.0402 0 0135 0.4684 0 2246 0.1104 0 2246 0.1260 0.0597 HIGH 0 9839 0.1112 0.0711 0 0284 0.8319 0.4711 0 2587 0.4711 0.2937 0.1466 WGHTD AVE 0.7072 0 0732 0.0488 0 0174 0.5672 0 2903 0.1492 0 2903 0.1700 0.0821

Taylor LOW 0.7218 0 0772 0.0502 0 0189 0.5894 0 3125 0.1660 0 3125 0.1883 0.0939 HIGH 1 0996 0.1369 0.0805 0 0382 0.9748 0 5963 0 3621 0 5963 0.4016 0.2328 WGHTD AVE 0 8813 0.1022 0.0628 0 0274 0.7566 0.4425 0 2577 0.4425 0.2885 0.1579

Union LOW 0 6736 0 0668 0.0452 0 0151 0.5250 0 2517 0.1225 0 2517 0.1401 0.0653 HIGH 0 9535 0.1020 0.0668 0 0248 0.7807 0.4164 0 2177 0.4164 0.2486 0.1185 WGHTD AVE 0 9344 0 0997 0.0653 0 0241 0.7635 0.4052 0 2112 0.4052 0.2412 0.1148

Volusia LOW 2 0724 0.1199 0.0792 0 0280 1.8694 1.4417 0 8872 1.4417 1.0567 0.3773 HIGH 3 5267 0 2979 0.1453 0 0926 3.4874 2 9148 2 0510 2 9148 2.3192 1.1796 WGHTD AVE 2 8311 0.1856 0.1122 0 0486 2.6437 2.1142 1 3719 2.1142 1.6007 0.6575

Wakulla LOW 0.7826 0 0826 0.0535 0 0199 0.6331 0 3294 0.1720 0 3294 0.1955 0.0960 HIGH 1.4338 0.1907 0.1072 0 0570 1.3414 0 8961 0 5791 0 8961 0.6393 0.3833 WGHTD AVE 0 9261 0.1012 0.0636 0 0255 0.7729 0.4390 0 2493 0.4390 0.2801 0.1498

Walton LOW 1 2795 0.1529 0.0968 0 0419 1.1202 0 6714 0 3920 0 6714 0.4393 0.2378 HIGH 2 8183 0.4663 0.1975 0.1666 3.0121 2 3774 1 8096 2 3774 1.9343 1.3818 WGHTD AVE 1 8021 0 2517 0.1297 0 0758 1.7295 1 2031 0 8169 1 2031 0.8913 0.5688

Washington LOW 1.1307 0.1253 0.0816 0 0320 0.9505 0 5337 0 2910 0 5337 0.3309 0.1633 HIGH 1 8330 0 2569 0.1396 0 0809 1.7641 1 2198 0 8179 1 2198 0.8945 0.5636 WGHTD AVE 1.4157 0.1730 0.1058 0 0487 1.2658 0.7893 0.4738 0.7893 0.5307 0.2900

STATEWIDE LOW 0 5378 0 0517 0.0360 0 0115 0.4131 0.1914 0 0921 0.1914 0.1050 0.0498 HIGH 11.1642 1.7971 0.3275 0.7054 13 0137 12.1735 10.3662 12.1735 10.9509 8.2199 WGHTD AVE 4.1626 0 3625 0.1612 0.1155 4.1130 3.4621 2.4727 3.4621 2.7791 1.4793

*Includes contents and A.L.E.

FPHLM V3.0 2008 296 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - MOBILE HOMES $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Alachua LOW 3.4893 0.3201 0.0628 0.1119 3 2369 2 8128 1 8910 3 2369 2 8128 1 8910 HIGH 4.8586 0.5444 0.0813 0.2005 4 8072 4 2833 3.1079 4 8072 4 2833 3.1079 WGHTD AVE 4.0836 0.4082 0.0708 0.1462 3 9060 3.4340 2 3849 3 9060 3.4340 2 3849

Baker LOW 1.8391 0.1272 0.0367 0.0384 1 5540 1 2934 0.7501 1 5540 1 2934 0.7501 HIGH 3.2685 0.2974 0.0587 0.1028 3 0219 2 6208 1.7473 3 0219 2 6208 1.7473 WGHTD AVE 2.8742 0.2534 0.0529 0.0853 2 6271 2 2636 1.4779 2 6271 2 2636 1.4779

Bay LOW 5.2362 0.5717 0.0875 0.2111 5.1479 4 5763 3 2946 5.1479 4 5763 3 2946 HIGH 17.2773 3.9429 0.2089 1.5389 21.5975 20.6605 18.4348 21.5975 20.6605 18.4348 WGHTD AVE 8.6133 1.3542 0.1230 0.5192 9.4635 8.7520 7.1067 9.4635 8.7520 7.1067

Bradford LOW 3.3309 0.2876 0.0602 0.0981 3 0475 2 6312 1.7204 3 0475 2 6312 1.7204 HIGH 4.3992 0.4513 0.0753 0.1630 4 2444 3.7462 2 6346 4 2444 3.7462 2 6346 WGHTD AVE 3.4601 0.3054 0.0623 0.1055 3.1858 2.7586 1 8236 3.1858 2.7586 1 8236

Brevard LOW 9.0022 1.3042 0.1066 0.5172 9 8601 9 0540 7.1350 9 8601 9 0540 7.1350 HIGH 23.9266 5.1833 0.2286 2.0672 29.6967 28.3745 25.0296 29.6967 28.3745 25.0296 WGHTD AVE 13.2039 2.2450 0.1406 0.9090 15.2058 14.2012 11.7438 15.2058 14.2012 11.7438

Broward LOW 20.8229 4.2090 0.1977 1.7323 25.4043 24.1290 20.9442 25.4043 24.1290 20.9442 HIGH 32.5370 7.7568 0.2787 3.1837 42.0072 40.5287 36.7255 42.0072 40.5287 36.7255 WGHTD AVE 24.5582 5.2804 0.2234 2.1721 30.5920 29.2322 25.7927 30.5920 29.2322 25.7927

Calhoun LOW 4.0656 0.3831 0.0715 0.1363 3 8256 3 3436 2 2787 3 8256 3 3436 2 2787 HIGH 5.8498 0.7063 0.0947 0.2663 5 9423 5 3469 3 9921 5 9423 5 3469 3 9921 WGHTD AVE 4.6400 0.4686 0.0792 0.1705 4.4682 3 9455 2.7759 4.4682 3 9455 2.7759

Charlotte LOW 14.1266 2.2809 0.1498 0.9301 16.1866 15.1288 12.4919 16.1866 15.1288 12.4919 HIGH 25.8957 5.9680 0.2274 2.4021 32.9927 31.7288 28.4967 32.9927 31.7288 28.4967 WGHTD AVE 16.4876 2.8895 0.1689 1.1849 19.3442 18.2076 15.3598 19.3442 18.2076 15.3598

Citrus LOW 6.3724 0.7080 0.0821 0.2771 6 5483 5 8802 4 2979 6 5483 5 8802 4 2979 HIGH 9.8866 1.4434 0.1149 0.5806 10.9500 10.1147 8 0742 10.9500 10.1147 8 0742 WGHTD AVE 8.6227 1.1499 0.1038 0.4615 9 3109 8 5251 6 6251 9 3109 8 5251 6 6251

Clay LOW 3.1530 0.2664 0.0576 0.0899 2 8712 2.4692 1 5956 2 8712 2.4692 1 5956 HIGH 4.9517 0.5487 0.0830 0.2020 4 8889 4 3534 3.1515 4 8889 4 3534 3.1515 WGHTD AVE 3.6769 0.3496 0.0647 0.1234 3.4558 3 0181 2 0557 3.4558 3 0181 2 0557

Collier LOW 17.9400 3.0513 0.1857 1.2496 20.8667 19.5916 16.4103 20.8667 19.5916 16.4103 HIGH 31.9900 7.4602 0.2755 3.0191 40.9841 39.4831 35.6161 40.9841 39.4831 35.6161 WGHTD AVE 22.4562 4.4343 0.2141 1.8116 27.3047 25.9660 22.5690 27.3047 25.9660 22.5690

Columbia LOW 1.9718 0.1339 0.0395 0.0403 1 6663 1 3870 0 8014 1 6663 1 3870 0 8014 HIGH 3.5893 0.3469 0.0627 0.1221 3 3913 2 9668 2 0261 3 3913 2 9668 2 0261 WGHTD AVE 3.2525 0.2988 0.0587 0.1031 3 0217 2 6224 1.7469 3 0217 2 6224 1.7469

De Soto LOW 12.4117 1.7792 0.1439 0.7219 13.7162 12.6725 10.1194 13.7162 12.6725 10.1194 HIGH 15.4563 2.6339 0.1662 1.0725 17.9350 16.8283 14.0811 17.9350 16.8283 14.0811 WGHTD AVE 13.3102 2.0597 0.1499 0.8367 15.0121 13.9594 11.3747 15.0121 13.9594 11.3747

Dixie LOW 3.8562 0.4185 0.0655 0.1514 3.7659 3 3375 2 3803 3.7659 3 3375 2 3803 HIGH 7.6698 1.2308 0.1124 0.4721 8.4824 7 8473 6 3801 8.4824 7 8473 6 3801 WGHTD AVE 4.3809 0.5268 0.0723 0.1960 4.4123 3 9527 2 9187 4.4123 3 9527 2 9187

*Includes contents and A.L.E.

FPHLM V3.0 2008 297 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - MOBILE HOMES $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Duval LOW 1.7621 0.1183 0.0360 0.0352 1.4781 1 2257 0.7003 1.4781 1 2257 0.7003 HIGH 6.0189 1.0758 0.0864 0.4102 6 8182 6 3439 5 2725 6 8182 6 3439 5 2725 WGHTD AVE 3.1819 0.3102 0.0563 0.1083 2 9842 2 6047 1.7777 2 9842 2 6047 1.7777

Escambia LOW 7.5296 1.2235 0.1102 0.4789 8 3810 7.7729 6 3769 8 3810 7.7729 6 3769 HIGH 20.9183 5.1756 0.2385 2.0145 26.9483 25.9655 23.6067 26.9483 25.9655 23.6067 WGHTD AVE 12.0545 2.4112 0.1579 0.9334 14.3701 13.5831 11.7359 14.3701 13.5831 11.7359

Flagler LOW 8.0097 1.0771 0.0967 0.4304 8 6467 7 9094 6.1397 8 6467 7 9094 6.1397 HIGH 12.5353 2.2886 0.1317 0.9214 14.7705 13.8992 11.7523 14.7705 13.8992 11.7523 WGHTD AVE 10.0328 1.6832 0.1055 0.6754 11.4675 10.6693 8.7278 11.4675 10.6693 8.7278

Franklin LOW 8.4490 1.3899 0.1260 0.5392 9.4244 8.7471 7.1881 9.4244 8.7471 7.1881 HIGH 13.7937 3.0870 0.1759 1.2011 17.0621 16.2766 14.4457 17.0621 16.2766 14.4457 WGHTD AVE 11.3079 2.2887 0.1540 0.8963 13.5049 12.7664 11.0547 13.5049 12.7664 11.0547

Gadsen LOW 2.8224 0.2142 0.0530 0.0702 2.4934 2.1199 1 3135 2.4934 2.1199 1 3135 HIGH 4.4688 0.4477 0.0764 0.1625 4 2978 3.7894 2 6509 4 2978 3.7894 2 6509 WGHTD AVE 3.1087 0.2595 0.0568 0.0883 2 8219 2.4276 1 5661 2 8219 2.4276 1 5661

Gilchrist LOW 3.3646 0.3117 0.0596 0.1085 3.1409 2.7344 1 8380 3.1409 2.7344 1 8380 HIGH 5.0621 0.6116 0.0826 0.2279 5.1345 4 6170 3.4400 5.1345 4 6170 3.4400 WGHTD AVE 4.5257 0.5174 0.0753 0.1885 4 5034 4 0210 2 9329 4 5034 4 0210 2 9329

Glades LOW 14.3247 2.1039 0.1644 0.8545 15.9263 14.7406 11.8411 15.9263 14.7406 11.8411 HIGH 15.0953 2.2981 0.1708 0.9328 16.9467 15.7330 12.7539 16.9467 15.7330 12.7539 WGHTD AVE 15.0713 2.2919 0.1706 0.9304 16.9149 15.7021 12.7254 16.9149 15.7021 12.7254

Gulf LOW 5.8431 0.7533 0.0960 0.2843 6 0106 5.4344 4.1354 6 0106 5.4344 4.1354 HIGH 8.8832 1.4410 0.1332 0.5597 9 8648 9.1447 7.4899 9 8648 9.1447 7.4899 WGHTD AVE 6.9638 1.0016 0.1079 0.3809 7.4103 6.7818 5 3535 7.4103 6.7818 5 3535

Hamilton LOW 1.8481 0.1319 0.0367 0.0405 1 5743 1 3142 0.7689 1 5743 1 3142 0.7689 HIGH 2.9284 0.2656 0.0529 0.0914 2.7073 2 3468 1 5577 2.7073 2 3468 1 5577 WGHTD AVE 2.3712 0.1942 0.0448 0.0642 2.1169 1 8069 1.1416 2.1169 1 8069 1.1416

Hardee LOW 12.1542 1.7206 0.1417 0.6982 13.3855 12.3535 9 8337 13.3855 12.3535 9 8337 HIGH 14.6185 2.3969 0.1603 0.9772 16.7778 15.6898 12.9986 16.7778 15.6898 12.9986 WGHTD AVE 12.8611 1.8749 0.1482 0.7647 14.2842 13.2218 10.6201 14.2842 13.2218 10.6201

Hendry LOW 16.1157 2.5394 0.1728 1.0373 18.3319 17.0911 14.0191 18.3319 17.0911 14.0191 HIGH 19.3221 3.3926 0.1973 1.3891 22.6784 21.3495 18.0246 22.6784 21.3495 18.0246 WGHTD AVE 18.2076 3.0688 0.1894 1.2494 21.1053 19.7921 16.5187 21.1053 19.7921 16.5187

Hernando LOW 7.4898 0.8538 0.0952 0.3370 7.7615 6 9989 5.1855 7.7615 6 9989 5.1855 HIGH 10.4789 1.5958 0.1195 0.6435 11.7352 10.8783 8.7848 11.7352 10.8783 8.7848 WGHTD AVE 8.9395 1.2478 0.1055 0.5001 9.7629 8 9706 7 0507 9.7629 8 9706 7 0507

Highlands LOW 10.8109 1.3336 0.1328 0.5358 11.4818 10.4628 8 0017 11.4818 10.4628 8 0017 HIGH 14.6558 2.2028 0.1667 0.8964 16.4055 15.2191 12.3084 16.4055 15.2191 12.3084 WGHTD AVE 12.0152 1.6068 0.1441 0.6508 13.0349 11.9710 9 3844 13.0349 11.9710 9 3844

Hillsborough LOW 8.9868 1.2550 0.1062 0.5057 9 8300 9 0441 7.1405 9 8300 9 0441 7.1405 HIGH 17.9157 3.6970 0.1750 1.4990 21.9970 20.9594 18.3620 21.9970 20.9594 18.3620 WGHTD AVE 12.0657 1.9622 0.1336 0.7938 13.7797 12.8610 10.6023 13.7797 12.8610 10.6023

Holmes LOW 5.1279 0.5753 0.0852 0.2146 5.1088 4 5581 3 3228 5.1088 4 5581 3 3228 HIGH 7.2471 1.0553 0.1104 0.4066 7.7752 7.1250 5 6366 7.7752 7.1250 5 6366 WGHTD AVE 5.7153 0.7201 0.0921 0.2706 5 8583 5 2868 3 9953 5 8583 5 2868 3 9953

*Includes contents and A.L.E.

FPHLM V3.0 2008 298 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - MOBILE HOMES $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Indian River LOW 13.1099 2.0695 0.1473 0.8291 14.7953 13.7342 11.1389 14.7953 13.7342 11.1389 HIGH 23.3504 5.2962 0.2184 2.1087 29.4541 28.2248 25.1332 29.4541 28.2248 25.1332 WGHTD AVE 15.8031 2.8051 0.1673 1.1255 18.4592 17.3244 14.5206 18.4592 17.3244 14.5206

Jackson LOW 2.6931 0.1912 0.0516 0.0605 2 3317 1 9640 1.1761 2 3317 1 9640 1.1761 HIGH 5.4955 0.6545 0.0897 0.2468 5 5497 4 9853 3.7144 5 5497 4 9853 3.7144 WGHTD AVE 4.2683 0.4152 0.0747 0.1508 4 0645 3 5702 2.4720 4 0645 3 5702 2.4720

Jefferson LOW 2.4823 0.2228 0.0456 0.0755 2 2702 1 9570 1 2782 2 2702 1 9570 1 2782 HIGH 3.1736 0.3496 0.0559 0.1258 3 0756 2.7215 1 9398 3 0756 2.7215 1 9398 WGHTD AVE 2.5852 0.2375 0.0469 0.0815 2 3818 2 0614 1 3651 2 3818 2 0614 1 3651

Lafayette LOW 2.7609 0.2618 0.0498 0.0907 2 5689 2 2316 1.4978 2 5689 2 2316 1.4978 HIGH 3.4400 0.3508 0.0598 0.1247 3 2906 2 8921 2 0109 3 2906 2 8921 2 0109 WGHTD AVE 3.4209 0.3470 0.0596 0.1236 3 2672 2 8707 1 9942 3 2672 2 8707 1 9942

Lake LOW 5.5881 0.5125 0.0755 0.1946 5 5732 4 9204 3 3547 5 5732 4 9204 3 3547 HIGH 12.9722 2.0765 0.1441 0.8414 14.7548 13.7517 11.2831 14.7548 13.7517 11.2831 WGHTD AVE 9.2545 1.1920 0.1126 0.4765 9 9042 9 0397 6 9545 9 9042 9 0397 6 9545

Lee LOW 15.9529 2.6773 0.1662 1.0916 18.4774 17.3226 14.4337 18.4774 17.3226 14.4337 HIGH 26.7834 5.9981 0.2381 2.4321 33.8690 32.5309 29.1001 33.8690 32.5309 29.1001 WGHTD AVE 18.4275 3.3054 0.1843 1.3540 21.7642 20.5336 17.4363 21.7642 20.5336 17.4363

Leon LOW 1.9680 0.1373 0.0389 0.0419 1 6740 1 3972 0 8150 1 6740 1 3972 0 8150 HIGH 4.0037 0.4221 0.0682 0.1525 3 8892 3.4404 2.4355 3 8892 3.4404 2.4355 WGHTD AVE 3.3419 0.3207 0.0588 0.1126 3.1439 2.7463 1 8692 3.1439 2.7463 1 8692

Levy LOW 3.4000 0.3180 0.0603 0.1113 3.1851 2.7767 1 8745 3.1851 2.7767 1 8745 HIGH 6.6547 0.9356 0.1030 0.3557 7 0618 6.4484 5 0433 7 0618 6.4484 5 0433 WGHTD AVE 5.2036 0.6378 0.0843 0.2399 5 2882 4.7586 3 5604 5 2882 4.7586 3 5604

Liberty LOW 3.8841 0.3463 0.0690 0.1214 3 6111 3.1394 2 0926 3 6111 3.1394 2 0926 HIGH 4.1690 0.4102 0.0721 0.1478 3 9656 3.4794 2 3980 3 9656 3.4794 2 3980 WGHTD AVE 3.9643 0.3693 0.0697 0.1309 3.7281 3 2559 2 2054 3.7281 3 2559 2 2054

Madison LOW 2.0421 0.1591 0.0390 0.0511 1.7859 1 5101 0 9261 1.7859 1 5101 0 9261 HIGH 3.0121 0.2964 0.0533 0.1037 2 8350 2.4739 1 6829 2 8350 2.4739 1 6829 WGHTD AVE 2.6125 0.2355 0.0471 0.0801 2 3914 2 0651 1 3580 2 3914 2 0651 1 3580

Manatee LOW 12.8499 2.1785 0.1339 0.8886 14.9166 13.9940 11.6955 14.9166 13.9940 11.6955 HIGH 22.2970 5.0201 0.2002 2.0201 28.1584 27.0105 24.1040 28.1584 27.0105 24.1040 WGHTD AVE 14.2307 2.5726 0.1439 1.0481 16.8192 15.8598 13.4637 16.8192 15.8598 13.4637

Marion LOW 4.1450 0.3181 0.0604 0.1146 3 8831 3 3451 2.1263 3 8831 3 3451 2.1263 HIGH 9.6459 1.3663 0.1130 0.5497 10.5925 9.7576 7.7274 10.5925 9.7576 7.7274 WGHTD AVE 7.1801 0.8215 0.0913 0.3226 7.4369 6.7015 4 9526 7.4369 6.7015 4 9526

Martin LOW 16.2410 2.6276 0.1718 1.0893 18.5737 17.3185 14.2215 18.5737 17.3185 14.2215 HIGH 28.4921 6.3549 0.2545 2.6190 35.9627 34.4941 30.7609 35.9627 34.4941 30.7609 WGHTD AVE 20.8032 3.7102 0.1981 1.5697 24.5058 23.1610 19.7978 24.5058 23.1610 19.7978

Miami-Dade LOW 19.7340 3.9745 0.1869 1.6387 24.0547 22.8417 19.8116 24.0547 22.8417 19.8116 HIGH 40.0738 10.2301 0.3275 4.1885 52.9989 51.4323 47.3540 52.9989 51.4323 47.3540 WGHTD AVE 26.3924 5.9109 0.2349 2.4148 33.3572 31.9951 28.5196 33.3572 31.9951 28.5196

Monroe LOW 22.8512 4.7854 0.2059 2.0888 28.5119 27.3154 24.2544 28.5119 27.3154 24.2544 HIGH 38.1638 9.5743 0.3086 4.0939 50.4018 48.9356 45.1247 50.4018 48.9356 45.1247 WGHTD AVE 34.3098 8.5846 0.2715 3.6103 45.1361 43.7545 40.1765 45.1361 43.7545 40.1765

Nassau LOW 2.1029 0.1790 0.0400 0.0591 1 8799 1 6046 1 0188 1 8799 1 6046 1 0188 HIGH 3.4767 0.4283 0.0586 0.1562 3.4811 3.1097 2 2909 3.4811 3.1097 2 2909 WGHTD AVE 2.5478 0.2397 0.0466 0.0818 2 3522 2 0353 1 3527 2 3522 2 0353 1 3527

*Includes contents and A.L.E.

FPHLM V3.0 2008 299 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - MOBILE HOMES $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Okaloosa LOW 7.3593 1.1387 0.1097 0.4429 8 0555 7.4317 6 0043 8 0555 7.4317 6 0043 HIGH 15.7314 3.3964 0.1963 1.3300 19.2973 18.3750 16.1884 19.2973 18.3750 16.1884 WGHTD AVE 10.9447 2.1050 0.1449 0.8026 12.8350 12.0795 10.3170 12.8350 12.0795 10.3170

Okeechobee LOW 13.0692 1.8467 0.1529 0.7448 14.3422 13.2073 10.4545 14.3422 13.2073 10.4545 HIGH 14.5669 2.2165 0.1652 0.8958 16.3252 15.1388 12.2322 16.3252 15.1388 12.2322 WGHTD AVE 14.1161 2.0615 0.1625 0.8328 15.6505 14.4656 11.5720 15.6505 14.4656 11.5720

Orange LOW 7.7020 0.8271 0.0993 0.3250 7 8802 7 0725 5.1500 7 8802 7 0725 5.1500 HIGH 12.5483 1.9502 0.1413 0.7870 14.1381 13.1322 10.6681 14.1381 13.1322 10.6681 WGHTD AVE 9.8784 1.2983 0.1177 0.5233 10.6327 9.7240 7 5273 10.6327 9.7240 7 5273

Osceola LOW 8.9905 1.0486 0.1129 0.4190 9.4092 8 5262 6.4055 9.4092 8 5262 6.4055 HIGH 11.8525 1.6892 0.1386 0.6797 13.0381 12.0153 9 5246 13.0381 12.0153 9 5246 WGHTD AVE 10.6331 1.3995 0.1284 0.5640 11.4722 10.5078 8.1716 11.4722 10.5078 8.1716

Palm Beach LOW 17.1209 2.8696 0.1772 1.1871 19.7717 18.4887 15.3132 19.7717 18.4887 15.3132 HIGH 32.8068 7.4990 0.2896 3.0877 41.8035 40.2129 36.1215 41.8035 40.2129 36.1215 WGHTD AVE 20.8912 4.0657 0.2054 1.6586 25.2005 23.8483 20.4615 25.2005 23.8483 20.4615

Pasco LOW 8.4051 0.9767 0.1024 0.3892 8.7872 7 9575 5 9630 8.7872 7 9575 5 9630 HIGH 11.5147 1.9545 0.1303 0.7887 13.2337 12.3851 10.3014 13.2337 12.3851 10.3014 WGHTD AVE 9.8960 1.4261 0.1160 0.5721 10.9183 10.0732 8 0175 10.9183 10.0732 8 0175

Pinellas LOW 10.3584 1.6123 0.1131 0.6525 11.7062 10.8801 8 8478 11.7062 10.8801 8 8478 HIGH 24.6352 6.0204 0.2147 2.3987 31.8031 30.6902 27.8635 31.8031 30.6902 27.8635 WGHTD AVE 12.0728 2.0920 0.1257 0.8467 14.0586 13.1875 11.0309 14.0586 13.1875 11.0309

Polk LOW 9.4758 1.1197 0.1174 0.4482 10.0020 9 0790 6 8371 10.0020 9 0790 6 8371 HIGH 15.8206 2.7639 0.1686 1.1221 18.4596 17.3389 14.5654 18.4596 17.3389 14.5654 WGHTD AVE 11.5345 1.6289 0.1342 0.6589 12.6849 11.6940 9 2761 12.6849 11.6940 9 2761

Putnam LOW 3.9239 0.3659 0.0691 0.1288 3 6824 3 2145 2.1808 3 6824 3 2145 2.1808 HIGH 5.4544 0.6687 0.0888 0.2507 5 5499 4 9960 3.7440 5 5499 4 9960 3.7440 WGHTD AVE 5.0138 0.5772 0.0834 0.2139 5 0012 4.4699 3 2770 5 0012 4.4699 3 2770

St. Johns LOW 3.2292 0.2938 0.0582 0.1011 2 9806 2 5822 1.7153 2 9806 2 5822 1.7153 HIGH 7.1208 1.1543 0.1045 0.4416 7 8722 7 2773 5 9117 7 8722 7 2773 5 9117 WGHTD AVE 4.8983 0.6464 0.0787 0.2391 5 0433 4 5532 3.4562 5 0433 4 5532 3.4562

St. Lucie LOW 14.5469 2.2291 0.1576 0.9220 16.3868 15.2065 12.3028 16.3868 15.2065 12.3028 HIGH 21.0830 4.0607 0.2046 1.6783 25.3893 24.0421 20.6616 25.3893 24.0421 20.6616 WGHTD AVE 19.6638 3.6519 0.1933 1.5223 23.4243 22.1192 18.8543 23.4243 22.1192 18.8543

Santa Rosa LOW 7.1279 1.0726 0.1075 0.4162 7.7285 7.1075 5 6894 7.7285 7.1075 5 6894 HIGH 15.8590 3.4755 0.1961 1.3610 19.5683 18.6624 16.5074 19.5683 18.6624 16.5074 WGHTD AVE 10.6131 2.0247 0.1440 0.7909 12.4336 11.6855 9 9422 12.4336 11.6855 9 9422

Sarasota LOW 12.2634 2.0492 0.1288 0.8335 14.1638 13.2626 11.0232 14.1638 13.2626 11.0232 HIGH 17.5159 3.2816 0.1728 1.3385 20.9599 19.8427 17.0215 20.9599 19.8427 17.0215 WGHTD AVE 15.7782 2.8697 0.1586 1.1708 18.7173 17.6717 15.0396 18.7173 17.6717 15.0396

Seminole LOW 8.3302 0.9722 0.1049 0.3849 8 6862 7 8509 5 8571 8 6862 7 8509 5 8571 HIGH 10.3667 1.4838 0.1212 0.5936 11.3897 10.4845 8 2928 11.3897 10.4845 8 2928 WGHTD AVE 9.1647 1.1355 0.1119 0.4540 9.7153 8 8381 6.7265 9.7153 8 8381 6.7265

Sumter LOW 7.6794 0.8810 0.0973 0.3469 7 9629 7.1782 5 3104 7 9629 7.1782 5 3104 HIGH 10.0531 1.3568 0.1197 0.5454 10.9055 10.0037 7 8143 10.9055 10.0037 7 8143 WGHTD AVE 9.5322 1.2722 0.1139 0.5098 10.2977 9.4263 7 3169 10.2977 9.4263 7 3169

*Includes contents and A.L.E.

FPHLM V3.0 2008 300 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - MOBILE HOMES $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Suwanee LOW 2.0690 0.1557 0.0402 0.0493 1.7930 1 5103 0 9146 1.7930 1 5103 0 9146 HIGH 4.2187 0.4609 0.0711 0.1680 4.1416 3 6786 2 6387 4.1416 3 6786 2 6387 WGHTD AVE 2.6244 0.2306 0.0483 0.0786 2 3916 2 0602 1 3447 2 3916 2 0602 1 3447

Taylor LOW 2.8196 0.2772 0.0502 0.0969 2 6496 2 3109 1 5716 2 6496 2 3109 1 5716 HIGH 5.0648 0.6712 0.0805 0.2508 5 2500 4.7537 3 6283 5 2500 4.7537 3 6283 WGHTD AVE 3.9303 0.4794 0.0648 0.1768 3 9553 3 5461 2 6310 3 9553 3 5461 2 6310

Union LOW 2.3177 0.1656 0.0452 0.0516 1 9951 1 6752 0 9965 1 9951 1 6752 0 9965 HIGH 3.8091 0.3644 0.0668 0.1287 3 5866 3.1342 2.1390 3 5866 3.1342 2.1390 WGHTD AVE 3.6015 0.3372 0.0640 0.1180 3 3667 2 9324 1 9802 3 3667 2 9324 1 9802

Volusia LOW 6.0261 0.6380 0.0792 0.2485 6.1119 5.4590 3 9260 6.1119 5.4590 3 9260 HIGH 14.0758 2.6636 0.1453 1.0719 16.7761 15.8407 13.5230 16.7761 15.8407 13.5230 WGHTD AVE 9.4072 1.3534 0.1096 0.5364 10.3466 9 5223 7 5272 10.3466 9 5223 7 5272

Wakulla LOW 2.9751 0.2730 0.0535 0.0935 2.7436 2 3746 1 5723 2.7436 2 3746 1 5723 HIGH 7.2100 1.1180 0.1072 0.4288 7 9016 7 2911 5 8770 7 9016 7 2911 5 8770 WGHTD AVE 3.2745 0.3265 0.0568 0.1140 3 0921 2.7044 1 8555 3 0921 2.7044 1 8555

Walton LOW 6.1197 0.8160 0.0968 0.3114 6 3754 5.7858 4.4533 6 3754 5.7858 4.4533 HIGH 15.7296 3.3690 0.1975 1.3196 19.2433 18.3121 16.1056 19.2433 18.3121 16.1056 WGHTD AVE 7.5852 1.1583 0.1097 0.4535 8 2437 7 5927 6.1028 8 2437 7 5927 6.1028

Washington LOW 4.8510 0.5335 0.0816 0.1967 4.7660 4 2352 3 0514 4.7660 4 2352 3 0514 HIGH 9.8511 1.6849 0.1396 0.6559 11.1525 10.3903 8 6130 11.1525 10.3903 8 6130 WGHTD AVE 6.7910 0.9464 0.1056 0.3640 7.1905 6 5583 5.1140 7.1905 6 5583 5.1140

STATEWIDE LOW 1.7621 0.1183 0.0360 0.0352 1.4781 1 2257 0.7003 1.4781 1 2257 0.7003 HIGH 40.0738 10.2301 0.3275 4.1885 52.9989 51.4323 47.3540 52.9989 51.4323 47.3540 WGHTD AVE 11.6363 1.9297 0.1236 0.7884 13.3171 12.4172 10.2201 13.3171 12.4172 10.2201

*Includes contents and A.L.E.

FPHLM V3.0 2008 301 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - Renters -- FRAME $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Alachua LOW 0.1893 0 0454 0.0593 0.0543 0.0514 0.0633 0 0593 0 0533 HIGH 0.2544 0 0652 0.0952 0.0868 0.0807 0.1017 0 0952 0 0849 WGHTD AVE 0.2158 0 0532 0.0729 0.0666 0.0626 0.0778 0 0729 0 0653

Baker LOW 0.1084 0 0243 0.0293 0.0272 0.0263 0.0311 0 0293 0 0268 HIGH 0.1822 0 0441 0.0603 0.0555 0.0524 0.0642 0 0603 0 0544 WGHTD AVE 0.1755 0 0430 0.0574 0.0528 0.0499 0.0611 0 0574 0 0518

Bay LOW 0.2763 0 0713 0.1045 0.0955 0.0888 0.1114 0.1045 0 0934 HIGH 1.4189 0 5171 1.4292 1.3434 1.2024 1.4828 1.4292 1 3114 WGHTD AVE 0.5808 0.1824 0.3982 0.3676 0.3296 0.4190 0 3982 0 3579

Bradford LOW 0.1847 0 0437 0.0557 0.0510 0.0486 0.0594 0 0557 0 0501 HIGH 0.2349 0 0587 0.0825 0.0754 0.0706 0.0881 0 0825 0 0739 WGHTD AVE 0.1896 0 0447 0.0581 0.0533 0.0506 0.0620 0 0581 0 0523

Brevard LOW 0.4143 0.1170 0.2361 0.2180 0.1966 0.2487 0 2361 0 2125 HIGH 1.6928 0 5796 1.7223 1.6249 1.4548 1.7822 1.7223 1 5873 WGHTD AVE 0.5939 0.1746 0.3966 0.3675 0.3294 0.4162 0 3966 0 3580

Broward LOW 1.6582 0 5814 1.6702 1.5724 1.4047 1.7306 1 6702 1 5351 HIGH 3.3546 1 2130 3.7890 3.5911 3.2189 3.9072 3.7890 3 5114 WGHTD AVE 2.1266 0.7594 2.2447 2.1190 1.8952 2.3215 2 2447 2 0699

Calhoun LOW 0.2185 0 0536 0.0714 0.0653 0.0616 0.0762 0 0714 0 0640 HIGH 0.3038 0 0815 0.1268 0.1157 0.1067 0.1351 0.1268 0.1130 WGHTD AVE 0.2539 0 0634 0.0892 0.0814 0.0762 0.0953 0 0892 0 0797

Charlotte LOW 0.5810 0.1711 0.3527 0.3241 0.2909 0.3726 0 3527 0 3154 HIGH 1.5938 0 5511 1.5719 1.4740 1.3126 1.6331 1 5719 1.4372 WGHTD AVE 0.7975 0 2373 0.5874 0.5439 0.4854 0.6163 0 5874 0 5295

Citrus LOW 0.2745 0 0688 0.1076 0.0986 0.0912 0.1143 0.1076 0 0964 HIGH 0.4331 0.1226 0.2418 0.2226 0.2009 0.2554 0 2418 0 2169 WGHTD AVE 0.3686 0 0994 0.1816 0.1668 0.1517 0.1923 0.1816 0.1627

Clay LOW 0.1745 0 0412 0.0528 0.0485 0.0461 0.0563 0 0528 0 0476 HIGH 0.2614 0 0671 0.0992 0.0906 0.0842 0.1058 0 0992 0 0886 WGHTD AVE 0.2001 0 0492 0.0685 0.0628 0.0589 0.0730 0 0685 0 0615

Collier LOW 0.7114 0 2127 0.4366 0.3998 0.3574 0.4622 0.4366 0 3885 HIGH 1.7059 0 5954 1.6158 1.5040 1.3307 1.6869 1 6158 1.4632 WGHTD AVE 0.9454 0 3054 0.7111 0.6561 0.5829 0.7479 0.7111 0 6377

Columbia LOW 0.1157 0 0259 0.0305 0.0283 0.0274 0.0324 0 0305 0 0279 HIGH 0.1984 0 0491 0.0693 0.0636 0.0596 0.0737 0 0693 0 0623 WGHTD AVE 0.1796 0 0437 0.0589 0.0541 0.0510 0.0628 0 0589 0 0531

De Soto LOW 0.5124 0.1421 0.2627 0.2404 0.2176 0.2786 0 2627 0 2341 HIGH 0.6743 0 2041 0.4394 0.4043 0.3616 0.4634 0.4394 0 3933 WGHTD AVE 0.5682 0.1642 0.3290 0.3022 0.2719 0.3477 0 3290 0 2942

Dixie LOW 0.2122 0 0548 0.0843 0.0774 0.0718 0.0896 0 0843 0 0757 HIGH 0.4495 0.1386 0.2941 0.2722 0.2454 0.3091 0 2941 0 2654 WGHTD AVE 0.2731 0 0774 0.1320 0.1213 0.1110 0.1397 0.1320 0.1184

*Includes contents and A.L.E.

