Freeport and Vicinity, Texas Hurricane-Flood Protection ... · Freeport, Texas, Brazoria County,...

i

Freeport and Vicinity, Texas Hurricane-Flood Protection

Draft Feasibility Report

®

TABLE OF CONTENTS

U.S. Army Engineer District, Galveston Southwestern Division

May 2005

i

Syllabus This study addresses a request by the non-Federal sponsor, the Velasco Drainage District (VDD), to reevaluate the Freeport Hurricane-Flood Protection Levee System using current data and design criteria. The request was precipitated by a risk analysis study performed by Dow Chemical Company that expressed concern that circumstances, conditions, and technology changes since the original 1958 feasibility study warranted reevaluation of the project. The Freeport Hurricane-Flood Protection levee system was authorized through the Flood Control Act of October 23, 1962, Public Law 87-874. The system included the design and construction of new levees, improvements to existing levees, drainage control structures, pump stations, and a tide gate. The existing Freeport Hurricane-Flood protection system was designed based on complete blocked conditions. Pumps were established for interior rainfall runoff and possible levee overtopping flood storage. Since then, the non-Federal sponsor has upgraded the pump capacity for the interior areas and currently maintains the levee and its interior facilities. In July 1998, the Risk Control Consulting (RCC) Group of J&H Marsh & McLennan delivered a report commissioned by the Dow Chemical Company titled, Hurricane Vulnerability and Impact Analysis for the Freeport, Texas Facility. As part of their evaluation, RCC produced a series of surge maps defining the maximum envelope of water (MEOW) for Saffir-Simpson scale category 1 through 5 hurricanes. The extent of inland penetration by hurricane induced storm surge flooding was based on the Federal Emergency Management Agency SLOSH (Sea, Lake, and Overland Surges from Hurricanes) model. The SLOSH model is a stochastic model that is not endorsed by USACE for determining frequency analysis for design of a hurricane flood protection system. The inherent deficiencies in the SLOSH model results notwithstanding, Dow used the results to generate a table of maximum water levels for Category 3-5 hurricanes at Oyster Creek City Hall, Freeport City Hall, Clute City Hall, Lake Jackson City Hall, Jones Creek City Hall, Brazosport College, and Angleton City Hall. Reported maximum storm surge water levels ranged from a low of 0 feet (Freeport City Hall, Category 3 storm) to a high of 20 feet (Oyster Creek City Hall, Category 5 storm). This information was distributed to the general public and resulted in significant level of concern expressed in a series of newspaper articles in the local press. Following the distribution of the Dow Chemical Company report, the non-Federal sponsor submitted a letter to the Galveston District seeking assistance under Section 216 of the 1970 Flood Control Act which provides for review of completed USACE projects that may have changed because of physical or economic reasons. In order to evaluate the level of protection afforded by the Freeport Hurricane-Flood Protection Levee system, the feasibility study investigated the following factors: 1) survey and review existing project conditions; 2) update

ii

the hurricane surge elevation versus frequency for the project area utilizing USACE approved storm surge modeling methods; and 3) determine changed conditions and effects on current design criteria. During the initial stage of the study, a topographic survey of the existing levee system was performed and compared with the design elevations and a review of the structural integrity of the levees. Investigations included reviewing original geotechnical information, design methods, as-built drawings, findings of periodic inspections of the levee protection system and conduction additional geotechnical investigations. Following the establishment of existing levee elevations and conditions, a risk-based storm surge frequency-of-occurrence (storm impact frequency analysis) was performed. The finite element numerical hydrodynamic model Advanced Circulation (ADCIRC), (Leutich, R.A., Westerink, J.J., and Scheffner, N.W., 1992) was used to numerically simulate the propagation of tides and storms across the Gulf of Mexico into the study area. The model was verified by comparing model-generated tide time series with the corresponding time series reconstructed from existing harmonic analyses and on-site measurements of water surface elevation. The results of the storm surge frequency-of-occurrence provide estimates of the various indices for engineering performance based on existing USACE guidance and procedures. Analyses were limited to engineering performance or non-economic related performance for the feasibility study. Analyses performed used current state-of-the-art numerical models with current methods and current data. Based upon survey results, the study indicated that no overtopping of the levee system would occur during the 100-year storm plus tide event. The report documents that the Freeport Hurricane-Flood Protection levee system maintains a storm surge capacity of a 1 per cent exceedance probability (100-year) and that the levee system provides sufficient protection for an extreme 1percent event (100-year) against storm surge plus wave run-up. Based upon these results, the non-Federal sponsor (VDD) has requested that no additional work be performed on the feasibility study and this report concludes that No Action is the recommended plan.

iii

TABLE OF CONTENTS

Section…………………………………………………………………………………Page

I. INTRODUCTION…………………………………………………………………..…1

Purpose and Authority…………………………………………………………….2

Description of the Study Area…………………………………………………….3

Project Area Description………………………………………………………….8

Non-Federal Sponsor Coordination…………………………………………….…8

Prior Authorizations…………………………………………………………….…9

Study and Report Process………………………………………………………..11

II. PROBLEM IDENTIFICATION……………………………………………………..12

Hurricane-Flood Damage Protection…………………………………………….12

III. FORMULATION OBJECTIVES, CONSTRAINTS, AND CRITERIA……………15

National Objectives………………………………………………………………15

Planning Objectives……………………………………………………………...15

Planning Constraints……………………………………………………………..16

Technical Criteria………………………………………………………………...16

Economic Criteria………………………………………………………………..17

Environmental Criteria…………………………………………………………...17

Social and Other Criteria………………………………………………………....18

Plan Formulation Rationale……………………………………………………....18

IV. PLAN FORMULATION………..……………………… ………………………….19

Without Project Condition/No Action…………………………………………...19

V. PUBLIC COORDINATION………………………………………………………… 25

VI. RECOMMENDATION……………………………………………………………..26

iv

TABLES

No. Title Pages

1 Freeport Hurricane-Flood Protection Levee System Layout…………………….20

2 Levee Analysis Breakdown versus Modeled Storm Surge……………………….24

FIGURES

No. Title Pages

1 Map of the Freeport Hurricane-Flood Protection Levee System……………..….2

2 Location of stations used fro input to EST model……………………………….22

3 200-year storm surge values, Freeport Hurricane-Flood Protection Levee ……..22

4 100-year storm surge values, Freeport Hurricane-Flood Protection Levee ……..23

5 50-year storm surge values, Freeport Hurricane-Flood Protection Levee ………23

APPENDICES

A – Engineering Appendix

B – Risk Assessment for Exposure to Hurricane Conditions at Freeport, Texas

C – Non-Federal Sponsor Request for Study Termination

D – Cost Share Allocation

FREEPORT AND VICINITY, TEXAS HURRICANE-FLOOD PROTECTION

FEASIBILITY REPORT

I. INTRODUCTION

The purpose of the Freeport and Vicinity, Texas, Hurricane-Flood Protection feasibility study is to address a request by the non-Federal sponsor, the Velasco Drainage District (VDD), to reevaluate the levee system using current data and design criteria. The study area is located in Freeport, Texas, Brazoria County, approximately 48 miles southwest of Galveston, Texas. The existing project was completed in 1982 and consists of a system of levees and pump stations that protect an area of approximately 42 square miles (Figure 1). Freeport is an important industrial center and deepwater port on the Texas coast. The principal sources of income are derived from processing petroleum and petroleum by-products. Brazoria County claims to house the world’s largest chemical complex with Dow Chemical being the primary employer. The population of Freeport and vicinity was 110,363 according to the 2000 Census Report. A partial, conservative estimate of property value within the study area in 2000 is estimated at well over $502 billion. In a letter dated October 6, 1998, the VDD requested a review of the Freeport Hurricane Levee Project under Section 216 of the 1970 Flood Control Act. The request was precipitated by a risk analysis study performed by Dow Chemical Company. Based upon the results of the Dow study, the non-Federal sponsor expressed concern that circumstances, conditions, and technology changes since the original 1958 study (with an addendum dated 1966/1967) warranted reevaluation of the project. The Dow study indicated a significant increase in property values for both industrial and residential property since the 1958 U.S. Army Corps of Engineers (USACE) study as well as a difference of opinion among the technical modeling community as to the extent of storm surge to be predicted for the study area. An Initial Appraisal Report completed in May 1999 recommended that a reconnaissance level study new start be initiated in FY 2000. The 905(b) Analysis, completed in April 2002, determined that an investigation of the existing levee system was warranted based on the large economic infrastructure that could potentially be impacted by overtopping of the levees. The reconnaissance report recommended that a feasibility study be initiated. This current comprehensive hurricane-flood protection study investigates the feasibility of improving the Freeport and vicinity hurricane-flood protection levee system. This section of the report identifies the study authority, scope, participants and coordination, related studies, and study process. The study area is shown on Figure 1.

2

Figure 1. Map of the Freeport Hurricane-Flood Protection Levee System.

PURPOSE AND AUTHORITY

The purpose of this study is to evaluate the level of hurricane-flood protection provided by the existing Freeport Hurricane-Flood Protection levee and to evaluate alternative designs in the event that the existing system is deficient. Because of concerns raised by the Dow report, the Project Delivery Team determined early in the study process that the initial study efforts should focus on establishing the existing and future without-project conditions in order to assess hurricane-flood protection competence.

3

The authority for the USACE to conduct this study comes from Section 216 of the 1970 Flood Control Act which reads:

“The Secretary of the Army, acting through the Chief of Engineers, is authorized to review the operation of projects the construction of which has been completed and which were constructed by the Corps of Engineers in the interest of navigation, flood control, water supply, and related purposes, when found advisable due to significant changed physical or economic conditions, and to report thereon to Congress with recommendations on the advisability of modifying the structures or their operation, and for improving the quality of the environment in the overall public interest.”