FPHLM V3.0 2008 302 Form A-6A, Output Ranges LOSS COSTS PER $1,000 (2002 FHCF exposure data) PERSONAL RESIDENTIAL - Renters -- FRAME Rev. 11-1-07 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Duval LOW 0.1040 0 0232 0.0273 0.0254 0.0245 0.0290 0 0273 0 0250 HIGH 0.4036 0.1311 0.3127 0.2919 0.2624 0.3264 0 3127 0 2849 WGHTD AVE 0.2044 0 0519 0.0789 0.0725 0.0675 0.0837 0 0789 0 0710

Escambia LOW 0.3816 0.1160 0.2186 0.1997 0.1796 0.2319 0 2186 0.1941 HIGH 2.0046 0.7474 2.1820 2.0604 1.8471 2.2563 2.1820 2 0132 WGHTD AVE 0.8620 0 3005 0.7483 0.6966 0.6211 0.7818 0.7483 0 6785

Flagler LOW 0.3414 0 0921 0.1656 0.1519 0.1380 0.1755 0.1656 0.1480 HIGH 0.6261 0.1973 0.4735 0.4395 0.3921 0.4959 0.4735 0.4279 WGHTD AVE 0.4429 0.1279 0.2771 0.2559 0.2296 0.2916 0 2771 0 2492

Franklin LOW 0.5018 0.1581 0.3359 0.3107 0.2797 0.3532 0 3359 0 3028 HIGH 1.1751 0.4251 1.1722 1.1026 0.9880 1.2157 1.1722 1 0766 WGHTD AVE 0.6627 0 2005 0.5246 0.4891 0.4392 0.5481 0 5246 0.4770

Gadsen LOW 0.1589 0 0368 0.0446 0.0410 0.0394 0.0475 0 0446 0 0404 HIGH 0.2372 0 0595 0.0822 0.0751 0.0705 0.0878 0 0822 0 0736 WGHTD AVE 0.1736 0 0415 0.0518 0.0475 0.0453 0.0552 0 0518 0 0467

Gilchrist LOW 0.1860 0 0453 0.0618 0.0567 0.0534 0.0658 0 0618 0 0556 HIGH 0.2688 0 0718 0.1149 0.1050 0.0967 0.1222 0.1149 0.1026 WGHTD AVE 0.2550 0 0656 0.1054 0.0964 0.0890 0.1121 0.1054 0 0942

Glades LOW 0.5830 0.1627 0.2991 0.2732 0.2468 0.3177 0 2991 0 2658 HIGH 0.6264 0.1786 0.3430 0.3139 0.2827 0.3635 0 3430 0 3054 WGHTD AVE 0.6243 0.1786 0.3414 0.3124 0.2814 0.3618 0 3414 0 3040

Gulf LOW 0.3186 0 0883 0.1478 0.1353 0.1238 0.1570 0.1478 0.1320 HIGH 0.5111 0.1590 0.3273 0.3019 0.2719 0.3449 0 3273 0 2941 WGHTD AVE 0.4221 0.1235 0.2430 0.2237 0.2024 0.2566 0 2430 0 2180

Hamilton LOW 0.1078 0 0242 0.0287 0.0266 0.0257 0.0305 0 0287 0 0262 HIGH 0.1634 0 0396 0.0536 0.0493 0.0465 0.0571 0 0536 0 0483 WGHTD AVE 0.1462 0 0363 0.0463 0.0425 0.0403 0.0492 0 0463 0 0418

Hardee LOW 0.4924 0.1351 0.2416 0.2205 0.1999 0.2568 0 2416 0 2147 HIGH 0.6172 0.1825 0.3746 0.3437 0.3081 0.3960 0 3746 0 3343 WGHTD AVE 0.5219 0.1457 0.2720 0.2487 0.2246 0.2887 0 2720 0 2420

Hendry LOW 0.6370 0.1840 0.3590 0.3285 0.2951 0.3805 0 3590 0 3194 HIGH 0.8270 0 2536 0.5578 0.5139 0.4587 0.5877 0 5578 0.4998 WGHTD AVE 0.7352 0 2158 0.4573 0.4198 0.3756 0.4831 0.4573 0.4083

Hernando LOW 0.3152 0 0795 0.1253 0.1146 0.1057 0.1333 0.1253 0.1120 HIGH 0.4523 0.1305 0.2691 0.2479 0.2226 0.2838 0 2691 0 2413 WGHTD AVE 0.3906 0.1070 0.2038 0.1872 0.1693 0.2156 0 2038 0.1824

Highlands LOW 0.4381 0.1133 0.1822 0.1661 0.1524 0.1942 0.1822 0.1620 HIGH 0.5943 0.1674 0.3117 0.2846 0.2567 0.3310 0 3117 0 2769 WGHTD AVE 0.4940 0.1335 0.2322 0.2120 0.1928 0.2470 0 2322 0 2065

Hillsborough LOW 0.3736 0.1023 0.1850 0.1693 0.1534 0.1964 0.1850 0.1649 HIGH 0.8974 0 2980 0.7492 0.6947 0.6162 0.7847 0.7492 0 6757 WGHTD AVE 0.4976 0.1477 0.3071 0.2824 0.2530 0.3242 0 3071 0 2747

Holmes LOW 0.2635 0 0686 0.1007 0.0918 0.0853 0.1075 0.1007 0 0898 HIGH 0.3819 0.1114 0.2021 0.1852 0.1679 0.2143 0 2021 0.1804 WGHTD AVE 0.2916 0 0752 0.1233 0.1126 0.1037 0.1313 0.1233 0.1099

*Includes contents and A.L.E.

FPHLM V3.0 2008 303 Form A-6A, Output Ranges LOSS COSTS PER $1,000 (2002 FHCF exposure data) PERSONAL RESIDENTIAL - Renters -- FRAME Rev. 11-1-07 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Indian River LOW 0.6004 0.1766 0.3730 0.3442 0.3088 0.3928 0 3730 0 3352 HIGH 1.7766 0 6085 1.8205 1.7188 1.5396 1.8827 1 8205 1 6794 WGHTD AVE 0.8473 0 2587 0.6510 0.6057 0.5409 0.6807 0 6510 0 5902

Jackson LOW 0.1540 0 0352 0.0416 0.0384 0.0371 0.0443 0 0416 0 0379 HIGH 0.2828 0 0750 0.1146 0.1045 0.0966 0.1222 0.1146 0.1021 WGHTD AVE 0.2198 0 0544 0.0740 0.0677 0.0637 0.0790 0 0740 0 0664

Jefferson LOW 0.1410 0 0340 0.0463 0.0426 0.0402 0.0492 0 0463 0 0419 HIGH 0.1787 0 0467 0.0754 0.0695 0.0642 0.0798 0 0754 0 0680 WGHTD AVE 0.1438 0 0346 0.0481 0.0443 0.0417 0.0511 0 0481 0 0435

Lafayette LOW 0.1550 0 0380 0.0533 0.0490 0.0460 0.0567 0 0533 0 0481 HIGH 0.1914 0 0484 0.0722 0.0663 0.0618 0.0766 0 0722 0 0649 WGHTD AVE 0.1914 0 0484 0.0722 0.0663 0.0618 0.0766 0 0722 0 0649

Lake LOW 0.2430 0 0573 0.0767 0.0704 0.0665 0.0818 0 0767 0 0690 HIGH 0.5581 0.1638 0.3345 0.3072 0.2756 0.3535 0 3345 0 2989 WGHTD AVE 0.4108 0.1129 0.1937 0.1772 0.1612 0.2056 0.1937 0.1727

Lee LOW 0.6630 0.1964 0.3960 0.3626 0.3248 0.4193 0 3960 0 3526 HIGH 1.4280 0.4896 1.3229 1.2332 1.0949 1.3801 1 3229 1 2006 WGHTD AVE 0.8150 0 2474 0.5745 0.5299 0.4725 0.6047 0 5745 0 5154

Leon LOW 0.1150 0 0258 0.0307 0.0284 0.0274 0.0326 0 0307 0 0280 HIGH 0.2171 0 0552 0.0814 0.0745 0.0694 0.0866 0 0814 0 0729 WGHTD AVE 0.1807 0 0441 0.0605 0.0555 0.0522 0.0645 0 0605 0 0544

Levy LOW 0.1862 0 0453 0.0609 0.0558 0.0526 0.0650 0 0609 0 0547 HIGH 0.3616 0.1038 0.1907 0.1752 0.1591 0.2018 0.1907 0.1708 WGHTD AVE 0.2729 0 0729 0.1178 0.1078 0.0991 0.1252 0.1178 0.1052

Liberty LOW 0.2108 0 0510 0.0659 0.0603 0.0572 0.0703 0 0659 0 0592 HIGH 0.2243 0 0559 0.0801 0.0734 0.0686 0.0853 0 0801 0 0719 WGHTD AVE 0.2137 0 0526 0.0700 0.0641 0.0605 0.0746 0 0700 0 0628

Madison LOW 0.1169 0 0268 0.0328 0.0302 0.0290 0.0348 0 0328 0 0297 HIGH 0.1672 0 0413 0.0582 0.0534 0.0500 0.0619 0 0582 0 0523 WGHTD AVE 0.1544 0 0368 0.0504 0.0462 0.0436 0.0536 0 0504 0 0454

Manatee LOW 0.5298 0.1592 0.3365 0.3092 0.2765 0.3553 0 3365 0 3006 HIGH 1.3450 0.4606 1.2995 1.2183 1.0859 1.3504 1 2995 1.1881 WGHTD AVE 0.6684 0 2048 0.4904 0.4535 0.4041 0.5151 0.4904 0.4412

Marion LOW 0.1871 0 0416 0.0525 0.0484 0.0463 0.0558 0 0525 0 0476 HIGH 0.4117 0.1143 0.2152 0.1975 0.1788 0.2278 0 2152 0.1924 WGHTD AVE 0.3139 0 0792 0.1265 0.1158 0.1068 0.1344 0.1265 0.1132

Martin LOW 0.9335 0 3057 0.7377 0.6855 0.6120 0.7720 0.7377 0 6677 HIGH 2.5695 0 9213 2.7726 2.6187 2.3422 2.8660 2.7726 2 5583 WGHTD AVE 1.3447 0.4731 1.2483 1.1695 1.0443 1.2983 1 2483 1.1407

Miami-Dade LOW 1.5006 0 5273 1.4881 1.3977 1.2459 1.5442 1.4881 1 3636 HIGH 4.8376 1.7530 5.7157 5.4433 4.9006 5.8748 5.7157 5 3301 WGHTD AVE 2.5696 0 9183 2.7962 2.6411 2.3608 2.8901 2.7962 2 5799

Monroe LOW 1.5257 0 5757 1.5418 1.4448 1.2885 1.6029 1 5418 1.4088 HIGH 4.4110 1 6907 5.3132 5.0643 4.5728 5.4591 5 3132 4 9613 WGHTD AVE 3.2999 1 0929 3.7929 3.6000 3.2392 3.9083 3.7929 3 5225

Nassau LOW 0.1235 0 0295 0.0406 0.0375 0.0357 0.0432 0 0406 0 0369 HIGH 0.2077 0 0572 0.1058 0.0982 0.0900 0.1113 0.1058 0 0960 WGHTD AVE 0.1789 0 0460 0.0759 0.0700 0.0647 0.0802 0 0759 0 0685

*Includes contents and A.L.E.

FPHLM V3.0 2008 304 Form A-6A, Output Ranges LOSS COSTS PER $1,000 (2002 FHCF exposure data) PERSONAL RESIDENTIAL - Renters -- FRAME Rev. 11-1-07 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Okaloosa LOW 0.3797 0.1134 0.2103 0.1924 0.1736 0.2231 0 2103 0.1872 HIGH 1.0561 0 3781 0.9551 0.8892 0.7912 0.9976 0 9551 0 8658 WGHTD AVE 0.8647 0 3003 0.7353 0.6829 0.6082 0.7695 0.7353 0 6648

Okeechobee LOW 0.5396 0.1480 0.2670 0.2441 0.2212 0.2835 0 2670 0 2377 HIGH 0.6182 0.1766 0.3449 0.3163 0.2848 0.3650 0 3449 0 3079 WGHTD AVE 0.5942 0.1664 0.3152 0.2886 0.2604 0.3341 0 3152 0 2809

Orange LOW 0.3213 0 0790 0.1170 0.1068 0.0993 0.1247 0.1170 0.1045 HIGH 0.5567 0.1622 0.3347 0.3083 0.2771 0.3530 0 3347 0 3002 WGHTD AVE 0.3956 0.1036 0.1731 0.1582 0.1446 0.1841 0.1731 0.1542

Osceola LOW 0.3645 0 0920 0.1407 0.1281 0.1183 0.1502 0.1407 0.1251 HIGH 0.5028 0.1397 0.2615 0.2398 0.2168 0.2770 0 2615 0 2335 WGHTD AVE 0.3924 0.1006 0.1634 0.1490 0.1367 0.1741 0.1634 0.1454

Palm Beach LOW 1.0678 0 3545 0.9017 0.8421 0.7529 0.9404 0 9017 0 8211 HIGH 3.2448 1.1629 3.5998 3.4116 3.0626 3.7128 3 5998 3 3366 WGHTD AVE 1.7384 0 5720 1.7150 1.6155 1.4459 1.7767 1.7150 1 5778

Pasco LOW 0.3449 0 0872 0.1356 0.1237 0.1142 0.1446 0.1356 0.1208 HIGH 0.5265 0.1600 0.3542 0.3270 0.2923 0.3726 0 3542 0 3182 WGHTD AVE 0.4415 0.1257 0.2528 0.2325 0.2092 0.2671 0 2528 0 2264

Pinellas LOW 0.4347 0.1258 0.2561 0.2357 0.2119 0.2704 0 2561 0 2295 HIGH 1.7709 0 6224 1.8666 1.7577 1.5666 1.9331 1 8666 1.7154 WGHTD AVE 0.6589 0.1919 0.5000 0.4641 0.4139 0.5236 0 5000 0.4519

Polk LOW 0.3833 0 0977 0.1500 0.1366 0.1261 0.1602 0.1500 0.1334 HIGH 0.7055 0 2163 0.4760 0.4385 0.3914 0.5015 0.4760 0.4265 WGHTD AVE 0.4669 0.1281 0.2299 0.2101 0.1905 0.2441 0 2299 0 2046

Putnam LOW 0.2127 0 0517 0.0690 0.0631 0.0596 0.0736 0 0690 0 0619 HIGH 0.2837 0 0754 0.1184 0.1080 0.0995 0.1262 0.1184 0.1055 WGHTD AVE 0.2595 0 0671 0.1003 0.0915 0.0849 0.1069 0.1003 0 0894

St. Johns LOW 0.1815 0 0441 0.0610 0.0561 0.0529 0.0648 0 0610 0 0551 HIGH 0.4072 0.1253 0.2587 0.2386 0.2145 0.2725 0 2587 0 2324 WGHTD AVE 0.3008 0 0847 0.1567 0.1444 0.1313 0.1656 0.1567 0.1408

St. Lucie LOW 0.8123 0 2609 0.6113 0.5676 0.5077 0.6404 0 6113 0 5530 HIGH 1.5420 0 5354 1.4888 1.3980 1.2485 1.5456 1.4888 1 3642 WGHTD AVE 1.1559 0 3834 1.0098 0.9438 0.8428 1.0522 1 0098 0 9202

Santa Rosa LOW 0.3640 0.1066 0.1916 0.1750 0.1584 0.2035 0.1916 0.1704 HIGH 1.0965 0 3955 1.0157 0.9471 0.8431 1.0598 1 0157 0 9224 WGHTD AVE 0.8104 0 2746 0.6811 0.6330 0.5643 0.7126 0 6811 0 6163

Sarasota LOW 0.5143 0.1540 0.3247 0.2980 0.2667 0.3431 0 3247 0 2898 HIGH 0.7877 0 2499 0.5900 0.5458 0.4864 0.6195 0 5900 0 5310 WGHTD AVE 0.6380 0.1981 0.4469 0.4122 0.3676 0.4704 0.4469 0.4009

Seminole LOW 0.3499 0 0890 0.1423 0.1302 0.1199 0.1514 0.1423 0.1272 HIGH 0.4571 0.1281 0.2495 0.2297 0.2075 0.2635 0 2495 0 2238 WGHTD AVE 0.3755 0 0973 0.1588 0.1451 0.1331 0.1689 0.1588 0.1416

Sumter LOW 0.3254 0 0822 0.1301 0.1188 0.1094 0.1385 0.1301 0.1160 HIGH 0.4160 0.1127 0.2015 0.1844 0.1675 0.2138 0 2015 0.1797 WGHTD AVE 0.3687 0 0947 0.1620 0.1482 0.1355 0.1722 0.1620 0.1446

*Includes contents and A.L.E.

FPHLM V3.0 2008 305 Form A-6A, Output Ranges LOSS COSTS PER $1,000 (2002 FHCF exposure data) PERSONAL RESIDENTIAL - Renters -- FRAME Rev. 11-1-07 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Suwanee LOW 0.1194 0 0272 0.0333 0.0308 0.0296 0.0354 0 0333 0 0303 HIGH 0.2280 0 0587 0.0887 0.0812 0.0753 0.0943 0 0887 0 0794 WGHTD AVE 0.1444 0 0342 0.0451 0.0415 0.0393 0.0479 0 0451 0 0407

Taylor LOW 0.1573 0 0389 0.0554 0.0509 0.0476 0.0589 0 0554 0 0498 HIGH 0.2888 0 0818 0.1515 0.1398 0.1275 0.1599 0.1515 0.1365 WGHTD AVE 0.2258 0 0608 0.1069 0.0985 0.0902 0.1131 0.1069 0 0962

Union LOW 0.1344 0 0304 0.0366 0.0338 0.0326 0.0388 0 0366 0 0333 HIGH 0.2076 0 0508 0.0696 0.0638 0.0600 0.0741 0 0696 0 0625 WGHTD AVE 0.1834 0 0437 0.0585 0.0537 0.0508 0.0622 0 0585 0 0527

Volusia LOW 0.2570 0 0628 0.0941 0.0861 0.0800 0.1001 0 0941 0 0843 HIGH 0.7406 0 2395 0.6104 0.5689 0.5072 0.6372 0 6104 0 5543 WGHTD AVE 0.4237 0.1182 0.2347 0.2161 0.1951 0.2478 0 2347 0 2106

Wakulla LOW 0.1678 0 0408 0.0564 0.0518 0.0488 0.0599 0 0564 0 0508 HIGH 0.4095 0.1239 0.2524 0.2331 0.2105 0.2659 0 2524 0 2272 WGHTD AVE 0.1918 0 0456 0.0721 0.0664 0.0618 0.0765 0 0721 0 0650

Walton LOW 0.3172 0 0883 0.1474 0.1347 0.1230 0.1567 0.1474 0.1313 HIGH 1.0515 0 3755 0.9446 0.8791 0.7822 0.9869 0 9446 0 8559 WGHTD AVE 0.6085 0 2007 0.4439 0.4106 0.3674 0.4664 0.4439 0 3997

Washington LOW 0.2563 0 0662 0.0973 0.0889 0.0827 0.1037 0 0973 0 0870 HIGH 0.5541 0.1767 0.3726 0.3436 0.3082 0.3925 0 3726 0 3345 WGHTD AVE 0.3780 0.1087 0.1942 0.1777 0.1613 0.2061 0.1942 0.1731

STATEWIDE LOW 0.1040 0 0232 0.0273 0.0254 0.0245 0.0290 0 0273 0 0250 HIGH 4.8376 1.7530 5.7157 5.4433 4.9006 5.8748 5.7157 5 3301 WGHTD AVE 0.5101 0.1469 0.3486 0.3238 0.2910 0.3654 0 3486 0 3157

*Includes contents and A.L.E.

FPHLM V3.0 2008 306 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - Renters -- MASONRY 0% $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Alachua LOW 0.1869 0 0445 0.0565 0.0516 0.0491 0 0604 0 0565 0 0507 HIGH 0.2487 0 0631 0.0884 0.0804 0 0750 0 0946 0 0884 0 0786 WGHTD AVE 0.2122 0 0518 0.0685 0.0624 0 0589 0 0733 0 0685 0 0612

Baker LOW 0.1076 0 0241 0.0285 0.0264 0 0255 0 0303 0 0285 0 0260 HIGH 0.1791 0 0430 0.0566 0.0519 0 0491 0 0603 0 0566 0 0509 WGHTD AVE 0.1734 0 0415 0.0540 0.0495 0 0470 0 0576 0 0540 0 0486

Bay LOW 0.2702 0 0690 0.0970 0.0884 0 0825 0.1037 0 0970 0 0864 HIGH 1.1956 0.4402 1.1377 1.0593 0 9372 1.1875 1.1377 1 0307 WGHTD AVE 0.5973 0.1647 0.4034 0.3707 0.3304 0.4257 0.4034 0 3603

Bradford LOW 0.1824 0 0430 0.0533 0.0488 0 0466 0 0570 0 0533 0 0479 HIGH 0.2303 0 0570 0.0771 0.0702 0 0660 0 0824 0 0771 0 0688 WGHTD AVE 0.1843 0 0432 0.0539 0.0494 0 0471 0 0577 0 0539 0 0485

Brevard LOW 0.3608 0 0968 0.1740 0.1597 0.1447 0.1842 0.1740 0.1556 HIGH 1.2625 0.4295 1.1750 1.0976 0 9722 1 2236 1.1750 1 0688 WGHTD AVE 0.5108 0.1437 0.2962 0.2726 0 2443 0 3125 0 2962 0 2652

Broward LOW 1.0141 0 3432 0.8552 0.7954 0.7051 0 8937 0 8552 0.7740 HIGH 1.9487 0.7049 1.9835 1.8579 1 6449 2 0610 1 9835 1 8098 WGHTD AVE 1.3103 0.4503 1.1959 1.1155 0 9879 1 2467 1.1959 1 0860

Calhoun LOW 0.2153 0 0525 0.0676 0.0617 0 0585 0 0723 0 0676 0 0606 HIGH 0.2951 0 0780 0.1159 0.1054 0 0975 0.1239 0.1159 0.1029 WGHTD AVE 0.2458 0 0602 0.0818 0.0745 0 0701 0 0876 0 0818 0 0730

Charlotte LOW 0.4983 0.1392 0.2560 0.2337 0 2107 0 2719 0 2560 0 2273 HIGH 1.1980 0.4096 1.0715 0.9949 0 8780 1.1204 1 0715 0 9672 WGHTD AVE 0.5934 0.1725 0.3490 0.3195 0 2856 0 3695 0 3490 0 3105

Citrus LOW 0.2534 0 0606 0.0866 0.0793 0 0742 0 0922 0 0866 0 0777 HIGH 0.3790 0.1017 0.1790 0.1638 0.1487 0.1899 0.1790 0.1596 WGHTD AVE 0.3288 0 0844 0.1377 0.1260 0.1155 0.1464 0.1377 0.1230

Clay LOW 0.1725 0 0405 0.0505 0.0462 0 0441 0 0539 0 0505 0 0454 HIGH 0.2553 0 0649 0.0919 0.0837 0 0780 0 0983 0 0919 0 0818 WGHTD AVE 0.1984 0 0487 0.0646 0.0590 0 0556 0 0690 0 0646 0 0579

Collier LOW 0.6071 0.1717 0.3147 0.2862 0 2573 0 3349 0 3147 0 2781 HIGH 1.2950 0.4441 1.1002 1.0140 0 8913 1.1562 1.1002 0 9839 WGHTD AVE 0.7584 0 2316 0.4845 0.4431 0 3938 0 5129 0.4845 0.4302

Columbia LOW 0.1151 0 0257 0.0299 0.0276 0 0268 0 0318 0 0299 0 0273 HIGH 0.1948 0 0478 0.0650 0.0594 0 0559 0 0693 0 0650 0 0583 WGHTD AVE 0.1739 0 0424 0.0545 0.0499 0 0473 0 0581 0 0545 0 0490

De Soto LOW 0.4547 0.1193 0.1973 0.1798 0.1641 0 2101 0.1973 0.1752 HIGH 0.5702 0.1639 0.3153 0.2881 0 2584 0 3344 0 3153 0 2800 WGHTD AVE 0.5050 0.1407 0.2531 0.2312 0 2087 0 2688 0 2531 0 2249

Dixie LOW 0.2066 0 0528 0.0774 0.0708 0 0658 0 0825 0 0774 0 0693 HIGH 0.4151 0.1265 0.2500 0.2297 0 2063 0 2642 0 2500 0 2235 WGHTD AVE 0.2319 0 0602 0.0933 0.0852 0 0787 0 0994 0 0933 0 0833

*Includes contents and A.L.E.

FPHLM V3.0 2008 307 Form A-6A, Output Ranges LOSS COSTS PER $1,000 (2002 FHCF exposure data) PERSONAL RESIDENTIAL - Renters -- MASONRY Rev. 11-1-07 0% $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Duval LOW 0.1034 0 0230 0.0267 0.0248 0 0240 0 0284 0 0267 0 0244 HIGH 0.3611 0.1162 0.2576 0.2385 0 2129 0 2704 0 2576 0 2321 WGHTD AVE 0.1983 0 0497 0.0723 0.0663 0 0618 0 0770 0 0723 0 0649

Escambia LOW 0.3612 0.1079 0.1927 0.1752 0.1577 0 2053 0.1927 0.1703 HIGH 1.6370 0 6213 1.7000 1.5894 1.4059 1.7687 1.7000 1 5474 WGHTD AVE 0.8025 0 2701 0.6636 0.6136 0 5432 0 6964 0 6636 0 5964

Flagler LOW 0.3059 0 0782 0.1258 0.1149 0.1054 0.1338 0.1258 0.1121 HIGH 0.5111 0.1546 0.3330 0.3065 0 2730 0 3510 0 3330 0 2980 WGHTD AVE 0.3680 0 0997 0.1901 0.1742 0.1571 0 2013 0.1901 0.1696

Franklin LOW 0.4629 0.1438 0.2855 0.2621 0 2352 0 3018 0 2855 0 2550 HIGH 0.9904 0 3609 0.9304 0.8668 0.7679 0 9708 0 9304 0 8436 WGHTD AVE 0.6490 0 2136 0.5017 0.4642 0.4130 0 5266 0 5017 0.4516

Gadsen LOW 0.1578 0 0365 0.0433 0.0398 0 0384 0 0462 0 0433 0 0392 HIGH 0.2331 0 0580 0.0772 0.0704 0 0663 0 0827 0 0772 0 0690 WGHTD AVE 0.1685 0 0401 0.0484 0.0444 0 0426 0 0517 0 0484 0 0437

Gilchrist LOW 0.1831 0 0443 0.0583 0.0534 0 0505 0 0623 0 0583 0 0524 HIGH 0.2603 0 0688 0.1045 0.0952 0 0879 0.1116 0.1045 0 0930 WGHTD AVE 0.2518 0 0664 0.0996 0.0907 0 0839 0.1063 0 0996 0 0886

Glades LOW 0.5165 0.1361 0.2239 0.2037 0.1856 0 2388 0 2239 0.1983 HIGH 0.5488 0.1479 0.2536 0.2310 0 2095 0 2700 0 2536 0 2248 WGHTD AVE 0.5488 0.1479 0.2536 0.2310 0 2095 0 2700 0 2536 0 2248

Gulf LOW 0.3070 0 0839 0.1332 0.1214 0.1114 0.1420 0.1332 0.1184 HIGH 0.4746 0.1455 0.2804 0.2568 0 2307 0 2969 0 2804 0 2499 WGHTD AVE 0.4272 0.1197 0.2362 0.2162 0.1949 0 2504 0 2362 0 2104

Hamilton LOW 0.1072 0 0240 0.0280 0.0259 0 0251 0 0297 0 0280 0 0255 HIGH 0.1611 0 0387 0.0507 0.0465 0 0440 0 0541 0 0507 0 0456 WGHTD AVE 0.1477 0 0363 0.0450 0.0413 0.0393 0 0480 0 0450 0 0406

Hardee LOW 0.4399 0.1141 0.1828 0.1663 0.1522 0.1951 0.1828 0.1621 HIGH 0.5296 0.1483 0.2717 0.2478 0.2232 0 2888 0 2717 0 2410 WGHTD AVE 0.4523 0.1182 0.1928 0.1754 0.1602 0 2056 0.1928 0.1709

Hendry LOW 0.5547 0.1514 0.2638 0.2401 0 2172 0 2810 0 2638 0 2336 HIGH 0.6942 0 2024 0.3982 0.3642 0.3258 0.4220 0 3982 0 3539 WGHTD AVE 0.6410 0.1809 0.3445 0.3145 0.2822 0 3657 0 3445 0 3056

Hernando LOW 0.2911 0 0699 0.1001 0.0915 0 0853 0.1067 0.1001 0 0895 HIGH 0.3939 0.1071 0.1962 0.1796 0.1620 0 2080 0.1962 0.1748 WGHTD AVE 0.3625 0 0974 0.1703 0.1558 0.1413 0.1808 0.1703 0.1517

Highlands LOW 0.4017 0 0986 0.1436 0.1308 0.1215 0.1535 0.1436 0.1278 HIGH 0.5243 0.1395 0.2320 0.2110 0.1919 0 2475 0 2320 0 2054 WGHTD AVE 0.4387 0.1115 0.1725 0.1571 0.1445 0.1841 0.1725 0.1532

Hillsborough LOW 0.3335 0 0864 0.1399 0.1275 0.1166 0.1490 0.1399 0.1243 HIGH 0.7114 0 2284 0.5190 0.4769 0.4216 0 5471 0 5190 0.4629 WGHTD AVE 0.4256 0.1193 0.2216 0.2025 0.1822 0 2352 0 2216 0.1969

Holmes LOW 0.2573 0 0663 0.0933 0.0848 0 0790 0 0999 0 0933 0 0829 HIGH 0.3640 0.1048 0.1797 0.1639 0.1487 0.1912 0.1797 0.1596 WGHTD AVE 0.2759 0 0722 0.1065 0.0969 0.0897 0.1138 0.1065 0 0946

*Includes contents and A.L.E.

FPHLM V3.0 2008 308 Form A-6A, Output Ranges LOSS COSTS PER $1,000 (2002 FHCF exposure data) PERSONAL RESIDENTIAL - Renters -- MASONRY Rev. 11-1-07 0% $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Indian River LOW 0.5134 0.1436 0.2702 0.2476 0 2230 0 2861 0 2702 0 2410 HIGH 1.3214 0.4504 1.2406 1.1598 1 0272 1 2912 1 2406 1.1295 WGHTD AVE 0.6922 0 2076 0.4604 0.4247 0 3781 0.4845 0.4604 0.4130

Jackson LOW 0.1533 0 0350 0.0407 0.0375 0 0364 0 0434 0 0407 0 0370 HIGH 0.2751 0 0721 0.1051 0.0956 0 0886 0.1125 0.1051 0 0934 WGHTD AVE 0.2223 0 0551 0.0728 0.0664 0 0626 0 0779 0 0728 0 0651

Jefferson LOW 0.1388 0 0332 0.0437 0.0401 0 0380 0 0465 0 0437 0 0394 HIGH 0.1731 0 0447 0.0684 0.0627 0 0581 0 0726 0 0684 0 0613 WGHTD AVE 0.1408 0 0343 0.0453 0.0416 0 0393 0 0482 0 0453 0 0408

Lafayette LOW 0.1523 0 0370 0.0501 0.0459 0 0433 0 0533 0 0501 0 0450 HIGH 0.1871 0 0469 0.0669 0.0612 0 0572 0 0712 0 0669 0 0600 WGHTD AVE 0.1777 0 0427 0.0618 0.0567 0 0530 0 0659 0 0618 0 0555

Lake LOW 0.2277 0 0524 0.0659 0.0605 0 0578 0 0702 0 0659 0 0595 HIGH 0.4806 0.1336 0.2433 0.2222 0 2003 0 2585 0 2433 0 2161 WGHTD AVE 0.3663 0 0937 0.1451 0.1324 0.1217 0.1546 0.1451 0.1292

Lee LOW 0.5630 0.1599 0.2878 0.2620 0 2361 0 3063 0 2878 0 2547 HIGH 1.0937 0 3671 0.9043 0.8349 0.7364 0 9496 0 9043 0 8108 WGHTD AVE 0.6599 0.1943 0.3922 0.3587 0 3203 0.4155 0 3922 0 3485

Leon LOW 0.1144 0 0257 0.0299 0.0277 0 0268 0 0318 0 0299 0 0273 HIGH 0.2123 0 0535 0.0755 0.0689 0 0644 0 0806 0 0755 0 0674 WGHTD AVE 0.1825 0 0441 0.0593 0.0542 0 0512 0 0633 0 0593 0 0532

Levy LOW 0.1835 0 0443 0.0576 0.0527 0.0499 0 0616 0 0576 0 0517 HIGH 0.3438 0 0973 0.1681 0.1536 0.1395 0.1787 0.1681 0.1496 WGHTD AVE 0.2487 0 0654 0.0977 0.0891 0 0824 0.1043 0 0977 0 0871

Liberty LOW 0.2083 0 0502 0.0630 0.0576 0 0548 0 0674 0 0630 0 0565 HIGH 0.2209 0 0545 0.0747 0.0682 0 0640 0 0797 0 0747 0 0669 WGHTD AVE 0.2106 0 0509 0.0660 0.0603 0 0572 0 0705 0 0660 0 0592

Madison LOW 0.1159 0 0264 0.0316 0.0291 0 0280 0 0337 0 0316 0 0287 HIGH 0.1643 0 0402 0.0546 0.0500 0 0471 0 0582 0 0546 0 0490 WGHTD AVE 0.1495 0 0348 0.0464 0.0425 0 0403 0 0494 0 0464 0 0417

Manatee LOW 0.4496 0.1282 0.2419 0.2207 0.1981 0 2568 0 2419 0 2145 HIGH 1.0191 0 3438 0.8882 0.8247 0.7289 0 9290 0 8882 0 8020 WGHTD AVE 0.5618 0.1562 0.3554 0.3261 0 2903 0 3755 0 3554 0 3168

Marion LOW 0.1796 0 0387 0.0464 0.0430 0 0414 0 0493 0 0464 0 0424 HIGH 0.3643 0 0958 0.1610 0.1471 0.1341 0.1712 0.1610 0.1434 WGHTD AVE 0.2915 0 0703 0.1023 0.0936 0 0872 0.1090 0.1023 0 0916

Martin LOW 0.6376 0.1903 0.3758 0.3454 0 3098 0 3970 0 3758 0 3360 HIGH 1.5221 0 5375 1.4352 1.3395 1.1857 1.4955 1.4352 1 3040 WGHTD AVE 0.8423 0 2597 0.6072 0.5619 0 5004 0 6373 0 6072 0 5467

Miami-Dade LOW 0.9153 0 3087 0.7489 0.6946 0 6151 0.7841 0.7489 0 6755 HIGH 2.8084 1 0323 3.0917 2.9122 2 5860 3.1997 3 0917 2 8408 WGHTD AVE 1.6336 0 5457 1.5800 1.4765 1 3065 1 6446 1 5800 1.4376

Monroe LOW 1.3850 0 5281 1.3518 1.2575 1.1117 1.4116 1 3518 1 2231 HIGH 3.5941 1.4108 4.2207 3.9854 3 5441 4 3605 4 2207 3 8899 WGHTD AVE 2.8259 1.1571 3.2105 3.0207 2.6808 3 3252 3 2105 2 9456

Nassau LOW 0.1211 0 0287 0.0383 0.0354 0 0335 0 0407 0 0383 0 0348 HIGH 0.1968 0 0534 0.0919 0.0847 0 0775 0 0971 0 0919 0 0827 WGHTD AVE 0.1737 0 0423 0.0688 0.0631 0 0584 0 0730 0 0688 0 0617

*Includes contents and A.L.E.

FPHLM V3.0 2008 309 Form A-6A, Output Ranges LOSS COSTS PER $1,000 (2002 FHCF exposure data) PERSONAL RESIDENTIAL - Renters -- MASONRY Rev. 11-1-07 0% $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Okaloosa LOW 0.3606 0.1058 0.1861 0.1694 0.1530 0.1981 0.1861 0.1647 HIGH 0.9322 0 3336 0.7941 0.7337 0 6482 0 8337 0.7941 0.7127 WGHTD AVE 0.8343 0 2732 0.6788 0.6263 0 5541 0.7133 0 6788 0 6085

Okeechobee LOW 0.4805 0.1247 0.2013 0.1833 0.1676 0 2146 0 2013 0.1787 HIGH 0.5389 0.1458 0.2538 0.2315 0 2098 0 2700 0 2538 0 2254 WGHTD AVE 0.5250 0.1397 0.2368 0.2158 0.1962 0 2521 0 2368 0 2102

Orange LOW 0.2999 0 0705 0.0956 0.0874 0 0822 0.1021 0 0956 0 0856 HIGH 0.4798 0.1327 0.2442 0.2235 0 2017 0 2589 0 2442 0 2176 WGHTD AVE 0.3573 0 0890 0.1336 0.1219 0.1127 0.1425 0.1336 0.1190

Osceola LOW 0.3376 0 0810 0.1131 0.1030 0 0963 0.1209 0.1131 0.1008 HIGH 0.4450 0.1171 0.1956 0.1786 0.1628 0 2081 0.1956 0.1740 WGHTD AVE 0.3711 0 0924 0.1367 0.1245 0.1153 0.1460 0.1367 0.1216

Palm Beach LOW 0.7142 0 2193 0.4651 0.4295 0 3842 0.4892 0.4651 0.4181 HIGH 1.9287 0 6863 1.9138 1.7936 1 5917 1 9883 1 9138 1.7479 WGHTD AVE 1.1251 0 3745 0.9561 0.8907 0.7907 0 9979 0 9561 0 8672

Pasco LOW 0.3188 0 0768 0.1086 0.0991 0 0925 0.1160 0.1086 0 0969 HIGH 0.4417 0.1279 0.2528 0.2317 0 2074 0 2675 0 2528 0 2253 WGHTD AVE 0.3865 0.1057 0.1921 0.1757 0.1587 0 2038 0.1921 0.1710

Pinellas LOW 0.3754 0.1031 0.1871 0.1711 0.1546 0.1986 0.1871 0.1665 HIGH 1.2993 0.4572 1.2652 1.1794 1 0396 1 3188 1 2652 1.1472 WGHTD AVE 0.4942 0.1417 0.3081 0.2830 0 2523 0 3253 0 3081 0 2750

Polk LOW 0.3531 0 0857 0.1204 0.1097 0.1024 0.1287 0.1204 0.1073 HIGH 0.5919 0.1726 0.3395 0.3105 0 2777 0 3598 0 3395 0 3017 WGHTD AVE 0.4176 0.1086 0.1743 0.1588 0.1453 0.1858 0.1743 0.1547

Putnam LOW 0.2097 0 0506 0.0653 0.0596 0 0565 0 0699 0 0653 0 0585 HIGH 0.2753 0 0723 0.1081 0.0983 0 0908 0.1156 0.1081 0 0960 WGHTD AVE 0.2553 0 0656 0.0942 0.0857 0 0796 0.1007 0 0942 0 0837

St. Johns LOW 0.1785 0 0430 0.0573 0.0526 0 0497 0 0610 0 0573 0 0516 HIGH 0.3807 0.1158 0.2250 0.2063 0.1851 0 2380 0 2250 0 2006 WGHTD AVE 0.3031 0 0842 0.1532 0.1403 0.1272 0.1624 0.1532 0.1367

St. Lucie LOW 0.5688 0.1656 0.3158 0.2901 0 2613 0 3338 0 3158 0 2824 HIGH 0.9652 0 3192 0.7626 0.7076 0 6284 0.7985 0.7626 0 6885 WGHTD AVE 0.7639 0 2378 0.5266 0.4866 0.4340 0 5534 0 5266 0.4735

Santa Rosa LOW 0.3471 0.1002 0.1704 0.1550 0.1404 0.1817 0.1704 0.1508 HIGH 0.9610 0 3471 0.8394 0.7764 0 6858 0 8804 0 8394 0.7544 WGHTD AVE 0.8359 0 2485 0.6886 0.6363 0.5629 0.7230 0 6886 0 6183

Sarasota LOW 0.4369 0.1242 0.2344 0.2137 0.1921 0 2490 0 2344 0 2078 HIGH 0.6430 0.1954 0.4141 0.3799 0.3382 0.4376 0.4141 0 3691 WGHTD AVE 0.5320 0.1578 0.3197 0.2926 0.2612 0 3384 0 3197 0 2843

Seminole LOW 0.3215 0 0779 0.1128 0.1031 0 0960 0.1203 0.1128 0.1008 HIGH 0.4011 0.1067 0.1854 0.1697 0.1542 0.1967 0.1854 0.1654 WGHTD AVE 0.3461 0 0853 0.1266 0.1155 0.1071 0.1351 0.1266 0.1129

Sumter LOW 0.2999 0 0722 0.1034 0.0944 0.0880 0.1104 0.1034 0 0924 HIGH 0.3742 0 0956 0.1529 0.1395 0.1279 0.1629 0.1529 0.1361 WGHTD AVE 0.3490 0 0882 0.1358 0.1239 0.1141 0.1448 0.1358 0.1210

*Includes contents and A.L.E.