DESCRIPTION OF THE STUDY AREA The Freeport Hurricane-Flood Protection levee system is located on the central portion of the Texas coast approximately 48 miles southwest of Galveston, Texas. The study area is located in Brazoria County.

Physiography The study area is located in the southern portion of Brazoria County, Texas, on the Gulf of Mexico at the mouth of the Brazos River, as shown on Figure 1. The study area lies in a low coastal plain dissected by streams, canals, and waterways. The land surface elevation varies from 3 to 4 feet (NAVD 88) along the coast to greater than 15 feet (NAVD 88) about 15 miles inland. The project area is bounded generally by the Brazos River on the west, Oyster Creek on the north and east, and the Gulf Intracoastal Waterway (GIWW) on the south. The study area is located in the West Gulf Coast subdivision of the Atlantic and Gulf Coast Plains geomorphic province. The region is characterized as Quarternary (Holocene and Recent) Alluvium containing thick deposits of clay, silt, and sand overlying the Pleistocene Beaumont Formation. These formations are up to several hundred feet thick, and are comprised mainly of stream channel, levee, and backswamp deposits associated with numerous current and former river channels and bayous. This land area, located along the Old Brazos River, the Dow Barge Canal system, and the GIWW, which all join the Freeport Harbor Channel (FHC) and the Gulf of Mexico, has undergone significant land and watercourse alteration due to industrial development.

4

Climate The climate of the study area is humid subtropical with warm to hot summers and mild winters. The dominant air mass in summer is marine tropical in which sea breezes moderate afternoon heat. Occasional showers or thunderstorms are common during this season. Winters are mild with considerable day-to-day variation between the marine tropical air mass and modified continental polar air masses. Periods of freezing temperatures are infrequent and usually last no longer than two or three days. Rainfall averages about 50 inches annually at Freeport. The annual rainfall distribution is greater for the early summer and fall periods and least for winter and late summer. Two principal wind regimes dominate the area and include persistent, southeasterly winds occurring from March through November and strong, short-lived northerly winds from December through February. Severe weather occurs periodically in the area in the form of thunderstorms, tornadoes, and tropical storms or hurricanes. Fish and Wildlife Resources The study area contains estuarine, upland, and wetland areas that support a varied population of fish and wildlife resources. The area contains an abundance of game and non-game wildlife resources. The area also supports a productive sport and commercial fishery and provides recreational opportunities that are intensively utilized during the year. Aquatic Resources Aquatic resources in the project area include the open waters of the Old Brazos River Channel (present-day FHC), the GIWW, Oyster Creek and intermittent tributaries of Salt Bayou. These open waters are part of the Brazos River estuary. Aquatic organisms in the project area reflect the great diversity of fish and invertebrate resources found in the surrounding coastal waters of the Gulf of Mexico. The ecologic stability of aquatic resources, notably abundance and diversity of fish and invertebrate species, present in the project area is dependent, in large part, on salinity, substrate, and vegetation. The EPA characterizes nearly one half of the Brazos River estuary as experiencing hypoxia (dissolved oxygen between 0 and 2 milligrams per liter (mg/l)), with all of the area having high sediment contaminants and degraded benthos. Benthic organisms are the largest and most diverse group of organisms inhabiting the Brazos River estuary system. Benthic invertebrates are an important food source for fish and larger

5

invertebrates in the Brazos River estuary and benthic organisms in the system include polychaetes, mollusks, and arthropods. The most important commercially harvested species that inhabit estuarine waters, for at least a portion of their life, are brown and white shrimp, Southern flounder, and blue crabs. With the combination of saline Gulf waters entering from the FHC and freshwater from the Brazos River and Oyster Creek, the area within the Brazos River estuary is important to the overall commercial and recreational fishery of Texas, despite its relatively small size. Sport fishing occurs as bank or boat fishing in areas accessible from the GIWW. Recreational landings consist typically of shrimp, crabs, bait fish (including sub-adult striped mullet), spotted seatrout, redfish, sand seatrout, Southern flounder, black drum, and red snapper. Terrestrial Resources Upland habitat types occurring in the study area include scrub/shrub and grassland/upland pasture. Scrub/shrub habitat is typically characterized as woody vegetation that is less than six feet in height. In the study area this vegetation includes: cordgrass, salt cedar, bigleaf sumpweed, goldenrod, sea myrtle, rattlebush, and ragweed. Mammals commonly associated with this habitat include coyote, eastern cottontail, opossum, raccoon, cotton rat, striped skunk, and white-tail deer. Bird species commonly associated with this community type include but are not limited to: American crow, kites, vultures, gulls, Carolina wren, starlings, northern cardinal, orioles, warblers, sparrows, Western kingbird, American kestrel, indigo bunting, northern mockingbird, hawks, owls, blue jay, woodpeckers, and thrushers. Grassland/upland pasture habitat is characterized by grasses interspersed with woody vegetation, with a few stands of trees. Vegetation of this habitat includes: Bermuda grass, Mexican hat, pricklypear cactus, spotted beebalm, gulf cordgrass, annual ragweed, bahia grass, St. Augustine grass, false indigo, and fox-tail bristle grass. Mammals commonly encountered in the grassland/upland pasture habitat include: bobcat, common gray fox, coyote, eastern cottontail, fulvous harvest mouse, raccoon, hispid cotton rat, nine-banded armadillo, striped skunk, and white-tailed deer. Over 50 species of reptiles and about 20 species of amphibians are common to the upland habitat types in the study area. Wetland Resources The wetland types present in the study area include estuarine and palustrine emergent marsh. Estuarine marsh is defined as tidal wetlands that are usually semi-enclosed by land, but have open, partly obstructed or sporadic access to the open ocean, and in which ocean water is at least occasionally diluted by freshwater runoff from land. This community type is often referred to as

6

“intertidal,” and in the study area is connected to the GIWW. The primary factor influencing species composition in estuarine marsh communities is salinity, with salt marsh having higher salinity and intermediate marsh influenced more by freshwater. The dominant species in the Brazos River estuarine habitat include: saltmeadow cordgrass (Spartina patens), gulf cordgrass (Spartina spartinae), saltgrass (Distichlis spicata), salt cedar (Tamarix ramosissima), glasswort (Baccharis spp.), seaside goldenrod (Solidago sempervirens), sand spikerush (Eleocharis montevidensi), Carolina wolfberry (Lycium carolinium), sea purslane (Sesivium maritimum), and bulrush (Scirpus spp.). Palustrine marsh is defined as all non-tidal wetlands dominated by trees, shrubs, persistent emergents, emergent mosses or lichens, and all such wetlands that occur in tidal areas where salinity due to ocean-derived salts is below 0.5 percent. Palustrine wetlands have traditionally been called marshes, swamps, bogs, fens, and prairies, and include small, shallow permanent and intermittent waterbodies often referred to as ponds. Palustrine wetlands occur throughout the Freeport Hurricane-Flood Protection project area and typical vegetation in this habitat includes: leafy three-square (Scirpus maritimus), bigleaf sumpweed (Iva frutescens), frog-fruit (Phyla lanceolata), alligator weed (Alternanthera philoxeroides), rattle-bush (Sesbania drummondii) green flatsedge (Cyperus virens), small-fruit spikerush (Eleocharis microcarpa), needle-rush (Juncus effuses), grassleaf rush (Juncus marginatus), horned beakrusy (Rhynchospora corniculata), retrorse flatsedge (Cyperus retrorsus), and sedge (Carex spp.). Threatened and Endangered Species There are several species that may occur in the project study area that are listed by the U.S. Fish and Wildlife Service and National Marine Fisheries Service as threatened and endangered. They are protected under provisions of the Endangered Species Act of 1973, as amended. Four bird species, the bald eagle, the brown pelican, the piping plover, and the whooping crane are federally listed in Brazoria County. Five marine reptile species, the Atlantic hawksbill sea turtle, the Kemp’s ridley sea turtle, the leatherback sea turtle, the green sea turtle, and the loggerhead sea turtle are federally listed in Brazoria County. Of the fives marine reptile species, the Kemp’s ridley is the only sea turtle where nesting is known to occur in the general area, however, suitable nesting habitat is not available in the immediate project area. The following species are on the State of Texas Protected Nongame list and have the potential to occur in the Freeport Hurricane-Flood Protection study area: reddish egret, sooty tern, swallow-tail kite, white-faced ibis, white-tailed hawk, wood stork, and American/Arctic peregrine falcon.

7

Cultural Resources Past studies have located several historic period sites, 41BO116, 41BO123, 41BO125, and 41BO135 near the mouth of the Brazos River. Three of these sites, 41BO123, 41BO125, and 41BO135 are related to the old towns of Velasco and Quintana. The remaining site, 41BO116 is a gun emplacement, which dates from the American Civil War to World War II. Several inventories for marine resources have also been conducted. None of the anomalies identified during these surveys were believed to be associated with historic vessels. One sunken vessel, the blockade-runner Acadia is thought to be located north of the east Freeport Harbor jetty. Socioeconomic Considerations The Freeport Hurricane-Flood Protection project is located within the greater Brazosport area located in south Texas at the mouth of the Brazos and San Bernard Rivers. The Brazosport area is comprised of nine cities: Brazoria, Clute, Freeport, Jones Creek, Lake Jackson, Oyster Creek, Quintana, Richwood, and Surfside Beach. The economy of the Brazosport area is based in chemical manufacturing, petro-chemical processing, varied other manufacturing, offshore extraction support complexes, deep-water port activities, tourism, and commercial and recreational fishing. The deepwater Port of Freeport handles large volumes of commodities including crude petroleum and petroleum products, petro-chemical products, and agricultural products. The Brazosport area is a popular recreational area and tourism is an important aspect of the local economy. Surfside and Quintana contain miles of beaches and represent a popular destination for greater Houston residents. Fishing, boating, and other water related activities are very popular and the Brazosport area has a fairly large sport fishing fleet. The diversity of coastal habitats in the Brazosport area supports a large variety of shore birds while the large number of adjacent shallow bays provide excellent habitat for waterfowl. The area is popular for waterfowl hunting as well as bird watching activities. Land use within the Brazosport area is divided principally among industrial land, range-pasture land, urban-residential and urban-commercial land, recreational land, park and recreational facilities, and marshlands. Water use includes: mineral production, commercial and recreational fishing, recreation, and transportation. The Brazosport area comprises the southeastern portion of Brazoria County that has a total area of 1,422 square miles and a 2000 population of 241,767. This represents a 26 percent increase over the 1990 population of 191,707. Brazoria County has increased in population at a rate

8

comparable to the State, however the population density of Brazoria County is now more than double that of the state average population density. The largest employers in Brazoria County include the Dow Chemical Company, the Texas Department of Criminal Justice, and a combination of public school districts. The 1999 per capita income in Brazoria County ($20,021) was similar to, but slightly higher than, the 1999 state ($19,617) per capita income.