FPHLM V3.0 2008 310 Form A-6A, Output Ranges LOSS COSTS PER $1,000 (2002 FHCF exposure data) PERSONAL RESIDENTIAL - Renters -- MASONRY Rev. 11-1-07 0% $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Suwanee LOW 0.1184 0 0269 0.0323 0.0298 0 0287 0 0343 0 0323 0 0293 HIGH 0.2223 0 0567 0.0818 0.0746 0 0695 0 0873 0 0818 0 0730 WGHTD AVE 0.1445 0 0345 0.0439 0.0403 0 0383 0 0467 0 0439 0 0396

Taylor LOW 0.1544 0 0379 0.0519 0.0476 0 0447 0 0553 0 0519 0 0466 HIGH 0.2739 0 0763 0.1324 0.1215 0.1107 0.1404 0.1324 0.1185 WGHTD AVE 0.2077 0 0568 0.0907 0.0831 0 0763 0 0963 0 0907 0 0812

Union LOW 0.1335 0 0302 0.0356 0.0329 0 0318 0 0379 0 0356 0 0324 HIGH 0.2040 0 0495 0.0652 0.0596 0 0563 0 0697 0 0652 0 0585 WGHTD AVE 0.1722 0 0396 0.0515 0.0472 0 0449 0 0549 0 0515 0 0464

Volusia LOW 0.2397 0 0560 0.0767 0.0703 0 0661 0 0817 0 0767 0 0689 HIGH 0.5958 0.1852 0.4257 0.3933 0 3491 0.4473 0.4257 0 3824 WGHTD AVE 0.3902 0 0979 0.1884 0.1727 0.1564 0.1996 0.1884 0.1683

Wakulla LOW 0.1653 0 0399 0.0532 0.0488 0 0461 0 0567 0 0532 0 0479 HIGH 0.3814 0.1139 0.2165 0.1985 0.1788 0 2293 0 2165 0.1933 WGHTD AVE 0.2494 0 0480 0.1089 0.0999 0 0914 0.1155 0.1089 0 0975

Walton LOW 0.3058 0 0838 0.1330 0.1210 0.1108 0.1419 0.1330 0.1180 HIGH 0.9326 0 3332 0.7905 0.7302 0 6453 0 8299 0.7905 0.7094 WGHTD AVE 0.7673 0 2163 0.6049 0.5579 0.4942 0 6360 0 6049 0 5421

Washington LOW 0.2505 0 0640 0.0902 0.0822 0 0767 0 0964 0 0902 0 0804 HIGH 0.5139 0.1617 0.3209 0.2940 0 2631 0 3395 0 3209 0 2858 WGHTD AVE 0.3551 0 0998 0.1700 0.1550 0.1409 0.1810 0.1700 0.1509

STATEWIDE LOW 0.1034 0 0230 0.0267 0.0248 0 0240 0 0284 0 0267 0 0244 HIGH 3.5941 1.4108 4.2207 3.9854 3 5441 4 3605 4 2207 3 8899 WGHTD AVE 0.7151 0 2149 0.5294 0.4910 0.4372 0 5547 0 5294 0.4780

*Includes contents and A.L.E.

FPHLM V3.0 2008 311 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - Condo Owners -- FRAME $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Alachua LOW 0 0924 0.1893 0.0454 0.1227 0 0636 0 0555 0.1227 0.0636 0.0555 HIGH 0.1153 0 2544 0.0652 0.1782 0.1036 0 0896 0.1782 0.1036 0.0896 WGHTD AVE 0 0991 0 2103 0.0516 0.1397 0 0758 0 0659 0.1397 0.0758 0.0659

Baker LOW 0 0568 0.1084 0.0243 0 0665 0 0309 0 0276 0.0665 0.0309 0.0276 HIGH 0 0876 0.1822 0.0441 0.1206 0 0648 0 0570 0.1206 0.0648 0.0570 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Bay LOW 0.1258 0 2763 0.0713 0.1944 0.1136 0 0986 0.1944 0.1136 0.0986 HIGH 0 3655 1.4189 0.5171 1.7850 1 6026 1.4489 1.7850 1.6026 1.4489 WGHTD AVE 0 2341 0 6857 0.2341 0.7246 0 5889 0 5196 0.7246 0.5889 0.5196

Bradford LOW 0 0895 0.1847 0.0437 0.1174 0 0594 0 0520 0.1174 0.0594 0.0520 HIGH 0.1085 0 2349 0.0587 0.1594 0 0892 0 0776 0.1594 0.0892 0.0776 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Brevard LOW 0 2959 0.4143 0.1170 0.4272 0 2677 0 2339 0.4272 0.2677 0.2339 HIGH 0 6267 1 6928 0.5796 2 2378 1 9456 1.7673 2.2378 1.9456 1.7673 WGHTD AVE 0 3779 0 6040 0.1788 0 6694 0.4680 0.4119 0.6694 0.4680 0.4119

Broward LOW 0 6805 1 6582 0.5814 2 2054 1 8964 1.7105 2.2054 1.8964 1.7105 HIGH 1 0147 3 3546 1.2130 4 6941 4 2770 3 9116 4.6941 4.2770 3.9116 WGHTD AVE 0.7961 2.1806 0.7744 2 9668 2 6181 2 3762 2.9668 2.6181 2.3762

Calhoun LOW 0.1043 0 2185 0.0536 0.1436 0 0766 0 0668 0.1436 0.0766 0.0668 HIGH 0.1309 0 3038 0.0815 0 2235 0.1385 0.1200 0.2235 0.1385 0.1200 WGHTD AVE 0.1215 0 2670 0.0000 0.1830 0.1042 0 0902 0.1830 0.1042 0.0902

Charlotte LOW 0.4027 0 5810 0.1711 0 6231 0.4023 0 3481 0.6231 0.4023 0.3481 HIGH 0 6265 1 5938 0.5511 2 0905 1.7845 1 6013 2.0905 1.7845 1.6013 WGHTD AVE 0.4628 1 0801 0.3013 1 2055 0 9435 0 8363 1.2055 0.9435 0.8363

Citrus LOW 0 2382 0 2745 0.0688 0 2525 0.1215 0.1048 0.2525 0.1215 0.1048 HIGH 0 3234 0.4331 0.1226 0.4506 0 2758 0 2390 0.4506 0.2758 0.2390 WGHTD AVE 0 2894 0 3651 0.0994 0 3610 0 2034 0.1759 0.3610 0.2034 0.1759

Clay LOW 0 0859 0.1745 0.0412 0.1112 0 0564 0 0494 0.1112 0.0564 0.0494 HIGH 0.1180 0 2614 0.0671 0.1842 0.1079 0 0936 0.1842 0.1079 0.0936 WGHTD AVE 0 0944 0.1991 0.0492 0.1339 0 0735 0 0642 0.1339 0.0735 0.0642

Collier LOW 0.4888 0.7114 0.2127 0.7691 0.4998 0.4288 0.7691 0.4998 0.4288 HIGH 0.7237 1.7059 0.5954 2 2108 1 8447 1 6303 2.2108 1.8447 1.6303 WGHTD AVE 0 5471 0 9482 0.3026 1.1075 0 8121 0.7061 1.1075 0.8121 0.7061

Columbia LOW 0 0604 0.1157 0.0259 0 0701 0 0321 0 0286 0.0701 0.0321 0.0286 HIGH 0 0924 0.1984 0.0491 0.1341 0 0748 0 0655 0.1341 0.0748 0.0655 WGHTD AVE 0 0798 0.1678 0.0395 0.1069 0 0553 0 0486 0.1069 0.0553 0.0486

De Soto LOW 0.4024 0 5124 0.1421 0 5201 0 3001 0 2575 0.5201 0.3001 0.2575 HIGH 0.4550 0 6743 0.2041 0.7494 0 5034 0.4354 0.7494 0.5034 0.4354 WGHTD AVE 0.4203 0 5841 0.1724 0 6226 0 3937 0 3399 0.6226 0.3937 0.3399

Dixie LOW 0 0951 0 2122 0.0548 0.1527 0 0919 0 0803 0.1527 0.0919 0.0803 HIGH 0.1631 0.4495 0.1386 0.4270 0 3276 0 2893 0.4270 0.3276 0.2893 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

*Includes contents and A.L.E.

FPHLM V3.0 2008 312 Form A-6A, Output Ranges LOSS COSTS PER $1,000 (2002 FHCF exposure data) PERSONAL RESIDENTIAL - Condo Owners -- FRAME Rev. 11-1-07 $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Duval LOW 0 0548 0.1040 0.0232 0 0630 0 0287 0 0257 0.0630 0.0287 0.0257 HIGH 0.1333 0.4036 0.1311 0.4258 0 3492 0 3125 0.4258 0.3492 0.3125 WGHTD AVE 0 0946 0 2084 0.0529 0.1477 0 0872 0 0764 0.1477 0.0872 0.0764

Escambia LOW 0.1482 0 3816 0.1160 0 3370 0 2432 0 2100 0.3370 0.2432 0.2100 HIGH 0.4621 2 0046 0.7474 2 6557 2.4465 2 2299 2.6557 2.4465 2.2299 WGHTD AVE 0 2844 1 0585 0.3646 1 2446 1 0900 0 9768 1.2446 1.0900 0.9768

Flagler LOW 0 2737 0 3414 0.0921 0 3370 0.1887 0.1624 0.3370 0.1887 0.1624 HIGH 0 3585 0 6261 0.1973 0.7286 0 5388 0.4746 0.7286 0.5388 0.4746 WGHTD AVE 0 2998 0.4818 0.1377 0 5174 0 3532 0 3088 0.5174 0.3532 0.3088

Franklin LOW 0.1780 0 5018 0.1581 0.4830 0 3744 0 3301 0.4830 0.3744 0.3301 HIGH 0 3045 1.1751 0.4251 1.4665 1 3139 1.1899 1.4665 1.3139 1.1899 WGHTD AVE 0 2737 1 0182 0.3481 1 2228 1 0803 0 9757 1.2228 1.0803 0.9757

Gadsen LOW 0 0796 0.1589 0.0368 0 0980 0 0471 0 0416 0.0980 0.0471 0.0416 HIGH 0.1101 0 2372 0.0595 0.1598 0 0888 0 0772 0.1598 0.0888 0.0772 WGHTD AVE 0 0863 0 0000 0.0000 0.1154 0 0600 0 0525 0.1154 0.0600 0.0525

Gilchrist LOW 0 0880 0.1860 0.0453 0.1230 0 0664 0 0582 0.1230 0.0664 0.0582 HIGH 0.1161 0 2688 0.0718 0 2007 0.1259 0.1093 0.2007 0.1259 0.1093 WGHTD AVE 0 0000 0 2688 0.0000 0 2007 0.1259 0.1093 0.2007 0.1259 0.1093

Glades LOW 0.4574 0 5830 0.1627 0 5929 0 3426 0 2924 0.5929 0.3426 0.2924 HIGH 0.4742 0 6264 0.1786 0 6515 0 3928 0 3366 0.6515 0.3928 0.3366 WGHTD AVE 0.4742 0 6264 0.0000 0 6515 0 3928 0 3366 0.6515 0.3928 0.3366

Gulf LOW 0.1335 0 3186 0.0883 0 2481 0.1624 0.1412 0.2481 0.1624 0.1412 HIGH 0.1859 0 5111 0.1590 0.4793 0 3646 0 3201 0.4793 0.3646 0.3201 WGHTD AVE 0.1859 0 5111 0.1590 0.4793 0 3646 0 3201 0.4793 0.3646 0.3201

Hamilton LOW 0 0562 0.1078 0.0242 0 0656 0 0302 0 0269 0.0656 0.0302 0.0269 HIGH 0 0782 0.1634 0.0396 0.1076 0 0576 0 0505 0.1076 0.0576 0.0505 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Hardee LOW 0 3955 0.4924 0.1351 0.4931 0 2764 0 2358 0.4931 0.2764 0.2358 HIGH 0.4383 0 6172 0.1825 0 6679 0.4294 0 3695 0.6679 0.4294 0.3695 WGHTD AVE 0.4081 0 5295 0.1495 0 5454 0 3222 0 2755 0.5454 0.3222 0.2755

Hendry LOW 0.4747 0 6370 0.1840 0 6712 0.4118 0 3524 0.6712 0.4118 0.3524 HIGH 0 5416 0 8270 0.2536 0 9305 0 6391 0 5538 0.9305 0.6391 0.5538 WGHTD AVE 0 5192 0.7741 0.2275 0 8460 0 5643 0.4872 0.8460 0.5643 0.4872

Hernando LOW 0 2739 0 3152 0.0795 0 2912 0.1419 0.1219 0.2912 0.1419 0.1219 HIGH 0 3339 0.4523 0.1305 0.4791 0 3074 0 2666 0.4791 0.3074 0.2666 WGHTD AVE 0 3102 0 3898 0.1056 0.4041 0 2399 0 2074 0.4041 0.2399 0.2074

Highlands LOW 0 3776 0.4381 0.1133 0.4142 0 2074 0.1769 0.4142 0.2074 0.1769 HIGH 0.4610 0 5943 0.1674 0 6098 0 3573 0 3047 0.6098 0.3573 0.3047 WGHTD AVE 0.4152 0.4966 0.1322 0.4929 0 2690 0 2297 0.4929 0.2690 0.2297

Hillsborough LOW 0 2979 0 3736 0.1023 0 3736 0 2113 0.1811 0.3736 0.2113 0.1811 HIGH 0.4711 0 8974 0.2980 1.1034 0 8579 0.7519 1.1034 0.8579 0.7519 WGHTD AVE 0 3429 0.4882 0.1441 0 5250 0 3392 0 2928 0.5250 0.3392 0.2928

Holmes LOW 0.1195 0 2635 0.0686 0.1867 0.1095 0 0948 0.1867 0.1095 0.0948 HIGH 0.1548 0 3819 0.1114 0 3218 0 2241 0.1946 0.3218 0.2241 0.1946 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

*Includes contents and A.L.E.

FPHLM V3.0 2008 313 Form A-6A, Output Ranges LOSS COSTS PER $1,000 (2002 FHCF exposure data) PERSONAL RESIDENTIAL - Condo Owners -- FRAME Rev. 11-1-07 $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Indian River LOW 0.4061 0 6004 0.1766 0 6436 0.4241 0 3698 0.6436 0.4241 0.3698 HIGH 0 6416 1.7766 0.6085 2 3552 2 0557 1 8705 2.3552 2.0557 1.8705 WGHTD AVE 0.4880 0 9302 0.2834 1 0886 0 8313 0.7378 1.0886 0.8313 0.7378

Jackson LOW 0 0786 0.1540 0.0352 0 0936 0 0438 0 0389 0.0936 0.0438 0.0389 HIGH 0.1264 0 2828 0.0750 0 2063 0.1252 0.1083 0.2063 0.1252 0.1083 WGHTD AVE 0 0862 0 0000 0.0000 0.1051 0 0502 0 0443 0.1051 0.0502 0.0443

Jefferson LOW 0 0681 0.1410 0.0340 0 0931 0 0497 0 0438 0.0931 0.0497 0.0438 HIGH 0 0819 0.1787 0.0467 0.1329 0 0824 0 0724 0.1329 0.0824 0.0724 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Lafayette LOW 0 0739 0.1550 0.0380 0.1045 0 0575 0 0505 0.1045 0.0575 0.0505 HIGH 0 0879 0.1914 0.0484 0.1344 0 0784 0 0686 0.1344 0.0784 0.0686 WGHTD AVE 0 0879 0.1914 0.0000 0.1344 0 0784 0 0686 0.1344 0.0784 0.0686

Lake LOW 0 2190 0 2430 0.0573 0 2104 0 0857 0 0742 0.2104 0.0857 0.0742 HIGH 0 3992 0 5581 0.1638 0 5993 0 3832 0 3302 0.5993 0.3832 0.3302 WGHTD AVE 0 3590 0.4702 0.1350 0.4815 0 2854 0 2454 0.4815 0.2854 0.2454

Lee LOW 0.4444 0 6630 0.1964 0.7161 0.4546 0 3895 0.7161 0.4546 0.3895 HIGH 0 6297 1.4280 0.4896 1 8283 1 5076 1 3368 1.8283 1.5076 1.3368 WGHTD AVE 0 5060 0 9250 0.2730 1 0366 0.7584 0 6620 1.0366 0.7584 0.6620

Leon LOW 0 0598 0.1150 0.0258 0 0699 0 0323 0 0288 0.0699 0.0323 0.0288 HIGH 0 0985 0 2171 0.0552 0.1518 0 0884 0 0770 0.1518 0.0884 0.0770 WGHTD AVE 0 0921 0.1986 0.0495 0.1349 0 0757 0 0661 0.1349 0.0757 0.0661

Levy LOW 0 0870 0.1862 0.0453 0.1218 0 0654 0 0571 0.1218 0.0654 0.0571 HIGH 0.1452 0 3616 0.1038 0 3028 0 2110 0.1843 0.3028 0.2110 0.1843 WGHTD AVE 0.1312 0 3360 0.0980 0 2899 0 2075 0.1821 0.2899 0.2075 0.1821

Liberty LOW 0.1007 0 2108 0.0510 0.1354 0 0704 0 0615 0.1354 0.0704 0.0615 HIGH 0.1046 0 2243 0.0559 0.1520 0 0867 0 0757 0.1520 0.0867 0.0757 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Madison LOW 0 0598 0.1169 0.0268 0 0725 0 0347 0 0307 0.0725 0.0347 0.0307 HIGH 0 0790 0.1672 0.0413 0.1133 0 0629 0 0550 0.1133 0.0629 0.0550 WGHTD AVE 0 0790 0.1672 0.0413 0.1133 0 0629 0 0550 0.1133 0.0629 0.0550

Manatee LOW 0 3553 0 5298 0.1592 0 5788 0 3844 0 3322 0.5788 0.3844 0.3322 HIGH 0 5490 1 3450 0.4606 1.7453 1.4747 1 3226 1.7453 1.4747 1.3226 WGHTD AVE 0.4206 0 8741 0.2600 1 0140 0.7871 0 6951 1.0140 0.7871 0.6951

Marion LOW 0.1781 0.1871 0.0416 0.1531 0 0580 0 0508 0.1531 0.0580 0.0508 HIGH 0 3192 0.4117 0.1143 0.4187 0 2455 0 2117 0.4187 0.2455 0.2117 WGHTD AVE 0 2852 0 3331 0.0848 0 3130 0.1585 0.1363 0.3130 0.1585 0.1363

Martin LOW 0 5503 0 9335 0.3057 1.1167 0 8425 0.7405 1.1167 0.8425 0.7405 HIGH 0 8807 2 5695 0.9213 3 5203 3.1357 2 8495 3.5203 3.1357 2.8495 WGHTD AVE 0 6592 1 3695 0.4697 1.7679 1.4542 1 2990 1.7679 1.4542 1.2990

Miami-Dade LOW 0 6369 1 5006 0.5273 1 9855 1 6927 1 5196 1.9855 1.6927 1.5196 HIGH 1 2354 4 8376 1.7530 6 8797 6.4193 5 9288 6.8797 6.4193 5.9288 WGHTD AVE 0 9271 2 8654 1.0692 4 0250 3 6376 3 3174 4.0250 3.6376 3.3174

Monroe LOW 0.7199 1 5257 0.5757 2 0996 1.7802 1 5789 2.0996 1.7802 1.5789 HIGH 1 2136 4.4110 1.6907 6.4357 6 0079 5 5438 6.4357 6.0079 5.5438 WGHTD AVE 1 0733 3 0164 1.2614 4.4576 4 0581 3.7035 4.4576 4.0581 3.7035

Nassau LOW 0 0600 0.1235 0.0295 0 0821 0 0433 0 0383 0.0821 0.0433 0.0383 HIGH 0 0872 0 2077 0.0572 0.1703 0.1164 0.1034 0.1703 0.1164 0.1034 WGHTD AVE 0 0807 0.1836 0.0483 0.1380 0 0870 0 0766 0.1380 0.0870 0.0766

*Includes contents and A.L.E.

FPHLM V3.0 2008 314 Form A-6A, Output Ranges LOSS COSTS PER $1,000 (2002 FHCF exposure data) PERSONAL RESIDENTIAL - Condo Owners -- FRAME Rev. 11-1-07 $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Okaloosa LOW 0.1496 0 3797 0.1134 0 3282 0 2335 0 2022 0.3282 0.2335 0.2022 HIGH 0 3058 1 0561 0.3781 1 2421 1 0745 0 9555 1.2421 1.0745 0.9555 WGHTD AVE 0 2795 0 9989 0.3342 1.1317 0 9706 0 8616 1.1317 0.9706 0.8616

Okeechobee LOW 0.4189 0 5396 0.1480 0 5345 0 3037 0 2605 0.5345 0.3037 0.2605 HIGH 0.4499 0 6182 0.1766 0 6396 0 3928 0 3387 0.6396 0.3928 0.3387 WGHTD AVE 0.4421 0 5784 0.1607 0 5804 0 3367 0 2889 0.5804 0.3367 0.2889

Orange LOW 0 2861 0 3213 0.0790 0 2892 0.1322 0.1133 0.2892 0.1322 0.1133 HIGH 0 3966 0 5567 0.1622 0 5961 0 3822 0 3317 0.5961 0.3822 0.3317 WGHTD AVE 0 3321 0 3929 0.1029 0 3758 0.1948 0.1667 0.3758 0.1948 0.1667

Osceola LOW 0 3225 0 3645 0.0920 0 3362 0.1597 0.1361 0.3362 0.1597 0.1361 HIGH 0 3924 0 5028 0.1397 0 5107 0 2986 0 2568 0.5107 0.2986 0.2568 WGHTD AVE 0 3452 0.4026 0.1046 0 3832 0.1948 0.1661 0.3832 0.1948 0.1661

Palm Beach LOW 0 5798 1 0678 0.3545 1 3106 1 0265 0 9117 1.3106 1.0265 0.9117 HIGH 1 0248 3 2448 1.1629 4.4921 4 0585 3.7131 4.4921 4.0585 3.7131 WGHTD AVE 0.7318 1.7686 0.6237 2 3677 2 0357 1 8417 2.3677 2.0357 1.8417

Pasco LOW 0 2856 0 3449 0.0872 0 3192 0.1538 0.1316 0.3192 0.1538 0.1316 HIGH 0 3629 0 5265 0.1600 0 5849 0.4037 0 3520 0.5849 0.4037 0.3520 WGHTD AVE 0 3255 0.4414 0.1264 0.4621 0 2867 0 2477 0.4621 0.2867 0.2477

Pinellas LOW 0 3059 0.4347 0.1258 0.4585 0 2915 0 2529 0.4585 0.2915 0.2529 HIGH 0 5998 1.7709 0.6224 2 3896 2.1142 1 9132 2.3896 2.1142 1.9132 WGHTD AVE 0 3602 0 6515 0.2021 0.7545 0 5622 0.4955 0.7545 0.5622 0.4955

Polk LOW 0 3337 0 3833 0.0977 0 3569 0.1703 0.1452 0.3569 0.1703 0.1452 HIGH 0.4618 0.7055 0.2163 0.7937 0 5454 0.4725 0.7937 0.5454 0.4725 WGHTD AVE 0 3737 0.4615 0.1258 0.4599 0 2558 0 2185 0.4599 0.2558 0.2185

Putnam LOW 0.1006 0 2127 0.0517 0.1392 0 0741 0 0646 0.1392 0.0741 0.0646 HIGH 0.1242 0 2837 0.0754 0 2098 0.1296 0.1122 0.2098 0.1296 0.1122 WGHTD AVE 0.1206 0 2757 0.0720 0.1980 0.1194 0.1033 0.1980 0.1194 0.1033

St. Johns LOW 0 0870 0.1815 0.0441 0.1210 0 0656 0 0577 0.1210 0.0656 0.0577 HIGH 0.1517 0.4072 0.1253 0 3816 0 2887 0 2532 0.3816 0.2887 0.2532 WGHTD AVE 0.1269 0 3185 0.0914 0 2700 0.1904 0.1670 0.2700 0.1904 0.1670

St. Lucie LOW 0 5048 0 8123 0.2609 0 9508 0 6982 0 6126 0.9508 0.6982 0.6126 HIGH 0 6797 1 5420 0.5354 2 0094 1 6901 1 5182 2.0094 1.6901 1.5182 WGHTD AVE 0 6236 1 3373 0.4608 1 6789 1 3755 1 2287 1.6789 1.3755 1.2287

Santa Rosa LOW 0.1485 0 3640 0.1066 0 3066 0 2126 0.1836 0.3066 0.2126 0.1836 HIGH 0 3099 1 0965 0.3955 1 3099 1.1426 1 0186 1.3099 1.1426 1.0186 WGHTD AVE 0 2976 1 0465 0.3749 1 2390 1 0774 0 9614 1.2390 1.0774 0.9614

Sarasota LOW 0 3444 0 5143 0.1540 0 5604 0 3708 0 3200 0.5604 0.3708 0.3200 HIGH 0.4583 0.7877 0.2499 0 9196 0 6734 0 5891 0.9196 0.6734 0.5891 WGHTD AVE 0 3953 0 6420 0.1980 0.7265 0 5114 0.4448 0.7265 0.5114 0.4448

Seminole LOW 0 2948 0 3499 0.0890 0 3224 0.1603 0.1382 0.3224 0.1603 0.1382 HIGH 0 3362 0.4571 0.1281 0.4664 0 2832 0 2461 0.4664 0.2832 0.2461 WGHTD AVE 0 3152 0 3791 0.0982 0 3547 0.1808 0.1554 0.3547 0.1808 0.1554

Sumter LOW 0 2821 0 3254 0.0822 0 3010 0.1476 0.1262 0.3010 0.1476 0.1262 HIGH 0 3391 0.4160 0.1127 0.4115 0 2299 0.1971 0.4115 0.2299 0.1971 WGHTD AVE 0 3323 0.4135 0.1125 0.4088 0 2277 0.1953 0.4088 0.2277 0.1953

*Includes contents and A.L.E.

FPHLM V3.0 2008 315 Form A-6A, Output Ranges LOSS COSTS PER $1,000 (2002 FHCF exposure data) PERSONAL RESIDENTIAL - Condo Owners -- FRAME Rev. 11-1-07 $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Suwanee LOW 0 0615 0.1194 0.0272 0 0740 0 0353 0 0313 0.0740 0.0353 0.0313 HIGH 0.1022 0 2280 0.0587 0.1623 0 0965 0 0841 0.1623 0.0965 0.0841 WGHTD AVE 0 0989 0.1748 0.0526 0.1394 0 0806 0 0704 0.1394 0.0806 0.0704

Taylor LOW 0 0744 0.1573 0.0389 0.1073 0 0599 0 0524 0.1073 0.0599 0.0524 HIGH 0.1155 0 2888 0.0818 0 2400 0.1672 0.1473 0.2400 0.1672 0.1473 WGHTD AVE 0.1155 0 2888 0.0818 0 2400 0.1672 0.1473 0.2400 0.1672 0.1473

Union LOW 0 0688 0.1344 0.0304 0 0821 0 0386 0 0343 0.0821 0.0386 0.0343 HIGH 0 0984 0 2076 0.0508 0.1381 0 0749 0 0655 0.1381 0.0749 0.0655 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Volusia LOW 0 2294 0 2570 0.0628 0 2301 0.1060 0 0913 0.2301 0.1060 0.0913 HIGH 0.4064 0.7406 0.2395 0 8970 0 6951 0 6164 0.8970 0.6951 0.6164 WGHTD AVE 0 3125 0.4661 0.1272 0.4818 0 3102 0 2707 0.4818 0.3102 0.2707

Wakulla LOW 0 0805 0.1678 0.0408 0.1119 0 0607 0 0533 0.1119 0.0607 0.0533 HIGH 0.1525 0.4095 0.1239 0 3751 0 2807 0 2470 0.3751 0.2807 0.2470 WGHTD AVE 0.1062 0 2413 0.0794 0.1974 0.1322 0.1162 0.1974 0.1322 0.1162

Walton LOW 0.1340 0 3172 0.0883 0 2480 0.1621 0.1404 0.2480 0.1621 0.1404 HIGH 0 3103 1 0515 0.3755 1 2334 1 0635 0 9446 1.2334 1.0635 0.9446 WGHTD AVE 0 2684 0 6896 0.2769 0.7939 0 6551 0 5788 0.7939 0.6551 0.5788

Washington LOW 0.1173 0 2563 0.0662 0.1810 0.1058 0 0918 0.1810 0.1058 0.0918 HIGH 0.1964 0 5541 0.1767 0 5365 0.4162 0 3649 0.5365 0.4162 0.3649 WGHTD AVE 0.1173 0 2563 0.0662 0.1810 0.1058 0 0918 0.1810 0.1058 0.0918

STATEWIDE LOW 0 0548 0.1040 0.0232 0 0631 0 0287 0 0257 0.0631 0.0287 0.0257 HIGH 1 2354 4 8376 1.7530 6 8797 6.4193 5 9288 6.8797 6.4193 5.9288 WGHTD AVE 0.4129 0 8906 0.2819 1 0614 0 8511 0.7606 1.0614 0.8511 0.7606

*Includes contents and A.L.E.

FPHLM V3.0 2008 316 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - Condo Owners -- MASONRY $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Alachua LOW 0 0898 0.1869 0.0445 0.1183 0 0601 0 0525 0.1183 0.0601 0.0525 HIGH 0.1111 0 2487 0.0631 0.1685 0 0954 0 0824 0.1685 0.0954 0.0824 WGHTD AVE 0 0978 0 2106 0.0515 0.1368 0 0728 0 0633 0.1368 0.0728 0.0633

Baker LOW 0 0556 0.1076 0.0241 0 0651 0 0299 0 0267 0.0651 0.0299 0.0267 HIGH 0 0850 0.1791 0.0430 0.1154 0 0604 0 0530 0.1154 0.0604 0.0530 WGHTD AVE 0 0850 0 0000 0.0000 0.1154 0 0604 0 0530 0.1154 0.0604 0.0530

Bay LOW 0.1212 0 2702 0.0690 0.1839 0.1046 0 0906 0.1839 0.1046 0.0906 HIGH 0 3264 1.1956 0.4402 1.4538 1 2782 1.1376 1.4538 1.2782 1.1376 WGHTD AVE 0 2205 0 6592 0.2175 0 6714 0 5371 0.4695 0.6714 0.5371 0.4695

Bradford LOW 0 0871 0.1824 0.0430 0.1136 0 0565 0 0495 0.1136 0.0565 0.0495 HIGH 0.1048 0 2303 0.0570 0.1516 0 0827 0 0718 0.1516 0.0827 0.0718 WGHTD AVE 0 0874 0 2063 0.0504 0.1248 0 0641 0 0560 0.1248 0.0641 0.0560

Brevard LOW 0 2644 0 3608 0.0968 0 3455 0.1961 0.1696 0.3455 0.1961 0.1696 HIGH 0 5351 1 2625 0.4295 1 6021 1 3338 1.1908 1.6021 1.3338 1.1908 WGHTD AVE 0 3467 0 5469 0.1604 0 5782 0 3838 0 3341 0.5782 0.3838 0.3341

Broward LOW 0 5241 1 0141 0.3432 1 2436 0 9759 0 8637 1.2436 0.9759 0.8637 HIGH 0.7492 1 9487 0.7049 2 6153 2 2618 2 0284 2.6153 2.2618 2.0284 WGHTD AVE 0 6117 1 3461 0.4656 1.7222 1.4193 1 2645 1.7222 1.4193 1.2645

Calhoun LOW 0.1011 0 2153 0.0525 0.1380 0 0721 0 0628 0.1380 0.0721 0.0628 HIGH 0.1255 0 2951 0.0780 0 2087 0.1256 0.1085 0.2087 0.1256 0.1085 WGHTD AVE 0.1096 0 2153 0.0000 0.1553 0 0840 0 0729 0.1553 0.0840 0.0729

Charlotte LOW 0 3565 0.4983 0.1392 0.4953 0 2902 0 2486 0.4953 0.2902 0.2486 HIGH 0 5280 1.1980 0.4096 1.4994 1 2199 1 0761 1.4994 1.2199 1.0761 WGHTD AVE 0 3982 0 6007 0.1771 0 6403 0.4145 0 3571 0.6403 0.4145 0.3571

Citrus LOW 0 2147 0 2534 0.0606 0 2189 0 0965 0 0834 0.2189 0.0965 0.0834 HIGH 0 2872 0 3790 0.1017 0 3658 0 2024 0.1742 0.3658 0.2024 0.1742 WGHTD AVE 0 2634 0 3338 0.0857 0 3101 0.1602 0.1380 0.3101 0.1602 0.1380

Clay LOW 0 0836 0.1725 0.0405 0.1076 0 0535 0 0470 0.1076 0.0535 0.0470 HIGH 0.1137 0 2553 0.0649 0.1739 0 0992 0 0859 0.1739 0.0992 0.0859 WGHTD AVE 0 0924 0 2004 0.0492 0.1317 0 0714 0 0622 0.1317 0.0714 0.0622

Collier LOW 0.4313 0 6071 0.1717 0 6069 0 3576 0 3040 0.6069 0.3576 0.3040 HIGH 0 6093 1 2950 0.4441 1 5893 1 2565 1 0919 1.5893 1.2565 1.0919 WGHTD AVE 0.4798 0.7838 0.2402 0 8561 0 5824 0.4998 0.8561 0.5824 0.4998

Columbia LOW 0 0593 0.1151 0.0257 0 0689 0 0313 0 0279 0.0689 0.0313 0.0279 HIGH 0 0894 0.1948 0.0478 0.1280 0 0697 0 0609 0.1280 0.0697 0.0609 WGHTD AVE 0 0790 0.1640 0.0388 0.1039 0 0529 0 0464 0.1039 0.0529 0.0464

De Soto LOW 0 3579 0.4547 0.1193 0.4284 0 2230 0.1906 0.4284 0.2230 0.1906 HIGH 0 3994 0 5702 0.1639 0 5870 0 3589 0 3074 0.5870 0.3589 0.3074 WGHTD AVE 0 3808 0 5210 0.1490 0 5243 0 3070 0 2630 0.5243 0.3070 0.2630

Dixie LOW 0 0916 0 2066 0.0528 0.1435 0 0838 0 0730 0.1435 0.0838 0.0730 HIGH 0.1529 0.4151 0.1265 0 3741 0 2774 0 2423 0.3741 0.2774 0.2423 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

*Includes contents and A.L.E.

FPHLM V3.0 2008 317 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - Condo Owners -- MASONRY $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Duval LOW 0 0538 0.1034 0.0230 0 0620 0 0280 0 0250 0.0620 0.0280 0.0250 HIGH 0.1238 0 3611 0.1162 0 3615 0 2871 0 2536 0.3615 0.2871 0.2536 WGHTD AVE 0 0911 0 2037 0.0510 0.1400 0 0808 0 0706 0.1400 0.0808 0.0706

Escambia LOW 0.1396 0 3612 0.1079 0 3036 0 2127 0.1826 0.3036 0.2127 0.1826 HIGH 0.4040 1 6370 0.6213 2.1128 1 9121 1.7139 2.1128 1.9121 1.7139 WGHTD AVE 0 2807 0 9935 0.3555 1.1716 1 0161 0 9004 1.1716 1.0161 0.9004

Flagler LOW 0 2446 0 3059 0.0782 0 2808 0.1417 0.1216 0.2808 0.1417 0.1216 HIGH 0 3133 0 5111 0.1546 0 5540 0 3782 0 3284 0.5540 0.3782 0.3284 WGHTD AVE 0 2806 0.4409 0.1252 0.4529 0 2926 0 2533 0.4529 0.2926 0.2533

Franklin LOW 0.1662 0.4629 0.1438 0.4221 0 3167 0 2764 0.4221 0.3167 0.2764 HIGH 0 2718 0 9904 0.3609 1.1913 1 0444 0 9309 1.1913 1.0444 0.9309 WGHTD AVE 0 2263 0 6836 0.2521 0.7761 0 6500 0 5755 0.7761 0.6500 0.5755

Gadsen LOW 0 0776 0.1578 0.0365 0 0958 0 0456 0 0403 0.0958 0.0456 0.0403 HIGH 0.1064 0 2331 0.0580 0.1526 0 0828 0 0719 0.1526 0.0828 0.0719 WGHTD AVE 0 0792 0.1611 0.0375 0 0986 0 0477 0 0421 0.0986 0.0477 0.0421

Gilchrist LOW 0 0853 0.1831 0.0443 0.1180 0 0624 0 0545 0.1180 0.0624 0.0545 HIGH 0.1112 0 2603 0.0688 0.1869 0.1137 0 0984 0.1869 0.1137 0.0984 WGHTD AVE 0 0853 0.1831 0.0443 0.1180 0 0624 0 0545 0.1180 0.0624 0.0545

Glades LOW 0.4064 0 5165 0.1361 0.4870 0 2535 0 2156 0.4870 0.2535 0.2156 HIGH 0.4202 0 5488 0.1479 0 5285 0 2875 0 2450 0.5285 0.2875 0.2450 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Gulf LOW 0.1275 0 3070 0.0839 0 2289 0.1453 0.1257 0.2289 0.1453 0.1257 HIGH 0.1740 0.4746 0.1455 0.4219 0 3107 0 2701 0.4219 0.3107 0.2701 WGHTD AVE 0.1275 0.4734 0.0000 0 3952 0 2878 0 2502 0.3952 0.2878 0.2502

Hamilton LOW 0 0551 0.1072 0.0240 0 0643 0 0294 0 0261 0.0643 0.0294 0.0261 HIGH 0 0759 0.1611 0.0387 0.1034 0 0542 0 0475 0.1034 0.0542 0.0475 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Hardee LOW 0 3523 0.4399 0.1141 0.4091 0 2067 0.1760 0.4091 0.2067 0.1760 HIGH 0 3862 0 5296 0.1483 0 5304 0 3090 0 2638 0.5304 0.3090 0.2638 WGHTD AVE 0 3523 0 0000 0.0000 0.4091 0 2067 0.1760 0.4091 0.2067 0.1760

Hendry LOW 0.4199 0 5547 0.1514 0 5408 0 2997 0 2549 0.5408 0.2997 0.2549 HIGH 0.4743 0 6942 0.2024 0.7235 0.4535 0 3889 0.7235 0.4535 0.3889 WGHTD AVE 0.4623 0 6673 0.1879 0 6834 0.4193 0 3589 0.6834 0.4193 0.3589

Hernando LOW 0 2460 0 2911 0.0699 0 2524 0.1118 0 0962 0.2524 0.1118 0.0962 HIGH 0 2959 0 3939 0.1071 0 3833 0 2225 0.1913 0.3833 0.2225 0.1913 WGHTD AVE 0 2715 0 3510 0.0931 0 3346 0.1804 0.1550 0.3346 0.1804 0.1550

Highlands LOW 0 3387 0.4017 0.0986 0 3553 0.1611 0.1377 0.3553 0.1611 0.1377 HIGH 0.4090 0 5243 0.1395 0.4983 0 2631 0 2235 0.4983 0.2631 0.2235 WGHTD AVE 0 3628 0.4371 0.1108 0.4013 0.1951 0.1665 0.4013 0.1951 0.1665

Hillsborough LOW 0 2656 0 3335 0.0864 0 3098 0.1580 0.1349 0.3098 0.1580 0.1349 HIGH 0.4048 0.7114 0.2284 0 8186 0 5932 0 5124 0.8186 0.5932 0.5124 WGHTD AVE 0 3090 0.4305 0.1211 0.4329 0 2567 0 2198 0.4329 0.2567 0.2198

Holmes LOW 0.1149 0 2573 0.0663 0.1762 0.1008 0 0870 0.1762 0.1008 0.0870 HIGH 0.1468 0 3640 0.1048 0 2930 0.1978 0.1708 0.2930 0.1978 0.1708 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

*Includes contents and A.L.E.