PROJECT AREA DESCRIPTION The Freeport Hurricane-Flood Protection levee system was authorized through the Flood Control Act of October 23, 1962, Public Law 87-874. The system included the design and construction of new levees, improvements to existing levees, drainage control structures, pump stations, and a tide gate. The design of these features was documented in Design Memorandum (DM) Nos. 1 through 11 dated from June 1965 to July 1975. These documents are held in the Galveston District’s Library at 2000 Fort Point Road. The Freeport Hurricane-Flood Protection system included modifying 30 miles of existing levees, constructing 4 miles of new levees and wave barriers, modifying 1,584 feet of existing floodwalls and constructing 4.3 miles of new floodwalls. Over 14 miles of existing levees were incorporated into the system. The existing Federal levee terminates near Dow Plant B on the west side of the project and near Lake Barbara at Oyster Creek on the east side (Figure 1). The project currently ties into high ground inland in a horseshoe shape and does not completely enclose the area. Local land subsidence after original construction was estimated at generally less than a foot in the majority of the system. New earth levees were constructed and existing levees were modified (strengthened and enlarged) for the Freeport Hurricane-Flood Protection system. Generally the levees were designed with 3(H):1(V) interior slope angles and exterior slope angles ranging from 6:1 where wave attack was severe and 3:1 elsewhere. The levee crown width ranged from 12 to 20 feet and the crest elevations ranged from 15 to 21 feet above Mean Sea Level (MSL). Where excess settlement was anticipated, the levees were overbuilt accordingly. In some areas, existing property restrictions necessitated the use of floodwalls. An inverted “T” design was used for these floodwalls, which typically do not exceed 5 feet in height. In most cases the levees where turfed with Bermuda grass. Where vehicular traffic was anticipated, a 20-foot crown width was maintained, and the crown was topped with a flexible, bituminous surface treaded road. This surface treatment also acted to stabilize the moisture content of the levee and reduce cracking during dry weather. NON-FEDERAL SPONSOR AND COORDINATION The District Engineer, Galveston District, USACE, is responsible for the overall management of the study and report preparation. The VDD is the non-Federal sponsor for the study.

9

PRIOR AUTHORIZATIONS The Galveston District completed the Interim Report on Hurricane Survey of Freeport and Vicinity, Texas on May 15, 1958 to determine the need and economic feasibility for providing hurricane flood protection to Freeport and vicinity, Texas. At that time, local interests had constructed a system of storm protection levees to protect the city of Freeport, the communities of Lake Jackson, Clute, and Oyster Creek, and the adjacent areas on the east side of the Brazos River from storm tides and Brazos River floods. The extant levee protection system measured more than 50 miles including 3.5 miles of federally-constructed levees along the Brazos River diversion channel west of Freeport, which local interests had enlarged. The interim report for Freeport and Vicinity, Texas, recommended that a plan be authorized to provide improvements consisting principally of:

a. Rehabilitation, enlargement, and extension of existing earthen levees protecting the Freeport, Oyster Creek, Lake Barbara, Clute and Jackson areas.

b. Construction of additional pumping stations in the city of Freeport for adequate drainage of enclosed improved areas.

c. Reservation of storage capacities in ponding areas. Local interests would prevent encroachment on ponding areas that would reduce the storage capacities unless equivalent pumping capacities are provided at their expense.

The Freeport Hurricane-Flood Protection Levee System was authorized through the Flood Control Act of October 23, 1962, Public Law 87-874. The system included the design and construction of new levees, improvements to existing levees, drainage control structures, pump stations, and a tide gate. The design of these features was documented in the following Design Memoranda:

Design Memo No. Description Date Approved

1 Extension of Oyster Aug 1964 Creek Levee 2A Hydrology (Surge & Waves) Feb 1965 2A Hydrology (Surge & Waves) Jun 1965 revised

10

Supp No.1 Hydrology (Surge & Waves) May 1966 2A 2B Hydrology & Hydraulics Nov 1965 (Interior Drainage) Supp No 1 Hydrology & Hydraulics Apr 1966 2B (Interior Drainage) Supp No 2 Hydrology & Hydraulics Nov 1966 2B (Interior Drainage) Supp No 3 Hydrology & Hydraulics Apr 1967 2B (Interior Drainage) Supp No 4 Hydrology & Hydraulics Jun 1968 2B (Interior Drainage) 3 East & Oyster Creek Levees Nov 1965 4 General Design Memorandum Jun 1966 5 South Levee and East Bank Nov 1966 Brazos Rive Levee Supp No 1 South Levee Drainage Structure Mar 1967 5 & Portable Electric Gate Operators Supp No 2 Floodwall Stations 161+00 to Oct 1967 5 167+00 East Bank Brazos River Levee 6 East Bank Brazos River Levee Apr 1975 7 Freeport Pump Station Jun 1967 8 Old River South Levee Stations Oct 1967 139+00 to 278+00 & Wave Barrier 8A Tide Gate and Old River Levees Oct 1972 Supp No 1 Tide Gate and Old River Levees Nov 1974 8A 9 Old River North Levee Jan 1972 10 Velasco Pump Station May 1969

11

10 Velasco Pump Station (Revised) Nov 1970 11 Dow Barge Canal & Turning Basin Jul 1975 12 Corrosion Control Jun 1967 STUDY AND REPORT PROCESS In April 2002, the Galveston District completed a 905(b)(1) Analysis for the Freeport and Vicinity, Texas Hurricane-Flood Protection project. The report concluded that the nature and scope of the potential problem, the complexity of the potential issues, and the high level of public interest demonstrate the need for closer study of the Freeport Hurricane-Flood Protection system. This feasibility study follows the recommendations given in the 905(b)(1) Analysis. It includes a detailed analysis of the existing conditions of the Freeport Hurricane-Flood Protection levee system as well as a comprehensive analysis of storm damage potential using state-of-the-art computer modeling. Results of this study form the basis for selection of the recommended plan. The USACE study process provides for a systematic preparation and evaluation of alternate plans which address study area problems and opportunities. This process involves all six functional planning steps: Specify Problems and Opportunities Inventory and Forecast Conditions Formulate Alternative Plans Evaluate Effects of Alternative Plans Compare Alternative Plans Select Recommended Plan The earlier 905(b) Analysis emphasized identification of problems and opportunities. Emphasis in this Feasibility Report is on establishing existing and future conditions, assessment of impacts, and selection of a recommended plan.

12

II. PROBLEM IDENTIFICATION HURRICANE-FLOOD DAMAGE PROTECTION In July 1998, the Risk Control Consulting (RCC) Group of J&H Marsh & McLennan delivered a report commissioned by the Dow Chemical Company titled, Hurricane Vulnerability and Impact Analysis for the Freeport, Texas Facility. RCC was commissioned by the Dow Chemical Company to quantify the hurricane exposure and loss potential at Dow’s Freeport, Texas facility. The report was intended for risk management purposes only. RCC used their Insurance/Investment Risk Assessment System (IRAS) model to evaluate the exposure and loss potential from both historic and stochastic hurricanes. All known past hurricanes were simulated to determine what impact they would have on the Dow Chemical facility if they were to occur today. Additionally, theoretically possible hurricanes were simulated to determine what impact they could have on the Dow facility. Based on the IRAS simulations, the worst-case historic event would result in an aggregate loss equivalent to approximately 6 percent of Dow’s total facility valuation. RCC reported that the simulation results for stochastic or theoretical hurricanes resulted in a loss equivalent to 30 percent of the total facility valuation. As part of their evaluation, RCC produced a series of surge maps defining the maximum envelope of water (MEOW) for Saffir-Simpson scale category 1 through 5 hurricanes. The extent of inland penetration by hurricane induced storm surge flooding was based on the Federal Emergency Management Agency SLOSH (Sea, Lake, and Overland Surges from Hurricanes) model and a 1993 report by Texas A&M University entitled Storm Atlas – Brazoria, Galveston, and Harris Counties, that is also based on the SLOSH model. It must be noted that the SLOSH model is a stochastic model that is not endorsed by USACE for determining frequency analysis for design of a hurricane flood protection system. The results of the SLOSH model are based upon a Joint Probability Method concept in which a family of storms are built through a random combination of storm parameters. Because some of these combinations of events have a negligible chance of occurring, i.e., small pressure deficit and large maximum winds, the resulting analysis tends to be very conservative and may be unrealistic. The inherent deficiencies in the SLOSH model results notwithstanding, Dow used the results to generate a table of maximum water levels for Category 3-5 hurricanes at Oyster Creek City Hall, Freeport City Hall, Clute City Hall, Lake Jackson City Hall, Jones Creek City Hall, Brazosport College, and Angleton City Hall. Reported maximum storm surge water levels ranged from a low of 0 feet (Freeport City Hall, Category 3 storm) to a high of 20 feet (Oyster Creek City Hall,

13

Category 5 storm). This information was distributed to the general public and resulted in significant level of concern expressed in a series of newspaper articles in the local press. Following the distribution of the Dow Chemical Company report, the non-Federal sponsor, VDD, submitted a letter to the Galveston District seeking assistance under Section 216 of the 1970 Flood Control Act which provides for review of completed USACE projects that may have changed because of physical or economic reasons. In their request, VDD cited six major changes that have occurred since the 1958 completion of the Hurricane Survey of Freeport and Vicinity Texas which justify the review: 1) industrial and residential property values have significantly increased, possibly 10 to 100 fold; 2) there has been significant advancement in computer and modeling technology; 3) there is approximately an additional 40 years of actual hurricane data and analysis available; 4) the Brazos River Harbor and Navigation District and the USACE harbor dredging projects have significantly reduced the ponding area outlined in the 1958 study; 5) the VDD has added significant pumping capacity (3,000,000 gpm) relative to the recommended capacity in the study; and 6) possible increased subsidence in the local coastal plain. In regard to the issues raised by the VDD and the Dow Chemical Company report, the preliminary screening performed during the 905(b) Analysis identified three potentially significant problems related to the Freeport Hurricane Flood Protection levee system:

1) The non-Federal sponsor has not performed levee elevation surveys, and thus it is

not possible to accurately compare the Design Memorandum authorized protection levels to the existing protection levels.

2) There may be the potential for wave overtopping and interior drainage and ponding problems at the Oyster Creek Levee location.

3) The original intent of the project was to provide a protection level based on a Standard Project Hurricane, which evolved into a revised, but USACE-approved Hurricane Surge Elevation Design. The analysis of hurricane surge elevations included the analysis of historical flood elevations in the area of Freeport and Galveston, Texas. The highest flood elevation of record was near 14.5 feet at Galveston in 1900, which is below the Freeport levee protection elevation. However, with more storm data available now than in the early 1960’s, the stage frequency curve may have changed slightly.

Based upon the identification of these potentially significant problems, the 905(b) Analysis recommended future analysis focus on the following factors: 1) survey and review existing project conditions; 2) update the hurricane surge elevation versus frequency for the project area utilizing USACE approved storm surge modeling methods; and 3) determine changed conditions

14

and effects on current design criteria. This Feasibility Report documents the methods and results of efforts to resolve these three problem factors.

15

III. FORMULATION OBJECTIVES, CONSTRAINTS, AND CRITERIA NATIONAL OBJECTIVES The fundamental national objective of Federal participation in water resources development projects is to assure that an optimum contribution is made to the welfare of all people. The Water Resources Council’s Economic and Environmental Principles and Guidelines for Water and Related Land Resources Implementation Studies dated March 1983 and the National Environmental Policy Act of 1969 (NEPA) provide the basis for Federal policy for planning Federal water resources projects. These authorities have established the procedures for formulation and evaluation of water resources projects. Additional policies and regulations, derived from executive and legislative authority, further define the criteria for assessment of plan impacts, risk analysis, review and coordination procedures, and project implementation. Current Federal policy dictates that National Economic Development (NED) is the primary national objective in water resources planning. NED objectives stress increasing the value of the Nation’s output of goods and services and improving economic efficiency on a national level. Planning objectives designed to improve NED are concerned with the value of increased output of goods and services resulting from external economics associated with a plan. The Federal objective of water and related land resources planning is to contribute to NED in a manner that is consistent with protecting the Nation’s environment. Consequently, the resource’s condition should be more desirable with the selected plan than under the without-project condition. National objectives are designed to assure systematic interdisciplinary planning, assessment, and evaluation of plans addressing natural, cultural, and environmental concerns, which will be responsive to Federal laws and regulations. PLANNING OBJECTIVES The primary objective of Federal storm damage reduction activities is to contribute to the Nation’s economy while protecting the Nation’s environmental resources in accordance with existing laws, regulation, and executive orders. Single purpose shore protection projects are formulated to provide hurricane and storm damage reduction. Highest priority is for reducing damages to existing development. Plan formulation activities were pursued with the following objectives in mind:

16

1) To evaluate the level of protection provided by the existing Federal project, and

2) Dependent upon the results of (1), to provide a level of protection consistent with the current economics of the project area.

PLANNING CONSTRAINTS Plans must be formulated with regard to addressing the problems and needs of the area, taking into consideration future without-project conditions. The plans should identify tangible and intangible benefits and costs from economic, environmental, social, and regional perspectives. Institutional implementation constraints should be identified. The formulation framework requires the systematic preparation and evaluation of alternative solutions to the recognized water resource related problems within the study area. The process also requires that impacts of the proposed action be measured and results displayed or accounted for in terms of contributions to: NED, Environmental Quality, Regional Economic Development, and Other Social Effects. Interaction with other interests must be maintained throughout the planning process to avoid duplication of effort, minimize conflicts, obtain consistency, and assure completeness. The following constraints apply to this feasibility study:

• Fish and wildlife habitat affected by the project plan should be preserved, if possible; • The study process and plans developed must comply with Federal laws and policies; and • Alternative plans that resolve problems in one area should not create or amplify problems

in other areas. Current guidance specifies that the Federal objective of planning is to contribute to NED consistent with protecting the Nation’s environment. The following general criteria are applicable to all water resource studies and have been used to guide the formulation of this study. Technical, economic, environmental, and social criteria have been established to guide the project development process. These criteria are discussed below. TECHNICAL CRITERIA Technical criteria are the minimum parameters of project design that ensure that the project function and structure meet the needs of the customer. Technical criteria significant to the Freeport Hurricane Flood Protection system include: forcing function criteria, configuration criteria, geotechnical criteria and maintenance criteria. These criteria are partly interrelated and partly independent of each other. Forcing function criteria must always be seen in relation to configuration and geotechnical criteria. For example, the “design storm” and the “Standard

17

Project Hurricane” are obsolete concepts. Short- and long-term wave statistics are needed, and the hydrodynamics of wave interaction with its surroundings must be known in detail. The parameters of wave height, wave period, storm duration, and surge water level are quasi-independent components of a storm whose effect on a project design must be understood to adequately examine alternatives. The frequency of occurrence (or the return period) of an event (such as a storm) is a primary component of wave (forcing function) criteria. ECONOMIC CRITERIA The economic criteria require that tangible benefits attributable to a project exceed project costs. Project benefits and costs are reduced to average annual equivalent values and related in a ratio of benefits to costs (Benefits-to-Costs ratio or BCR). This ratio must exceed unity to meet the NED objective. Selected plans, whether structural, nonstructural, or a combination of both, should maximize excess benefits over costs; however, unquantifiable features must be addressed subjectively. These criteria are used to develop plans that achieve the objective of NED and provide a base condition for consideration of economically unquantifiable factors which may impact project proposals. All structural and nonstructural measures for hurricane storm damage reduction projects should be evaluated using the appropriate period of analysis and the currently applicable interest. Total annual costs should include amounts for operation, maintenance, major replacements, and mitigation, as well as amortization and interest on the investment. ENVIRONMENTAL CRITERIA The general environmental criteria for water resource projects are identified in Federal environmental statutes, executive orders, and planning guidelines. It is the national policy that fish and wildlife resource conservation be given equal consideration with other study purposes in the formulation and evaluation of alternative plans. The basic guidance during planning studies is to assure that care is taken to preserve and protect significant ecological, aesthetic, and cultural values, and to conserve natural resources. These efforts should provide the means to maintain and restore, as applicable, the desirable qualities of the human and natural environment. Alternative plans formulated to provide hurricane storm damage protection should avoid damaging the environment to the extent practicable and contain measures to minimize or mitigate unavoidable environmental damages. Particular emphasis is placed on the following:

18

• Protection, preservation, and improvement of the existing fish and wildlife resources along with the protection and preservation of estuaries and wetland habitats and water quality;

• Consideration in the project design of the least disruptive construction techniques and methods;

• Mitigation for project-related unavoidable impacts by minimizing, rectifying, reducing or eliminating, compensating, replacing, or substituting resources;

• Preservation of significant historical and archaeological resources through avoidance of effects. This is the preferable action to any other form of mitigation since these are finite, non-renewable resources.

SOCIAL AND OTHER CRITERIA Plans proposed for implementation should have an overall favorable impact on the social well being of affected interests, and have overall public acceptance. Structural and nonstructural alternatives must reflect close coordination with interested Federal and State agencies and the affected public. The effects of these measures on the environment must be carefully identified and compared with technical, economic, and social considerations and evaluated in light of public input. PLAN FORMULATION RATIONALE The rationale for formulating and developing alternative solutions is discussed in the following paragraph. The planning framework requires the systematic preparation and evaluation of alternative ways of addressing problems, needs, concerns, and opportunities while considering environmental factors. The criteria and broad planning objectives previously identified form the basis for subsequent plan formulation, screening, and ultimately plan selection. The planning process for this study has been driven by the overall objective of providing hurricane storm damage reduction benefits for Freeport, Texas, and the surrounding community. The first phase of this process was to establish the existing conditions of the Federal project. Next, the expected future without-project conditions were developed to evaluate the level of protection provided by the existing Federal project and for comparison with other alternatives.