FPHLM V3.0 2008 318 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - Condo Owners -- MASONRY $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Indian River LOW 0 3599 0 5134 0.1436 0 5102 0 3055 0 2635 0.5102 0.3055 0.2635 HIGH 0 5412 1 3214 0.4504 1 6836 1.4083 1 2592 1.6836 1.4083 1.2592 WGHTD AVE 0.4425 0.7900 0.2411 0 8856 0 6397 0 5600 0.8856 0.6397 0.5600

Jackson LOW 0 0769 0.1533 0.0350 0 0920 0 0427 0 0379 0.0920 0.0427 0.0379 HIGH 0.1213 0 2751 0.0721 0.1934 0.1140 0 0983 0.1934 0.1140 0.0983 WGHTD AVE 0 0852 0.1800 0.0420 0.1078 0 0517 0 0455 0.1078 0.0517 0.0455

Jefferson LOW 0 0663 0.1388 0.0332 0 0893 0 0466 0 0410 0.0893 0.0466 0.0410 HIGH 0 0792 0.1731 0.0447 0.1238 0 0742 0 0649 0.1238 0.0742 0.0649 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Lafayette LOW 0 0717 0.1523 0.0370 0 0999 0 0536 0 0470 0.0999 0.0536 0.0470 HIGH 0 0849 0.1871 0.0469 0.1273 0 0721 0 0630 0.1273 0.0721 0.0630 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Lake LOW 0.1983 0 2277 0.0524 0.1898 0 0725 0 0631 0.1898 0.0725 0.0631 HIGH 0 3524 0.4806 0.1336 0.4777 0 2765 0 2364 0.4777 0.2765 0.2364 WGHTD AVE 0 3068 0 3838 0.0985 0 3539 0.1784 0.1526 0.3539 0.1784 0.1526

Lee LOW 0 3921 0 5630 0.1599 0 5690 0 3275 0 2786 0.5690 0.3275 0.2786 HIGH 0 5342 1 0937 0.3671 1 3241 1 0315 0 8997 1.3241 1.0315 0.8997 WGHTD AVE 0.4442 0 6953 0.2063 0.7380 0.4838 0.4155 0.7380 0.4838 0.4155

Leon LOW 0 0586 0.1144 0.0257 0 0687 0 0314 0 0280 0.0687 0.0314 0.0280 HIGH 0 0950 0 2123 0.0535 0.1436 0 0814 0 0707 0.1436 0.0814 0.0707 WGHTD AVE 0 0885 0.1927 0.0478 0.1271 0 0693 0 0604 0.1271 0.0693 0.0604

Levy LOW 0 0845 0.1835 0.0443 0.1169 0 0614 0 0537 0.1169 0.0614 0.0537 HIGH 0.1378 0 3438 0.0973 0 2745 0.1849 0.1603 0.2745 0.1849 0.1603 WGHTD AVE 0.1086 0 2722 0.0612 0.1914 0.1199 0.1041 0.1914 0.1199 0.1041

Liberty LOW 0 0977 0 2083 0.0502 0.1308 0 0669 0 0585 0.1308 0.0669 0.0585 HIGH 0.1012 0 2209 0.0545 0.1444 0 0803 0 0699 0.1444 0.0803 0.0699 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Madison LOW 0 0585 0.1159 0.0264 0 0707 0 0333 0 0295 0.0707 0.0333 0.0295 HIGH 0 0766 0.1643 0.0402 0.1082 0 0586 0 0512 0.1082 0.0586 0.0512 WGHTD AVE 0 0619 0.1270 0.0299 0 0801 0 0404 0 0356 0.0801 0.0404 0.0356

Manatee LOW 0 3134 0.4496 0.1282 0.4549 0 2747 0 2349 0.4549 0.2747 0.2349 HIGH 0.4651 1 0191 0.3438 1 2582 1 0106 0 8913 1.2582 1.0106 0.8913 WGHTD AVE 0 3657 0 6831 0.1966 0.7466 0 5368 0.4661 0.7466 0.5368 0.4661

Marion LOW 0.1626 0.1796 0.0387 0.1412 0 0507 0 0446 0.1412 0.0507 0.0446 HIGH 0 2841 0 3643 0.0958 0 3440 0.1819 0.1560 0.3440 0.1819 0.1560 WGHTD AVE 0 2663 0 3252 0.0813 0 2919 0.1399 0.1202 0.2919 0.1399 0.1202

Martin LOW 0.4440 0 6376 0.1903 0 6681 0.4264 0 3696 0.6681 0.4264 0.3696 HIGH 0 6615 1 5221 0.5375 1 9638 1 6357 1.4575 1.9638 1.6357 1.4575 WGHTD AVE 0 5359 0 9883 0.3142 1.1576 0 8753 0.7713 1.1576 0.8753 0.7713

Miami-Dade LOW 0.4895 0 9153 0.3087 1.1081 0 8548 0.7524 1.1081 0.8548 0.7524 HIGH 0 8949 2 8084 1.0323 3 9031 3 5146 3.1863 3.9031 3.5146 3.1863 WGHTD AVE 0 6990 1 8861 0.6566 2.4793 2.1405 1 9197 2.4793 2.1405 1.9197

Monroe LOW 0 6946 1 3850 0.5281 1 8785 1 5605 1 3660 1.8785 1.5605 1.3660 HIGH 1.1164 3 5941 1.4108 5 2250 4 8015 4 3581 5.2250 4.8015 4.3581 WGHTD AVE 1 0052 2 9197 1.1379 4.1902 3.7790 3.4042 4.1902 3.7790 3.4042

Nassau LOW 0 0585 0.1211 0.0287 0 0782 0 0409 0 0362 0.0782 0.0409 0.0362 HIGH 0 0835 0.1968 0.0534 0.1535 0.1006 0 0885 0.1535 0.1006 0.0885 WGHTD AVE 0 0776 0.1770 0.0459 0.1274 0 0774 0 0678 0.1274 0.0774 0.0678

*Includes contents and A.L.E.

FPHLM V3.0 2008 319 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - Condo Owners -- MASONRY $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Okaloosa LOW 0.1413 0 3606 0.1058 0 2970 0 2050 0.1765 0.2970 0.2050 0.1765 HIGH 0 2768 0 9322 0.3336 1 0522 0 8909 0.7824 1.0522 0.8909 0.7824 WGHTD AVE 0 2642 0 9008 0.3185 0 9951 0 8377 0.7349 0.9951 0.8377 0.7349

Okeechobee LOW 0 3748 0.4805 0.1247 0.4425 0 2269 0.1935 0.4425 0.2269 0.1935 HIGH 0.4002 0 5389 0.1458 0 5170 0 2869 0 2454 0.5170 0.2869 0.2454 WGHTD AVE 0 3948 0 5128 0.1348 0.4782 0 2507 0 2139 0.4782 0.2507 0.2139

Orange LOW 0 2564 0 2999 0.0705 0 2545 0.1064 0 0916 0.2545 0.1064 0.0916 HIGH 0 3507 0.4798 0.1327 0.4762 0 2768 0 2381 0.4762 0.2768 0.2381 WGHTD AVE 0 2976 0 3588 0.0894 0 3210 0.1511 0.1294 0.3210 0.1511 0.1294

Osceola LOW 0 2900 0 3376 0.0810 0 2924 0.1264 0.1082 0.2924 0.1264 0.1082 HIGH 0 3501 0.4450 0.1171 0.4194 0 2211 0.1892 0.4194 0.2211 0.1892 WGHTD AVE 0 3081 0 3704 0.0913 0 3285 0.1518 0.1297 0.3285 0.1518 0.1297

Palm Beach LOW 0.4650 0.7142 0.2193 0.7782 0 5281 0.4620 0.7782 0.5281 0.4620 HIGH 0.7681 1 9287 0.6863 2 5471 2.1777 1 9565 2.5471 2.1777 1.9565 WGHTD AVE 0 5776 1 2005 0.4024 1.4771 1.1818 1 0517 1.4771 1.1818 1.0517

Pasco LOW 0 2548 0 3188 0.0768 0 2770 0.1214 0.1042 0.2770 0.1214 0.1042 HIGH 0 3217 0.4417 0.1279 0.4551 0 2869 0 2471 0.4551 0.2869 0.2471 WGHTD AVE 0 2785 0 3863 0.1075 0 3809 0 2224 0.1911 0.3809 0.2224 0.1911

Pinellas LOW 0 2712 0 3754 0.1031 0 3672 0 2117 0.1819 0.3672 0.2117 0.1819 HIGH 0.4991 1 2993 0.4572 1 6909 1.4396 1 2794 1.6909 1.4396 1.2794 WGHTD AVE 0 3283 0 5561 0.1658 0 6185 0.4363 0 3799 0.6185 0.4363 0.3799

Polk LOW 0 2986 0 3531 0.0857 0 3087 0.1346 0.1152 0.3087 0.1346 0.1152 HIGH 0.4046 0 5919 0.1726 0 6168 0 3867 0 3315 0.6168 0.3867 0.3315 WGHTD AVE 0 3331 0.4147 0.1070 0 3836 0.1923 0.1639 0.3836 0.1923 0.1639

Putnam LOW 0 0972 0 2097 0.0506 0.1338 0 0697 0 0607 0.1338 0.0697 0.0607 HIGH 0.1191 0 2753 0.0723 0.1958 0.1175 0.1013 0.1958 0.1175 0.1013 WGHTD AVE 0.1135 0 2481 0.0629 0.1707 0 0971 0 0840 0.1707 0.0971 0.0840

St. Johns LOW 0 0845 0.1785 0.0430 0.1158 0 0612 0 0538 0.1158 0.0612 0.0538 HIGH 0.1425 0 3807 0.1158 0 3399 0 2496 0 2170 0.3399 0.2496 0.2170 WGHTD AVE 0.1260 0 3207 0.0931 0 2678 0.1871 0.1627 0.2678 0.1871 0.1627

St. Lucie LOW 0.4111 0 5688 0.1656 0 5819 0 3577 0 3101 0.5819 0.3577 0.3101 HIGH 0 5294 0 9652 0.3192 1.1456 0 8684 0.7658 1.1456 0.8684 0.7658 WGHTD AVE 0.4933 0 8271 0.2615 0 9393 0 6753 0 5918 0.9393 0.6753 0.5918

Santa Rosa LOW 0.1407 0 3471 0.1002 0 2791 0.1877 0.1612 0.2791 0.1877 0.1612 HIGH 0 2795 0 9610 0.3471 1.1031 0 9422 0 8291 1.1031 0.9422 0.8291 WGHTD AVE 0 2659 0 8976 0.3198 1 0124 0 8578 0.7544 1.0124 0.8578 0.7544

Sarasota LOW 0 3041 0.4369 0.1242 0.4411 0 2660 0 2272 0.4411 0.2660 0.2272 HIGH 0 3994 0 6430 0.1954 0 6985 0.4714 0.4068 0.6985 0.4714 0.4068 WGHTD AVE 0 3492 0 5380 0.1598 0 5692 0 3698 0 3180 0.5692 0.3698 0.3180

Seminole LOW 0 2663 0 3215 0.0779 0 2782 0.1258 0.1082 0.2782 0.1258 0.1082 HIGH 0 3003 0.4011 0.1067 0 3803 0 2089 0.1800 0.3803 0.2089 0.1800 WGHTD AVE 0 2847 0 3474 0.0856 0 3052 0.1416 0.1214 0.3052 0.1416 0.1214

Sumter LOW 0 2539 0 2999 0.0722 0 2603 0.1157 0 0992 0.2603 0.1157 0.0992 HIGH 0 3028 0 3742 0.0956 0 3427 0.1725 0.1475 0.3427 0.1725 0.1475 WGHTD AVE 0 2935 0 3632 0.0928 0 3326 0.1649 0.1410 0.3326 0.1649 0.1410

*Includes contents and A.L.E.

FPHLM V3.0 2008 320 Form A-6A, Output Ranges (2002 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - Condo Owners -- MASONRY $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Suwanee LOW 0 0601 0.1184 0.0269 0 0722 0 0340 0 0302 0.0722 0.0340 0.0302 HIGH 0 0984 0 2223 0.0567 0.1529 0 0884 0 0768 0.1529 0.0884 0.0768 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Taylor LOW 0 0722 0.1544 0.0379 0.1023 0 0557 0 0487 0.1023 0.0557 0.0487 HIGH 0.1100 0 2739 0.0763 0 2164 0.1453 0.1269 0.2164 0.1453 0.1269 WGHTD AVE 0.1100 0 2739 0.0763 0 2164 0.1453 0.1269 0.2164 0.1453 0.1269

Union LOW 0 0674 0.1335 0.0302 0 0805 0 0374 0 0333 0.0805 0.0374 0.0333 HIGH 0 0954 0 2040 0.0495 0.1319 0 0698 0 0609 0.1319 0.0698 0.0609 WGHTD AVE 0 0674 0.1335 0.0302 0 0805 0 0374 0 0333 0.0805 0.0374 0.0333

Volusia LOW 0 2072 0 2397 0.0560 0 2024 0 0853 0 0737 0.2024 0.0853 0.0737 HIGH 0 3527 0 5958 0.1852 0 6716 0.4847 0.4234 0.6716 0.4847 0.4234 WGHTD AVE 0 3059 0.4611 0.1304 0.4764 0 3035 0 2630 0.4764 0.3035 0.2630

Wakulla LOW 0 0783 0.1653 0.0399 0.1074 0 0569 0 0499 0.1074 0.0569 0.0499 HIGH 0.1434 0 3814 0.1139 0 3315 0 2397 0 2088 0.3315 0.2397 0.2088 WGHTD AVE 0.1434 0 3814 0.1139 0 3315 0 2397 0 2088 0.3315 0.2397 0.2088

Walton LOW 0.1280 0 3058 0.0838 0 2289 0.1451 0.1252 0.2289 0.1451 0.1252 HIGH 0 2818 0 9326 0.3332 1 0513 0 8875 0.7789 1.0513 0.8875 0.7789 WGHTD AVE 0 2555 0.7677 0.2783 0 8544 0.7066 0 6191 0.8544 0.7066 0.6191

Washington LOW 0.1131 0 2505 0.0640 0.1711 0 0973 0 0843 0.1711 0.0973 0.0843 HIGH 0.1833 0 5139 0.1617 0.4729 0 3563 0 3095 0.4729 0.3563 0.3095 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

STATEWIDE LOW 0 0538 0.1034 0.0230 0 0620 0 0280 0 0250 0.0620 0.0280 0.0250 HIGH 1.1164 3 5941 1.4108 5 2250 4 8015 4 3581 5.2250 4.8015 4.3581 WGHTD AVE 0.4886 1 0686 0.3314 1 2574 0 9958 0 8829 1.2574 0.9958 0.8829

*Includes contents and A.L.E.

FPHLM V3.0 2008 321 Form A-6A: Output Ranges (2002 Weights), Item C: Zip codes with losses but no exposure:

32142 32198 32290 32395 32885 32891 32893 33199 33261 33299 33650 33663 33697

FPHLM V3.0 2008 322 Form A-6A: Output Ranges (2002 Weights), Item D: Zip codes with exposure but no losses:

32151 32230 32335 32573 32574 32575 32576 32581 32582 32589 32592 32593 32594 32595 32596 32597 32598 32613 33044 33188 33192

FPHLM V3.0 2008 323 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 Personal Residential -- Owners -- FRAME $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Alachua LOW 0 9243 0 0947 0.0628 0 0227 0.7528 0.4034 0.2115 0.4034 0 2434 0.1118 HIGH 1.1533 0.1272 0.0813 0 0326 0.9834 0.5745 0.3211 0 5745 0 3667 0.1783 WGHTD AVE 1 0219 0.1096 0.0709 0 0272 0.8534 0.4795 0.2589 0.4795 0 2972 0.1395

Baker LOW 0 5677 0 0542 0.0367 0 0122 0.4400 0.2112 0.1043 0 2112 0.1197 0.0554 HIGH 0 8755 0 0911 0.0587 0 0220 0.7155 0.3858 0.2052 0 3858 0 2350 0.1118 WGHTD AVE 0 8370 0 0865 0.0560 0 0208 0.6814 0.3639 0.1921 0 3639 0 2202 0.1041

Bay LOW 1 2576 0.1381 0.0875 0 0357 1.0702 0.6235 0.3484 0 6235 0 3977 0.1942 HIGH 3 6548 0.7095 0.2089 0 2585 4.1917 3.5535 2.9320 3 5535 3 0789 2.4173 WGHTD AVE 2 0947 0 2903 0.1450 0 0932 2.0577 1.5020 1.0580 1 5020 1.1521 0.7511

Bradford LOW 0 8953 0 0923 0.0602 0 0219 0.7272 0.3866 0.1998 0 3866 0 2305 0.1038 HIGH 1 0851 0.1174 0.0753 0 0293 0.9116 0.5180 0.2823 0 5180 0 3237 0.1533 WGHTD AVE 0 9202 0 0950 0.0622 0 0226 0.7489 0.4001 0.2078 0.4001 0 2396 0.1088

Brevard LOW 2 9589 0 2072 0.1066 0 0585 2.8071 2.2855 1.5104 2 2855 1.7523 0.7646 HIGH 6 2668 0 8464 0.2286 0 2898 6.8328 6.1104 4.8555 6.1104 5 2520 3.5138 WGHTD AVE 3 8013 0 2926 0.1373 0 0875 3.6995 3.0876 2.1173 3 0876 2.4223 1.1518

Broward LOW 6 8054 0 8291 0.1977 0 2907 7.3574 6.5943 5.1156 6 5943 5 5907 3.5343 HIGH 10.1466 1 6773 0.2787 0 6065 11 8495 10 9922 9.2009 10.9922 9.7784 7.1512 WGHTD AVE 7 9885 1.1009 0.2280 0 3963 8.9107 8.1025 6.4798 8.1025 7 0023 4.6999

Calhoun LOW 1 0433 0.1093 0.0715 0 0268 0.8610 0.4733 0.2507 0.4733 0 2887 0.1327 HIGH 1 3090 0.1519 0.0947 0 0407 1.1416 0.6890 0.3981 0 6890 0.4504 0.2322 WGHTD AVE 1.1547 0.1250 0.0811 0 0315 0.9734 0.5570 0.3031 0 5570 0 3481 0.1631

Charlotte LOW 4 0272 0 2905 0.1498 0 0856 3.9168 3.2827 2.2385 3 2827 2 5681 1.1724 HIGH 6 2648 0.7969 0.2274 0 2755 6.8339 6.1052 4.8003 6.1052 5 2136 3.3840 WGHTD AVE 4 8308 0.4030 0.1809 0.1307 4.8457 4.1644 3.0028 4.1644 3 3702 1.7887

Citrus LOW 2 3820 0.1372 0.0821 0 0344 2.1982 1.7629 1.1126 1.7629 1 3163 0.4921 HIGH 3 2340 0 2165 0.1149 0 0613 3.0989 2.5735 1.7228 2 5735 1 9914 0.8674 WGHTD AVE 2 9064 0.1816 0.1019 0 0489 2.7401 2.2442 1.4665 2 2442 1.7113 0.6999

Clay LOW 0 8589 0 0872 0.0576 0 0206 0.6933 0.3644 0.1882 0 3644 0 2168 0.0989 HIGH 1.1803 0.1307 0.0830 0 0336 1.0077 0.5900 0.3309 0 5900 0 3776 0.1847 WGHTD AVE 0 9476 0.1000 0.0647 0 0246 0.7838 0.4329 0.2335 0.4329 0 2674 0.1274

Collier LOW 4 8882 0 3557 0.1857 0.1064 4.7733 4.0132 2.7511 4 0132 3.1495 1.4568 HIGH 7 2367 0 8529 0.2755 0 2977 7.8072 6.9535 5.3906 6 9535 5 8863 3.6842 WGHTD AVE 5 5240 0.4751 0.2108 0.1526 5.5755 4.7959 3.4445 4.7959 3 8724 2.0217

Columbia LOW 0 6045 0 0578 0.0395 0 0129 0.4691 0.2258 0.1105 0 2258 0.1272 0.0575 HIGH 0 9238 0 0992 0.0627 0 0245 0.7680 0.4277 0.2317 0.4277 0 2651 0.1274 WGHTD AVE 0 8647 0 0910 0.0585 0 0221 0.7103 0.3854 0.2048 0 3854 0 2348 0.1109

De Soto LOW 4 0237 0 2562 0.1439 0 0711 3.8368 3.1812 2.1148 3.1812 2.4517 1.0376 HIGH 4 5496 0 3372 0.1662 0.1021 4.4714 3.7899 2.6342 3.7899 3 0004 1.4322 WGHTD AVE 4.1876 0 2877 0.1511 0 0832 4.0480 3.3883 2.3013 3 3883 2 6450 1.1902

Dixie LOW 0 9509 0.1061 0.0655 0 0274 0.8098 0.4717 0.2677 0.4717 0 3038 0.1546 HIGH 1 6311 0 2247 0.1124 0 0693 1.5693 1.1029 0.7543 1.1029 0 8256 0.5241 WGHTD AVE 1.1306 0.1359 0.0781 0 0376 1.0052 0.6306 0.3871 0 6306 0.4325 0.2431

*Includes contents and A.L.E.

FPHLM V3.0 2008 324 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 Personal Residential -- Owners -- FRAME $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Duval LOW 0 5482 0 0520 0.0360 0 0116 0.4234 0.2013 0.0983 0 2013 0.1130 0.0517 HIGH 1 3334 0 2018 0.0864 0 0656 1.3248 0.9644 0.7115 0 9644 0.7615 0.5427 WGHTD AVE 0 9007 0 0979 0.0617 0 0247 0.7520 0.4211 0.2347 0.4211 0 2658 0.1359

Escambia LOW 1.4819 0.1908 0.1102 0 0580 1.3922 0.9455 0.6171 0 9455 0 6825 0.4046 HIGH 4 6206 1 0023 0.2385 0 3737 5.5827 4.9318 4.2443 4 9318 4.4115 3.6365 WGHTD AVE 2 5487 0.4196 0.1661 0.1460 2.7087 2.1423 1.6497 2.1423 1.7579 1.2807

Flagler LOW 2.7374 0.1707 0.0967 0 0461 2.5761 2.1036 1.3713 2.1036 1 6015 0.6543 HIGH 3 5853 0 3130 0.1317 0 0987 3.5877 3.0485 2.1659 3 0485 2.4439 1.2676 WGHTD AVE 2 9976 0 2236 0.1104 0 0655 2.9014 2.4091 1.6345 2.4091 1 8777 0.8689

Franklin LOW 1.7797 0 2509 0.1260 0 0791 1.7322 1.2306 0.8486 1 2306 0 9261 0.5964 HIGH 3 0451 0 5875 0.1759 0 2125 3.4619 2.9048 2.3923 2 9048 2 5094 1.9805 WGHTD AVE 2 2762 0 3783 0.1475 0.1307 2.3997 1.8703 1.4317 1 8703 1 5259 1.1138

Gadsen LOW 0.7958 0 0794 0.0530 0 0184 0.6356 0.3241 0.1620 0 3241 0.1875 0.0827 HIGH 1.1013 0.1186 0.0764 0 0297 0.9246 0.5250 0.2840 0 5250 0 3265 0.1523 WGHTD AVE 0 8381 0 0856 0.0565 0 0204 0.6777 0.3564 0.1824 0 3564 0 2106 0.0945

Gilchrist LOW 0 8799 0 0930 0.0596 0 0227 0.7251 0.3968 0.2112 0 3968 0 2424 0.1139 HIGH 1.1609 0.1344 0.0826 0 0359 1.0166 0.6214 0.3629 0 6214 0.4110 0.2118 WGHTD AVE 1 0799 0.1222 0.0754 0 0319 0.9312 0.5556 0.3184 0 5556 0 3616 0.1831

Glades LOW 4 5736 0 2915 0.1644 0 0813 4.3644 3.6205 2.4095 3 6205 2.7921 1.1856 HIGH 4.7418 0 3132 0.1708 0 0893 4.5545 3.7965 2.5507 3.7965 2 9446 1.2849 WGHTD AVE 4.7349 0 3123 0.1704 0 0889 4.5463 3.7889 2.5446 3.7889 2 9380 1.2806

Gulf LOW 1 3350 0.1593 0.0960 0 0442 1.1831 0.7338 0.4405 0.7338 0.4941 0.2695 HIGH 1 8586 0 2556 0.1332 0 0795 1.7895 1.2541 0.8507 1 2541 0 9319 0.5867 WGHTD AVE 1.7840 0 2331 0.1282 0 0711 1.6879 1.1669 0.7819 1.1669 0 8585 0.5335

Hamilton LOW 0 5619 0 0539 0.0367 0 0121 0.4363 0.2100 0.1032 0 2100 0.1186 0.0543 HIGH 0.7819 0 0817 0.0529 0 0198 0.6398 0.3452 0.1829 0 3452 0 2097 0.0991 WGHTD AVE 0 6830 0 0687 0.0451 0 0160 0.5457 0.2818 0.1453 0 2818 0.1667 0.0778

Hardee LOW 3 9552 0 2462 0.1417 0 0676 3.7582 3.1081 2.0564 3.1081 2 3885 0.9948 HIGH 4 3828 0 3086 0.1603 0 0912 4.2682 3.5959 2.4655 3 5959 2 8235 1.2993 WGHTD AVE 4 0572 0 2574 0.1459 0 0717 3.8699 3.2110 2.1373 3 2110 2.4766 1.0498

Hendry LOW 4.7473 0 3185 0.1728 0 0920 4.5794 3.8307 2.5940 3 8307 2 9851 1.3274 HIGH 5.4157 0.4135 0.1973 0.1268 5.3453 4.5397 3.1699 4 5397 3 6040 1.7503 WGHTD AVE 5 0792 0 3620 0.1848 0.1071 4.9492 4.1723 2.8698 4.1723 3 2822 1.5263

Hernando LOW 2.7395 0.1576 0.0952 0 0397 2.5436 2.0479 1.3006 2 0479 1 5349 0.5788 HIGH 3 3393 0 2261 0.1195 0 0653 3.2091 2.6704 1.7973 2 6704 2 0730 0.9158 WGHTD AVE 3 0171 0.1952 0.1062 0 0536 2.8640 2.3589 1.5600 2 3589 1 8116 0.7658

Highlands LOW 3.7763 0 2191 0.1328 0 0567 3.5310 2.8800 1.8582 2 8800 2.1800 0.8466 HIGH 4 6097 0 2972 0.1667 0 0837 4.4133 3.6720 2.4547 3 6720 2 8396 1.2188 WGHTD AVE 4 0884 0 2515 0.1463 0 0683 3.8741 3.1930 2.0996 3.1930 2.4446 1.0028

Hillsborough LOW 2 9792 0.1868 0.1062 0 0511 2.8209 2.3206 1.5293 2 3206 1.7786 0.7393 HIGH 4.7113 0.4487 0.1750 0.1490 4.8527 4.2234 3.1255 4 2234 3.4735 1.9529 WGHTD AVE 3 5189 0 2538 0.1281 0 0758 3.4244 2.8727 1.9692 2 8727 2 2545 1.0467

Holmes LOW 1.1945 0.1317 0.0852 0 0343 1.0247 0.6058 0.3406 0 6058 0 3892 0.1886 HIGH 1 5478 0.1909 0.1104 0 0557 1.4169 0.9311 0.5888 0 9311 0 6561 0.3741 WGHTD AVE 1 2842 0.1461 0.0914 0 0396 1.1211 0.6828 0.3987 0 6828 0.4517 0.2322

*Includes contents and A.L.E.

FPHLM V3.0 2008 325 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 Personal Residential -- Owners -- FRAME $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Indian River LOW 4 0612 0 3002 0.1473 0 0883 3.9313 3.2682 2.2086 3 2682 2 5420 1.1593 HIGH 6.4165 0 8883 0.2184 0 3042 7.0856 6.3460 5.0573 6 3460 5.4647 3.6825 WGHTD AVE 4 8783 0.4419 0.1784 0.1430 4.9254 4.2125 3.0270 4 2125 3.4012 1.8144

Jackson LOW 0.7861 0 0770 0.0516 0 0176 0.6192 0.3082 0.1518 0 3082 0.1756 0.0772 HIGH 1 2637 0.1414 0.0897 0 0375 1.0937 0.6570 0.3770 0 6570 0.4289 0.2144 WGHTD AVE 1 0368 0.1086 0.0716 0 0265 0.8565 0.4736 0.2523 0.4736 0 2902 0.1343

Jefferson LOW 0 6811 0 0705 0.0456 0 0170 0.5524 0.2926 0.1551 0 2926 0.1770 0.0858 HIGH 0 8187 0 0894 0.0559 0 0234 0.6816 0.3945 0.2287 0 3945 0 2575 0.1377 WGHTD AVE 0 6932 0 0722 0.0464 0 0175 0.5646 0.3019 0.1614 0 3019 0.1840 0.0900

Lafayette LOW 0.7391 0 0775 0.0498 0 0190 0.6058 0.3281 0.1770 0 3281 0 2018 0.0989 HIGH 0 8787 0 0957 0.0598 0 0242 0.7371 0.4176 0.2325 0.4176 0 2643 0.1326 WGHTD AVE 0 8748 0 0952 0.0596 0 0241 0.7334 0.4151 0.2309 0.4151 0 2626 0.1316

Lake LOW 2.1902 0.1215 0.0755 0 0286 1.9931 1.5725 0.9724 1 5725 1.1593 0.4114 HIGH 3 9921 0 2790 0.1441 0 0819 3.8696 3.2444 2.2114 3 2444 2 5380 1.1550 WGHTD AVE 3 2395 0.1934 0.1143 0 0513 3.0338 2.4767 1.6042 2.4767 1 8788 0.7456

Lee LOW 4.4442 0 3315 0.1662 0 0982 4.3516 3.6643 2.5188 3 6643 2 8806 1.3462 HIGH 6 2974 0.7140 0.2381 0 2448 6.7236 5.9550 4.5703 5 9550 5 0092 3.0669 WGHTD AVE 5 2726 0.4392 0.2003 0.1430 5.3052 4.5639 3.2822 4 5639 3 6881 1.9341

Leon LOW 0 5975 0 0575 0.0389 0 0129 0.4649 0.2249 0.1105 0 2249 0.1272 0.0578 HIGH 0 9848 0.1086 0.0682 0 0276 0.8342 0.4811 0.2677 0.4811 0 3054 0.1499 WGHTD AVE 0 8000 0 0819 0.0538 0 0196 0.6488 0.3491 0.1839 0 3491 0 2111 0.0987

Levy LOW 0 8702 0 0931 0.0603 0 0227 0.7162 0.3881 0.2052 0 3881 0 2347 0.1115 HIGH 1.4524 0.1808 0.1030 0 0519 1.3269 0.8678 0.5470 0 8678 0 6097 0.3479 WGHTD AVE 1.1966 0.1384 0.0846 0 0374 1.0482 0.6436 0.3826 0 6436 0.4308 0.2306

Liberty LOW 1 0071 0.1054 0.0690 0 0255 0.8275 0.4501 0.2343 0.4501 0 2706 0.1215 HIGH 1 0456 0.1122 0.0721 0 0280 0.8728 0.4897 0.2665 0.4897 0 3048 0.1472 WGHTD AVE 1 0102 0.1066 0.0694 0 0260 0.8342 0.4585 0.2416 0.4585 0 2784 0.1272

Madison LOW 0 5980 0 0584 0.0390 0 0134 0.4702 0.2333 0.1171 0 2333 0.1345 0.0619 HIGH 0.7896 0 0836 0.0533 0 0206 0.6509 0.3566 0.1934 0 3566 0 2204 0.1080 WGHTD AVE 0.7347 0 0761 0.0492 0 0183 0.5975 0.3186 0.1685 0 3186 0.1928 0.0920

Manatee LOW 3 5529 0 2649 0.1339 0 0796 3.4766 2.9239 2.0131 2 9239 2 3004 1.0797 HIGH 5.4896 0 6725 0.2002 0 2303 5.9208 5.2511 4.0836 5 2511 4.4525 2.8331 WGHTD AVE 4 0435 0 3385 0.1528 0.1074 4.0429 3.4567 2.4725 3.4567 2.7832 1.4496

Marion LOW 1.7807 0 0936 0.0604 0 0208 1.5846 1.2162 0.7248 1 2162 0 8766 0.2845 HIGH 3.1919 0 2058 0.1130 0 0571 3.0369 2.5079 1.6632 2 5079 1 9296 0.8173 WGHTD AVE 2.7604 0.1615 0.0940 0 0410 2.5604 2.0678 1.3174 2 0678 1 5530 0.5928

Martin LOW 5 5029 0.4668 0.1718 0.1529 5.5300 4.7683 3.3573 4.7683 3 8094 1.9207 HIGH 8 8069 1 2848 0.2545 0.4607 9.9439 9.0835 7.3514 9 0835 7 9084 5.4443 WGHTD AVE 6 3155 0 6358 0.1932 0 2159 6.5592 5.7662 4.2537 5.7662 4.7392 2.6749

Miami-Dade LOW 6 3691 0.7503 0.1869 0 2637 6.8431 6.1184 4.7117 6.1184 5.1637 3.2098 HIGH 12.3543 2.4188 0.3275 0 8765 15 0084 14.1356 12 2773 14.1356 12.8762 10.1047 WGHTD AVE 8 3652 1 2131 0.2408 0.4478 9.4765 8.6752 7.0289 8 6752 7 5597 5.1899

Monroe LOW 7.1985 0.7628 0.2059 0 2879 7.7536 7.0540 5.5083 7 0540 6 0115 3.7381 HIGH 12.1357 2 2055 0.3086 0 8453 14 5620 13.7846 12 0217 13.7846 12.5941 9.8970 WGHTD AVE 9 6682 1 5247 0.2574 0 5603 11.1948 10.4356 8.7298 10.4356 9 2845 6.7193

Nassau LOW 0 5997 0 0617 0.0400 0 0148 0.4799 0.2456 0.1301 0 2456 0.1471 0.0754 HIGH 0 8718 0.1039 0.0586 0 0286 0.7626 0.4645 0.2882 0.4645 0 3193 0.1884 WGHTD AVE 0.7626 0 0844 0.0520 0 0214 0.6392 0.3609 0.2069 0 3609 0 2321 0.1259

*Includes contents and A.L.E.

FPHLM V3.0 2008 326 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 Personal Residential -- Owners -- FRAME $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Okaloosa LOW 1.4958 0.1898 0.1097 0 0567 1.3875 0.9250 0.5953 0 9250 0 6597 0.3870 HIGH 3 0581 0 5280 0.1963 0.1891 3.3402 2.7105 2.1297 2.7105 2 2611 1.6775 WGHTD AVE 2 5120 0 3951 0.1687 0.1373 2.6263 2.0478 1.5481 2 0478 1 6577 1.1778

Okeechobee LOW 4.1892 0 2698 0.1529 0 0740 3.9651 3.2470 2.1249 3 2470 2.4772 1.0210 HIGH 4.4990 0 3091 0.1652 0 0883 4.3178 3.5766 2.3871 3 5766 2.7616 1.2010 WGHTD AVE 4.4508 0 2969 0.1636 0 0835 4.2500 3.5072 2.3226 3 5072 2 6954 1.1460

Orange LOW 2 8607 0.1607 0.0993 0 0395 2.6596 2.1350 1.3431 2.1350 1 5915 0.5808 HIGH 3 9662 0 2783 0.1413 0 0811 3.8348 3.2051 2.1760 3 2051 2 5012 1.1339 WGHTD AVE 3.4115 0 2064 0.1208 0 0553 3.2095 2.6258 1.7080 2 6258 1 9970 0.8003

Osceola LOW 3 2248 0.1823 0.1129 0 0460 2.9954 2.4274 1.5492 2.4274 1 8254 0.6883 HIGH 3 9236 0 2514 0.1386 0 0698 3.7164 3.0706 2.0343 3 0706 2 3612 0.9978 WGHTD AVE 3.4711 0 2057 0.1217 0 0535 3.2536 2.6565 1.7195 2 6565 2 0146 0.7924

Palm Beach LOW 5.7979 0 5339 0.1772 0.1772 5.9093 5.1337 3.6847 5.1337 4.1492 2.2020 HIGH 10.2483 1 6224 0.2896 0 5815 11 8186 10 8977 8.9881 10.8977 9 6036 6.8432 WGHTD AVE 6 9414 0 8469 0.2115 0 2886 7.5089 6.6949 5.1187 6 6949 5 6250 3.4541

Pasco LOW 2 8557 0.1725 0.1024 0 0436 2.7207 2.2343 1.4563 2 2343 1.7160 0.6482 HIGH 3 6290 0 2632 0.1303 0 0800 3.4848 2.9082 1.9709 2 9082 2 2671 1.0520 WGHTD AVE 3 2898 0 2275 0.1189 0 0653 3.1692 2.6347 1.7743 2 6347 2 0455 0.9088

Pinellas LOW 3 0593 0 2173 0.1131 0 0629 2.9494 2.4477 1.6486 2.4477 1 8999 0.8499 HIGH 5 9985 0 8854 0.2147 0 3112 6.7554 6.1095 4.9691 6.1095 5 3293 3.7285 WGHTD AVE 3 6565 0 3158 0.1375 0 0991 3.6580 3.1138 2.2199 3.1138 2 5015 1.3033

Polk LOW 3 3369 0.1916 0.1174 0 0488 3.1180 2.5352 1.6250 2 5352 1 9115 0.7304 HIGH 4 6183 0 3527 0.1686 0.1082 4.5540 3.8624 2.6976 3 8624 3 0664 1.4913 WGHTD AVE 3.7575 0 2352 0.1345 0 0647 3.5712 2.9507 1.9481 2 9507 2 2647 0.9403

Putnam LOW 1 0061 0.1064 0.0691 0 0258 0.8337 0.4582 0.2425 0.4582 0 2793 0.1281 HIGH 1 2420 0.1418 0.0888 0 0377 1.0833 0.6583 0.3824 0 6583 0.4337 0.2210 WGHTD AVE 1.1677 0.1298 0.0824 0 0336 1.0008 0.5903 0.3338 0 5903 0 3803 0.1879

St. Johns LOW 0 8701 0 0908 0.0582 0 0220 0.7110 0.3831 0.2046 0 3831 0 2339 0.1127 HIGH 1 5166 0 2036 0.1045 0 0627 1.4439 1.0024 0.6803 1 0024 0.7454 0.4691 WGHTD AVE 1.1495 0.1362 0.0802 0 0376 1.0139 0.6299 0.3879 0 6299 0.4320 0.2468

St. Lucie LOW 5 0478 0.4061 0.1576 0.1304 5.0205 4.3016 2.9842 4 3016 3.4060 1.6530 HIGH 6.7974 0.7710 0.2046 0 2677 7.2366 6.4350 4.8837 6.4350 5 3820 3.2452 WGHTD AVE 5 8722 0 5532 0.1806 0.1850 6.0163 5.2470 3.7991 5 2470 4 2635 2.3055

Santa Rosa LOW 1.4854 0.1820 0.1075 0 0533 1.3626 0.8990 0.5687 0 8990 0 6342 0.3590 HIGH 3 0990 0 5482 0.1961 0.1978 3.4253 2.8113 2.2319 2 8113 2 3645 1.7743 WGHTD AVE 2 3797 0 3724 0.1591 0.1274 2.4730 1.9157 1.4426 1 9157 1 5454 1.0957

Sarasota LOW 3.4445 0 2571 0.1288 0 0770 3.3643 2.8231 1.9379 2 8231 2 2170 1.0366 HIGH 4 5825 0 3938 0.1728 0.1250 4.6151 3.9581 2.8312 3 9581 3.1878 1.6510 WGHTD AVE 4 0263 0 3264 0.1520 0.1019 4.0083 3.4105 2.4007 3.4105 2.7199 1.3527

Seminole LOW 2 9479 0.1750 0.1049 0 0445 2.7302 2.1906 1.3853 2.1906 1 6371 0.6155 HIGH 3 3624 0 2285 0.1212 0 0641 3.2004 2.6270 1.7350 2 6270 2 0149 0.8614 WGHTD AVE 3.1343 0.1900 0.1123 0 0493 2.9242 2.3643 1.5106 2 3643 1.7782 0.6855

Sumter LOW 2 8209 0.1627 0.0973 0 0411 2.6107 2.1018 1.3326 2.1018 1 5739 0.5926 HIGH 3 3910 0 2080 0.1197 0 0563 3.2008 2.6290 1.7191 2 6290 2 0060 0.8143 WGHTD AVE 2 8682 0.1664 0.1000 0 0427 2.6612 2.1473 1.3661 2.1473 1 6113 0.6121

*Includes contents and A.L.E.