19

IV. PLAN FORMULATION WITHOUT-PROJECT CONDITION/NO ACTION The USACE planning guidance requires analysis of a “without-project” plan as one of the alternatives. Also, to comply with the requirements of NEPA, a No Action plan must be included in the alternatives array. The “without-project” plan is synonymous with the No Action Plan. The “without-project” plan also forms the basis for analysis with other alternative plans. A necessary first requisite to assess the without-project plan was to establish the existing conditions of the Freeport Hurricane-Flood Protection levee system. This required a topographic survey of the existing levee system and comparison with the design elevations as well as a review of the structural integrity of the levees. The future without-project conditions were established by deriving the stage-frequency relationships for storm surge frequency-of-occurrence including the wave run-up and levee overtopping with associated risk and uncertainty. The risk-based analysis is not only a requirement, but is critical for engineering performance evaluations and design of USACE levees. The Freeport Hurricane-Flood Protection levee is an embankment whose primary purpose is to furnish flood protection from seasonal high water and which may be subject to water loading for periods of a few days or a few weeks a year. The Freeport Hurricane-Flood Protection levee meets normal flood levee protection design requirements: (a) levee embankment may become saturated for only a short period of time beyond the limit of capillary saturation, (b) levee alignment is dictated primarily by flood protection requirements, and (c) borrow is generally obtained from shallow pits or from channels excavated adjacent to the levee. The Freeport Hurricane-Flood Protection levee system consists of 16 segments that vary in length from less than 500 feet to greater than 29,000 feet in length (Table 1). Topographic survey profiles were collected for each levee segment to establish the existing average levee elevation for that segment (Table 1). Site visits were also conducted and are documented in the Engineering Appendix (Appendix A). Target elevations were estimated and compared to survey profiles collected on existing levees and referenced to original levee design documents (reports, cut-sheets, etc.) to verify that average height requirements were met. As noted in the figures, maps and tables of deterministic analysis data (Appendix A), it was determined that any settlement, shrinkage, cracking, geologic subsidence, and previous construction tolerances had not significantly undermined or impacted the levee system overall height protection requirement along the perimeter.

20

Table 1.

Freeport Hurricane-Flood Protection Levee System Layout.

Station Limits Levee Name Length Sta. Sta. Year Built Average

Elev. (ft) Levee Crown

(ft) EAST BANK BRAZOS RIVER LEVEE (1) 13,810 0+00 138+10 - 16 12 WASTE WATER CANAL LEVEES (1) 511 0+00 5+11 - 20 12 WASTE WATER CANAL LEVEES (2) 385 0+00 3+85 - 20 12 EAST BANK BRAZOS RIVER LEVEE(2) 25,269 128+00 380+69 1970 17-19.5 12-20 SOUTH STORM LEVEE 18,795 0+00 187+95 1968 17.5-21.25 20 NORTH WAVE BARRIER 6,687 0+00 66+87 1972 16.5-18.5 20 SOUTH WAVE BARRIER 2,574 66+87 92+61 1972 21.5 20 OLD RIVER SOUTH LEVEE 14,533 133+00 278+33 1972 16.0-17 12 OLD RIVER SOUTH LEVEE NEW 4,389 0+00 43+89 - 16.0-17 12 OLD RIVER NORTH LEVEE 19,128 127+00 318+28 1973 16.5-17 16-20 OLD RIVER NORTH LEVEE BULKHEAD 5,374 0+00 53+74 - 16.5-17 16-20 DOW BARGE CANALS SOUTH LEVEE 31,828 0+00 318+28 - 16-19 20 DOW BARGE CANALS NORTH LEVEE 29,199 0+00 291+99 - 16-19 20 EAST STORM LEVEE 18,900 11+00 200+00 1968 22 20 OYSTER CREEK LEVEE 29,354 200+00 493+54 1967 15.5-22 20

EXTENSION OF OYSTER CREEK LEVEE 11,217 669+00 781+17 1965 16.5 20

In order to establish the without-project conditions, a risk-based storm surge frequency-of-occurrence (storm impact frequency analysis) was performed. The results of this comprehensive analysis of storm damage potential are documented in Risk Assessment for Exposure to Hurricane Conditions at Freeport, Texas (Appendix B). The finite element numerical hydrodynamic model Advanced Circulation (ADCIRC), (Leutich, R.A., Westerink, J.J., and Scheffner, N.W., 1992) was used to numerically simulate the propagation of tides and storms across the Gulf of Mexico into the study area. The model was verified by comparing model-generated tide time series with the corresponding time series reconstructed from existing harmonic analyses and on-site measurements of surface elevation. Storm event simulations were verified by comparing simulated results of water surface elevation with archived storm measurements. The study used the model to reproduce all 27 historic storm events that significantly impacted the study area since 1886. In order to insure that the most severe events have been included for all open coast stations, simulations also included 10 hypothetical events that could likely occur. The storm surges were compiled with tide conditions to represent high, slack and low water for neap, spring and mid range tides for use with the statistical analysis procedure, Empirical Simulations Technique (EST). EST used the historic and hypothetical storm events to generate a large population of life-cycle databases that were used to compute mean value maximum storm

21

surge elevation frequency relationships. Frequency computations were made at 35 levee locations in the Freeport area (Figure 2). The maximum storm surge plus tide value for the 200-year storm was found to be 17.9 feet (5.47 m) and occurred in Oyster Creek (Figure 3). For the 100-year storm the maximum storm surge plus tide value of 13.1 feet (3.98 m) also occurred in Oyster Creek (Figure 4), as did the 9.5 feet (2.90 m) maximum for the 50-year storm (Figure 5). None of the storm surge plus tide values were greater than the current height of the levee system. Wave conditions at the toe of the levee included predicting maximum waves generated by the storm at an appropriate offshore position and determining the transformation of those waves as they approach the levee. The initial wave transformation was performed using the STWAVE (Steady-state spectral WAVE) model. Results from STWAVE were used to compute the boundary conditions for the nonlinear model COULWAVE (Cornell University Long and Intermediate Wave Modeling Package) that modeled wave transformation when the storm surge was high enough to overtop the deltaic headlands; a condition that would allow the offshore waves to attack the levees. COULWAVE also computed the wave run-up on the levees.

The results documented in the Engineering Appendix (Appendix A) and the hurricane risk

assessment (Appendix B) provide estimates of the various indices for engineering performance

based on existing USACE guidance and procedures. Analyses were limited to engineering

performance or non-economic related performance for the feasibility study. Analyses performed

used current state-of-the-art numerical models with current methods and current data.

Appendices A and B document that the Freeport Hurricane-Flood Protection levee system

maintains a storm surge capacity of a 1 per cent exceedance probability (100-year). Appendix B

concludes that the levee system provides sufficient protection for an extreme 1per cent event

(100-year) against storm surge plus wave run-up. Results are summarized in Table 2.

22

Figure 2. Location of stations used for input to EST model.

Figure 3. 200-year storm surge values, Freeport Hurricane-Flood Protection Levee.

GULF OF MEXICO

BRAZOS RIVER

FREEPORT

Values are in meters

23

Figure 4. 100-year storm surge values, Freeport Hurricane-Flood Protection Levee.

Figure 5. 50-year storm surge values, Freeport Hurricane Flood Protection Levee.



24

Table 2 Levee Analysis Breakdown versus Modeled Storm Surge

Station Limits H&H Risk Assessment Report

Data (ft)

Levee Name Length

(ft) Sta. Sta. Average Elev. (ft)

50-Year Storm Surge

100-Year Storm Surge


East Bank Brazos River Levee - 1 13,810 0+00 138+10 16 7.0-9.0 11.0-13.0 15.0-17 Wastewater Canal Levees -1 511 0+00 5+11 20 7.0-9.0 11.0-13.0 15.0-17 Wastewater Canal Levees -2 385 0+00 3+85 20 7.0-9.0 11.0-13.0 15.0-17 East Bank Brazos River Levee - 2 25,269 128+00 380+69 17-19.5 7.0-9.0 11.0-13.0 15.0-17 South Storm Levee 18,795 0+00 187+95 17.5-21.25 7.0-9.0 11.0-13.0 15.0-17 North Wave Barrier 6,687 0+00 66+87 17.5-18.5 7.0-9.0 11.0-13.0 15.0-17 South Wave Barrier 2,574 66+87 92+61 21.5 7.0-9.0 11.0-13.0 15.0-17 Ole River South Levee 14,533 133+00 278+33 16.0-17 7.0-9.0 11.0-13.0 15.0-17 Old River South Levee New 4,389 0+00 43+89 16.0-17 7.0-9.0 11.0-13.0 15.0-17 Ole River North Levee 19,128 127+00 318+28 16.5-17 7.0-9.0 11.0-13.0 15.0-17 Ole River North Levee Bulkhead 5,374 0+00 53+74 16.5-17 7.0-9.0 11.0-13.0 15.0-17 Dow Barge Canals South Levee 31,828 0+00 318+28 16-19 7.0-9.0 11.0-13.0 15.0-17 Dow Barge Canals North Levee 29,199 0+00 291+99 16-19 7.0-9.0 11.0-13.0 15.0-17 East Storm Levee 18,900 11+00 200+00 22 7.0-9.0 11.0-13.0 15.0-17 Oyster Creek Levee 29,354 200+00 493+54 15.5-22.0 7.0-9.0 11.0-13.0 15.0-17 Extension of Oyster Creek Levee 11,217 669+00 781+17 16.5 7.0-9.0 11.0-13.0 15.0-17

25

V. PUBLIC COORDINATION On June 3, 2004, a public meeting was conducted in conjunction with the Brazosport Emergency Planning Committee (BEPC) at the annual Hurricane Preparedness Seminar in Lake Jackson, Texas. During the meeting, Dr. Jeff Waters of the Galveston District provided an overview of the feasibility study process and the plan formulation objectives, constraints, and criteria. Dr. Billy Edge (Texas A&M University) followed with a detailed presentation on the results of the ADCIRC study, Risk Assessment for Exposure to Hurricane Conditions at Freeport, Texas. The floor was then opened to questions and comments from the public. Only two comments were received related to the feasibility study and both were complimentary of the study process and results. The remaining questions and comments were in reference to general evacuation procedures during tropical storm threats and were referred to public officials from the BEPC for response.

26

FREEPORT AND VICINITY, TEXAS HURRICANE-FLOOD PROTECTION

ENGINEERING APPENDIX 03 August 2004

1

1. Purpose. This appendix presents the technical considerations and evaluations made in developing and analyzing the engineering features that resulted in the decision to take no further action in studying or modifying the Freeport Hurricane Flood Protection levee system. A report, Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas, dated February 29, 2004, prepared by Dr. Billy Edge for the Velasco Drainage District, together with land surveys completed by the U.S. Army Corps of Engineers, Galveston District, were the primary sources of information used in this study. Because the local sponsor is satisfied that information from Dr. Edge’s report along with land surveys is adequate to make the determination of no further action, the Feasibility Study was not completed, and this appendix only documents work that was done and is not considered a formal Engineering Appendix. 2. Civil 2.1 Summary The Freeport Hurricane Flood Levee Protection Feasibility Study included performing an analysis of the existing elevations, slope and relief of the levee systems and relating this to the as-built conditions. The existing federal levee terminates near Dow Plant B on the west side of the project and near Lake Barbara at Oyster Creek on the east side (Map 1). The project currently ties into high ground inland in a horseshoe shape and does not completely ring the area. Local land subsidence after original construction was estimated at generally less than a foot in the majority of the system. Utilizing updated topographic surveys about the perimeter, this analysis focused on comparing the surges or predictive flow over the top of the levee by stillwater surge for the given storm frequencies. The Freeport Hurricane Flood levee protection elevations meet the original design and satisfy to a reasonable extent the modeled storm surge elevations for the mean 50-year, 100-year and 200-year storms, as defined in Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas, approximately at or above the elevations specified by the original design. This analysis on the levee system presents an overall representation of elevation characteristics and generalized data tables along with photo displays to identify the levee system in sufficient enough detail to determine no immediate upgrade to meet overtopping at modeled 100-yr surge is recommended at present. Table 1 summarizes the individual levees that make up the Freeport Hurricane Flood Protection Levee System. Table 2 through Table 13 present specific features of each of these levees. Common features of this analysis included the interpretation and use of general existing survey levee profiles and other acceptable data collection techniques to assess the problem of determining the amount of significant settlement to the levee system (if any) and making a learned recommendation based on analytical findings. Since the results from this analysis, as mentioned, found no significant changes to the levee heights (elevations per 500-1000 linear feet) in the system, no further action was required. Based on this analysis and results from a Project Delivery Team (PDT) site visit indicate that the levee has been kept in adequate Operations and Maintenance condition allowing the levee system to continue to satisfactorily meet and/or exceed as-built engineering performance design criteria as it relates to elevations.

2

This analysis of the overall levee system integrity focused only on plan/profile requirements and historical as-built conditions.

MAP 1. Freeport Hurricane Flood Protection Levee System

2.2 General Considerations The Freeport Levee as used herein is defined as an embankment whose primary purpose is to furnish flood protections from seasonal high water and which may be subject to water loading for periods of a few days or weeks a year. Freeport meets normal levee flood protection design requirements: (a) levee embankment may become saturated for only a short period of time beyond the limit of capillary saturation, (b) levee alignment is dictated primarily by flood protection requirements, and (c) borrow is generally obtained from shallow pits or from channels excavated adjacent to the levee. Levees are broadly classified according to the area they protect or according to use. An urban levee, such as the FHP levee, is one that provides protection from flooding in communities, including industrial, commercial, and residential (EM 1110-2-1913). The principal causes of

3

levee failure are: overtopping, surface erosion, internal erosion (piping), and slides within the levee embankment or foundation soils. Given the satisfactory condition of the Freeport levee as noted from site visits (Photos 1-9) and interpretive review of the survey profiles, the summary of findings together with the summary of storm surge elevations from Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas are listed in Table 14. Recommended corrective actions consist primarily of monitoring condition of access and service roads where problem areas are likely to develop (along SSTML) and an adaptive plan may typically be employed to ensure continued levee elevation requirements are met. This should include the following: 2.2.1 Settlement Long-term monitoring of settlement of the levees due to foundation consolidation may be performed annually. The data should be compared with the expected design rate (less than 1 foot per year). Settlement monitoring of the levees should be continued annually until the analyses of the survey data shows that the rate of and amount of settlement are within design expectations. Should the rates and amount of settlement prove unacceptable then corrective measures may be recommended and action taken. 2.2.2 Annual Inspections A walkover inspection of the levees should be performed on a regular basis. The frequency would be based upon determining that the performance of the levee features (including hard structures, culverts, pipelines), and site in general, are in accordance with design expectations (average elevation minimum 16 feet MSL). The inspection should look for erosion problems and evidence of burrowing mammals. 2.2.3 Maintenance This will consist of corrective action in response to problems identified when monitoring levee conditions. Actions could include adding material to compensate for excessive settling or erosion, repair of storm damage, reinforcing the levee surface to withstand erosion in problem areas (to the minimum extent necessary), repair drainage structures, or control of burrowing rodents. Any rodent control efforts will need to be carefully planned and executed to avoid negative impacts to adjacent habitats. 2.3 Conclusion In conclusion, this above subjective analysis summarizes the nodal (metric) elevations around the perimeter as reported in the Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas, by Dr. Billy Edge and relates this data to the average existing heights of the Freeport Levee system as determined by survey. It concludes that the existing heights are within a satisfactory tolerance for the 50, 100 and 200- year storm surges, as defined in Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas, respectively. This analysis study provides a general framework detailing levee performance as it relates to the elevations.

4

The fundamental technical basis for the study is documented in the Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas report and consisted of calculations for waves, wave runup, and overtopping elevations. Using information presented in this report, average target elevations were estimated and compared to survey profiles collected on existing levee and referenced to original levee design documents (reports, cut-sheets, etc.) to verify average height requirements were met. As noted in the figures, maps and tables of deterministic analysis data, it was obvious by general accounts any settlement, shrinkage, cracking, geological subsidence, and previous construction tolerances had not significantly undermined or impacted the levee system overall height protection requirement along the perimeter. The individual characteristics of each levee reach meets sufficient protection levels against the required 100-year storm surge, as defined in Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas. 3. Geotechnical 3.1 Purpose The objective of this section was to review the status of the Freeport Hurricane Flood Levee Protection System in light of the findings presented in the recent Risk Assessment for Exposure to Hurricane Conditions at Freeport, Texas, dated February 29, 2004, by Dr. Billy L. Edge of the Texas Engineering Experiment Station (Texas A&M University). This included reviewing original geotechnical information, design methods, as-built drawings, and findings of periodic inspections of the levee protection system and conducting additional geotechnical investigations and studies if warranted. In conjunction with the risk assessment study, the Galveston District conducted profile surveys of the levee protection system in 2003. Based on survey results, the study indicated that no overtopping of the levee system would occur during the 100-year storm plus tide event, as defined in Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas. As a result of the conclusions of the aforementioned study, review of existing data, design methods, as-built drawings, and periodic inspection reports, no additional geotechnical investigations or studies were deemed necessary. The following paragraphs summarize the geotechnical aspects of the Freeport Hurricane Flood Levee Protection System. 3.2 Design Memoranda The Freeport Hurricane Flood Levee Protection (HFP) System was authorized through the Flood Control Act of 23 October 1962, Public Law 87-874. The systems included the design and construction of new levees, improvements to existing levees, drainage control structures, pump stations, and a tide gate. The design of these features was documented in Design Memorandum (DM) Nos. 1 through 11 dated from June 1965 to July 1975. These documents are held in the District’s Library at 2000 Fort Point Road and were reviewed as part of this study. In particular, the geotechnical content of the following documents was reviewed:

DM No. 1 – Extension of Oyster Creek Levee DM No. 3 – East and Oyster Creek Levees

5

DM No. 4 – General Design Memorandum DM No. 5 – South Levee and East Bank Brazos River Levee DM No. 6 – East Bank Brazos River Levee DM No. 7 – Freeport Pumping Station DM No. 8 – Old River South Levee, DM No. 11 – Dow Barge Canal Levees and East Levee

As identified in DM No. 4, the Freeport Hurricane Flood Protection System included modifying 30 miles of existing levees, constructing 4 miles of new levees and wave barriers, modifying 1,584 feet of existing floodwalls and constructing 4.3 miles of new floodwalls. Over 14 miles of existing levees were incorporated into the system. 3.3 Geologic Conditions

Freeport is located in the Gulf Coastal Plain Physiographic Province. Specifically, Freeport is sited within the deltaic flood plain of the Brazos River. During the Pleistocene age, as a result in a drop in sea level during glacial advances, the river eroded a wide valley into the underlying Beaumont Formation. With the subsequent rise in sea level with during periods of glacial retreat, the eroded valley was flooded and filled with recent sediments consisting primarily of clays, silts, and sands. The thickness of these recent sediments generally exceeds 50 feet and was not penetrated during the numerous soil boring explorations taken for the various studies for the Hurricane Flood Protection System (HFP). The HFP system traverses a variety of geologic features including the Bryan Mound, a post-Pleistocene age piercement type salt dome and a topographic high located along the South Levee; marshlands located in the Recent sediments; and natural levees of the old Brazos River. 3.3.1 Stratigraphy

Numerous soil borings, both undisturbed and auger borings were taken for the project between 1964 and 1975. Generally these borings disclosed near surface (less than 15 feet) soils that consisted of soft clays sands and silts in the Southern portion of the project while the near surface soils in the Northern portion of the project consisted of medium to very stiff clays, dense sands and silts. Generally throughout the project, the soils below about 15 feet consisted of stiff to very stiff reddish brown clay. 3.3.2 Shear Strength Determination

Soil strengths for foundation and levee design were made on typical materials in the project area. Methods for determining these strengths included direct shear, triaxial “R” and “Q” tests, and unconfined compression tests. Test results were plotted based on similar consistencies. Design strengths were determined by the dividing lined between the lower 1/3 and upper 2/3 of the points plotted. The friction angle for long-term strength values (zero cohesion) ranged from 20 degrees to 28 degrees. Cohesion for short-term strength values, end of construction case, (zero friction angle) ranged from 200 to 800 pounds per square foot.