FPHLM V3.0 2008 327 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 Personal Residential -- Owners -- FRAME $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Suwanee LOW 0 6147 0 0597 0.0402 0 0136 0.4818 0.2374 0.1188 0 2374 0.1365 0.0629 HIGH 1 0217 0.1140 0.0711 0 0294 0.8724 0.5105 0.2879 0 5105 0 3277 0.1634 WGHTD AVE 0.7482 0 0765 0.0500 0 0182 0.6053 0.3221 0.1702 0 3221 0.1948 0.0928

Taylor LOW 0.7444 0 0786 0.0502 0 0195 0.6132 0.3357 0.1828 0 3357 0 2081 0.1029 HIGH 1.1551 0.1444 0.0805 0 0409 1.0391 0.6593 0.4141 0 6593 0.4588 0.2703 WGHTD AVE 0 9427 0.1092 0.0646 0 0295 0.8200 0.4990 0.3010 0.4990 0 3367 0.1876

Union LOW 0 6883 0 0672 0.0452 0 0152 0.5397 0.2658 0.1315 0 2658 0.1515 0.0683 HIGH 0 9837 0.1038 0.0668 0 0254 0.8123 0.4470 0.2393 0.4470 0 2745 0.1293 WGHTD AVE 0 9641 0.1015 0.0653 0 0248 0.7945 0.4351 0.2322 0.4351 0 2664 0.1252

Volusia LOW 2 2937 0.1285 0.0792 0 0314 2.1008 1.6711 1.0431 1 6711 1 2393 0.4499 HIGH 4 0644 0 3703 0.1453 0.1198 4.1207 3.5445 2.5709 3 5445 2 8790 1.5577 WGHTD AVE 3.1593 0 2088 0.1123 0 0581 3.0032 2.4758 1.6448 2.4758 1 9063 0.8212

Wakulla LOW 0 8054 0 0839 0.0535 0 0204 0.6569 0.3525 0.1883 0 3525 0 2150 0.1043 HIGH 1 5252 0 2048 0.1072 0 0620 1.4500 1.0029 0.6695 1 0029 0.7373 0.4523 WGHTD AVE 0 9537 0.1005 0.0662 0 0266 0.8021 0.4702 0.2744 0.4702 0 3087 0.1659

Walton LOW 1 3398 0.1586 0.0968 0 0442 1.1868 0.7364 0.4436 0.7364 0.4973 0.2717 HIGH 3.1032 0 5257 0.1975 0.1877 3.3741 2.7359 2.1394 2.7359 2 2770 1.6703 WGHTD AVE 2.1358 0 3102 0.1463 0.1032 2.1360 1.5826 1.1411 1 5826 1 2341 0.8328

Washington LOW 1.1730 0.1281 0.0816 0 0331 0.9955 0.5774 0.3232 0 5774 0 3685 0.1814 HIGH 1 9642 0 2771 0.1396 0 0884 1.9206 1.3738 0.9515 1 3738 1 0376 0.6691 WGHTD AVE 1.4978 0.1816 0.1069 0 0520 1.3569 0.8767 0.5429 0 8767 0 6079 0.3367

STATEWIDE LOW 0 5482 0 0520 0.0360 0 0116 0.4234 0.2013 0.0983 0 2013 0.1130 0.0517 HIGH 12.3543 2.4188 0.3275 0 8765 15 0084 14.1356 12 2773 14.1356 12.8762 10.1047 WGHTD AVE 2 9476 0 2661 0.1267 0 0834 2.8481 2.3279 1.6486 2 3279 1 8468 1.0062

*Includes contents and A.L.E.

FPHLM V3.0 2008 328 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL -Owners -- MASONRY 0% 0% $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Alachua LOW 0 8981 0 0934 0.0628 0 0222 0.7258 0 3773 0.1935 0 3773 0.2216 0.1037 HIGH 1.1107 0.1244 0.0813 0 0316 0.9383 0 5307 0 2889 0 5307 0.3290 0.1604 WGHTD AVE 0 9779 0.1062 0.0700 0 0260 0.8079 0.4378 0 2298 0.4378 0.2628 0.1243

Baker LOW 0 5565 0 0538 0.0367 0 0121 0.4287 0 2004 0 0974 0 2004 0.1109 0.0530 HIGH 0 8502 0 0896 0.0587 0 0215 0.6890 0 3601 0.1870 0 3601 0.2133 0.1028 WGHTD AVE 0 8168 0 0856 0.0562 0 0204 0.6595 0 3418 0.1764 0 3418 0.2013 0.0966

Bay LOW 1 2116 0.1351 0.0875 0 0345 1.0214 0 5762 0 3135 0 5762 0.3570 0.1749 HIGH 3 2644 0 5978 0.2089 0 2201 3.6547 3 0201 2.4271 3 0201 2.5612 1.9541 WGHTD AVE 1 8752 0 2542 0.1405 0 0805 1.7973 1 2514 0 8412 1 2514 0.9219 0.5756

Bradford LOW 0 8707 0 0912 0.0602 0 0215 0.7019 0 3622 0.1835 0 3622 0.2105 0.0970 HIGH 1 0484 0.1151 0.0753 0 0285 0.8729 0.4805 0 2554 0.4805 0.2918 0.1393 WGHTD AVE 0 8971 0 0940 0.0625 0 0222 0.7246 0 3760 0.1914 0 3760 0.2193 0.1017

Brevard LOW 2 6438 0.1804 0.1066 0 0484 2.4577 1 9386 1 2519 1 9386 1.4616 0.6084 HIGH 5 3507 0 6313 0.2286 0 2148 5.5796 4 8622 3.7422 4 8622 4.0899 2.5791 WGHTD AVE 3 3783 0 2508 0.1384 0 0719 3.2236 2 6137 1.7501 2 6137 2.0162 0.9133

Broward LOW 5 2412 0 5070 0.1977 0.1716 5.3590 4 6029 3 3705 4 6029 3.7579 2.1116 HIGH 7.4921 0 9744 0.2787 0 3525 8.2480 7.4005 5 8986 7.4005 6.3736 4.2648 WGHTD AVE 6 0385 0 6477 0.2268 0 2261 6.3393 5 5386 4.1830 5 5386 4.6106 2.7659

Calhoun LOW 1 0114 0.1077 0.0715 0 0262 0.8280 0.4414 0 2286 0.4414 0.2619 0.1224 HIGH 1 2550 0.1475 0.0947 0 0390 1.0832 0 6321 0 3546 0 6321 0.4005 0.2061 WGHTD AVE 1.1183 0.1230 0.0811 0 0306 0.9344 0 5185 0 2752 0 5185 0.3150 0.1488

Charlotte LOW 3 5647 0 2492 0.1498 0 0696 3.4004 2.7697 1 8427 2.7697 2.1296 0.9233 HIGH 5 2802 0 5990 0.2274 0 2048 5.5861 4 8628 3 6994 4 8628 4.0613 2.4757 WGHTD AVE 3 9652 0 3044 0.1701 0 0903 3.8608 3.1935 2.1870 3.1935 2.4990 1.1738

Citrus LOW 2.1466 0.1267 0.0821 0 0303 1.9502 1 5168 0 9425 1 5168 1.1186 0.4091 HIGH 2 8715 0.1895 0.1149 0 0509 2.7018 2.1791 1.4260 2.1791 1.6592 0.6909 WGHTD AVE 2 6318 0.1655 0.1034 0 0425 2.4414 1 9433 1 2477 1 9433 1.4623 0.5801

Clay LOW 0 8356 0 0862 0.0576 0 0203 0.6698 0 3417 0.1729 0 3417 0.1980 0.0923 HIGH 1.1366 0.1276 0.0830 0 0325 0.9612 0 5448 0 2975 0 5448 0.3387 0.1661 WGHTD AVE 0 9253 0 0993 0.0654 0 0243 0.7599 0.4077 0 2150 0.4077 0.2452 0.1181

Collier LOW 4 3130 0 3036 0.1857 0 0859 4.1298 3 3739 2 2539 3 3739 2.6005 1.1391 HIGH 6 0929 0 6475 0.2755 0 2220 6.3891 5 5422 4.1521 5 5422 4.5847 2.6836 WGHTD AVE 4.7928 0 3858 0.2109 0.1185 4.7265 3 9526 2.7539 3 9526 3.1261 1.5305

Columbia LOW 0 5925 0 0576 0.0395 0 0129 0.4573 0 2144 0.1033 0 2144 0.1181 0.0553 HIGH 0 8944 0 0974 0.0627 0 0239 0.7372 0 3978 0 2104 0 3978 0.2397 0.1164 WGHTD AVE 0 8392 0 0895 0.0585 0 0216 0.6834 0 3596 0.1868 0 3596 0.2131 0.1022

De Soto LOW 3 5794 0 2273 0.1439 0 0596 3.3556 2.7034 1.7590 2.7034 2.0516 0.8332 HIGH 3 9941 0 2851 0.1662 0 0820 3.8476 3.1700 2.1453 3.1700 2.4639 1.1128 WGHTD AVE 3.7632 0 2569 0.1539 0 0713 3.5837 2 9231 1 9470 2 9231 2.2498 0.9761

Dixie LOW 0 9163 0.1033 0.0655 0 0264 0.7724 0.4353 0 2405 0.4353 0.2722 0.1386 HIGH 1 5289 0 2075 0.1124 0 0632 1.4458 0 9814 0 6497 0 9814 0.7131 0.4416 WGHTD AVE 1 0175 0.1180 0.0725 0 0309 0.8741 0 5132 0 2943 0 5132 0.3313 0.1751

*Includes contents and A.L.E.

FPHLM V3.0 2008 329 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL -Owners -- MASONRY 0% 0% $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Duval LOW 0 5378 0 0517 0.0360 0 0115 0.4131 0.1914 0 0921 0.1914 0.1050 0.0498 HIGH 1 2377 0.1805 0.0864 0 0581 1.2017 0 8427 0 6009 0 8427 0.6457 0.4476 WGHTD AVE 0 8721 0 0955 0.0615 0 0238 0.7204 0 3901 0 2114 0 3901 0.2388 0.1222

Escambia LOW 1 3955 0.1806 0.1102 0 0540 1.2936 0 8488 0 5362 0 8488 0.5941 0.3456 HIGH 4 0399 0 8185 0.2385 0 3106 4.7592 4.1124 3.4544 4.1124 3.6084 2.8929 WGHTD AVE 2 3686 0 3841 0.1683 0.1331 2.4832 1 9167 1.4414 1 9167 1.5406 1.0997

Flagler LOW 2.4460 0.1530 0.0967 0 0391 2.2624 1.7923 1.1450 1.7923 1.3443 0.5290 HIGH 3.1334 0 2555 0.1317 0 0773 3.0601 2 5241 1.7407 2 5241 1.9826 0.9660 WGHTD AVE 2 5673 0.1796 0.1061 0 0493 2.4186 1 9370 1 2680 1 9370 1.4735 0.6286

Franklin LOW 1 6620 0 2314 0.1260 0 0719 1.5901 1 0908 0.7272 1 0908 0.7960 0.5003 HIGH 2.7181 0.4952 0.1759 0.1805 3.0135 2.4594 1 9699 2.4594 2.0766 1.5924 WGHTD AVE 2.1236 0 3400 0.1492 0.1156 2.1987 1 6729 1 2524 1 6729 1.3381 0.9590

Gadsen LOW 0.7762 0 0789 0.0530 0 0182 0.6165 0 3058 0.1502 0 3058 0.1726 0.0786 HIGH 1 0643 0.1165 0.0764 0 0290 0.8859 0.4876 0 2574 0.4876 0.2947 0.1389 WGHTD AVE 0 8153 0 0846 0.0563 0 0200 0.6541 0 3337 0.1673 0 3337 0.1920 0.0883

Gilchrist LOW 0 8535 0 0916 0.0596 0 0222 0.6976 0 3702 0.1926 0 3702 0.2200 0.1049 HIGH 1.1122 0.1302 0.0826 0 0344 0.9635 0 5696 0 3233 0 5696 0.3658 0.1875 WGHTD AVE 1 0314 0.1176 0.0724 0 0303 0.8767 0 5046 0 2807 0 5046 0.3182 0.1605

Glades LOW 4 0641 0 2583 0.1644 0 0681 3.8123 3 0724 2 0002 3 0724 2.3323 0.9489 HIGH 4 2021 0 2744 0.1708 0 0739 3.9648 3 2110 2.1078 3 2110 2.4498 1.0205 WGHTD AVE 4.1991 0 2740 0.1706 0 0738 3.9614 3 2079 2.1054 3 2079 2.4472 1.0189

Gulf LOW 1 2753 0.1535 0.0960 0 0419 1.1169 0 6692 0 3894 0 6692 0.4366 0.2362 HIGH 1.7403 0 2373 0.1332 0 0727 1.6485 1.1154 0.7316 1.1154 0.8036 0.4944 WGHTD AVE 1 6178 0 2119 0.1234 0 0632 1.5007 0 9914 0 6365 0 9914 0.7015 0.4226

Hamilton LOW 0 5508 0 0536 0.0367 0 0120 0.4251 0.1993 0 0964 0.1993 0.1099 0.0520 HIGH 0.7594 0 0805 0.0529 0 0194 0.6164 0 3226 0.1671 0 3226 0.1907 0.0915 WGHTD AVE 0 6779 0 0699 0.0466 0 0164 0.5410 0 2731 0.1384 0 2731 0.1579 0.0754

Hardee LOW 3 5226 0 2199 0.1417 0 0570 3.2922 2 6455 1.7141 2 6455 2.0025 0.8019 HIGH 3 8618 0 2648 0.1603 0 0741 3.6901 3 0216 2 0199 3 0216 2.3311 1.0184 WGHTD AVE 3 5926 0 2267 0.1452 0 0594 3.3663 2.7118 1.7639 2.7118 2.0575 0.8332

Hendry LOW 4.1988 0 2774 0.1728 0 0757 3.9775 3 2329 2.1372 3 2329 2.4770 1.0494 HIGH 4.7430 0 3471 0.1973 0.1012 4.5853 3.7845 2 5707 3.7845 2.9481 1.3516 WGHTD AVE 4 5332 0 3175 0.1872 0 0896 4.3413 3 5625 2 3952 3 5625 2.7578 1.2270

Hernando LOW 2.4603 0.1455 0.0952 0 0350 2.2542 1.7619 1.1013 1.7619 1.3042 0.4808 HIGH 2 9585 0.1969 0.1195 0 0536 2.7906 2 2548 1.4819 2 2548 1.7211 0.7247 WGHTD AVE 2.7527 0.1814 0.1099 0 0486 2.5828 2 0755 1 3554 2 0755 1.5779 0.6558

Highlands LOW 3 3871 0 2008 0.1328 0 0493 3.1195 2.4717 1 5673 2.4717 1.8465 0.6976 HIGH 4 0899 0 2622 0.1667 0 0697 3.8486 3.1112 2 0335 3.1112 2.3676 0.9721 WGHTD AVE 3 6448 0 2248 0.1459 0 0576 3.3940 2.7164 1.7489 2.7164 2.0482 0.8076

Hillsborough LOW 2 6563 0.1667 0.1062 0 0432 2.4726 1 9748 1 2745 1 9748 1.4907 0.5957 HIGH 4 0479 0 3557 0.1750 0.1142 4.0656 3.4405 2.4658 3.4405 2.7689 1.4588 WGHTD AVE 3.1138 0 2157 0.1287 0 0604 2.9660 2.4137 1 6081 2.4137 1.8575 0.8112

Holmes LOW 1.1490 0.1287 0.0852 0 0331 0.9763 0 5588 0 3059 0 5588 0.3487 0.1695 HIGH 1.4676 0.1820 0.1104 0 0524 1.3263 0 8423 0 5166 0 8423 0.5762 0.3232 WGHTD AVE 1 2191 0.1396 0.0907 0 0372 1.0501 0 6151 0 3470 0 6151 0.3931 0.1995

*Includes contents and A.L.E.

FPHLM V3.0 2008 330 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL -Owners -- MASONRY 0% 0% $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Indian River LOW 3 5991 0 2567 0.1473 0 0718 3.4128 2.7533 1 8145 2.7533 2.1039 0.9095 HIGH 5.4121 0 6607 0.2184 0 2252 5.7796 5 0452 3 8948 5 0452 4.2522 2.7018 WGHTD AVE 4 3088 0 3646 0.1809 0.1142 4.2495 3 5367 2.4741 3 5367 2.8031 1.4151

Jackson LOW 0.7688 0 0766 0.0516 0 0175 0.6020 0 2918 0.1414 0 2918 0.1623 0.0740 HIGH 1 2131 0.1376 0.0897 0 0361 1.0392 0 6041 0 3371 0 6041 0.3828 0.1910 WGHTD AVE 1 0018 0.1068 0.0712 0 0260 0.8213 0.4393 0 2279 0.4393 0.2611 0.1220

Jefferson LOW 0 6625 0 0694 0.0456 0 0166 0.5330 0 2738 0.1418 0 2738 0.1611 0.0791 HIGH 0.7924 0 0865 0.0559 0 0224 0.6537 0 3632 0 2046 0 3632 0.2300 0.1224 WGHTD AVE 0 6716 0 0707 0.0462 0 0170 0.5419 0 2804 0.1462 0 2804 0.1659 0.0820

Lafayette LOW 0.7174 0 0761 0.0498 0 0185 0.5830 0 3060 0.1612 0 3060 0.1830 0.0906 HIGH 0 8493 0 0936 0.0598 0 0235 0.7057 0 3871 0 2100 0 3871 0.2381 0.1199 WGHTD AVE 0 8481 0 0934 0.0597 0 0234 0.7045 0 3864 0 2096 0 3864 0.2376 0.1196

Lake LOW 1 9834 0.1139 0.0755 0 0262 1.7776 1 3588 0 8292 1 3588 0.9906 0.3479 HIGH 3 5237 0 2403 0.1441 0 0668 3.3508 2.7291 1 8141 2.7291 2.0978 0.9068 WGHTD AVE 3 0732 0.1905 0.1214 0 0487 2.8519 2 2747 1.4596 2 2747 1.7114 0.6733

Lee LOW 3 9209 0 2815 0.1662 0 0800 3.7627 3 0792 2 0627 3 0792 2.3775 1.0521 HIGH 5 3424 0 5469 0.2381 0.1836 5.5459 4.7828 3 5505 4.7828 3.9339 2.2551 WGHTD AVE 4 3959 0 3355 0.1900 0.1000 4.2868 3 5553 2.4375 3 5553 2.7845 1.3069

Leon LOW 0 5855 0 0572 0.0389 0 0128 0.4529 0 2135 0.1033 0 2135 0.1179 0.0554 HIGH 0 9499 0.1061 0.0682 0 0268 0.7971 0.4451 0 2413 0.4451 0.2745 0.1353 WGHTD AVE 0.7794 0 0812 0.0542 0 0194 0.6285 0 3285 0.1695 0 3285 0.1936 0.0922

Levy LOW 0 8445 0 0917 0.0603 0 0221 0.6894 0 3622 0.1870 0 3622 0.2129 0.1029 HIGH 1 3781 0.1719 0.1030 0 0486 1.2422 0.7849 0.4795 0.7849 0.5350 0.3001 WGHTD AVE 1.1434 0.1337 0.0849 0 0354 0.9892 0 5823 0 3318 0 5823 0.3745 0.1946

Liberty LOW 0 9775 0.1041 0.0690 0 0251 0.7972 0.4209 0 2145 0.4209 0.2463 0.1129 HIGH 1 0118 0.1105 0.0721 0 0273 0.8377 0.4558 0 2412 0.4558 0.2748 0.1337 WGHTD AVE 0 9797 0.1052 0.0693 0 0256 0.8029 0.4282 0 2207 0.4282 0.2530 0.1176

Madison LOW 0 5847 0 0579 0.0390 0 0132 0.4567 0 2204 0.1086 0 2204 0.1239 0.0586 HIGH 0.7656 0 0821 0.0533 0 0201 0.6257 0 3321 0.1757 0 3321 0.1995 0.0987 WGHTD AVE 0.7183 0 0755 0.0495 0 0181 0.5798 0 3003 0.1555 0 3003 0.1770 0.0857

Manatee LOW 3.1339 0 2248 0.1339 0 0641 3.0051 2.4554 1 6469 2.4554 1.8969 0.8425 HIGH 4 6508 0 5096 0.2002 0.1719 4.8653 4 2004 3.1605 4 2004 3.4830 2.0805 WGHTD AVE 3.4496 0 2686 0.1497 0 0805 3.3645 2.7855 1 9183 2.7855 2.1868 1.0458

Marion LOW 1 6256 0 0898 0.0604 0 0194 1.4258 1 0589 0 6258 1 0589 0.7564 0.2473 HIGH 2 8414 0.1822 0.1130 0 0479 2.6561 2.1299 1 3823 2.1299 1.6134 0.6554 WGHTD AVE 2.4762 0.1473 0.0948 0 0357 2.2600 1.7709 1.1092 1.7709 1.3126 0.4878

Martin LOW 4.4397 0 3188 0.1718 0 0951 4.2669 3 5110 2 3368 3 5110 2.7041 1.1912 HIGH 6 6145 0.7610 0.2545 0 2687 7.0449 6.1935 4.7441 6.1935 5.2006 3.2257 WGHTD AVE 4 9672 0.4063 0.1928 0.1292 4.9033 4.1179 2 8602 4.1179 3.2547 1.6060

Miami-Dade LOW 4 8946 0.4577 0.1869 0.1544 4.9735 4 2558 3 0851 4 2558 3.4531 1.8923 HIGH 8 9491 1.4042 0.3275 0 5162 10 3236 9.4621 7 8946 9.4621 8.3910 6.1510 WGHTD AVE 6 2144 0 6895 0.2382 0 2452 6.5969 5 8063 4.4372 5 8063 4.8697 2.9849

Monroe LOW 6 9457 0 6925 0.2059 0 2640 7.4068 6.7073 5.1270 6.7073 5.6425 3.3324 HIGH 11.1642 1.7971 0.3086 0.7054 13 0137 12.1735 10.3662 12.1735 10.9509 8.2199 WGHTD AVE 9 9364 1 5115 0.2785 0 5738 11.4788 10.6869 8 8611 10.6869 9.4559 6.7135

Nassau LOW 0 5847 0 0605 0.0400 0 0143 0.4637 0 2299 0.1188 0 2299 0.1336 0.0693 HIGH 0 8347 0 0984 0.0586 0 0267 0.7191 0.4219 0 2531 0.4219 0.2807 0.1627 WGHTD AVE 0.7256 0 0795 0.0511 0 0198 0.5976 0 3223 0.1775 0 3223 0.1991 0.1065

*Includes contents and A.L.E.

FPHLM V3.0 2008 331 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL -Owners -- MASONRY 0% 0% $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Okaloosa LOW 1.4135 0.1803 0.1097 0 0529 1.2938 0 8331 0 5192 0 8331 0.5762 0.3322 HIGH 2.7682 0.4661 0.1963 0.1668 2.9696 2 3435 1.7902 2 3435 1.9090 1.3787 WGHTD AVE 2 3013 0 3591 0.1689 0.1244 2.3727 1.7963 1 3191 1.7963 1.4179 0.9819

Okeechobee LOW 3.7485 0 2403 0.1529 0 0623 3.4867 2.7721 1.7777 2.7721 2.0836 0.8255 HIGH 4 0022 0 2695 0.1652 0 0729 3.7698 3 0325 1 9781 3 0325 2.3034 0.9554 WGHTD AVE 3 9659 0 2606 0.1629 0 0691 3.7168 2 9777 1 9282 2 9777 2.2518 0.9144

Orange LOW 2 5640 0.1500 0.0993 0 0352 2.3530 1 8422 1.1424 1 8422 1.3574 0.4870 HIGH 3 5071 0 2399 0.1413 0 0663 3.3260 2 6997 1.7884 2 6997 2.0708 0.8928 WGHTD AVE 3 0309 0.1850 0.1195 0 0467 2.8033 2 2249 1.4179 2 2249 1.6669 0.6435

Osceola LOW 2 9001 0.1688 0.1129 0 0405 2.6545 2 0892 1 3124 2 0892 1.5518 0.5722 HIGH 3 5011 0 2225 0.1386 0 0585 3.2566 2 6070 1 6900 2 6070 1.9736 0.7995 WGHTD AVE 3.1657 0.1923 0.1250 0 0482 2.9300 2 3276 1.4838 2 3276 1.7443 0.6709

Palm Beach LOW 4 6504 0 3571 0.1772 0.1097 4.5233 3.7536 2 5469 3.7536 2.9247 1.3634 HIGH 7 6807 0 9644 0.2896 0 3431 8.3645 7.4535 5 8512 7.4535 6.3574 4.1389 WGHTD AVE 5 5038 0 5395 0.2113 0.1806 5.6388 4 8321 3 5132 4 8321 3.9277 2.1786

Pasco LOW 2 5480 0.1594 0.1024 0 0384 2.3820 1 8980 1 2296 1 8980 1.4350 0.5384 HIGH 3 2170 0 2209 0.1303 0 0640 3.0337 2.4537 1 6219 2.4537 1.8791 0.8147 WGHTD AVE 2 8698 0.1968 0.1177 0 0542 2.7169 2.1925 1.4434 2.1925 1.6747 0.7130

Pinellas LOW 2.7123 0.1877 0.1131 0 0515 2.5691 2 0700 1 3613 2 0700 1.5797 0.6725 HIGH 4 9907 0 6496 0.2147 0 2286 5.4343 4.7935 3.7745 4.7935 4.0907 2.6991 WGHTD AVE 3 0883 0 2414 0.1328 0 0718 2.9964 2.4640 1 6877 2.4640 1.9273 0.9181

Polk LOW 2 9858 0.1766 0.1174 0 0429 2.7582 2.1783 1 3732 2.1783 1.6215 0.6042 HIGH 4 0456 0 2959 0.1686 0 0863 3.9067 3 2192 2.1870 3 2192 2.5075 1.1511 WGHTD AVE 3 3524 0 2101 0.1347 0 0547 3.1332 2 5147 1 6261 2 5147 1.9011 0.7591

Putnam LOW 0 9722 0.1048 0.0691 0 0253 0.8019 0.4275 0 2212 0.4275 0.2535 0.1182 HIGH 1.1906 0.1376 0.0888 0 0362 1.0275 0 6041 0 3410 0 6041 0.3863 0.1959 WGHTD AVE 1.1266 0.1272 0.0828 0 0327 0.9574 0 5479 0 3021 0 5479 0.3434 0.1701

St. Johns LOW 0 8450 0 0892 0.0582 0 0215 0.6848 0 3577 0.1866 0 3577 0.2124 0.1036 HIGH 1.4253 0.1904 0.1045 0 0579 1.3364 0 8967 0 5899 0 8967 0.6478 0.3990 WGHTD AVE 1.1720 0.1442 0.0856 0 0407 1.0416 0 6454 0 3975 0 6454 0.4404 0.2567

St. Lucie LOW 4.1105 0 2844 0.1576 0 0828 3.9190 3 2053 2.1085 3 2053 2.4512 1.0449 HIGH 5 2937 0.4826 0.2046 0.1596 5.3435 4 5490 3 2550 4 5490 3.6614 1.9493 WGHTD AVE 4 5476 0 3429 0.1758 0.1048 4.4116 3 6560 2.4713 3 6560 2.8422 1.3078

Santa Rosa LOW 1.4073 0.1735 0.1075 0 0501 1.2746 0 8128 0.4985 0 8128 0.5566 0.3098 HIGH 2.7953 0.4805 0.1961 0.1736 3.0333 2.4228 1 8709 2.4228 1.9909 1.4539 WGHTD AVE 2 2484 0 3515 0.1624 0.1204 2.3151 1.7535 1 2915 1.7535 1.3869 0.9657

Sarasota LOW 3 0411 0 2184 0.1288 0 0621 2.9103 2 3721 1 5865 2 3721 1.8293 0.8096 HIGH 3 9941 0 3215 0.1728 0 0977 3.9310 3 2779 2 2765 3 2779 2.5874 1.2594 WGHTD AVE 3 5850 0 2761 0.1543 0 0822 3.4939 2 8918 1 9813 2 8918 2.2636 1.0633

Seminole LOW 2 6629 0.1608 0.1049 0 0390 2.4279 1 8907 1.1774 1 8907 1.3957 0.5127 HIGH 3 0028 0 2006 0.1212 0 0534 2.8050 2 2344 1.4438 2 2344 1.6869 0.6900 WGHTD AVE 2 8289 0.1736 0.1126 0 0429 2.5979 2 0398 1 2821 2 0398 1.5147 0.5686

Sumter LOW 2 5392 0.1500 0.0973 0 0361 2.3138 1 8072 1.1277 1 8072 1.3364 0.4916 HIGH 3 0277 0.1871 0.1197 0 0478 2.8112 2 2424 1.4375 2 2424 1.6862 0.6601 WGHTD AVE 2 6315 0.1569 0.1027 0 0387 2.4098 1 8920 1.1884 1 8920 1.4048 0.5252

*Includes contents and A.L.E.

FPHLM V3.0 2008 332 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL -Owners -- MASONRY 0% 0% $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Suwanee LOW 0 6014 0 0592 0.0402 0 0135 0.4684 0 2246 0.1104 0 2246 0.1260 0.0597 HIGH 0 9839 0.1112 0.0711 0 0284 0.8319 0.4711 0 2587 0.4711 0.2937 0.1466 WGHTD AVE 0.7221 0 0748 0.0498 0 0178 0.5797 0 2984 0.1539 0 2984 0.1753 0.0847

Taylor LOW 0.7218 0 0772 0.0502 0 0189 0.5894 0 3125 0.1660 0 3125 0.1883 0.0939 HIGH 1 0996 0.1369 0.0805 0 0382 0.9748 0 5963 0 3621 0 5963 0.4016 0.2328 WGHTD AVE 0 8792 0.1021 0.0627 0 0272 0.7546 0.4411 0 2567 0.4411 0.2874 0.1573

Union LOW 0 6736 0 0668 0.0452 0 0151 0.5250 0 2517 0.1225 0 2517 0.1401 0.0653 HIGH 0 9535 0.1020 0.0668 0 0248 0.7807 0.4164 0 2177 0.4164 0.2486 0.1185 WGHTD AVE 0 9328 0 0995 0.0652 0 0241 0.7620 0.4043 0 2106 0.4043 0.2406 0.1145

Volusia LOW 2 0724 0.1199 0.0792 0 0280 1.8694 1.4417 0 8872 1.4417 1.0567 0.3773 HIGH 3 5267 0 2979 0.1453 0 0926 3.4874 2 9148 2 0510 2 9148 2.3192 1.1796 WGHTD AVE 2 8231 0.1831 0.1125 0 0479 2.6321 2.1030 1 3619 2.1030 1.5904 0.6490

Wakulla LOW 0.7826 0 0826 0.0535 0 0199 0.6331 0 3294 0.1720 0 3294 0.1955 0.0960 HIGH 1.4338 0.1907 0.1072 0 0570 1.3414 0 8961 0 5791 0 8961 0.6393 0.3833 WGHTD AVE 0 8876 0 0951 0.0632 0 0243 0.7351 0.4092 0 2281 0.4092 0.2569 0.1349

Walton LOW 1 2795 0.1529 0.0968 0 0419 1.1202 0 6714 0 3920 0 6714 0.4393 0.2378 HIGH 2 8183 0.4663 0.1975 0.1666 3.0121 2 3774 1 8096 2 3774 1.9343 1.3818 WGHTD AVE 1 9463 0 2757 0.1433 0 0891 1.9026 1 3589 0 9468 1 3589 1.0281 0.6744

Washington LOW 1.1307 0.1253 0.0816 0 0320 0.9505 0 5337 0 2910 0 5337 0.3309 0.1633 HIGH 1 8330 0 2569 0.1396 0 0809 1.7641 1 2198 0 8179 1 2198 0.8945 0.5636 WGHTD AVE 1.4164 0.1726 0.1063 0 0488 1.2659 0.7893 0.4736 0.7893 0.5306 0.2897

STATEWIDE LOW 0 5378 0 0517 0.0360 0 0115 0.4131 0.1914 0 0921 0.1914 0.1050 0.0498 HIGH 11.1642 1.7971 0.3275 0.7054 13 0137 12.1735 10.3662 12.1735 10.9509 8.2199 WGHTD AVE 4 0595 0 3434 0.1635 0.1060 3.9916 3 3458 2 3702 3 3458 2.6721 1.3936

*Includes contents and A.L.E.

FPHLM V3.0 2008 333 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - MOBILE HOMES $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Alachua LOW 3.4893 0.3201 0.0628 0.1119 3 2369 2 8128 1 8910 3 2369 2 8128 1 8910 HIGH 4.8586 0.5444 0.0813 0.2005 4 8072 4 2833 3.1079 4 8072 4 2833 3.1079 WGHTD AVE 4.0766 0.4063 0.0708 0.1452 3 8960 3.4247 2 3770 3 8960 3.4247 2 3770

Baker LOW 1.8391 0.1272 0.0367 0.0384 1 5540 1 2934 0.7501 1 5540 1 2934 0.7501 HIGH 3.2685 0.2974 0.0587 0.1028 3 0219 2 6208 1.7473 3 0219 2 6208 1.7473 WGHTD AVE 2.8496 0.2471 0.0524 0.0836 2 5917 2 2316 1.4538 2 5917 2 2316 1.4538

Bay LOW 5.2362 0.5717 0.0875 0.2111 5.1479 4 5763 3 2946 5.1479 4 5763 3 2946 HIGH 17.2773 3.9429 0.2089 1.5389 21.5975 20.6605 18.4348 21.5975 20.6605 18.4348 WGHTD AVE 8.5448 1.3301 0.1275 0.5256 9.4073 8 6989 7 0618 9.4073 8 6989 7 0618

Bradford LOW 3.3309 0.2876 0.0602 0.0981 3 0475 2 6312 1.7204 3 0475 2 6312 1.7204 HIGH 4.3992 0.4513 0.0753 0.1630 4 2444 3.7462 2 6346 4 2444 3.7462 2 6346 WGHTD AVE 3.4753 0.3074 0.0629 0.1067 3 2038 2.7754 1 8374 3 2038 2.7754 1 8374

Brevard LOW 9.0022 1.3042 0.1066 0.5172 9 8601 9 0540 7.1350 9 8601 9 0540 7.1350 HIGH 23.9266 5.1833 0.2286 2.0672 29.6967 28.3745 25.0296 29.6967 28.3745 25.0296 WGHTD AVE 13.2357 2.2442 0.1452 0.9105 15.2490 14.2443 11.7862 15.2490 14.2443 11.7862

Broward LOW 20.8229 4.2090 0.1977 1.7323 25.4043 24.1290 20.9442 25.4043 24.1290 20.9442 HIGH 32.5370 7.7568 0.2787 3.1837 42.0072 40.5287 36.7255 42.0072 40.5287 36.7255 WGHTD AVE 24.5724 5.2801 0.2243 2.1636 30.6042 29.2445 25.8051 30.6042 29.2445 25.8051

Calhoun LOW 4.0656 0.3831 0.0715 0.1363 3 8256 3 3436 2 2787 3 8256 3 3436 2 2787 HIGH 5.8498 0.7063 0.0947 0.2663 5 9423 5 3469 3 9921 5 9423 5 3469 3 9921 WGHTD AVE 4.5883 0.4642 0.0786 0.1698 4.4203 3 9008 2.7397 4.4203 3 9008 2.7397

Charlotte LOW 14.1266 2.2809 0.1498 0.9301 16.1866 15.1288 12.4919 16.1866 15.1288 12.4919 HIGH 25.8957 5.9680 0.2274 2.4021 32.9927 31.7288 28.4967 32.9927 31.7288 28.4967 WGHTD AVE 16.4020 2.8719 0.1678 1.1674 19.2239 18.0873 15.2403 19.2239 18.0873 15.2403

Citrus LOW 6.3724 0.7080 0.0821 0.2771 6 5483 5 8802 4 2979 6 5483 5 8802 4 2979 HIGH 9.8866 1.4434 0.1149 0.5806 10.9500 10.1147 8 0742 10.9500 10.1147 8 0742 WGHTD AVE 8.6626 1.1647 0.1043 0.4675 9 3756 8 5885 6 6846 9 3756 8 5885 6 6846

Clay LOW 3.1530 0.2664 0.0576 0.0899 2 8712 2.4692 1 5956 2 8712 2.4692 1 5956 HIGH 4.9517 0.5487 0.0830 0.2020 4 8889 4 3534 3.1515 4 8889 4 3534 3.1515 WGHTD AVE 3.6516 0.3447 0.0649 0.1214 3.4270 2 9913 2 0335 3.4270 2 9913 2 0335

Collier LOW 17.9400 3.0513 0.1857 1.2496 20.8667 19.5916 16.4103 20.8667 19.5916 16.4103 HIGH 31.9900 7.4602 0.2755 3.0191 40.9841 39.4831 35.6161 40.9841 39.4831 35.6161 WGHTD AVE 22.4907 4.4424 0.2153 1.8192 27.3616 26.0222 22.6232 27.3616 26.0222 22.6232

Columbia LOW 1.9718 0.1339 0.0395 0.0403 1 6663 1 3870 0 8014 1 6663 1 3870 0 8014 HIGH 3.5893 0.3469 0.0627 0.1221 3 3913 2 9668 2 0261 3 3913 2 9668 2 0261 WGHTD AVE 3.3016 0.3046 0.0591 0.1056 3 0714 2 6683 1.7830 3 0714 2 6683 1.7830

De Soto LOW 12.4117 1.7792 0.1439 0.7219 13.7162 12.6725 10.1194 13.7162 12.6725 10.1194 HIGH 15.4563 2.6339 0.1662 1.0725 17.9350 16.8283 14.0811 17.9350 16.8283 14.0811 WGHTD AVE 13.5716 2.1285 0.1519 0.8646 15.3656 14.3066 11.7028 15.3656 14.3066 11.7028

Dixie LOW 3.8562 0.4185 0.0655 0.1514 3.7659 3 3375 2 3803 3.7659 3 3375 2 3803 HIGH 7.6698 1.2308 0.1124 0.4721 8.4824 7 8473 6 3801 8.4824 7 8473 6 3801 WGHTD AVE 4.2783 0.5027 0.0711 0.1782 4 2723 3 8199 2 8034 4 2723 3 8199 2 8034

*Includes contents and A.L.E.

FPHLM V3.0 2008 334 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - MOBILE HOMES $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Duval LOW 1.7621 0.1183 0.0360 0.0352 1.4781 1 2257 0.7003 1.4781 1 2257 0.7003 HIGH 6.0189 1.0758 0.0864 0.4102 6 8182 6 3439 5 2725 6 8182 6 3439 5 2725 WGHTD AVE 3.1311 0.3019 0.0554 0.1048 2 9260 2 5501 1.7321 2 9260 2 5501 1.7321

Escambia LOW 7.5296 1.2235 0.1102 0.4789 8 3810 7.7729 6 3769 8 3810 7.7729 6 3769 HIGH 20.9183 5.1756 0.2385 2.0145 26.9483 25.9655 23.6067 26.9483 25.9655 23.6067 WGHTD AVE 12.0590 2.4027 0.1563 0.9322 14.3583 13.5726 11.7286 14.3583 13.5726 11.7286

Flagler LOW 8.0097 1.0771 0.0967 0.4304 8 6467 7 9094 6.1397 8 6467 7 9094 6.1397 HIGH 12.5353 2.2886 0.1317 0.9214 14.7705 13.8992 11.7523 14.7705 13.8992 11.7523 WGHTD AVE 10.0476 1.6675 0.1102 0.6681 11.4696 10.6710 8.7285 11.4696 10.6710 8.7285

Franklin LOW 8.4490 1.3899 0.1260 0.5392 9.4244 8.7471 7.1881 9.4244 8.7471 7.1881 HIGH 13.7937 3.0870 0.1759 1.2011 17.0621 16.2766 14.4457 17.0621 16.2766 14.4457 WGHTD AVE 11.5853 2.3562 0.1555 0.9036 13.8523 13.1095 11.3867 13.8523 13.1095 11.3867

Gadsen LOW 2.8224 0.2142 0.0530 0.0702 2.4934 2.1199 1 3135 2.4934 2.1199 1 3135 HIGH 4.4688 0.4477 0.0764 0.1625 4 2978 3.7894 2 6509 4 2978 3.7894 2 6509 WGHTD AVE 3.0810 0.2555 0.0567 0.0879 2.7947 2.4032 1 5479 2.7947 2.4032 1 5479

Gilchrist LOW 3.3646 0.3117 0.0596 0.1085 3.1409 2.7344 1 8380 3.1409 2.7344 1 8380 HIGH 5.0621 0.6116 0.0826 0.2279 5.1345 4 6170 3.4400 5.1345 4 6170 3.4400 WGHTD AVE 4.5100 0.5138 0.0741 0.1882 4.4808 3 9997 2 9147 4.4808 3 9997 2 9147

Glades LOW 14.3247 2.1039 0.1644 0.8545 15.9263 14.7406 11.8411 15.9263 14.7406 11.8411 HIGH 15.0953 2.2981 0.1708 0.9328 16.9467 15.7330 12.7539 16.9467 15.7330 12.7539 WGHTD AVE 15.0727 2.2923 0.1706 0.9307 16.9168 15.7040 12.7272 16.9168 15.7040 12.7272

Gulf LOW 5.8431 0.7533 0.0960 0.2843 6 0106 5.4344 4.1354 6 0106 5.4344 4.1354 HIGH 8.8832 1.4410 0.1332 0.5597 9 8648 9.1447 7.4899 9 8648 9.1447 7.4899 WGHTD AVE 6.8695 0.9711 0.1122 0.3795 7 3071 6 6825 5 2638 7 3071 6 6825 5 2638

Hamilton LOW 1.8481 0.1319 0.0367 0.0405 1 5743 1 3142 0.7689 1 5743 1 3142 0.7689 HIGH 2.9284 0.2656 0.0529 0.0914 2.7073 2 3468 1 5577 2.7073 2 3468 1 5577 WGHTD AVE 2.3615 0.1922 0.0445 0.0631 2.1027 1.7938 1.1313 2.1027 1.7938 1.1313

Hardee LOW 12.1542 1.7206 0.1417 0.6982 13.3855 12.3535 9 8337 13.3855 12.3535 9 8337 HIGH 14.6185 2.3969 0.1603 0.9772 16.7778 15.6898 12.9986 16.7778 15.6898 12.9986 WGHTD AVE 12.8278 1.8722 0.1483 0.7611 14.2525 13.1918 10.5946 14.2525 13.1918 10.5946

Hendry LOW 16.1157 2.5394 0.1728 1.0373 18.3319 17.0911 14.0191 18.3319 17.0911 14.0191 HIGH 19.3221 3.3926 0.1973 1.3891 22.6784 21.3495 18.0246 22.6784 21.3495 18.0246 WGHTD AVE 18.2374 3.0849 0.1893 1.2617 21.1627 19.8478 16.5699 21.1627 19.8478 16.5699

Hernando LOW 7.4898 0.8538 0.0952 0.3370 7.7615 6 9989 5.1855 7.7615 6 9989 5.1855 HIGH 10.4789 1.5958 0.1195 0.6435 11.7352 10.8783 8.7848 11.7352 10.8783 8.7848 WGHTD AVE 9.0173 1.2628 0.1065 0.5067 9 8584 9 0628 7.1340 9 8584 9 0628 7.1340

Highlands LOW 10.8109 1.3336 0.1328 0.5358 11.4818 10.4628 8 0017 11.4818 10.4628 8 0017 HIGH 14.6558 2.2028 0.1667 0.8964 16.4055 15.2191 12.3084 16.4055 15.2191 12.3084 WGHTD AVE 11.9863 1.6042 0.1428 0.6475 12.9997 11.9365 9 3526 12.9997 11.9365 9 3526

Hillsborough LOW 8.9868 1.2550 0.1062 0.5057 9 8300 9 0441 7.1405 9 8300 9 0441 7.1405 HIGH 17.9157 3.6970 0.1750 1.4990 21.9970 20.9594 18.3620 21.9970 20.9594 18.3620 WGHTD AVE 12.1898 1.9922 0.1346 0.8086 13.9464 13.0236 10.7537 13.9464 13.0236 10.7537

Holmes LOW 5.1279 0.5753 0.0852 0.2146 5.1088 4 5581 3 3228 5.1088 4 5581 3 3228 HIGH 7.2471 1.0553 0.1104 0.4066 7.7752 7.1250 5 6366 7.7752 7.1250 5 6366 WGHTD AVE 5.7186 0.7140 0.0924 0.2668 5 8486 5 2772 3 9860 5 8486 5 2772 3 9860

*Includes contents and A.L.E.

FPHLM V3.0 2008 335 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - MOBILE HOMES $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Indian River LOW 13.1099 2.0695 0.1473 0.8291 14.7953 13.7342 11.1389 14.7953 13.7342 11.1389 HIGH 23.3504 5.2962 0.2184 2.1087 29.4541 28.2248 25.1332 29.4541 28.2248 25.1332 WGHTD AVE 15.8122 2.8250 0.1681 1.1293 18.5095 17.3759 14.5742 18.5095 17.3759 14.5742

Jackson LOW 2.6931 0.1912 0.0516 0.0605 2 3317 1 9640 1.1761 2 3317 1 9640 1.1761 HIGH 5.4955 0.6545 0.0897 0.2468 5 5497 4 9853 3.7144 5 5497 4 9853 3.7144 WGHTD AVE 4.2911 0.4175 0.0746 0.1507 4 0840 3 5880 2.4855 4 0840 3 5880 2.4855

Jefferson LOW 2.4823 0.2228 0.0456 0.0755 2 2702 1 9570 1 2782 2 2702 1 9570 1 2782 HIGH 3.1736 0.3496 0.0559 0.1258 3 0756 2.7215 1 9398 3 0756 2.7215 1 9398 WGHTD AVE 2.5718 0.2375 0.0468 0.0819 2 3728 2 0536 1 3600 2 3728 2 0536 1 3600

Lafayette LOW 2.7609 0.2618 0.0498 0.0907 2 5689 2 2316 1.4978 2 5689 2 2316 1.4978 HIGH 3.4400 0.3508 0.0598 0.1247 3 2906 2 8921 2 0109 3 2906 2 8921 2 0109 WGHTD AVE 3.4341 0.3497 0.0598 0.1243 3 2835 2 8856 2 0058 3 2835 2 8856 2 0058

Lake LOW 5.5881 0.5125 0.0755 0.1946 5 5732 4 9204 3 3547 5 5732 4 9204 3 3547 HIGH 12.9722 2.0765 0.1441 0.8414 14.7548 13.7517 11.2831 14.7548 13.7517 11.2831 WGHTD AVE 9.1390 1.1625 0.1111 0.4665 9.7487 8 8902 6 8212 9.7487 8 8902 6 8212

Lee LOW 15.9529 2.6773 0.1662 1.0916 18.4774 17.3226 14.4337 18.4774 17.3226 14.4337 HIGH 26.7834 5.9981 0.2381 2.4321 33.8690 32.5309 29.1001 33.8690 32.5309 29.1001 WGHTD AVE 18.4587 3.3135 0.1855 1.3494 21.8068 20.5754 17.4759 21.8068 20.5754 17.4759

Leon LOW 1.9680 0.1373 0.0389 0.0419 1 6740 1 3972 0 8150 1 6740 1 3972 0 8150 HIGH 4.0037 0.4221 0.0682 0.1525 3 8892 3.4404 2.4355 3 8892 3.4404 2.4355 WGHTD AVE 3.3981 0.3312 0.0597 0.1162 3 2123 2 8117 1 9259 3 2123 2 8117 1 9259

Levy LOW 3.4000 0.3180 0.0603 0.1113 3.1851 2.7767 1 8745 3.1851 2.7767 1 8745 HIGH 6.6547 0.9356 0.1030 0.3557 7 0618 6.4484 5 0433 7 0618 6.4484 5 0433 WGHTD AVE 5.2627 0.6475 0.0856 0.2451 5 3597 4 8260 3 6179 5 3597 4 8260 3 6179

Liberty LOW 3.8841 0.3463 0.0690 0.1214 3 6111 3.1394 2 0926 3 6111 3.1394 2 0926 HIGH 4.1690 0.4102 0.0721 0.1478 3 9656 3.4794 2 3980 3 9656 3.4794 2 3980 WGHTD AVE 3.9628 0.3682 0.0694 0.1298 3.7241 3 2525 2 2033 3.7241 3 2525 2 2033

Madison LOW 2.0421 0.1591 0.0390 0.0511 1.7859 1 5101 0 9261 1.7859 1 5101 0 9261 HIGH 3.0121 0.2964 0.0533 0.1037 2 8350 2.4739 1 6829 2 8350 2.4739 1 6829 WGHTD AVE 2.5964 0.2345 0.0471 0.0800 2 3791 2 0540 1 3493 2 3791 2 0540 1 3493

Manatee LOW 12.8499 2.1785 0.1339 0.8886 14.9166 13.9940 11.6955 14.9166 13.9940 11.6955 HIGH 22.2970 5.0201 0.2002 2.0201 28.1584 27.0105 24.1040 28.1584 27.0105 24.1040 WGHTD AVE 14.3264 2.5917 0.1450 1.0595 16.9416 15.9785 13.5727 16.9416 15.9785 13.5727

Marion LOW 4.1450 0.3181 0.0604 0.1146 3 8831 3 3451 2.1263 3 8831 3 3451 2.1263 HIGH 9.6459 1.3663 0.1130 0.5497 10.5925 9.7576 7.7274 10.5925 9.7576 7.7274 WGHTD AVE 7.2224 0.8280 0.0919 0.3272 7.4875 6.7502 4 9963 7.4875 6.7502 4 9963

Martin LOW 16.2410 2.6276 0.1718 1.0893 18.5737 17.3185 14.2215 18.5737 17.3185 14.2215 HIGH 28.4921 6.3549 0.2545 2.6190 35.9627 34.4941 30.7609 35.9627 34.4941 30.7609 WGHTD AVE 19.3072 3.4838 0.1945 1.4304 22.8000 21.4729 18.1630 22.8000 21.4729 18.1630

Miami-Dade LOW 19.7340 3.9745 0.1869 1.6387 24.0547 22.8417 19.8116 24.0547 22.8417 19.8116 HIGH 40.0738 10.2301 0.3275 4.1885 52.9989 51.4323 47.3540 52.9989 51.4323 47.3540 WGHTD AVE 26.9378 6.0263 0.2382 2.4804 34.0614 32.6937 29.1977 34.0614 32.6937 29.1977

Monroe LOW 22.8512 4.7854 0.2059 2.0888 28.5119 27.3154 24.2544 28.5119 27.3154 24.2544 HIGH 38.1638 9.5743 0.3086 4.0939 50.4018 48.9356 45.1247 50.4018 48.9356 45.1247 WGHTD AVE 34.7279 8.5949 0.2838 3.6755 45.6462 44.2592 40.6659 45.6462 44.2592 40.6659

Nassau LOW 2.1029 0.1790 0.0400 0.0591 1 8799 1 6046 1 0188 1 8799 1 6046 1 0188 HIGH 3.4767 0.4283 0.0586 0.1562 3.4811 3.1097 2 2909 3.4811 3.1097 2 2909 WGHTD AVE 2.5506 0.2383 0.0467 0.0817 2 3507 2 0337 1 3508 2 3507 2 0337 1 3508

*Includes contents and A.L.E.

FPHLM V3.0 2008 336 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - MOBILE HOMES $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Okaloosa LOW 7.3593 1.1387 0.1097 0.4429 8 0555 7.4317 6 0043 8 0555 7.4317 6 0043 HIGH 15.7314 3.3964 0.1963 1.3300 19.2973 18.3750 16.1884 19.2973 18.3750 16.1884 WGHTD AVE 10.1699 1.8989 0.1394 0.7594 11.8455 11.1170 9.4232 11.8455 11.1170 9.4232

Okeechobee LOW 13.0692 1.8467 0.1529 0.7448 14.3422 13.2073 10.4545 14.3422 13.2073 10.4545 HIGH 14.5669 2.2165 0.1652 0.8958 16.3252 15.1388 12.2322 16.3252 15.1388 12.2322 WGHTD AVE 14.1239 2.0661 0.1625 0.8342 15.6635 14.4783 11.5839 15.6635 14.4783 11.5839

Orange LOW 7.7020 0.8271 0.0993 0.3250 7 8802 7 0725 5.1500 7 8802 7 0725 5.1500 HIGH 12.5483 1.9502 0.1413 0.7870 14.1381 13.1322 10.6681 14.1381 13.1322 10.6681 WGHTD AVE 9.8746 1.2986 0.1194 0.5232 10.6371 9.7285 7 5316 10.6371 9.7285 7 5316

Osceola LOW 8.9905 1.0486 0.1129 0.4190 9.4092 8 5262 6.4055 9.4092 8 5262 6.4055 HIGH 11.8525 1.6892 0.1386 0.6797 13.0381 12.0153 9 5246 13.0381 12.0153 9 5246 WGHTD AVE 10.7306 1.4191 0.1290 0.5717 11.5903 10.6211 8 2719 11.5903 10.6211 8 2719

Palm Beach LOW 17.1209 2.8696 0.1772 1.1871 19.7717 18.4887 15.3132 19.7717 18.4887 15.3132 HIGH 32.8068 7.4990 0.2896 3.0877 41.8035 40.2129 36.1215 41.8035 40.2129 36.1215 WGHTD AVE 21.0811 4.0939 0.2048 1.6788 25.4146 24.0607 20.6682 25.4146 24.0607 20.6682

Pasco LOW 8.4051 0.9767 0.1024 0.3892 8.7872 7 9575 5 9630 8.7872 7 9575 5 9630 HIGH 11.5147 1.9545 0.1303 0.7887 13.2337 12.3851 10.3014 13.2337 12.3851 10.3014 WGHTD AVE 10.0420 1.4627 0.1163 0.5852 11.1084 10.2595 8.1928 11.1084 10.2595 8.1928

Pinellas LOW 10.3584 1.6123 0.1131 0.6525 11.7062 10.8801 8 8478 11.7062 10.8801 8 8478 HIGH 24.6352 6.0204 0.2147 2.3987 31.8031 30.6902 27.8635 31.8031 30.6902 27.8635 WGHTD AVE 12.0016 2.0675 0.1248 0.8357 13.9506 13.0815 10.9306 13.9506 13.0815 10.9306

Polk LOW 9.4758 1.1197 0.1174 0.4482 10.0020 9 0790 6 8371 10.0020 9 0790 6 8371 HIGH 15.8206 2.7639 0.1686 1.1221 18.4596 17.3389 14.5654 18.4596 17.3389 14.5654 WGHTD AVE 11.6306 1.6499 0.1364 0.6699 12.8137 11.8195 9 3927 12.8137 11.8195 9 3927

Putnam LOW 3.9239 0.3659 0.0691 0.1288 3 6824 3 2145 2.1808 3 6824 3 2145 2.1808 HIGH 5.4544 0.6687 0.0888 0.2507 5 5499 4 9960 3.7440 5 5499 4 9960 3.7440 WGHTD AVE 5.0364 0.5814 0.0839 0.2149 5 0279 4.4952 3 2990 5 0279 4.4952 3 2990

St. Johns LOW 3.2292 0.2938 0.0582 0.1011 2 9806 2 5822 1.7153 2 9806 2 5822 1.7153 HIGH 7.1208 1.1543 0.1045 0.4416 7 8722 7 2773 5 9117 7 8722 7 2773 5 9117 WGHTD AVE 4.9196 0.6387 0.0786 0.2297 5 0378 4 5457 3.4440 5 0378 4 5457 3.4440

St. Lucie LOW 14.5469 2.2291 0.1576 0.9220 16.3868 15.2065 12.3028 16.3868 15.2065 12.3028 HIGH 21.0830 4.0607 0.2046 1.6783 25.3893 24.0421 20.6616 25.3893 24.0421 20.6616 WGHTD AVE 19.6929 3.6819 0.1934 1.5211 23.4882 22.1812 18.9114 23.4882 22.1812 18.9114

Santa Rosa LOW 7.1279 1.0726 0.1075 0.4162 7.7285 7.1075 5 6894 7.7285 7.1075 5 6894 HIGH 15.8590 3.4755 0.1961 1.3610 19.5683 18.6624 16.5074 19.5683 18.6624 16.5074 WGHTD AVE 10.7040 2.0153 0.1456 0.7981 12.5090 11.7584 10.0090 12.5090 11.7584 10.0090

Sarasota LOW 12.2634 2.0492 0.1288 0.8335 14.1638 13.2626 11.0232 14.1638 13.2626 11.0232 HIGH 17.5159 3.2816 0.1728 1.3385 20.9599 19.8427 17.0215 20.9599 19.8427 17.0215 WGHTD AVE 15.7350 2.8604 0.1582 1.1685 18.6631 17.6184 14.9889 18.6631 17.6184 14.9889

Seminole LOW 8.3302 0.9722 0.1049 0.3849 8 6862 7 8509 5 8571 8 6862 7 8509 5 8571 HIGH 10.3667 1.4838 0.1212 0.5936 11.3897 10.4845 8 2928 11.3897 10.4845 8 2928 WGHTD AVE 9.1760 1.1396 0.1129 0.4554 9.7359 8 8579 6.7439 9.7359 8 8579 6.7439

Sumter LOW 7.6794 0.8810 0.0973 0.3469 7 9629 7.1782 5 3104 7 9629 7.1782 5 3104 HIGH 10.0531 1.3568 0.1197 0.5454 10.9055 10.0037 7 8143 10.9055 10.0037 7 8143 WGHTD AVE 9.5139 1.2639 0.1142 0.5069 10.2679 9 3972 7 2899 10.2679 9 3972 7 2899

*Includes contents and A.L.E.

FPHLM V3.0 2008 337 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - MOBILE HOMES $0 $0 $0 DEDUCTIBLE $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE APPURTENANT ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS STRUCTURE LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Suwanee LOW 2.0690 0.1557 0.0402 0.0493 1.7930 1 5103 0 9146 1.7930 1 5103 0 9146 HIGH 4.2187 0.4609 0.0711 0.1680 4.1416 3 6786 2 6387 4.1416 3 6786 2 6387 WGHTD AVE 2.6657 0.2352 0.0486 0.0797 2.4309 2 0961 1 3721 2.4309 2 0961 1 3721

Taylor LOW 2.8196 0.2772 0.0502 0.0969 2 6496 2 3109 1 5716 2 6496 2 3109 1 5716 HIGH 5.0648 0.6712 0.0805 0.2508 5 2500 4.7537 3 6283 5 2500 4.7537 3 6283 WGHTD AVE 3.9411 0.4803 0.0663 0.1755 3 9685 3 5583 2 6408 3 9685 3 5583 2 6408

Union LOW 2.3177 0.1656 0.0452 0.0516 1 9951 1 6752 0 9965 1 9951 1 6752 0 9965 HIGH 3.8091 0.3644 0.0668 0.1287 3 5866 3.1342 2.1390 3 5866 3.1342 2.1390 WGHTD AVE 3.6135 0.3377 0.0642 0.1177 3 3757 2 9408 1 9871 3 3757 2 9408 1 9871

Volusia LOW 6.0261 0.6380 0.0792 0.2485 6.1119 5.4590 3 9260 6.1119 5.4590 3 9260 HIGH 14.0758 2.6636 0.1453 1.0719 16.7761 15.8407 13.5230 16.7761 15.8407 13.5230 WGHTD AVE 9.3570 1.3312 0.1098 0.5308 10.2624 9.4390 7.4464 10.2624 9.4390 7.4464

Wakulla LOW 2.9751 0.2730 0.0535 0.0935 2.7436 2 3746 1 5723 2.7436 2 3746 1 5723 HIGH 7.2100 1.1180 0.1072 0.4288 7 9016 7 2911 5 8770 7 9016 7 2911 5 8770 WGHTD AVE 3.2231 0.3180 0.0566 0.1130 3 0375 2 6527 1 8111 3 0375 2 6527 1 8111

Walton LOW 6.1197 0.8160 0.0968 0.3114 6 3754 5.7858 4.4533 6 3754 5.7858 4.4533 HIGH 15.7296 3.3690 0.1975 1.3196 19.2433 18.3121 16.1056 19.2433 18.3121 16.1056 WGHTD AVE 7.8016 1.2294 0.1189 0.5006 8 6084 7 9496 6.4393 8 6084 7 9496 6.4393

Washington LOW 4.8510 0.5335 0.0816 0.1967 4.7660 4 2352 3 0514 4.7660 4 2352 3 0514 HIGH 9.8511 1.6849 0.1396 0.6559 11.1525 10.3903 8 6130 11.1525 10.3903 8 6130 WGHTD AVE 6.8181 0.9442 0.1056 0.3611 7 2040 6 5712 5.1253 7 2040 6 5712 5.1253

STATEWIDE LOW 1.7621 0.1183 0.0360 0.0352 1.4781 1 2257 0.7003 1.4781 1 2257 0.7003 HIGH 40.0738 10.2301 0.3275 4.1885 52.9989 51.4323 47.3540 52.9989 51.4323 47.3540 WGHTD AVE 11.3798 1.8482 0.1302 0.7093 12.9332 12.0404 9 8643 12.9332 12.0404 9 8643

*Includes contents and A.L.E.

FPHLM V3.0 2008 338 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - Renters -- FRAME $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Alachua LOW 0.1893 0 0454 0.0593 0.0543 0.0514 0.0633 0 0593 0 0533 HIGH 0.2544 0 0652 0.0952 0.0868 0.0807 0.1017 0 0952 0 0849 WGHTD AVE 0.2177 0 0537 0.0739 0.0675 0.0634 0.0789 0 0739 0 0661

Baker LOW 0.1084 0 0243 0.0293 0.0272 0.0263 0.0311 0 0293 0 0268 HIGH 0.1822 0 0441 0.0603 0.0555 0.0524 0.0642 0 0603 0 0544 WGHTD AVE 0.1650 0 0412 0.0533 0.0490 0.0464 0.0566 0 0533 0 0481

Bay LOW 0.2763 0 0713 0.1045 0.0955 0.0888 0.1114 0.1045 0 0934 HIGH 1.4189 0 5171 1.4292 1.3434 1.2024 1.4828 1.4292 1 3114 WGHTD AVE 0.5810 0.1798 0.3969 0.3664 0.3285 0.4177 0 3969 0 3567

Bradford LOW 0.1847 0 0437 0.0557 0.0510 0.0486 0.0594 0 0557 0 0501 HIGH 0.2349 0 0587 0.0825 0.0754 0.0706 0.0881 0 0825 0 0739 WGHTD AVE 0.1899 0 0454 0.0584 0.0536 0.0509 0.0623 0 0584 0 0526

Brevard LOW 0.4143 0.1170 0.2361 0.2180 0.1966 0.2487 0 2361 0 2125 HIGH 1.6928 0 5796 1.7223 1.6249 1.4548 1.7822 1.7223 1 5873 WGHTD AVE 0.5769 0.1662 0.3786 0.3507 0.3145 0.3975 0 3786 0 3417

Broward LOW 1.6582 0 5814 1.6702 1.5724 1.4047 1.7306 1 6702 1 5351 HIGH 3.3546 1 2130 3.7890 3.5911 3.2189 3.9072 3.7890 3 5114 WGHTD AVE 2.6202 0.7606 2.8529 2.7002 2.4192 2.9449 2 8529 2 6395

Calhoun LOW 0.2185 0 0536 0.0714 0.0653 0.0616 0.0762 0 0714 0 0640 HIGH 0.3038 0 0815 0.1268 0.1157 0.1067 0.1351 0.1268 0.1130 WGHTD AVE 0.2541 0 0632 0.0892 0.0814 0.0762 0.0953 0 0892 0 0797

Charlotte LOW 0.5810 0.1711 0.3527 0.3241 0.2909 0.3726 0 3527 0 3154 HIGH 1.5938 0 5511 1.5719 1.4740 1.3126 1.6331 1 5719 1.4372 WGHTD AVE 0.9439 0 2711 0.7543 0.7013 0.6253 0.7889 0.7543 0 6830

Citrus LOW 0.2745 0 0688 0.1076 0.0986 0.0912 0.1143 0.1076 0 0964 HIGH 0.4331 0.1226 0.2418 0.2226 0.2009 0.2554 0 2418 0 2169 WGHTD AVE 0.3579 0 0959 0.1729 0.1588 0.1445 0.1831 0.1729 0.1549

Clay LOW 0.1745 0 0412 0.0528 0.0485 0.0461 0.0563 0 0528 0 0476 HIGH 0.2614 0 0671 0.0992 0.0906 0.0842 0.1058 0 0992 0 0886 WGHTD AVE 0.1990 0 0486 0.0679 0.0622 0.0584 0.0723 0 0679 0 0610

Collier LOW 0.7114 0 2127 0.4366 0.3998 0.3574 0.4622 0.4366 0 3885 HIGH 1.7059 0 5954 1.6158 1.5040 1.3307 1.6869 1 6158 1.4632 WGHTD AVE 0.9560 0 3061 0.7204 0.6645 0.5904 0.7576 0.7204 0 6459

Columbia LOW 0.1157 0 0259 0.0305 0.0283 0.0274 0.0324 0 0305 0 0279 HIGH 0.1984 0 0491 0.0693 0.0636 0.0596 0.0737 0 0693 0 0623 WGHTD AVE 0.1774 0 0439 0.0582 0.0534 0.0504 0.0620 0 0582 0 0524

De Soto LOW 0.5124 0.1421 0.2627 0.2404 0.2176 0.2786 0 2627 0 2341 HIGH 0.6743 0 2041 0.4394 0.4043 0.3616 0.4634 0.4394 0 3933 WGHTD AVE 0.5677 0.1646 0.3282 0.3015 0.2713 0.3470 0 3282 0 2935

Dixie LOW 0.2122 0 0548 0.0843 0.0774 0.0718 0.0896 0 0843 0 0757 HIGH 0.4495 0.1386 0.2941 0.2722 0.2454 0.3091 0 2941 0 2654 WGHTD AVE 0.2570 0 0700 0.1153 0.1058 0.0971 0.1224 0.1153 0.1033

*Includes contents and A.L.E.

FPHLM V3.0 2008 339 Form A-6B, Output Ranges LOSS COSTS PER $1,000 (2007 FHCF exposure data) PERSONAL RESIDENTIAL - Renters -- FRAME Rev. 11-1-07 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Duval LOW 0.1040 0 0232 0.0273 0.0254 0.0245 0.0290 0 0273 0 0250 HIGH 0.4036 0.1311 0.3127 0.2919 0.2624 0.3264 0 3127 0 2849 WGHTD AVE 0.2020 0 0511 0.0776 0.0713 0.0664 0.0823 0 0776 0 0698

Escambia LOW 0.3816 0.1160 0.2186 0.1997 0.1796 0.2319 0 2186 0.1941 HIGH 2.0046 0.7474 2.1820 2.0604 1.8471 2.2563 2.1820 2 0132 WGHTD AVE 0.8531 0 2904 0.7345 0.6836 0.6095 0.7675 0.7345 0 6658

Flagler LOW 0.3414 0 0921 0.1656 0.1519 0.1380 0.1755 0.1656 0.1480 HIGH 0.6261 0.1973 0.4735 0.4395 0.3921 0.4959 0.4735 0.4279 WGHTD AVE 0.4462 0.1216 0.2770 0.2558 0.2296 0.2915 0 2770 0 2491

Franklin LOW 0.5018 0.1581 0.3359 0.3107 0.2797 0.3532 0 3359 0 3028 HIGH 1.1751 0.4251 1.1722 1.1026 0.9880 1.2157 1.1722 1 0766 WGHTD AVE 0.6581 0.1811 0.5144 0.4794 0.4304 0.5375 0 5144 0.4675

Gadsen LOW 0.1589 0 0368 0.0446 0.0410 0.0394 0.0475 0 0446 0 0404 HIGH 0.2372 0 0595 0.0822 0.0751 0.0705 0.0878 0 0822 0 0736 WGHTD AVE 0.1694 0 0407 0.0499 0.0459 0.0438 0.0532 0 0499 0 0451

Gilchrist LOW 0.1860 0 0453 0.0618 0.0567 0.0534 0.0658 0 0618 0 0556 HIGH 0.2688 0 0718 0.1149 0.1050 0.0967 0.1222 0.1149 0.1026 WGHTD AVE 0.2553 0 0657 0.1056 0.0966 0.0891 0.1124 0.1056 0 0944

Glades LOW 0.5830 0.1627 0.2991 0.2732 0.2468 0.3177 0 2991 0 2658 HIGH 0.6264 0.1786 0.3430 0.3139 0.2827 0.3635 0 3430 0 3054 WGHTD AVE 0.6264 0.1786 0.3430 0.3139 0.2827 0.3635 0 3430 0 3054

Gulf LOW 0.3186 0 0883 0.1478 0.1353 0.1238 0.1570 0.1478 0.1320 HIGH 0.5111 0.1590 0.3273 0.3019 0.2719 0.3449 0 3273 0 2941 WGHTD AVE 0.4691 0.1348 0.2848 0.2625 0.2368 0.3004 0 2848 0 2557

Hamilton LOW 0.1078 0 0242 0.0287 0.0266 0.0257 0.0305 0 0287 0 0262 HIGH 0.1634 0 0396 0.0536 0.0493 0.0465 0.0571 0 0536 0 0483 WGHTD AVE 0.1522 0 0357 0.0482 0.0443 0.0420 0.0513 0 0482 0 0435

Hardee LOW 0.4924 0.1351 0.2416 0.2205 0.1999 0.2568 0 2416 0 2147 HIGH 0.6172 0.1825 0.3746 0.3437 0.3081 0.3960 0 3746 0 3343 WGHTD AVE 0.5111 0.1415 0.2601 0.2376 0.2149 0.2762 0 2601 0 2313

Hendry LOW 0.6370 0.1840 0.3590 0.3285 0.2951 0.3805 0 3590 0 3194 HIGH 0.8270 0 2536 0.5578 0.5139 0.4587 0.5877 0 5578 0.4998 WGHTD AVE 0.7265 0 2149 0.4494 0.4124 0.3690 0.4749 0.4494 0.4010

Hernando LOW 0.3152 0 0795 0.1253 0.1146 0.1057 0.1333 0.1253 0.1120 HIGH 0.4523 0.1305 0.2691 0.2479 0.2226 0.2838 0 2691 0 2413 WGHTD AVE 0.3796 0.1016 0.1913 0.1756 0.1592 0.2026 0.1913 0.1711

Highlands LOW 0.4381 0.1133 0.1822 0.1661 0.1524 0.1942 0.1822 0.1620 HIGH 0.5943 0.1674 0.3117 0.2846 0.2567 0.3310 0 3117 0 2769 WGHTD AVE 0.4968 0.1359 0.2353 0.2149 0.1953 0.2502 0 2353 0 2093

Hillsborough LOW 0.3736 0.1023 0.1850 0.1693 0.1534 0.1964 0.1850 0.1649 HIGH 0.8974 0 2980 0.7492 0.6947 0.6162 0.7847 0.7492 0 6757 WGHTD AVE 0.5073 0.1485 0.3173 0.2919 0.2614 0.3347 0 3173 0 2840

Holmes LOW 0.2635 0 0686 0.1007 0.0918 0.0853 0.1075 0.1007 0 0898 HIGH 0.3819 0.1114 0.2021 0.1852 0.1679 0.2143 0 2021 0.1804 WGHTD AVE 0.3074 0 0809 0.1368 0.1251 0.1147 0.1455 0.1368 0.1220

*Includes contents and A.L.E.

FPHLM V3.0 2008 340 Form A-6B, Output Ranges LOSS COSTS PER $1,000 (2007 FHCF exposure data) PERSONAL RESIDENTIAL - Renters -- FRAME Rev. 11-1-07 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Indian River LOW 0.6004 0.1766 0.3730 0.3442 0.3088 0.3928 0 3730 0 3352 HIGH 1.7766 0 6085 1.8205 1.7188 1.5396 1.8827 1 8205 1 6794 WGHTD AVE 0.9644 0 2573 0.7722 0.7199 0.6425 0.8062 0.7722 0.7016

Jackson LOW 0.1540 0 0352 0.0416 0.0384 0.0371 0.0443 0 0416 0 0379 HIGH 0.2828 0 0750 0.1146 0.1045 0.0966 0.1222 0.1146 0.1021 WGHTD AVE 0.2160 0 0517 0.0709 0.0648 0.0612 0.0756 0 0709 0 0636

Jefferson LOW 0.1410 0 0340 0.0463 0.0426 0.0402 0.0492 0 0463 0 0419 HIGH 0.1787 0 0467 0.0754 0.0695 0.0642 0.0798 0 0754 0 0680 WGHTD AVE 0.1418 0 0341 0.0468 0.0431 0.0407 0.0497 0 0468 0 0423

Lafayette LOW 0.1550 0 0380 0.0533 0.0490 0.0460 0.0567 0 0533 0 0481 HIGH 0.1914 0 0484 0.0722 0.0663 0.0618 0.0766 0 0722 0 0649 WGHTD AVE 0.1914 0 0484 0.0722 0.0663 0.0618 0.0766 0 0722 0 0649

Lake LOW 0.2430 0 0573 0.0767 0.0704 0.0665 0.0818 0 0767 0 0690 HIGH 0.5581 0.1638 0.3345 0.3072 0.2756 0.3535 0 3345 0 2989 WGHTD AVE 0.4134 0.1114 0.1939 0.1773 0.1614 0.2059 0.1939 0.1728

Lee LOW 0.6630 0.1964 0.3960 0.3626 0.3248 0.4193 0 3960 0 3526 HIGH 1.4280 0.4896 1.3229 1.2332 1.0949 1.3801 1 3229 1 2006 WGHTD AVE 0.9004 0 2622 0.6725 0.6217 0.5537 0.7064 0 6725 0 6048

Leon LOW 0.1150 0 0258 0.0307 0.0284 0.0274 0.0326 0 0307 0 0280 HIGH 0.2171 0 0552 0.0814 0.0745 0.0694 0.0866 0 0814 0 0729 WGHTD AVE 0.1760 0 0436 0.0586 0.0537 0.0506 0.0624 0 0586 0 0527

Levy LOW 0.1862 0 0453 0.0609 0.0558 0.0526 0.0650 0 0609 0 0547 HIGH 0.3616 0.1038 0.1907 0.1752 0.1591 0.2018 0.1907 0.1708 WGHTD AVE 0.2700 0 0681 0.1161 0.1064 0.0979 0.1234 0.1161 0.1039

Liberty LOW 0.2108 0 0510 0.0659 0.0603 0.0572 0.0703 0 0659 0 0592 HIGH 0.2243 0 0559 0.0801 0.0734 0.0686 0.0853 0 0801 0 0719 WGHTD AVE 0.2125 0 0519 0.0683 0.0625 0.0592 0.0729 0 0683 0 0614

Madison LOW 0.1169 0 0268 0.0328 0.0302 0.0290 0.0348 0 0328 0 0297 HIGH 0.1672 0 0413 0.0582 0.0534 0.0500 0.0619 0 0582 0 0523 WGHTD AVE 0.1566 0 0381 0.0515 0.0473 0.0445 0.0548 0 0515 0 0463

Manatee LOW 0.5298 0.1592 0.3365 0.3092 0.2765 0.3553 0 3365 0 3006 HIGH 1.3450 0.4606 1.2995 1.2183 1.0859 1.3504 1 2995 1.1881 WGHTD AVE 0.7128 0 2096 0.5368 0.4970 0.4427 0.5632 0 5368 0.4837

Marion LOW 0.1871 0 0416 0.0525 0.0484 0.0463 0.0558 0 0525 0 0476 HIGH 0.4117 0.1143 0.2152 0.1975 0.1788 0.2278 0 2152 0.1924 WGHTD AVE 0.3083 0 0779 0.1230 0.1127 0.1040 0.1308 0.1230 0.1101

Martin LOW 0.9335 0 3057 0.7377 0.6855 0.6120 0.7720 0.7377 0 6677 HIGH 2.5695 0 9213 2.7726 2.6187 2.3422 2.8660 2.7726 2 5583 WGHTD AVE 1.3221 0.4301 1.1984 1.1210 1.0002 1.2478 1.1984 1 0930

Miami-Dade LOW 1.5006 0 5273 1.4881 1.3977 1.2459 1.5442 1.4881 1 3636 HIGH 4.8376 1.7530 5.7157 5.4433 4.9006 5.8748 5.7157 5 3301 WGHTD AVE 3.0942 0 9106 3.4465 3.2629 2.9218 3.5566 3.4465 3.1894

Monroe LOW 1.5257 0 5757 1.5418 1.4448 1.2885 1.6029 1 5418 1.4088 HIGH 4.4110 1 6907 5.3132 5.0643 4.5728 5.4591 5 3132 4 9613 WGHTD AVE 2.8884 1.1469 3.3055 3.1319 2.8134 3.4101 3 3055 3 0628

Nassau LOW 0.1235 0 0295 0.0406 0.0375 0.0357 0.0432 0 0406 0 0369 HIGH 0.2077 0 0572 0.1058 0.0982 0.0900 0.1113 0.1058 0 0960 WGHTD AVE 0.1767 0 0451 0.0743 0.0686 0.0634 0.0786 0 0743 0 0671

*Includes contents and A.L.E.