6

3.3.3 Slope Stability

Slope stability analyses were conducted for levee design using both computer and hand run methods. The computer program was developed by the Waterways Experiment Station and used a General Electric 225 computer. The computer program evaluated the circular arc modes of failure for both end of construction and long term loading conditions. Wedge type modes of failure were analyzed using hand methods for both end of construction and long term loading conditions. Generally a factor of safety of at least 1.25 was required for failure arcs passing thru the levee foundations. Factors of safety less that 1.25 were permitted for shallow arc failures of interior slopes where a slide would result in a relatively small volume of earth movement and would not impair the integrity of the flood protection system. 3.3.4 Bearing capacity

Terzaghi’s bearing capacity equation was used to design various pump stations and drainage control structures. Due to the large size of these structures and their lightly loaded condition, the bearing capacity of the native materials was generally suitable. However, at specific locations underlying clay material was undercut due to low coefficient of friction values to resist sliding. 3.3.5 Settlement

Terzaghi’s consolidation theory was used to analyze the anticipated settlement of the levee sections. Settlement values were computed using results of consolidation test on representative materials. Computed values ranged from a negligible amount where foundations were stiff to as much as 12 inches where soft foundations were encountered. 3.3.6 Seepage

Seepage beneath and through the levees was considered negligible for the project due to the relatively impervious clays that make up the foundations and levee backfill, the short duration of high tide, the low head, and the length of the seepage path. 3.3.7 Levees

New earth levees were constructed and existing levees were modified (strength and enlarged) for the HFP system. Generally the levees were designed with 3(H):1(V) interior slope angles and exterior slope angles ranging from 6:1 where wave attack was severe and 3:1 elsewhere. The levee crown width ranged from 12 to 20 feet and the crest elevations ranged from 15 to 21 feet above Mean Sea Level (MSL). Where excess settlement was anticipated, the levees were overbuilt accordingly. In some areas, existing property restrictions necessitated the use of floodwalls. An inverted “T” design was used for these floodwalls, which typically do not exceed 5 feet in height. In most cases the levees where turfed with Bermuda grass. Where vehicular

7

traffic was anticipated, a 20-foot crown width was maintained and the crown was topped with a flexible, bituminous surface treaded road. This surface treatment also acted to stabilize the moisture content of the levee and reduce cracking during dry weather. 3.3.8 Slope Protection

Graded riprap, in conjunction with blanket stone, was used throughout the project to provide energy dissipation and protect levee slopes from erosion by wave action. Slope protection was provided on various 3:1 and 6:1 slopes, at outlet structures and other drainage structures. 3.4 Periodic Inspection Reports Periodic inspections of the HFP system were conducted in accordance with ER 1110-2-100, Periodic Inspection and Continuing Evaluation of Completed Civil Works Structures, and were conducted on a frequency not exceeding 5 years. Reports of these inspections are held in the Operations Branch, Operations Division in the District’s Galveston office. Beginning in 1980, five inspections have been conducted to date. These reports were reviewed to identify geotechnical issues. A summary of findings is provided in the following sections. 3.4.1 Periodic inspection No. 1 (April 1980)

Construction of the HFP systems was 95% complete with completion due in the summer of 1980. The levees appeared in good condition. 3.4.2 Periodic inspection No. 2 (August 1984)

Levees appeared in good conditions with the following exceptions.

• North Wave Barrier Levee – severe erosion along GIWW, sponsor to repair. • Dow Barge Canal Turning Basin Levee – eroded toe, levee requires bank stabilization. • Old River South Levee – portion of the inside toe of levee is undercut needs repair. • South Levee and East Bank Brazos River Levee (Bryan Mound area) - has been degraded

due to erosion and vehicular traffic, needs to be restored. • Sheet pile wall at Dow Plant A Wastewater Structure – corroded sheeting needs

investigation and subsequent repair. 3.4.3 Periodic inspection No. 3 (September 1989)

Levees appeared well maintained and in good conditions. Previous deficiencies were repaired. Small sinkholes were noted on the Dow Plant A Levee. 3.4.5 Periodic inspection No. 4 (May 1994)

• East Storm Levee Pump Station – slope protection missing or displaced, needs repair. • Oyster Creek Levees – some sloughing and erosion noted.

8

• Old River North Levee – surface treatment in need of repair. • East Bank Brazos River Levee – erosion of toe resulting in some sloughing, slope

protection needed. 3.4.6 Periodic inspection No. 5 (June 1999)

• Dow Barge Canal South Levee – erosion of slopes noted and levee toe was being

repaired and shore protection installed. • Some re-turfing of levees being conducted as part of routine maintenance.

3.5 Conclusions This study was conducted to review the geotechnical aspects, both design and performance, of the existing Freeport Hurricane Flood Levee Protection System. Based on a review of the original design memoranda (DM), the methods used, including sampling and testing procedures, classification of materials, evaluation of shear strength, slope stability analysis, foundation design, and settlement analyses are consistent with today’s state of the art practice and are considered sound engineering practices. Furthermore, the performance of the facilities, as observed through the periodic inspections, supports the soundness of the original designs. Therefore, based on the conclusions of the Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas that the existing levee system will not be overtopped during the design storm, and the observations to date, no additional geotechnical engineering studies are required at this time.

4. Hydraulics and Hydrology This section of the Engineering Appendix introduces the hydrologic engineering goals and procedures for the Freeport and Vicinity Flood Damage Reduction Feasibility Study. 4.1 References

• “Risk Assessment for Exposure to Hurricane Conditions at Freeport, Texas,” Prepared for Vesasco Drainage District, June 3, 2004, by Billy L. Edge, Manish Aggarwal, Oscar Cruz-Castro, Robert E. Randall, Patrick J. Lynett, Department of Civil Engineering, Texas A&M University.

• U.S. Army Corps of Engineers. August 2002. “Coastal Engineering Manual Volume II,”

EM 1110-2-1100(Part II), Section II-5-5, Department of Army, Washington D.C.

• U.S. Army Corps of Engineers. January 1995. "Hydrologic Engineering Requirements for Flood Damage Reduction Studies,” Engineering Manual 1110-2-1419, Department of the Army, Washington D.C.

9

• U.S. Army Corps of Engineers. August 1996. "Risk-Based Analysis for Flood Damage Reduction Studies,” Engineering Manual 1110-2-1619, Department of the Army, Washington D.C.

4.2 Introduction The Freeport and Vicinity Flood Damage Reduction Feasibility Study is a re-evaluation of the Federal project that provides flood damage reduction in the vicinity of Freeport, Texas, and other communities including Oyster Creek, Lake Barbara, Clute, and Lake Jackson. Freeport is an important industrial center and a deep-water port. The Federal project consists of a system of levees and pumping stations that were last modified in 1982. The study originated from the non-Federal sponsor’s request to re-evaluate the existing levee system based on current data and design criteria. The non-Federal sponsor for the Feasibility Study is the Velasco Drainage District. The Velasco Drainage District contracted the Civil Engineering Department, University of Texas A&M to model the coastal areas for Freeport and vicinity and to perform storm surge and wave run-up analyses using risk-based procedures. The result was a detail report, entitled "Risk Assessment for Exposure to Hurricane Conditions at Freeport, Texas,” Prepare for Vesasco Drainage District, June 3, 2004, by Billy L. Edge, Manish Aggarwal, Oscar Cruz-Castro, Robert E. Randall, Patrick J. Lynett, Department of Civil Engineering, Texas A&M University. The report provides detail theory, analysis procedures, and results for the feasibility study. 4.3 Study Area The study area consists of about 42 square miles and is located in the vicinity of Freeport, Texas in Brazoria County. The Federal project includes a system of levees and pumping stations that provide flood damage reduction to the project area. The system of levees is referred as the Freeport Hurricane Levee and its interior areas are depicted in Figure 1. The main areas consist of the East area, the Velasco area, and the Freeport area. The East area includes the communities of Oyster Creek, Lake Barbara, Clute, Lake Jackson, and Dow Chemical Plant “B.” The Velasco area includes the formerly separate town of Velasco, and Dow Chemical Company Plant “A.” The Freeport area includes the portion of Freeport lying west of the Old Brazos River channel, and some shipping developments. 4.4 Hydrologic Objectives Hydrologic objectives encompass a risk-based framework as required by ER 1105-2-100, EM 1110-2-1419, and EM 1110-2-1619. The initial hydrologic goals were to derive stage-frequency relationships for storm surge frequency-of-occurrence including wave run-up and levee overtopping with associated risk and uncertainty. The risk-based analysis is not only a requirement, but is critical for engineering performance evaluations and design of Corps of Engineers levees. Another hydrologic objective for the feasibility study was to perform hydrologic analyses to evaluate potential flood damage reduction alternatives including the possibility of upgrading the levee system. Hydrologic analyses for this latter goal and for levee overtopping were however