FPHLM V3.0 2008 341 Form A-6B, Output Ranges LOSS COSTS PER $1,000 (2007 FHCF exposure data) PERSONAL RESIDENTIAL - Renters -- FRAME Rev. 11-1-07 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Okaloosa LOW 0.3797 0.1134 0.2103 0.1924 0.1736 0.2231 0 2103 0.1872 HIGH 1.0561 0 3781 0.9551 0.8892 0.7912 0.9976 0 9551 0 8658 WGHTD AVE 0.8271 0 2864 0.6925 0.6427 0.5724 0.7251 0 6925 0 6255

Okeechobee LOW 0.5396 0.1480 0.2670 0.2441 0.2212 0.2835 0 2670 0 2377 HIGH 0.6182 0.1766 0.3449 0.3163 0.2848 0.3650 0 3449 0 3079 WGHTD AVE 0.5983 0.1676 0.3198 0.2929 0.2642 0.3388 0 3198 0 2851

Orange LOW 0.3213 0 0790 0.1170 0.1068 0.0993 0.1247 0.1170 0.1045 HIGH 0.5567 0.1622 0.3347 0.3083 0.2771 0.3530 0 3347 0 3002 WGHTD AVE 0.4067 0.1036 0.1817 0.1661 0.1516 0.1931 0.1817 0.1619

Osceola LOW 0.3645 0 0920 0.1407 0.1281 0.1183 0.1502 0.1407 0.1251 HIGH 0.5028 0.1397 0.2615 0.2398 0.2168 0.2770 0 2615 0 2335 WGHTD AVE 0.3967 0.1027 0.1675 0.1528 0.1400 0.1784 0.1675 0.1490

Palm Beach LOW 1.0678 0 3545 0.9017 0.8421 0.7529 0.9404 0 9017 0 8211 HIGH 3.2448 1.1629 3.5998 3.4116 3.0626 3.7128 3 5998 3 3366 WGHTD AVE 2.4526 0 6791 2.6040 2.4625 2.2077 2.6901 2 6040 2.4070

Pasco LOW 0.3449 0 0872 0.1356 0.1237 0.1142 0.1446 0.1356 0.1208 HIGH 0.5265 0.1600 0.3542 0.3270 0.2923 0.3726 0 3542 0 3182 WGHTD AVE 0.4548 0.1323 0.2632 0.2419 0.2176 0.2780 0 2632 0 2355

Pinellas LOW 0.4347 0.1258 0.2561 0.2357 0.2119 0.2704 0 2561 0 2295 HIGH 1.7709 0 6224 1.8666 1.7577 1.5666 1.9331 1 8666 1.7154 WGHTD AVE 0.6196 0.1888 0.4596 0.4258 0.3797 0.4820 0.4596 0.4145

Polk LOW 0.3833 0 0977 0.1500 0.1366 0.1261 0.1602 0.1500 0.1334 HIGH 0.7055 0 2163 0.4760 0.4385 0.3914 0.5015 0.4760 0.4265 WGHTD AVE 0.4649 0.1265 0.2280 0.2084 0.1890 0.2422 0 2280 0 2029

Putnam LOW 0.2127 0 0517 0.0690 0.0631 0.0596 0.0736 0 0690 0 0619 HIGH 0.2837 0 0754 0.1184 0.1080 0.0995 0.1262 0.1184 0.1055 WGHTD AVE 0.2627 0 0688 0.1031 0.0941 0.0871 0.1099 0.1031 0 0919

St. Johns LOW 0.1815 0 0441 0.0610 0.0561 0.0529 0.0648 0 0610 0 0551 HIGH 0.4072 0.1253 0.2587 0.2386 0.2145 0.2725 0 2587 0 2324 WGHTD AVE 0.2911 0 0773 0.1466 0.1349 0.1230 0.1549 0.1466 0.1317

St. Lucie LOW 0.8123 0 2609 0.6113 0.5676 0.5077 0.6404 0 6113 0 5530 HIGH 1.5420 0 5354 1.4888 1.3980 1.2485 1.5456 1.4888 1 3642 WGHTD AVE 1.1839 0.4001 1.0491 0.9810 0.8760 1.0927 1 0491 0 9566

Santa Rosa LOW 0.3640 0.1066 0.1916 0.1750 0.1584 0.2035 0.1916 0.1704 HIGH 1.0965 0 3955 1.0157 0.9471 0.8431 1.0598 1 0157 0 9224 WGHTD AVE 0.7790 0 2535 0.6402 0.5943 0.5298 0.6702 0 6402 0 5786

Sarasota LOW 0.5143 0.1540 0.3247 0.2980 0.2667 0.3431 0 3247 0 2898 HIGH 0.7877 0 2499 0.5900 0.5458 0.4864 0.6195 0 5900 0 5310 WGHTD AVE 0.6392 0.1983 0.4477 0.4129 0.3682 0.4712 0.4477 0.4016

Seminole LOW 0.3499 0 0890 0.1423 0.1302 0.1199 0.1514 0.1423 0.1272 HIGH 0.4571 0.1281 0.2495 0.2297 0.2075 0.2635 0 2495 0 2238 WGHTD AVE 0.3772 0 0976 0.1604 0.1467 0.1345 0.1706 0.1604 0.1431

Sumter LOW 0.3254 0 0822 0.1301 0.1188 0.1094 0.1385 0.1301 0.1160 HIGH 0.4160 0.1127 0.2015 0.1844 0.1675 0.2138 0 2015 0.1797 WGHTD AVE 0.3485 0 0915 0.1488 0.1362 0.1249 0.1581 0.1488 0.1329

*Includes contents and A.L.E.

FPHLM V3.0 2008 342 Form A-6B, Output Ranges LOSS COSTS PER $1,000 (2007 FHCF exposure data) PERSONAL RESIDENTIAL - Renters -- FRAME Rev. 11-1-07 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Suwanee LOW 0.1194 0 0272 0.0333 0.0308 0.0296 0.0354 0 0333 0 0303 HIGH 0.2280 0 0587 0.0887 0.0812 0.0753 0.0943 0 0887 0 0794 WGHTD AVE 0.1445 0 0341 0.0449 0.0413 0.0392 0.0478 0 0449 0 0406

Taylor LOW 0.1573 0 0389 0.0554 0.0509 0.0476 0.0589 0 0554 0 0498 HIGH 0.2888 0 0818 0.1515 0.1398 0.1275 0.1599 0.1515 0.1365 WGHTD AVE 0.2064 0 0555 0.0936 0.0861 0.0791 0.0991 0 0936 0 0841

Union LOW 0.1344 0 0304 0.0366 0.0338 0.0326 0.0388 0 0366 0 0333 HIGH 0.2076 0 0508 0.0696 0.0638 0.0600 0.0741 0 0696 0 0625 WGHTD AVE 0.1848 0 0450 0.0594 0.0545 0.0516 0.0633 0 0594 0 0535

Volusia LOW 0.2570 0 0628 0.0941 0.0861 0.0800 0.1001 0 0941 0 0843 HIGH 0.7406 0 2395 0.6104 0.5689 0.5072 0.6372 0 6104 0 5543 WGHTD AVE 0.4225 0.1153 0.2319 0.2135 0.1928 0.2448 0 2319 0 2080

Wakulla LOW 0.1678 0 0408 0.0564 0.0518 0.0488 0.0599 0 0564 0 0508 HIGH 0.4095 0.1239 0.2524 0.2331 0.2105 0.2659 0 2524 0 2272 WGHTD AVE 0.2006 0 0453 0.0787 0.0724 0.0672 0.0833 0 0787 0 0709

Walton LOW 0.3172 0 0883 0.1474 0.1347 0.1230 0.1567 0.1474 0.1313 HIGH 1.0515 0 3755 0.9446 0.8791 0.7822 0.9869 0 9446 0 8559 WGHTD AVE 0.6672 0.1937 0.4999 0.4630 0.4137 0.5245 0.4999 0.4507

Washington LOW 0.2563 0 0662 0.0973 0.0889 0.0827 0.1037 0 0973 0 0870 HIGH 0.5541 0.1767 0.3726 0.3436 0.3082 0.3925 0 3726 0 3345 WGHTD AVE 0.3643 0.1050 0.1836 0.1680 0.1527 0.1949 0.1836 0.1637

STATEWIDE LOW 0.1040 0 0232 0.0273 0.0254 0.0245 0.0290 0 0273 0 0250 HIGH 4.8376 1.7530 5.7157 5.4433 4.9006 5.8748 5.7157 5 3301 WGHTD AVE 0.8203 0.1398 0.6590 0.6181 0.5546 0.6853 0 6590 0 6035

*Includes contents and A.L.E.

FPHLM V3.0 2008 343 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - Renters -- MASONRY 0% $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Alachua LOW 0.1869 0 0445 0.0565 0.0516 0.0491 0 0604 0 0565 0 0507 HIGH 0.2487 0 0631 0.0884 0.0804 0 0750 0 0946 0 0884 0 0786 WGHTD AVE 0.2131 0 0523 0.0689 0.0628 0 0593 0 0738 0 0689 0 0616

Baker LOW 0.1076 0 0241 0.0285 0.0264 0 0255 0 0303 0 0285 0 0260 HIGH 0.1791 0 0430 0.0566 0.0519 0 0491 0 0603 0 0566 0 0509 WGHTD AVE 0.1720 0 0410 0.0536 0.0492 0.0467 0 0572 0 0536 0 0483

Bay LOW 0.2702 0 0690 0.0970 0.0884 0 0825 0.1037 0 0970 0 0864 HIGH 1.1956 0.4402 1.1377 1.0593 0 9372 1.1875 1.1377 1 0307 WGHTD AVE 0.5353 0.1638 0.3397 0.3116 0 2785 0 3593 0 3397 0 3029

Bradford LOW 0.1824 0 0430 0.0533 0.0488 0 0466 0 0570 0 0533 0 0479 HIGH 0.2303 0 0570 0.0771 0.0702 0 0660 0 0824 0 0771 0 0688 WGHTD AVE 0.1861 0 0434 0.0548 0.0501 0 0478 0 0586 0 0548 0 0493

Brevard LOW 0.3608 0 0968 0.1740 0.1597 0.1447 0.1842 0.1740 0.1556 HIGH 1.2625 0.4295 1.1750 1.0976 0 9722 1 2236 1.1750 1 0688 WGHTD AVE 0.5045 0.1392 0.2877 0.2647 0 2374 0 3037 0 2877 0 2576

Broward LOW 1.0141 0 3432 0.8552 0.7954 0.7051 0 8937 0 8552 0.7740 HIGH 1.9487 0.7049 1.9835 1.8579 1 6449 2 0610 1 9835 1 8098 WGHTD AVE 1.3401 0.4487 1.2299 1.1476 1 0164 1 2818 1 2299 1.1173

Calhoun LOW 0.2153 0 0525 0.0676 0.0617 0 0585 0 0723 0 0676 0 0606 HIGH 0.2951 0 0780 0.1159 0.1054 0 0975 0.1239 0.1159 0.1029 WGHTD AVE 0.2418 0 0589 0.0799 0.0728 0 0685 0 0856 0 0799 0 0713

Charlotte LOW 0.4983 0.1392 0.2560 0.2337 0 2107 0 2719 0 2560 0 2273 HIGH 1.1980 0.4096 1.0715 0.9949 0 8780 1.1204 1 0715 0 9672 WGHTD AVE 0.5949 0.1762 0.3518 0.3220 0 2878 0 3724 0 3518 0 3129

Citrus LOW 0.2534 0 0606 0.0866 0.0793 0 0742 0 0922 0 0866 0 0777 HIGH 0.3790 0.1017 0.1790 0.1638 0.1487 0.1899 0.1790 0.1596 WGHTD AVE 0.3239 0 0842 0.1361 0.1245 0.1141 0.1446 0.1361 0.1215

Clay LOW 0.1725 0 0405 0.0505 0.0462 0 0441 0 0539 0 0505 0 0454 HIGH 0.2553 0 0649 0.0919 0.0837 0 0780 0 0983 0 0919 0 0818 WGHTD AVE 0.1983 0 0483 0.0646 0.0590 0 0556 0 0690 0 0646 0 0578

Collier LOW 0.6071 0.1717 0.3147 0.2862 0 2573 0 3349 0 3147 0 2781 HIGH 1.2950 0.4441 1.1002 1.0140 0 8913 1.1562 1.1002 0 9839 WGHTD AVE 0.7629 0 2326 0.4889 0.4472 0 3973 0 5175 0.4889 0.4341

Columbia LOW 0.1151 0 0257 0.0299 0.0276 0 0268 0 0318 0 0299 0 0273 HIGH 0.1948 0 0478 0.0650 0.0594 0 0559 0 0693 0 0650 0 0583 WGHTD AVE 0.1734 0 0420 0.0542 0.0497 0 0471 0 0579 0 0542 0 0488

De Soto LOW 0.4547 0.1193 0.1973 0.1798 0.1641 0 2101 0.1973 0.1752 HIGH 0.5702 0.1639 0.3153 0.2881 0 2584 0 3344 0 3153 0 2800 WGHTD AVE 0.5048 0.1397 0.2524 0.2306 0 2082 0 2681 0 2524 0 2244

Dixie LOW 0.2066 0 0528 0.0774 0.0708 0 0658 0 0825 0 0774 0 0693 HIGH 0.4151 0.1265 0.2500 0.2297 0 2063 0 2642 0 2500 0 2235 WGHTD AVE 0.2546 0 0675 0.1078 0.0984 0 0905 0.1148 0.1078 0 0961

*Includes contents and A.L.E.

FPHLM V3.0 2008 344 Form A-6B, Output Ranges LOSS COSTS PER $1,000 (2007 FHCF exposure data) PERSONAL RESIDENTIAL - Renters -- MASONRY Rev. 11-1-07 0% $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Duval LOW 0.1034 0 0230 0.0267 0.0248 0 0240 0 0284 0 0267 0 0244 HIGH 0.3611 0.1162 0.2576 0.2385 0 2129 0 2704 0 2576 0 2321 WGHTD AVE 0.1968 0 0491 0.0712 0.0652 0 0608 0 0758 0 0712 0 0638

Escambia LOW 0.3612 0.1079 0.1927 0.1752 0.1577 0 2053 0.1927 0.1703 HIGH 1.6370 0 6213 1.7000 1.5894 1.4059 1.7687 1.7000 1 5474 WGHTD AVE 0.7708 0 2644 0.6299 0.5821 0 5154 0 6614 0 6299 0 5657

Flagler LOW 0.3059 0 0782 0.1258 0.1149 0.1054 0.1338 0.1258 0.1121 HIGH 0.5111 0.1546 0.3330 0.3065 0 2730 0 3510 0 3330 0 2980 WGHTD AVE 0.3559 0 0944 0.1770 0.1621 0.1465 0.1876 0.1770 0.1579

Franklin LOW 0.4629 0.1438 0.2855 0.2621 0 2352 0 3018 0 2855 0 2550 HIGH 0.9904 0 3609 0.9304 0.8668 0.7679 0 9708 0 9304 0 8436 WGHTD AVE 0.8199 0 3128 0.7335 0.6821 0 6051 0.7667 0.7335 0 6638

Gadsen LOW 0.1578 0 0365 0.0433 0.0398 0 0384 0 0462 0 0433 0 0392 HIGH 0.2331 0 0580 0.0772 0.0704 0 0663 0 0827 0 0772 0 0690 WGHTD AVE 0.1666 0 0395 0.0479 0.0439 0 0421 0 0511 0 0479 0 0432

Gilchrist LOW 0.1831 0 0443 0.0583 0.0534 0 0505 0 0623 0 0583 0 0524 HIGH 0.2603 0 0688 0.1045 0.0952 0 0879 0.1116 0.1045 0 0930 WGHTD AVE 0.2323 0 0569 0.0866 0.0790 0 0734 0 0925 0 0866 0 0773

Glades LOW 0.5165 0.1361 0.2239 0.2037 0.1856 0 2388 0 2239 0.1983 HIGH 0.5488 0.1479 0.2536 0.2310 0 2095 0 2700 0 2536 0 2248 WGHTD AVE 0.5488 0.1479 0.2536 0.2310 0 2095 0 2700 0 2536 0 2248

Gulf LOW 0.3070 0 0839 0.1332 0.1214 0.1114 0.1420 0.1332 0.1184 HIGH 0.4746 0.1455 0.2804 0.2568 0 2307 0 2969 0 2804 0 2499 WGHTD AVE 0.4305 0.1236 0.2393 0.2190 0.1974 0 2537 0 2393 0 2132

Hamilton LOW 0.1072 0 0240 0.0280 0.0259 0 0251 0 0297 0 0280 0 0255 HIGH 0.1611 0 0387 0.0507 0.0465 0 0440 0 0541 0 0507 0 0456 WGHTD AVE 0.1539 0 0372 0.0477 0.0438 0 0415 0 0509 0 0477 0 0430

Hardee LOW 0.4399 0.1141 0.1828 0.1663 0.1522 0.1951 0.1828 0.1621 HIGH 0.5296 0.1483 0.2717 0.2478 0 2232 0 2888 0 2717 0 2410 WGHTD AVE 0.4552 0.1195 0.1960 0.1783 0.1627 0 2090 0.1960 0.1737

Hendry LOW 0.5547 0.1514 0.2638 0.2401 0 2172 0 2810 0 2638 0 2336 HIGH 0.6942 0 2024 0.3982 0.3642 0 3258 0.4220 0 3982 0 3539 WGHTD AVE 0.6586 0.1857 0.3613 0.3300 0 2958 0 3833 0 3613 0 3207

Hernando LOW 0.2911 0 0699 0.1001 0.0915 0 0853 0.1067 0.1001 0 0895 HIGH 0.3939 0.1071 0.1962 0.1796 0.1620 0 2080 0.1962 0.1748 WGHTD AVE 0.3617 0 0982 0.1701 0.1556 0.1412 0.1806 0.1701 0.1516

Highlands LOW 0.4017 0 0986 0.1436 0.1308 0.1215 0.1535 0.1436 0.1278 HIGH 0.5243 0.1395 0.2320 0.2110 0.1919 0 2475 0 2320 0 2054 WGHTD AVE 0.4400 0.1115 0.1733 0.1578 0.1451 0.1849 0.1733 0.1539

Hillsborough LOW 0.3335 0 0864 0.1399 0.1275 0.1166 0.1490 0.1399 0.1243 HIGH 0.7114 0 2284 0.5190 0.4769 0.4216 0 5471 0 5190 0.4629 WGHTD AVE 0.4305 0.1202 0.2258 0.2064 0.1856 0 2396 0 2258 0 2007

Holmes LOW 0.2573 0 0663 0.0933 0.0848 0 0790 0 0999 0 0933 0 0829 HIGH 0.3640 0.1048 0.1797 0.1639 0.1487 0.1912 0.1797 0.1596 WGHTD AVE 0.2756 0 0703 0.1053 0.0958 0 0888 0.1126 0.1053 0 0936

*Includes contents and A.L.E.

FPHLM V3.0 2008 345 Form A-6B, Output Ranges LOSS COSTS PER $1,000 (2007 FHCF exposure data) PERSONAL RESIDENTIAL - Renters -- MASONRY Rev. 11-1-07 0% $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Indian River LOW 0.5134 0.1436 0.2702 0.2476 0 2230 0 2861 0 2702 0 2410 HIGH 1.3214 0.4504 1.2406 1.1598 1 0272 1 2912 1 2406 1.1295 WGHTD AVE 0.7650 0 2050 0.5286 0.4882 0.4340 0 5555 0 5286 0.4748

Jackson LOW 0.1533 0 0350 0.0407 0.0375 0 0364 0 0434 0 0407 0 0370 HIGH 0.2751 0 0721 0.1051 0.0956 0 0886 0.1125 0.1051 0 0934 WGHTD AVE 0.2124 0 0519 0.0665 0.0607 0 0576 0 0712 0 0665 0 0596

Jefferson LOW 0.1388 0 0332 0.0437 0.0401 0 0380 0 0465 0 0437 0 0394 HIGH 0.1731 0 0447 0.0684 0.0627 0 0581 0 0726 0 0684 0 0613 WGHTD AVE 0.1410 0 0338 0.0451 0.0415 0 0392 0 0481 0 0451 0 0407

Lafayette LOW 0.1523 0 0370 0.0501 0.0459 0 0433 0 0533 0 0501 0 0450 HIGH 0.1871 0 0469 0.0669 0.0612 0 0572 0 0712 0 0669 0 0600 WGHTD AVE 0.1871 0 0469 0.0669 0.0612 0 0572 0 0712 0 0669 0 0600

Lake LOW 0.2277 0 0524 0.0659 0.0605 0 0578 0 0702 0 0659 0 0595 HIGH 0.4806 0.1336 0.2433 0.2222 0 2003 0 2585 0 2433 0 2161 WGHTD AVE 0.3709 0 0949 0.1481 0.1351 0.1242 0.1578 0.1481 0.1318

Lee LOW 0.5630 0.1599 0.2878 0.2620 0 2361 0 3063 0 2878 0 2547 HIGH 1.0937 0 3671 0.9043 0.8349 0.7364 0 9496 0 9043 0 8108 WGHTD AVE 0.6684 0.1954 0.3989 0.3648 0 3257 0.4226 0 3989 0 3544

Leon LOW 0.1144 0 0257 0.0299 0.0277 0 0268 0 0318 0 0299 0 0273 HIGH 0.2123 0 0535 0.0755 0.0689 0 0644 0 0806 0 0755 0 0674 WGHTD AVE 0.1779 0 0437 0.0575 0.0526 0 0497 0 0614 0 0575 0 0516

Levy LOW 0.1835 0 0443 0.0576 0.0527 0 0499 0 0616 0 0576 0 0517 HIGH 0.3438 0 0973 0.1681 0.1536 0.1395 0.1787 0.1681 0.1496 WGHTD AVE 0.2704 0 0724 0.1104 0.1006 0 0927 0.1178 0.1104 0 0982

Liberty LOW 0.2083 0 0502 0.0630 0.0576 0 0548 0 0674 0 0630 0 0565 HIGH 0.2209 0 0545 0.0747 0.0682 0 0640 0 0797 0 0747 0 0669 WGHTD AVE 0.2101 0 0505 0.0651 0.0595 0 0565 0 0696 0 0651 0 0584

Madison LOW 0.1159 0 0264 0.0316 0.0291 0 0280 0 0337 0 0316 0 0287 HIGH 0.1643 0 0402 0.0546 0.0500 0 0471 0 0582 0 0546 0 0490 WGHTD AVE 0.1463 0 0350 0.0453 0.0416 0 0394 0 0483 0 0453 0 0408

Manatee LOW 0.4496 0.1282 0.2419 0.2207 0.1981 0 2568 0 2419 0 2145 HIGH 1.0191 0 3438 0.8882 0.8247 0.7289 0 9290 0 8882 0 8020 WGHTD AVE 0.5692 0.1572 0.3615 0.3316 0 2951 0 3820 0 3615 0 3221

Marion LOW 0.1796 0 0387 0.0464 0.0430 0 0414 0 0493 0 0464 0 0424 HIGH 0.3643 0 0958 0.1610 0.1471 0.1341 0.1712 0.1610 0.1434 WGHTD AVE 0.2886 0 0704 0.1011 0.0925 0 0862 0.1077 0.1011 0 0905

Martin LOW 0.6376 0.1903 0.3758 0.3454 0 3098 0 3970 0 3758 0 3360 HIGH 1.5221 0 5375 1.4352 1.3395 1.1857 1.4955 1.4352 1 3040 WGHTD AVE 0.8151 0 2611 0.5782 0.5344 0.4760 0 6074 0 5782 0 5199

Miami-Dade LOW 0.9153 0 3087 0.7489 0.6946 0 6151 0.7841 0.7489 0 6755 HIGH 2.8084 1 0323 3.0917 2.9122 2 5860 3.1997 3 0917 2 8408 WGHTD AVE 1.7581 0 5273 1.7295 1.6178 1.4318 1.7989 1.7295 1 5755

Monroe LOW 1.3850 0 5281 1.3518 1.2575 1.1117 1.4116 1 3518 1 2231 HIGH 3.5941 1.4108 4.2207 3.9854 3 5441 4 3605 4 2207 3 8899 WGHTD AVE 2.8601 1 0612 3.2297 3.0391 2 6973 3 3448 3 2297 2 9636

Nassau LOW 0.1211 0 0287 0.0383 0.0354 0 0335 0 0407 0 0383 0 0348 HIGH 0.1968 0 0534 0.0919 0.0847 0 0775 0 0971 0 0919 0 0827 WGHTD AVE 0.1692 0 0414 0.0658 0.0604 0 0560 0 0698 0 0658 0 0591

*Includes contents and A.L.E.

FPHLM V3.0 2008 346 Form A-6B, Output Ranges LOSS COSTS PER $1,000 (2007 FHCF exposure data) PERSONAL RESIDENTIAL - Renters -- MASONRY Rev. 11-1-07 0% $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Okaloosa LOW 0.3606 0.1058 0.1861 0.1694 0.1530 0.1981 0.1861 0.1647 HIGH 0.9322 0 3336 0.7941 0.7337 0 6482 0 8337 0.7941 0.7127 WGHTD AVE 0.7727 0 2703 0.6182 0.5700 0 5046 0 6501 0 6182 0 5537

Okeechobee LOW 0.4805 0.1247 0.2013 0.1833 0.1676 0 2146 0 2013 0.1787 HIGH 0.5389 0.1458 0.2538 0.2315 0 2098 0 2700 0 2538 0 2254 WGHTD AVE 0.5224 0.1385 0.2335 0.2128 0.1936 0 2487 0 2335 0 2073

Orange LOW 0.2999 0 0705 0.0956 0.0874 0 0822 0.1021 0 0956 0 0856 HIGH 0.4798 0.1327 0.2442 0.2235 0 2017 0 2589 0 2442 0 2176 WGHTD AVE 0.3656 0 0909 0.1394 0.1272 0.1174 0.1487 0.1394 0.1242

Osceola LOW 0.3376 0 0810 0.1131 0.1030 0 0963 0.1209 0.1131 0.1008 HIGH 0.4450 0.1171 0.1956 0.1786 0.1628 0 2081 0.1956 0.1740 WGHTD AVE 0.3734 0 0931 0.1385 0.1262 0.1167 0.1479 0.1385 0.1232

Palm Beach LOW 0.7142 0 2193 0.4651 0.4295 0 3842 0.4892 0.4651 0.4181 HIGH 1.9287 0 6863 1.9138 1.7936 1 5917 1 9883 1 9138 1.7479 WGHTD AVE 1.2467 0 3695 1.0884 1.0154 0 9013 1.1349 1 0884 0 9888

Pasco LOW 0.3188 0 0768 0.1086 0.0991 0 0925 0.1160 0.1086 0 0969 HIGH 0.4417 0.1279 0.2528 0.2317 0 2074 0 2675 0 2528 0 2253 WGHTD AVE 0.3938 0.1079 0.1955 0.1788 0.1615 0 2075 0.1955 0.1740

Pinellas LOW 0.3754 0.1031 0.1871 0.1711 0.1546 0.1986 0.1871 0.1665 HIGH 1.2993 0.4572 1.2652 1.1794 1 0396 1 3188 1 2652 1.1472 WGHTD AVE 0.5016 0.1415 0.3156 0.2899 0 2584 0 3331 0 3156 0 2818

Polk LOW 0.3531 0 0857 0.1204 0.1097 0.1024 0.1287 0.1204 0.1073 HIGH 0.5919 0.1726 0.3395 0.3105 0 2777 0 3598 0 3395 0 3017 WGHTD AVE 0.4152 0.1077 0.1719 0.1566 0.1434 0.1834 0.1719 0.1527

Putnam LOW 0.2097 0 0506 0.0653 0.0596 0 0565 0 0699 0 0653 0 0585 HIGH 0.2753 0 0723 0.1081 0.0983 0 0908 0.1156 0.1081 0 0960 WGHTD AVE 0.2564 0 0664 0.0949 0.0863 0 0802 0.1015 0 0949 0 0843

St. Johns LOW 0.1785 0 0430 0.0573 0.0526 0 0497 0 0610 0 0573 0 0516 HIGH 0.3807 0.1158 0.2250 0.2063 0.1851 0 2380 0 2250 0 2006 WGHTD AVE 0.2878 0 0790 0.1398 0.1280 0.1164 0.1483 0.1398 0.1248

St. Lucie LOW 0.5688 0.1656 0.3158 0.2901 0 2613 0 3338 0 3158 0 2824 HIGH 0.9652 0 3192 0.7626 0.7076 0 6284 0.7985 0.7626 0 6885 WGHTD AVE 0.7241 0 2212 0.4788 0.4419 0 3947 0 5038 0.4788 0.4300

Santa Rosa LOW 0.3471 0.1002 0.1704 0.1550 0.1404 0.1817 0.1704 0.1508 HIGH 0.9610 0 3471 0.8394 0.7764 0 6858 0 8804 0 8394 0.7544 WGHTD AVE 0.7541 0 2495 0.6023 0.5558 0.4922 0 6331 0 6023 0 5400

Sarasota LOW 0.4369 0.1242 0.2344 0.2137 0.1921 0 2490 0 2344 0 2078 HIGH 0.6430 0.1954 0.4141 0.3799 0 3382 0.4376 0.4141 0 3691 WGHTD AVE 0.5375 0.1591 0.3246 0.2972 0 2652 0 3436 0 3246 0 2888

Seminole LOW 0.3215 0 0779 0.1128 0.1031 0 0960 0.1203 0.1128 0.1008 HIGH 0.4011 0.1067 0.1854 0.1697 0.1542 0.1967 0.1854 0.1654 WGHTD AVE 0.3462 0 0855 0.1272 0.1161 0.1076 0.1357 0.1272 0.1135

Sumter LOW 0.2999 0 0722 0.1034 0.0944 0 0880 0.1104 0.1034 0 0924 HIGH 0.3742 0 0956 0.1529 0.1395 0.1279 0.1629 0.1529 0.1361 WGHTD AVE 0.3300 0 0851 0.1249 0.1140 0.1054 0.1330 0.1249 0.1114

*Includes contents and A.L.E.

FPHLM V3.0 2008 347 Form A-6B, Output Ranges LOSS COSTS PER $1,000 (2007 FHCF exposure data) PERSONAL RESIDENTIAL - Renters -- MASONRY Rev. 11-1-07 0% $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Suwanee LOW 0.1184 0 0269 0.0323 0.0298 0 0287 0 0343 0 0323 0 0293 HIGH 0.2223 0 0567 0.0818 0.0746 0 0695 0 0873 0 0818 0 0730 WGHTD AVE 0.1482 0 0351 0.0453 0.0416 0 0395 0 0483 0 0453 0 0409

Taylor LOW 0.1544 0 0379 0.0519 0.0476 0 0447 0 0553 0 0519 0 0466 HIGH 0.2739 0 0763 0.1324 0.1215 0.1107 0.1404 0.1324 0.1185 WGHTD AVE 0.1978 0 0526 0.0839 0.0769 0 0707 0 0891 0 0839 0 0751

Union LOW 0.1335 0 0302 0.0356 0.0329 0 0318 0 0379 0 0356 0 0324 HIGH 0.2040 0 0495 0.0652 0.0596 0 0563 0 0697 0 0652 0 0585 WGHTD AVE 0.1827 0 0428 0.0560 0.0513 0 0487 0 0598 0 0560 0 0504

Volusia LOW 0.2397 0 0560 0.0767 0.0703 0 0661 0 0817 0 0767 0 0689 HIGH 0.5958 0.1852 0.4257 0.3933 0 3491 0.4473 0.4257 0 3824 WGHTD AVE 0.3780 0 0959 0.1771 0.1622 0.1472 0.1878 0.1771 0.1580

Wakulla LOW 0.1653 0 0399 0.0532 0.0488 0 0461 0 0567 0 0532 0 0479 HIGH 0.3814 0.1139 0.2165 0.1985 0.1788 0 2293 0 2165 0.1933 WGHTD AVE 0.2045 0 0484 0.0801 0.0735 0 0680 0 0851 0 0801 0 0719

Walton LOW 0.3058 0 0838 0.1330 0.1210 0.1108 0.1419 0.1330 0.1180 HIGH 0.9326 0 3332 0.7905 0.7302 0 6453 0 8299 0.7905 0.7094 WGHTD AVE 0.6235 0.1771 0.4428 0.4075 0 3623 0.4667 0.4428 0 3960

Washington LOW 0.2505 0 0640 0.0902 0.0822 0 0767 0 0964 0 0902 0 0804 HIGH 0.5139 0.1617 0.3209 0.2940 0 2631 0 3395 0 3209 0 2858 WGHTD AVE 0.3485 0 0996 0.1645 0.1499 0.1365 0.1752 0.1645 0.1460

STATEWIDE LOW 0.1034 0 0230 0.0267 0.0248 0 0240 0 0284 0 0267 0 0244 HIGH 3.5941 1.4108 4.2207 3.9854 3 5441 4 3605 4 2207 3 8899 WGHTD AVE 0.8276 0.1918 0.6271 0.5828 0 5184 0 6561 0 6271 0 5674

*Includes contents and A.L.E.

FPHLM V3.0 2008 348 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - Condo Owners -- FRAME $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Alachua LOW 0 0924 0.1893 0.0454 0.1227 0 0636 0 0555 0.1227 0.0636 0.0555 HIGH 0.1153 0 2544 0.0652 0.1782 0.1036 0 0896 0.1782 0.1036 0.0896 WGHTD AVE 0.1006 0 2130 0.0525 0.1425 0 0777 0 0675 0.1425 0.0777 0.0675

Baker LOW 0 0568 0.1084 0.0243 0 0665 0 0309 0 0276 0.0665 0.0309 0.0276 HIGH 0 0876 0.1822 0.0441 0.1206 0 0648 0 0570 0.1206 0.0648 0.0570 WGHTD AVE 0 0876 0.1822 0.0441 0.1206 0 0648 0 0570 0.1206 0.0648 0.0570

Bay LOW 0.1258 0 2763 0.0713 0.1944 0.1136 0 0986 0.1944 0.1136 0.0986 HIGH 0 3655 1.4189 0.5171 1.7850 1 6026 1.4489 1.7850 1.6026 1.4489 WGHTD AVE 0 2345 0 6875 0.2305 0.7288 0 5928 0 5231 0.7288 0.5928 0.5231

Bradford LOW 0 0895 0.1847 0.0437 0.1174 0 0594 0 0520 0.1174 0.0594 0.0520 HIGH 0.1085 0 2349 0.0587 0.1594 0 0892 0 0776 0.1594 0.0892 0.0776 WGHTD AVE 0.1015 0 0000 0.0000 0.1411 0 0758 0 0662 0.1411 0.0758 0.0662

Brevard LOW 0 2959 0.4143 0.1170 0.4272 0 2677 0 2339 0.4272 0.2677 0.2339 HIGH 0 6267 1 6928 0.5796 2 2378 1 9456 1.7673 2.2378 1.9456 1.7673 WGHTD AVE 0 3775 0 5975 0.1785 0 6622 0.4610 0.4055 0.6622 0.4610 0.4055

Broward LOW 0 6805 1 6582 0.5814 2 2054 1 8964 1.7105 2.2054 1.8964 1.7105 HIGH 1 0147 3 3546 1.2130 4 6941 4 2770 3 9116 4.6941 4.2770 3.9116 WGHTD AVE 0.7986 2 2200 0.7932 3 0091 2 6583 2.4131 3.0091 2.6583 2.4131

Calhoun LOW 0.1043 0 2185 0.0536 0.1436 0 0766 0 0668 0.1436 0.0766 0.0668 HIGH 0.1309 0 3038 0.0815 0 2235 0.1385 0.1200 0.2235 0.1385 0.1200 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Charlotte LOW 0.4027 0 5810 0.1711 0 6231 0.4023 0 3481 0.6231 0.4023 0.3481 HIGH 0 6265 1 5938 0.5511 2 0905 1.7845 1 6013 2.0905 1.7845 1.6013 WGHTD AVE 0.4730 1 2221 0.3639 1 3413 1 0722 0 9532 1.3413 1.0722 0.9532

Citrus LOW 0 2382 0 2745 0.0688 0 2525 0.1215 0.1048 0.2525 0.1215 0.1048 HIGH 0 3234 0.4331 0.1226 0.4506 0 2758 0 2390 0.4506 0.2758 0.2390 WGHTD AVE 0 2882 0 3601 0.0969 0 3560 0.1999 0.1728 0.3560 0.1999 0.1728

Clay LOW 0 0859 0.1745 0.0412 0.1112 0 0564 0 0494 0.1112 0.0564 0.0494 HIGH 0.1180 0 2614 0.0671 0.1842 0.1079 0 0936 0.1842 0.1079 0.0936 WGHTD AVE 0 0946 0 2009 0.0498 0.1359 0 0753 0 0657 0.1359 0.0753 0.0657

Collier LOW 0.4888 0.7114 0.2127 0.7691 0.4998 0.4288 0.7691 0.4998 0.4288 HIGH 0.7237 1.7059 0.5954 2 2108 1 8447 1 6303 2.2108 1.8447 1.6303 WGHTD AVE 0 5508 0 9702 0.3086 1.1322 0 8350 0.7265 1.1322 0.8350 0.7265

Columbia LOW 0 0604 0.1157 0.0259 0 0701 0 0321 0 0286 0.0701 0.0321 0.0286 HIGH 0 0924 0.1984 0.0491 0.1341 0 0748 0 0655 0.1341 0.0748 0.0655 WGHTD AVE 0 0800 0.1658 0.0382 0.1059 0 0547 0 0481 0.1059 0.0547 0.0481

De Soto LOW 0.4024 0 5124 0.1421 0 5201 0 3001 0 2575 0.5201 0.3001 0.2575 HIGH 0.4550 0 6743 0.2041 0.7494 0 5034 0.4354 0.7494 0.5034 0.4354 WGHTD AVE 0.4432 0 6277 0.1868 0 6936 0.4552 0 3934 0.6936 0.4552 0.3934

Dixie LOW 0 0951 0 2122 0.0548 0.1527 0 0919 0 0803 0.1527 0.0919 0.0803 HIGH 0.1631 0.4495 0.1386 0.4270 0 3276 0 2893 0.4270 0.3276 0.2893 WGHTD AVE 0.1272 0 3193 0.0870 0 2770 0.1980 0.1741 0.2770 0.1980 0.1741

*Includes contents and A.L.E.

FPHLM V3.0 2008 349 Form A-6B, Output Ranges LOSS COSTS PER $1,000 (2007 FHCF exposure data) PERSONAL RESIDENTIAL - Condo Owners -- FRAME Rev. 11-1-07 $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Duval LOW 0 0548 0.1040 0.0232 0 0630 0 0287 0 0257 0.0630 0.0287 0.0257 HIGH 0.1333 0.4036 0.1311 0.4258 0 3492 0 3125 0.4258 0.3492 0.3125 WGHTD AVE 0 0932 0 2060 0.0523 0.1456 0 0858 0 0753 0.1456 0.0858 0.0753

Escambia LOW 0.1482 0 3816 0.1160 0 3370 0 2432 0 2100 0.3370 0.2432 0.2100 HIGH 0.4621 2 0046 0.7474 2 6557 2.4465 2 2299 2.6557 2.4465 2.2299 WGHTD AVE 0 2822 1 0596 0.3673 1 2305 1 0764 0 9642 1.2305 1.0764 0.9642

Flagler LOW 0 2737 0 3414 0.0921 0 3370 0.1887 0.1624 0.3370 0.1887 0.1624 HIGH 0 3585 0 6261 0.1973 0.7286 0 5388 0.4746 0.7286 0.5388 0.4746 WGHTD AVE 0 3001 0.4833 0.1404 0 5148 0 3509 0 3068 0.5148 0.3509 0.3068

Franklin LOW 0.1780 0 5018 0.1581 0.4830 0 3744 0 3301 0.4830 0.3744 0.3301 HIGH 0 3045 1.1751 0.4251 1.4665 1 3139 1.1899 1.4665 1.3139 1.1899 WGHTD AVE 0 2871 1 0236 0.3729 1 2890 1.1438 1 0340 1.2890 1.1438 1.0340

Gadsen LOW 0 0796 0.1589 0.0368 0 0980 0 0471 0 0416 0.0980 0.0471 0.0416 HIGH 0.1101 0 2372 0.0595 0.1598 0 0888 0 0772 0.1598 0.0888 0.0772 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Gilchrist LOW 0 0880 0.1860 0.0453 0.1230 0 0664 0 0582 0.1230 0.0664 0.0582 HIGH 0.1161 0 2688 0.0718 0 2007 0.1259 0.1093 0.2007 0.1259 0.1093 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Glades LOW 0.4574 0 5830 0.1627 0 5929 0 3426 0 2924 0.5929 0.3426 0.2924 HIGH 0.4742 0 6264 0.1786 0 6515 0 3928 0 3366 0.6515 0.3928 0.3366 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Hardee LOW 0 3955 0.4924 0.1351 0.4931 0 2764 0 2358 0.4931 0.2764 0.2358 HIGH 0.4383 0 6172 0.1825 0 6679 0.4294 0 3695 0.6679 0.4294 0.3695 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Hendry LOW 0.4747 0 6370 0.1840 0 6712 0.4118 0 3524 0.6712 0.4118 0.3524 HIGH 0 5416 0 8270 0.2536 0 9305 0 6391 0 5538 0.9305 0.6391 0.5538 WGHTD AVE 0 5216 0.7821 0.2290 0 8557 0 5728 0.4948 0.8557 0.5728 0.4948

Hernando LOW 0 2739 0 3152 0.0795 0 2912 0.1419 0.1219 0.2912 0.1419 0.1219 HIGH 0 3339 0.4523 0.1305 0.4791 0 3074 0 2666 0.4791 0.3074 0.2666 WGHTD AVE 0 3080 0 3948 0.1077 0.4086 0 2438 0 2108 0.4086 0.2438 0.2108

Highlands LOW 0 3776 0.4381 0.1133 0.4142 0 2074 0.1769 0.4142 0.2074 0.1769 HIGH 0.4610 0 5943 0.1674 0 6098 0 3573 0 3047 0.6098 0.3573 0.3047 WGHTD AVE 0.4001 0.4805 0.1264 0.4701 0 2517 0 2150 0.4701 0.2517 0.2150

Hillsborough LOW 0 2979 0 3736 0.1023 0 3736 0 2113 0.1811 0.3736 0.2113 0.1811 HIGH 0.4711 0 8974 0.2980 1.1034 0 8579 0.7519 1.1034 0.8579 0.7519 WGHTD AVE 0 3446 0.4913 0.1455 0 5315 0 3454 0 2984 0.5315 0.3454 0.2984

Holmes LOW 0.1195 0 2635 0.0686 0.1867 0.1095 0 0948 0.1867 0.1095 0.0948 HIGH 0.1548 0 3819 0.1114 0 3218 0 2241 0.1946 0.3218 0.2241 0.1946 WGHTD AVE 0.1195 0 2635 0.0686 0.1867 0.1097 0 0949 0.1867 0.1097 0.0949

*Includes contents and A.L.E.

FPHLM V3.0 2008 350 Form A-6B, Output Ranges LOSS COSTS PER $1,000 (2007 FHCF exposure data) PERSONAL RESIDENTIAL - Condo Owners -- FRAME Rev. 11-1-07 $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Indian River LOW 0.4061 0 6004 0.1766 0 6436 0.4241 0 3698 0.6436 0.4241 0.3698 HIGH 0 6416 1.7766 0.6085 2 3552 2 0557 1 8705 2.3552 2.0557 1.8705 WGHTD AVE 0.4915 0 9311 0.2926 1 0971 0 8393 0.7450 1.0971 0.8393 0.7450

Jackson LOW 0 0786 0.1540 0.0352 0 0936 0 0438 0 0389 0.0936 0.0438 0.0389 HIGH 0.1264 0 2828 0.0750 0 2063 0.1252 0.1083 0.2063 0.1252 0.1083 WGHTD AVE 0 0862 0.1713 0.0396 0.1051 0 0502 0 0443 0.1051 0.0502 0.0443

Lafayette LOW 0 0739 0.1550 0.0380 0.1045 0 0575 0 0505 0.1045 0.0575 0.0505 HIGH 0 0879 0.1914 0.0484 0.1344 0 0784 0 0686 0.1344 0.0784 0.0686 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Lake LOW 0 2190 0 2430 0.0573 0 2104 0 0857 0 0742 0.2104 0.0857 0.0742 HIGH 0 3992 0 5581 0.1638 0 5993 0 3832 0 3302 0.5993 0.3832 0.3302 WGHTD AVE 0 3665 0.4797 0.1356 0.4934 0 2943 0 2530 0.4934 0.2943 0.2530

Lee LOW 0.4444 0 6630 0.1964 0.7161 0.4546 0 3895 0.7161 0.4546 0.3895 HIGH 0 6297 1.4280 0.4896 1 8283 1 5076 1 3368 1.8283 1.5076 1.3368 WGHTD AVE 0 5065 0 9449 0.2854 1 0450 0.7662 0 6689 1.0450 0.7662 0.6689

Leon LOW 0 0598 0.1150 0.0258 0 0699 0 0323 0 0288 0.0699 0.0323 0.0288 HIGH 0 0985 0 2171 0.0552 0.1518 0 0884 0 0770 0.1518 0.0884 0.0770 WGHTD AVE 0 0892 0.1909 0.0469 0.1278 0 0704 0 0615 0.1278 0.0704 0.0615

Levy LOW 0 0870 0.1862 0.0453 0.1218 0 0654 0 0571 0.1218 0.0654 0.0571 HIGH 0.1452 0 3616 0.1038 0 3028 0 2110 0.1843 0.3028 0.2110 0.1843 WGHTD AVE 0.1298 0 3355 0.0974 0 2882 0 2060 0.1807 0.2882 0.2060 0.1807

Madison LOW 0 0598 0.1169 0.0268 0 0725 0 0347 0 0307 0.0725 0.0347 0.0307 HIGH 0 0790 0.1672 0.0413 0.1133 0 0629 0 0550 0.1133 0.0629 0.0550 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Manatee LOW 0 3553 0 5298 0.1592 0 5788 0 3844 0 3322 0.5788 0.3844 0.3322 HIGH 0 5490 1 3450 0.4606 1.7453 1.4747 1 3226 1.7453 1.4747 1.3226 WGHTD AVE 0.4222 0 8951 0.2770 1 0230 0.7947 0.7018 1.0230 0.7947 0.7018

Marion LOW 0.1781 0.1871 0.0416 0.1531 0 0580 0 0508 0.1531 0.0580 0.0508 HIGH 0 3192 0.4117 0.1143 0.4187 0 2455 0 2117 0.4187 0.2455 0.2117 WGHTD AVE 0 2827 0 3343 0.0850 0 3114 0.1572 0.1352 0.3114 0.1572 0.1352

Martin LOW 0 5503 0 9335 0.3057 1.1167 0 8425 0.7405 1.1167 0.8425 0.7405 HIGH 0 8807 2 5695 0.9213 3 5203 3.1357 2 8495 3.5203 3.1357 2.8495 WGHTD AVE 0 6600 1 3582 0.4654 1.7685 1.4552 1 3000 1.7685 1.4552 1.3000

Miami-Dade LOW 0 6369 1 5006 0.5273 1 9855 1 6927 1 5196 1.9855 1.6927 1.5196 HIGH 1 2354 4 8376 1.7530 6 8797 6.4193 5 9288 6.8797 6.4193 5.9288 WGHTD AVE 0 9312 3 0973 1.1500 4 2127 3 8202 3.4882 4.2127 3.8202 3.4882

Monroe LOW 0.7199 1 5257 0.5757 2 0996 1.7802 1 5789 2.0996 1.7802 1.5789 HIGH 1 2136 4.4110 1.6907 6.4357 6 0079 5 5438 6.4357 6.0079 5.5438 WGHTD AVE 1 0428 3.1070 1.2694 4 5357 4.1330 3.7729 4.5357 4.1330 3.7729

Nassau LOW 0 0600 0.1235 0.0295 0 0821 0 0433 0 0383 0.0821 0.0433 0.0383 HIGH 0 0872 0 2077 0.0572 0.1703 0.1164 0.1034 0.1703 0.1164 0.1034 WGHTD AVE 0 0806 0.1834 0.0482 0.1378 0 0868 0 0765 0.1378 0.0868 0.0765

*Includes contents and A.L.E.

FPHLM V3.0 2008 351 Form A-6B, Output Ranges LOSS COSTS PER $1,000 (2007 FHCF exposure data) PERSONAL RESIDENTIAL - Condo Owners -- FRAME Rev. 11-1-07 $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Okaloosa LOW 0.1496 0 3797 0.1134 0 3282 0 2335 0 2022 0.3282 0.2335 0.2022 HIGH 0 3058 1 0561 0.3781 1 2421 1 0745 0 9555 1.2421 1.0745 0.9555 WGHTD AVE 0 2757 0 9780 0.3310 1 0913 0 9326 0 8272 1.0913 0.9326 0.8272

Orange LOW 0 2861 0 3213 0.0790 0 2892 0.1322 0.1133 0.2892 0.1322 0.1133 HIGH 0 3966 0 5567 0.1622 0 5961 0 3822 0 3317 0.5961 0.3822 0.3317 WGHTD AVE 0 3360 0 3996 0.1054 0 3844 0 2013 0.1723 0.3844 0.2013 0.1723

Osceola LOW 0 3225 0 3645 0.0920 0 3362 0.1597 0.1361 0.3362 0.1597 0.1361 HIGH 0 3924 0 5028 0.1397 0 5107 0 2986 0 2568 0.5107 0.2986 0.2568 WGHTD AVE 0 3350 0 3891 0.0996 0 3642 0.1808 0.1541 0.3642 0.1808 0.1541

Palm Beach LOW 0 5798 1 0678 0.3545 1 3106 1 0265 0 9117 1.3106 1.0265 0.9117 HIGH 1 0248 3 2448 1.1629 4.4921 4 0585 3.7131 4.4921 4.0585 3.7131 WGHTD AVE 0.7224 1.7805 0.6327 2 3569 2 0246 1 8314 2.3569 2.0246 1.8314

Pasco LOW 0 2856 0 3449 0.0872 0 3192 0.1538 0.1316 0.3192 0.1538 0.1316 HIGH 0 3629 0 5265 0.1600 0 5849 0.4037 0 3520 0.5849 0.4037 0.3520 WGHTD AVE 0 3270 0.4481 0.1300 0.4703 0 2930 0 2529 0.4703 0.2930 0.2529

Pinellas LOW 0 3059 0.4347 0.1258 0.4585 0 2915 0 2529 0.4585 0.2915 0.2529 HIGH 0 5998 1.7709 0.6224 2 3896 2.1142 1 9132 2.3896 2.1142 1.9132 WGHTD AVE 0 3628 0 6739 0.2119 0.7763 0 5824 0 5138 0.7763 0.5824 0.5138

Polk LOW 0 3337 0 3833 0.0977 0 3569 0.1703 0.1452 0.3569 0.1703 0.1452 HIGH 0.4618 0.7055 0.2163 0.7937 0 5454 0.4725 0.7937 0.5454 0.4725 WGHTD AVE 0 3719 0.4586 0.1250 0.4545 0 2510 0 2143 0.4545 0.2510 0.2143

Putnam LOW 0.1006 0 2127 0.0517 0.1392 0 0741 0 0646 0.1392 0.0741 0.0646 HIGH 0.1242 0 2837 0.0754 0 2098 0.1296 0.1122 0.2098 0.1296 0.1122 WGHTD AVE 0.1230 0 2779 0.0729 0 2021 0.1226 0.1061 0.2021 0.1226 0.1061

St. Johns LOW 0 0870 0.1815 0.0441 0.1210 0 0656 0 0577 0.1210 0.0656 0.0577 HIGH 0.1517 0.4072 0.1253 0 3816 0 2887 0 2532 0.3816 0.2887 0.2532 WGHTD AVE 0.1264 0 3173 0.0905 0 2684 0.1890 0.1658 0.2684 0.1890 0.1658

St. Lucie LOW 0 5048 0 8123 0.2609 0 9508 0 6982 0 6126 0.9508 0.6982 0.6126 HIGH 0 6797 1 5420 0.5354 2 0094 1 6901 1 5182 2.0094 1.6901 1.5182 WGHTD AVE 0 6195 1 3075 0.4491 1 6326 1 3315 1.1884 1.6326 1.3315 1.1884

Santa Rosa LOW 0.1485 0 3640 0.1066 0 3066 0 2126 0.1836 0.3066 0.2126 0.1836 HIGH 0 3099 1 0965 0.3955 1 3099 1.1426 1 0186 1.3099 1.1426 1.0186 WGHTD AVE 0 2980 1 0395 0.3751 1 2318 1 0702 0 9546 1.2318 1.0702 0.9546

Sarasota LOW 0 3444 0 5143 0.1540 0 5604 0 3708 0 3200 0.5604 0.3708 0.3200 HIGH 0.4583 0.7877 0.2499 0 9196 0 6734 0 5891 0.9196 0.6734 0.5891 WGHTD AVE 0 3939 0 6431 0.1982 0.7261 0 5115 0.4450 0.7261 0.5115 0.4450

Seminole LOW 0 2948 0 3499 0.0890 0 3224 0.1603 0.1382 0.3224 0.1603 0.1382 HIGH 0 3362 0.4571 0.1281 0.4664 0 2832 0 2461 0.4664 0.2832 0.2461 WGHTD AVE 0 3140 0 3779 0.0978 0 3533 0.1800 0.1548 0.3533 0.1800 0.1548

Sumter LOW 0 2821 0 3254 0.0822 0 3010 0.1476 0.1262 0.3010 0.1476 0.1262 HIGH 0 3391 0.4160 0.1127 0.4115 0 2299 0.1971 0.4115 0.2299 0.1971 WGHTD AVE 0 3279 0.4075 0.1093 0 3999 0 2211 0.1896 0.3999 0.2211 0.1896

*Includes contents and A.L.E.

FPHLM V3.0 2008 352 Form A-6B, Output Ranges LOSS COSTS PER $1,000 (2007 FHCF exposure data) PERSONAL RESIDENTIAL - Condo Owners -- FRAME Rev. 11-1-07 $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Suwanee LOW 0 0615 0.1194 0.0272 0 0740 0 0353 0 0313 0.0740 0.0353 0.0313 HIGH 0.1022 0 2280 0.0587 0.1623 0 0965 0 0841 0.1623 0.0965 0.0841 WGHTD AVE 0 0615 0.1194 0.0272 0 0740 0 0353 0 0313 0.0740 0.0353 0.0313

Taylor LOW 0 0744 0.1573 0.0389 0.1073 0 0599 0 0524 0.1073 0.0599 0.0524 HIGH 0.1155 0 2888 0.0818 0 2400 0.1672 0.1473 0.2400 0.1672 0.1473 WGHTD AVE 0 0957 0 2766 0.0774 0.1938 0.1305 0.1147 0.1938 0.1305 0.1147

Union LOW 0 0688 0.1344 0.0304 0 0821 0 0386 0 0343 0.0821 0.0386 0.0343 HIGH 0 0984 0 2076 0.0508 0.1381 0 0749 0 0655 0.1381 0.0749 0.0655 WGHTD AVE 0 0984 0 2076 0.0508 0.1381 0 0749 0 0655 0.1381 0.0749 0.0655

Volusia LOW 0 2294 0 2570 0.0628 0 2301 0.1060 0 0913 0.2301 0.1060 0.0913 HIGH 0.4064 0.7406 0.2395 0 8970 0 6951 0 6164 0.8970 0.6951 0.6164 WGHTD AVE 0 3149 0.4715 0.1306 0.4863 0 3141 0 2742 0.4863 0.3141 0.2742

Wakulla LOW 0 0805 0.1678 0.0408 0.1119 0 0607 0 0533 0.1119 0.0607 0.0533 HIGH 0.1525 0.4095 0.1239 0 3751 0 2807 0 2470 0.3751 0.2807 0.2470 WGHTD AVE 0.1177 0 2832 0.0837 0 2409 0.1685 0.1482 0.2409 0.1685 0.1482

Walton LOW 0.1340 0 3172 0.0883 0 2480 0.1621 0.1404 0.2480 0.1621 0.1404 HIGH 0 3103 1 0515 0.3755 1 2334 1 0635 0 9446 1.2334 1.0635 0.9446 WGHTD AVE 0 2663 0.7924 0.2870 0 8982 0.7520 0 6655 0.8982 0.7520 0.6655

Washington LOW 0.1173 0 2563 0.0662 0.1810 0.1058 0 0918 0.1810 0.1058 0.0918 HIGH 0.1964 0 5541 0.1767 0 5365 0.4162 0 3649 0.5365 0.4162 0.3649 WGHTD AVE 0.1711 0 3488 0.0662 0 3084 0 2170 0.1897 0.3084 0.2170 0.1897

STATEWIDE LOW 0 0548 0.1040 0.0232 0 0631 0 0287 0 0257 0.0631 0.0287 0.0257 HIGH 1 2354 4 8376 1.7530 6 8797 6.4193 5 9288 6.8797 6.4193 5.9288 WGHTD AVE 0 3801 0 8443 0.2728 0 9806 0.7826 0 6991 0.9806 0.7826 0.6991

*Includes contents and A.L.E.

FPHLM V3.0 2008 353 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - Condo Owners -- MASONRY $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Alachua LOW 0 0898 0.1869 0.0445 0.1183 0 0601 0 0525 0.1183 0.0601 0.0525 HIGH 0.1111 0 2487 0.0631 0.1685 0 0954 0 0824 0.1685 0.0954 0.0824 WGHTD AVE 0 0988 0 2141 0.0525 0.1393 0 0745 0 0647 0.1393 0.0745 0.0647

Baker LOW 0 0556 0.1076 0.0241 0 0651 0 0299 0 0267 0.0651 0.0299 0.0267 HIGH 0 0850 0.1791 0.0430 0.1154 0 0604 0 0530 0.1154 0.0604 0.0530 WGHTD AVE 0 0850 0.1791 0.0000 0.1154 0 0604 0 0530 0.1154 0.0604 0.0530

Bay LOW 0.1212 0 2702 0.0690 0.1839 0.1046 0 0906 0.1839 0.1046 0.0906 HIGH 0 3264 1.1956 0.4402 1.4538 1 2782 1.1376 1.4538 1.2782 1.1376 WGHTD AVE 0 2183 0 6586 0.2171 0 6667 0 5327 0.4657 0.6667 0.5327 0.4657

Bradford LOW 0 0871 0.1824 0.0430 0.1136 0 0565 0 0495 0.1136 0.0565 0.0495 HIGH 0.1048 0 2303 0.0570 0.1516 0 0827 0 0718 0.1516 0.0827 0.0718 WGHTD AVE 0 0934 0.1980 0.0476 0.1262 0 0654 0 0570 0.1262 0.0654 0.0570

Brevard LOW 0 2644 0 3608 0.0968 0 3455 0.1961 0.1696 0.3455 0.1961 0.1696 HIGH 0 5351 1 2625 0.4295 1 6021 1 3338 1.1908 1.6021 1.3338 1.1908 WGHTD AVE 0 3464 0 5488 0.1610 0 5783 0 3837 0 3340 0.5783 0.3837 0.3340

Broward LOW 0 5241 1 0141 0.3432 1 2436 0 9759 0 8637 1.2436 0.9759 0.8637 HIGH 0.7492 1 9487 0.7049 2 6153 2 2618 2 0284 2.6153 2.2618 2.0284 WGHTD AVE 0 6099 1 3527 0.4719 1.7242 1.4209 1 2659 1.7242 1.4209 1.2659

Calhoun LOW 0.1011 0 2153 0.0525 0.1380 0 0721 0 0628 0.1380 0.0721 0.0628 HIGH 0.1255 0 2951 0.0780 0 2087 0.1256 0.1085 0.2087 0.1256 0.1085 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Charlotte LOW 0 3565 0.4983 0.1392 0.4953 0 2902 0 2486 0.4953 0.2902 0.2486 HIGH 0 5280 1.1980 0.4096 1.4994 1 2199 1 0761 1.4994 1.2199 1.0761 WGHTD AVE 0.4005 0 6243 0.1856 0 6654 0.4376 0 3775 0.6654 0.4376 0.3775

Citrus LOW 0 2147 0 2534 0.0606 0 2189 0 0965 0 0834 0.2189 0.0965 0.0834 HIGH 0 2872 0 3790 0.1017 0 3658 0 2024 0.1742 0.3658 0.2024 0.1742 WGHTD AVE 0 2602 0 3326 0.0855 0 3073 0.1586 0.1365 0.3073 0.1586 0.1365

Clay LOW 0 0836 0.1725 0.0405 0.1076 0 0535 0 0470 0.1076 0.0535 0.0470 HIGH 0.1137 0 2553 0.0649 0.1739 0 0992 0 0859 0.1739 0.0992 0.0859 WGHTD AVE 0 0935 0 2012 0.0495 0.1329 0 0721 0 0628 0.1329 0.0721 0.0628

Collier LOW 0.4313 0 6071 0.1717 0 6069 0 3576 0 3040 0.6069 0.3576 0.3040 HIGH 0 6093 1 2950 0.4441 1 5893 1 2565 1 0919 1.5893 1.2565 1.0919 WGHTD AVE 0.4794 0.7846 0.2412 0 8562 0 5825 0.4999 0.8562 0.5825 0.4999

Columbia LOW 0 0593 0.1151 0.0257 0 0689 0 0313 0 0279 0.0689 0.0313 0.0279 HIGH 0 0894 0.1948 0.0478 0.1280 0 0697 0 0609 0.1280 0.0697 0.0609 WGHTD AVE 0 0798 0.1673 0.0397 0.1060 0 0543 0 0477 0.1060 0.0543 0.0477

De Soto LOW 0 3579 0.4547 0.1193 0.4284 0 2230 0.1906 0.4284 0.2230 0.1906 HIGH 0 3994 0 5702 0.1639 0 5870 0 3589 0 3074 0.5870 0.3589 0.3074 WGHTD AVE 0 3949 0 5541 0.1600 0 5695 0 3444 0 2950 0.5695 0.3444 0.2950

Dixie LOW 0 0916 0 2066 0.0528 0.1435 0 0838 0 0730 0.1435 0.0838 0.0730 HIGH 0.1529 0.4151 0.1265 0 3741 0 2774 0 2423 0.3741 0.2774 0.2423 WGHTD AVE 0.1130 0 3061 0.0732 0 2305 0.1562 0.1362 0.2305 0.1562 0.1362

*Includes contents and A.L.E.

FPHLM V3.0 2008 354 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - Condo Owners -- MASONRY $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Duval LOW 0 0538 0.1034 0.0230 0 0620 0 0280 0 0250 0.0620 0.0280 0.0250 HIGH 0.1238 0 3611 0.1162 0 3615 0 2871 0 2536 0.3615 0.2871 0.2536 WGHTD AVE 0 0898 0 2010 0.0505 0.1375 0 0790 0 0690 0.1375 0.0790 0.0690

Escambia LOW 0.1396 0 3612 0.1079 0 3036 0 2127 0.1826 0.3036 0.2127 0.1826 HIGH 0.4040 1 6370 0.6213 2.1128 1 9121 1.7139 2.1128 1.9121 1.7139 WGHTD AVE 0 2825 1 0070 0.3656 1.1918 1 0353 0 9178 1.1918 1.0353 0.9178

Flagler LOW 0 2446 0 3059 0.0782 0 2808 0.1417 0.1216 0.2808 0.1417 0.1216 HIGH 0 3133 0 5111 0.1546 0 5540 0 3782 0 3284 0.5540 0.3782 0.3284 WGHTD AVE 0 2660 0 3931 0.1113 0 3943 0 2431 0 2099 0.3943 0.2431 0.2099

Franklin LOW 0.1662 0.4629 0.1438 0.4221 0 3167 0 2764 0.4221 0.3167 0.2764 HIGH 0 2718 0 9904 0.3609 1.1913 1 0444 0 9309 1.1913 1.0444 0.9309 WGHTD AVE 0 2311 0 6929 0.2506 0.7898 0 6628 0 5871 0.7898 0.6628 0.5871

Gadsen LOW 0 0776 0.1578 0.0365 0 0958 0 0456 0 0403 0.0958 0.0456 0.0403 HIGH 0.1064 0 2331 0.0580 0.1526 0 0828 0 0719 0.1526 0.0828 0.0719 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Gulf LOW 0.1275 0 3070 0.0839 0 2289 0.1453 0.1257 0.2289 0.1453 0.1257 HIGH 0.1740 0.4746 0.1455 0.4219 0 3107 0 2701 0.4219 0.3107 0.2701 WGHTD AVE 0.1717 0.4709 0.1446 0.4142 0 3038 0 2641 0.4142 0.3038 0.2641

Hamilton LOW 0 0551 0.1072 0.0240 0 0643 0 0294 0 0261 0.0643 0.0294 0.0261 HIGH 0 0759 0.1611 0.0387 0.1034 0 0542 0 0475 0.1034 0.0542 0.0475 WGHTD AVE 0 0759 0.1611 0.0387 0.1034 0 0542 0 0475 0.1034 0.0542 0.0475

Hardee LOW 0 3523 0.4399 0.1141 0.4091 0 2067 0.1760 0.4091 0.2067 0.1760 HIGH 0 3862 0 5296 0.1483 0 5304 0 3090 0 2638 0.5304 0.3090 0.2638 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Hendry LOW 0.4199 0 5547 0.1514 0 5408 0 2997 0 2549 0.5408 0.2997 0.2549 HIGH 0.4743 0 6942 0.2024 0.7235 0.4535 0 3889 0.7235 0.4535 0.3889 WGHTD AVE 0.4675 0 6793 0.1945 0.7007 0.4340 0 3718 0.7007 0.4340 0.3718

Hernando LOW 0 2460 0 2911 0.0699 0 2524 0.1118 0 0962 0.2524 0.1118 0.0962 HIGH 0 2959 0 3939 0.1071 0 3833 0 2225 0.1913 0.3833 0.2225 0.1913 WGHTD AVE 0 2708 0 3521 0.0934 0 3347 0.1805 0.1551 0.3347 0.1805 0.1551

Highlands LOW 0 3387 0.4017 0.0986 0 3553 0.1611 0.1377 0.3553 0.1611 0.1377 HIGH 0.4090 0 5243 0.1395 0.4983 0 2631 0 2235 0.4983 0.2631 0.2235 WGHTD AVE 0 3604 0.4397 0.1116 0.4026 0.1963 0.1675 0.4026 0.1963 0.1675

Hillsborough LOW 0 2656 0 3335 0.0864 0 3098 0.1580 0.1349 0.3098 0.1580 0.1349 HIGH 0.4048 0.7114 0.2284 0 8186 0 5932 0 5124 0.8186 0.5932 0.5124 WGHTD AVE 0 3102 0.4323 0.1220 0.4353 0 2586 0 2215 0.4353 0.2586 0.2215

*Includes contents and A.L.E.

FPHLM V3.0 2008 355 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - Condo Owners -- MASONRY $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Indian River LOW 0 3599 0 5134 0.1436 0 5102 0 3055 0 2635 0.5102 0.3055 0.2635 HIGH 0 5412 1 3214 0.4504 1 6836 1.4083 1 2592 1.6836 1.4083 1.2592 WGHTD AVE 0.4417 0 8069 0.2505 0 8984 0 6511 0 5701 0.8984 0.6511 0.5701

Jackson LOW 0 0769 0.1533 0.0350 0 0920 0 0427 0 0379 0.0920 0.0427 0.0379 HIGH 0.1213 0 2751 0.0721 0.1934 0.1140 0 0983 0.1934 0.1140 0.0983 WGHTD AVE 0.1010 0.1934 0.0499 0.1301 0 0667 0 0582 0.1301 0.0667 0.0582

Lake LOW 0.1983 0 2277 0.0524 0.1898 0 0725 0 0631 0.1898 0.0725 0.0631 HIGH 0 3524 0.4806 0.1336 0.4777 0 2765 0 2364 0.4777 0.2765 0.2364 WGHTD AVE 0 3086 0 3854 0.0991 0 3564 0.1802 0.1541 0.3564 0.1802 0.1541

Lee LOW 0 3921 0 5630 0.1599 0 5690 0 3275 0 2786 0.5690 0.3275 0.2786 HIGH 0 5342 1 0937 0.3671 1 3241 1 0315 0 8997 1.3241 1.0315 0.8997 WGHTD AVE 0.4437 0.7034 0.2081 0.7394 0.4845 0.4160 0.7394 0.4845 0.4160

Leon LOW 0 0586 0.1144 0.0257 0 0687 0 0314 0 0280 0.0687 0.0314 0.0280 HIGH 0 0950 0 2123 0.0535 0.1436 0 0814 0 0707 0.1436 0.0814 0.0707 WGHTD AVE 0 0875 0.1905 0.0467 0.1248 0 0676 0 0589 0.1248 0.0676 0.0589

Levy LOW 0 0845 0.1835 0.0443 0.1169 0 0614 0 0537 0.1169 0.0614 0.0537 HIGH 0.1378 0 3438 0.0973 0 2745 0.1849 0.1603 0.2745 0.1849 0.1603 WGHTD AVE 0.1244 0 3177 0.0912 0 2607 0.1804 0.1570 0.2607 0.1804 0.1570

Madison LOW 0 0585 0.1159 0.0264 0 0707 0 0333 0 0295 0.0707 0.0333 0.0295 HIGH 0 0766 0.1643 0.0402 0.1082 0 0586 0 0512 0.1082 0.0586 0.0512 WGHTD AVE 0 0000 0 0000 0.0000 0 0000 0 0000 0 0000 0.0000 0.0000 0.0000

Manatee LOW 0 3134 0.4496 0.1282 0.4549 0 2747 0 2349 0.4549 0.2747 0.2349 HIGH 0.4651 1 0191 0.3438 1 2582 1 0106 0 8913 1.2582 1.0106 0.8913 WGHTD AVE 0 3667 0 6946 0.2115 0.7549 0 5442 0.4726 0.7549 0.5442 0.4726

Marion LOW 0.1626 0.1796 0.0387 0.1412 0 0507 0 0446 0.1412 0.0507 0.0446 HIGH 0 2841 0 3643 0.0958 0 3440 0.1819 0.1560 0.3440 0.1819 0.1560 WGHTD AVE 0 2671 0 3271 0.0819 0 2937 0.1412 0.1213 0.2937 0.1412 0.1213

Martin LOW 0.4440 0 6376 0.1903 0 6681 0.4264 0 3696 0.6681 0.4264 0.3696 HIGH 0 6615 1 5221 0.5375 1 9638 1 6357 1.4575 1.9638 1.6357 1.4575 WGHTD AVE 0 5366 0 9687 0.3179 1.1468 0 8650 0.7620 1.1468 0.8650 0.7620

Miami-Dade LOW 0.4895 0 9153 0.3087 1.1081 0 8548 0.7524 1.1081 0.8548 0.7524 HIGH 0 8949 2 8084 1.0323 3 9031 3 5146 3.1863 3.9031 3.5146 3.1863 WGHTD AVE 0.7049 1 9036 0.6858 2 5123 2.1712 1 9473 2.5123 2.1712 1.9473

Monroe LOW 0 6946 1 3850 0.5281 1 8785 1 5605 1 3660 1.8785 1.5605 1.3660 HIGH 1.1164 3 5941 1.4108 5 2250 4 8015 4 3581 5.2250 4.8015 4.3581 WGHTD AVE 0 9865 2 9734 1.1661 4 2521 3 8389 3.4597 4.2521 3.8389 3.4597

Nassau LOW 0 0585 0.1211 0.0287 0 0782 0 0409 0 0362 0.0782 0.0409 0.0362 HIGH 0 0835 0.1968 0.0534 0.1535 0.1006 0 0885 0.1535 0.1006 0.0885 WGHTD AVE 0 0776 0.1770 0.0459 0.1273 0 0773 0 0677 0.1273 0.0773 0.0677

*Includes contents and A.L.E.

FPHLM V3.0 2008 356 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - Condo Owners -- MASONRY $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Okaloosa LOW 0.1413 0 3606 0.1058 0 2970 0 2050 0.1765 0.2970 0.2050 0.1765 HIGH 0 2768 0 9322 0.3336 1 0522 0 8909 0.7824 1.0522 0.8909 0.7824 WGHTD AVE 0 2645 0 8851 0.3171 0 9815 0 8250 0.7236 0.9815 0.8250 0.7236

Okeechobee LOW 0 3748 0.4805 0.1247 0.4425 0 2269 0.1935 0.4425 0.2269 0.1935 HIGH 0.4002 0 5389 0.1458 0 5170 0 2869 0 2454 0.5170 0.2869 0.2454 WGHTD AVE 0 3949 0 5127 0.1347 0.4780 0 2505 0 2137 0.4780 0.2505 0.2137

Orange LOW 0 2564 0 2999 0.0705 0 2545 0.1064 0 0916 0.2545 0.1064 0.0916 HIGH 0 3507 0.4798 0.1327 0.4762 0 2768 0 2381 0.4762 0.2768 0.2381 WGHTD AVE 0 3009 0 3631 0.0907 0 3266 0.1550 0.1327 0.3266 0.1550 0.1327

Osceola LOW 0 2900 0 3376 0.0810 0 2924 0.1264 0.1082 0.2924 0.1264 0.1082 HIGH 0 3501 0.4450 0.1171 0.4194 0 2211 0.1892 0.4194 0.2211 0.1892 WGHTD AVE 0 2989 0 3550 0.0871 0 3106 0.1392 0.1190 0.3106 0.1392 0.1190

Palm Beach LOW 0.4650 0.7142 0.2193 0.7782 0 5281 0.4620 0.7782 0.5281 0.4620 HIGH 0.7681 1 9287 0.6863 2 5471 2.1777 1 9565 2.5471 2.1777 1.9565 WGHTD AVE 0 5734 1 2158 0.4149 1.4828 1.1867 1 0562 1.4828 1.1867 1.0562

Pasco LOW 0 2548 0 3188 0.0768 0 2770 0.1214 0.1042 0.2770 0.1214 0.1042 HIGH 0 3217 0.4417 0.1279 0.4551 0 2869 0 2471 0.4551 0.2869 0.2471 WGHTD AVE 0 2816 0 3888 0.1080 0 3840 0 2242 0.1926 0.3840 0.2242 0.1926

Pinellas LOW 0 2712 0 3754 0.1031 0 3672 0 2117 0.1819 0.3672 0.2117 0.1819 HIGH 0.4991 1 2993 0.4572 1 6909 1.4396 1 2794 1.6909 1.4396 1.2794 WGHTD AVE 0 3282 0 5641 0.1722 0 6272 0.4440 0 3866 0.6272 0.4440 0.3866

Polk LOW 0 2986 0 3531 0.0857 0 3087 0.1346 0.1152 0.3087 0.1346 0.1152 HIGH 0.4046 0 5919 0.1726 0 6168 0 3867 0 3315 0.6168 0.3867 0.3315 WGHTD AVE 0 3324 0.4148 0.1071 0 3826 0.1914 0.1631 0.3826 0.1914 0.1631

Putnam LOW 0 0972 0 2097 0.0506 0.1338 0 0697 0 0607 0.1338 0.0697 0.0607 HIGH 0.1191 0 2753 0.0723 0.1958 0.1175 0.1013 0.1958 0.1175 0.1013 WGHTD AVE 0.1114 0 2456 0.0626 0.1677 0 0948 0 0820 0.1677 0.0948 0.0820

St. Johns LOW 0 0845 0.1785 0.0430 0.1158 0 0612 0 0538 0.1158 0.0612 0.0538 HIGH 0.1425 0 3807 0.1158 0 3399 0 2496 0 2170 0.3399 0.2496 0.2170 WGHTD AVE 0.1288 0 3308 0.0947 0 2795 0.1972 0.1716 0.2795 0.1972 0.1716

St. Lucie LOW 0.4111 0 5688 0.1656 0 5819 0 3577 0 3101 0.5819 0.3577 0.3101 HIGH 0 5294 0 9652 0.3192 1.1456 0 8684 0.7658 1.1456 0.8684 0.7658 WGHTD AVE 0.4937 0 8152 0.2623 0 9298 0 6662 0 5837 0.9298 0.6662 0.5837

Santa Rosa LOW 0.1407 0 3471 0.1002 0 2791 0.1877 0.1612 0.2791 0.1877 0.1612 HIGH 0 2795 0 9610 0.3471 1.1031 0 9422 0 8291 1.1031 0.9422 0.8291 WGHTD AVE 0 2689 0 9071 0.3265 1 0307 0 8754 0.7705 1.0307 0.8754 0.7705

Sarasota LOW 0 3041 0.4369 0.1242 0.4411 0 2660 0 2272 0.4411 0.2660 0.2272 HIGH 0 3994 0 6430 0.1954 0 6985 0.4714 0.4068 0.6985 0.4714 0.4068 WGHTD AVE 0 3498 0 5378 0.1600 0 5705 0 3714 0 3196 0.5705 0.3714 0.3196

Seminole LOW 0 2663 0 3215 0.0779 0 2782 0.1258 0.1082 0.2782 0.1258 0.1082 HIGH 0 3003 0.4011 0.1067 0 3803 0 2089 0.1800 0.3803 0.2089 0.1800 WGHTD AVE 0 2836 0 3463 0.0853 0 3040 0.1410 0.1210 0.3040 0.1410 0.1210

Sumter LOW 0 2539 0 2999 0.0722 0 2603 0.1157 0 0992 0.2603 0.1157 0.0992 HIGH 0 3028 0 3742 0.0956 0 3427 0.1725 0.1475 0.3427 0.1725 0.1475 WGHTD AVE 0 2793 0 3436 0.0855 0 3073 0.1477 0.1265 0.3073 0.1477 0.1265

*Includes contents and A.L.E.

FPHLM V3.0 2008 357 Form A-6B, Output Ranges (2007 FHCF exposure data) LOSS COSTS PER $1,000 Rev. 11-1-07 PERSONAL RESIDENTIAL - Condo Owners -- MASONRY $0 $0 $0 DEDUCTIBLE $500 $1,000 $2,500 1% 2% 5% DEDUCTIBLE DEDUCTIBLE ADDITIONAL DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE DEDUCTIBLE COUNTY LOSS COSTS STRUCTURE CONTENTS LIVING EXPENSE TOTAL* TOTAL* TOTAL* TOTAL* TOTAL* TOTAL*

Union LOW 0 0674 0.1335 0.0302 0 0805 0 0374 0 0333 0.0805 0.0374 0.0333 HIGH 0 0954 0 2040 0.0495 0.1319 0 0698 0 0609 0.1319 0.0698 0.0609 WGHTD AVE 0 0954 0 2040 0.0495 0.1319 0 0698 0 0609 0.1319 0.0698 0.0609

Volusia LOW 0 2072 0 2397 0.0560 0 2024 0 0853 0 0737 0.2024 0.0853 0.0737 HIGH 0 3527 0 5958 0.1852 0 6716 0.4847 0.4234 0.6716 0.4847 0.4234 WGHTD AVE 0 3081 0.4669 0.1329 0.4848 0 3111 0 2697 0.4848 0.3111 0.2697

Wakulla LOW 0 0783 0.1653 0.0399 0.1074 0 0569 0 0499 0.1074 0.0569 0.0499 HIGH 0.1434 0 3814 0.1139 0 3315 0 2397 0 2088 0.3315 0.2397 0.2088 WGHTD AVE 0.1434 0 3483 0.1025 0 3044 0 2176 0.1896 0.3044 0.2176 0.1896

Walton LOW 0.1280 0 3058 0.0838 0 2289 0.1451 0.1252 0.2289 0.1451 0.1252 HIGH 0 2818 0 9326 0.3332 1 0513 0 8875 0.7789 1.0513 0.8875 0.7789 WGHTD AVE 0 2583 0 8024 0.2759 0 8831 0.7330 0 6424 0.8831 0.7330 0.6424

STATEWIDE LOW 0 0538 0.1034 0.0230 0 0620 0 0280 0 0250 0.0620 0.0280 0.0250 HIGH 1.1164 3 5941 1.4108 5 2250 4 8015 4 3581 5.2250 4.8015 4.3581 WGHTD AVE 0.4587 1 0236 0.3334 1.1653 0 9123 0 8075 1.1653 0.9123 0.8075

*Includes contents and A.L.E.

FPHLM V3.0 2008 358 Form A-6B: Output Ranges (2007 Weights), Item C: Zip codes with losses but no exposure:

32885 32886 32893 32898 33107 33119 33234 33245 33247 33255 33299 33340 33355 33630 33650 33662 33663 33664 33689 33690 33728 33729

FPHLM V3.0 2008 359 Form A-6B: Output Ranges (2007 Weights), Item D: Zip codes with exposure but no losses:

32723 32745 32896 33188 33542 33563 33575 33812 33966 33967 34637 34638 34714 34715

FPHLM V3.0 2008 360