10

not accomplished since the non-Federal sponsor used his prerogative to suspend the feasibility study as discussed in the main report. Thus, the hydrologic goal for this feasibility study is limited to the non-economic performance for the Freeport Hurricane Levee or in describing its engineering performance. 4.5 Models The non-Federal sponsor’s contractor to perform the core hydrologic engineering analyses for the feasibility study used several numerical models in the study. The following is just a brief mention of the models but Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas includes detail descriptions of all models and procedures. The Planetary Boundary Layer (PBL) model was used to develop wind and atmospheric data for the finite element numerical hydrodynamic model Advanced Circulation (ADCIRC). ADCIRC was used to generate and reproduce historical storms and also hypothetical storms. The model was well calibrated and reliable for the study. The Empirical Simulation Technique (EST) model was used for the storm impact frequency analysis. The EST is a current state-of-the-art model that basically uses a specific location database of historical storms and augments the database with other possible future storms for that site. The augmented hypothetical storms are derived from the historical storm database and modified so that, for example, they are slighted shifted in track, occur at different tides, and have a different strike location. Other models used for wave setup run-up analyses were the STWAVE and the COULWAVE models. These models, procedures, and their results are explained in detail in Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas. 4.6 Engineering Performance A risk-based storm surge frequency-of-occurrence analysis (storm impact frequency analysis), Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas, was performed as mentioned above by Texas A&M for this feasibility study. From the report’s results, different measures for engineering performance can be calculated. Tables 16 and 17 depict measures of engineering performance for only a few locations of Freeport Hurricane Levee corresponding to nodal Stations 2, 14, 26, and 33. Nodal Stations 2 and 14 represent some of the lower top of levee elevations. These and other nodal Stations and corresponding locations are depicted in Appendix C, Figure 3.1 of Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas. Nodal Stations 2, 14, 26, and 33 are located at Freeport Harbor, Dow Barge Canal North Levee, East Levee, and the South Levee, respectively. In Figure 1, the Dow Barge Canal North Levee is labeled “U”, East Levee is “B”, South Levee is “C”, and Freeport Harbor is in the area labeled “Y”. 4.6.1 Stage-Frequency Variance Uncertainty The results from the EST frequency analysis, Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas, include a plus one standard deviation representing an estimate of

11

error or variance of the computed stage-frequency. The referenced Coastal Engineering Manual (USACE, 2002) suggests that this “variation is well-suited to the development of design criteria requiring the quantification of the element of risk associated with the frequency predictions.” Utilizing these standard deviations from Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas and assuming a normal distribution for the uncertainty distribution, the following measures of engineering performance are estimated below. 4.6.2 Annual Exceedance Probability EM 1110-2-1619 defines expected annual exceedance probability, a key index of engineering performance, as a measure of the likelihood or probability of exceeding a specified stage in any given year. Stage-frequency for storm surge is presented in Appendix C, Storm Surge Frequency-of-Occurrence Relationships, of Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas. This appendix also has tables showing storm surge frequency-of-occurrence relationships in return-years. The standard deviation of variance is also included. Since engineering performance is usually described in percent probabilities, Table 15 shows a sample correlation of return-year versus percent-annual exceedance probability. Table 16 shows the measure of annual exceedance probability in percent for corresponding top of levee at nodal Stations 2, 14, 26, and 33. From Table 16, for example, at nodal Station 2, the annual exceedance probability is 0.58 percent. This implies, that the annual maximum surge elevation in any year has a 0.58-percent chance (0.0058 probability) of exceeding the elevation of the top of levee. Top of levee elevations for nodal Stations 26 and 33 are higher than those included in the stage-frequency tables in Appendix C of Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas. The maximum referenced frequency in the stage-frequency tables is the 200-yr (0.50% annual exceedance probability). To extrapolate values from the stage-frequency ratings beyond the 200-yr, the HEC-FDA program was used. The stage-frequency ratings were extrapolated using the graphical method in the “Exceedance Probability Functions with Uncertainty” module in HEC-FDA program. The program basically plotted and automatically extended the rating. From this, the corresponding exceedance probabilities for the top of levee were obtained. 4.6.3 Conditional Annual Non-Exceedance Probability Conditional annual non-exceedance probability (CNP) is an index of the likelihood that a specified target stage will not exceed, given the occurrence of a hydrometerological event. The target stage can be the top of levee elevation. Table 17 summarizes estimates of (CNP) for nodal Stations 2, 14, 26, and 33 using top of levee and storm surge corresponding to a 1%-event. From Table 17, for nodal Station 2 at Freeport Harbor, the conditional non-exceedance probability of storm surge corresponding to 1% exceedance probability event is 84%. That means, that the

12

storm surge of a 1%-event has a probability of 84% that it would not exceed the top of levee at nodal Station 2. Conversely, there is a 16% probability that the 1%-event storm surge would exceed the top of levee. At nodal Stations 26 and 33, the chances of the 1%-event storm surge exceeding the top of levee are very low. 4.6.4 Long-Term Risk Another index for engineering performance is long-term risk. It is defined as the probability of one or more exceedances of a selected target stage or capacity in a specified duration. The target stage can be the top of levee elevation. Equation 3-1 from EM 1110-2-1619 was used to calculate long-term risk for top of levee elevation for 25 and 50-year duration periods. For nodal Stations 2 and 14, the risk probability of storm surge exceeding the top of levee elevation one or more times is 13% and 24% for the 25 and 50 year duration periods, respectively. For nodal Stations 26 and 33, the risk probabilities are 7% and 14% for the same duration periods. 4.6.5 Levee and Interior Area Uncertainty In Corps of Engineers risk-based flood damage reduction studies dealing with existing and new levees not maintained or constructed to federal levee standards, EM 1110-2-1619 references uncertainty that arises as a consequence of:

1) Imperfect knowledge of how an existing levee will perform from a geotechnical standpoint. 2) Lack of ability to predict how interior water-control facilitates will perform. 3) Imperfect knowledge of the timeliness and thoroughness of closure of openings in an existing or new levee.

Since the initial goal of the feasibility study includes consideration for the levee’s structural and geotechnical integrity, the first referenced uncertainty dealing from a geotechnical standpoint is critical in determining the annual exceedance probability corresponding to possible levee failure. During the initial investigations, it was determined that the existing Freeport Hurricane Levee is a well-maintained Federal project levee and therefore meets all Corps standards. Therefore, in this study, the assumption used in describing the Freeport Hurricane Levee's engineering performance is that the levee is capable of being overtopped from storm surge and wave run-up before levee failure. The second and third referenced uncertainties dealing with interior area facilities were not investigated, as interior area analyses were not included in this study. However, as with all flood damage reduction analyses dealing with levees, the levee interior areas are susceptible to induced flooding from local drainage runoff, block conditions from levees, and levee overtopping. The existing Freeport Hurricane Levee system was designed based on complete blocked conditions. Pumps were established for interior rainfall runoff and possible levee overtopping flood storage. Since then, the non-Federal sponsor has upgraded the pump capacity for the interior areas and currently maintains the levee and its interior area facilities. As discussed previously, the study

13

was terminated in the initial phases of the feasibility study and no flood reduction measures were evaluated for the interior areas. 4.6.6 Consequences of Capacity Exceedance Another index used to describe engineering performance is consequences of capacity exceedance for alternative plans as required by EM 1110-2-1619. The respective EM states “regardless of the design capacity selected, the probability of design exceedance is never zero.” No formulation and alternative plans were analyzed for the Freeport and Vicinity Flood Damage Reduction Feasibility Study. No overtopping volume computations, levee breach, or complete levee failure analyses were performed for this study. However, for consequences of capacity exceedance from levee overtopping or complete levee failure, catastrophic results can be assumed to occur if the overtopping cannot be managed, or if there is a levee breach or levee failure. Significant damage can also be possible from blocked conditions if severe rainfall runoffs are not properly managed in the interior areas. Aside from this, it should also be assumed that under tropical or hurricane conditions, the damage from wind could be more catastrophic than from flooding conditions. 4.7 Wave Setup and Wave Run-up As discussed above, wave setup and wave run-up analyses were performed in detail and presented in Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas. In the original levee design, pumping stations were established for wave run-up and overtopping. Since then, the non-Federal sponsor has upgraded the pumping system. In Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas, the maximum wave elevations computed include the storm surge, tides, and wave setup and run-up. These levels are shown graphically in Figures 2-4. The mean 1%-event for these conditions did not overtop the levee system. Figure 34 in Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas indicates how much levee structure will remain above water for this event. The average is over three feet throughout the levee system. 4.8 Conclusions The results of this study provide estimates of the various indices for engineering performance based on new Corps of Engineers guidance and procedures. Analyses were limited to engineering performance or non-economic related performance for the feasibility study. Analyses performed used current state-of-the-art numerical models with current methods and current data. Based on the results from the analyses presented in Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas, and the above descriptions for engineering performance, the Freeport Hurricane Levee maintains a storm surge capacity of a 1% exceedance probability (100-yr). The Texas A&M report, Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas, provides a conclusion that the Freeport Hurricane Levee provides sufficient protection for an extreme 1%-event (100-yr) against storm surge plus wave run-up. The engineering performance probabilities described above for storm surge without wave run-up affirms the conclusion the Freeport Hurricane Levee has a high reliability of containing the storm

14

surge corresponding to a 1%-event. Its geotechnical and structural integrity would allow for wave overtopping if the levee is maintained to Corps of Engineers standards and criteria. 5. Project Cost Summary Project Cost not completed due to sponsor decision to end the study based on information in Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas. 6. Implementation Schedule Implementation Schedule not completed due to sponsor decision to end the study based on information in Risk Assessment for Exposure to Hurricane Conditions at Freeport Texas.

15

Table 1. Freeport Hurricane Flood Protection Levee System Layout

Station Limits Levee Name ID Length Sta. Sta. Plan

Sheets Profile Sheets

Year Built

Average Elev.

Levee Crown

EAST BANK BRAZOS RIVER LEVEE (1) EBBR_L1 13,810 0+00 138+10 10 4 - 16' 12' WASTE WATER CANAL LEVEES (1) WWCL_1 511 0+00 5+11 1 1 - 20' 12' WASTE WATER CANAL LEVEES (2) WWCL_2 385 0+00 3+85 1 1 - 20' 12' EAST BANK BRAZOS RIVER LEVEE(2) EBBR_L2 25,269 128+00 380+69 27 7 1970 17-19.5' 12'-20' SOUTH STORM LEVEE SSTML 18,795 0+00 187+95 13 5 1968 17.5-21.25 20' NORTH WAVE BARRIER NWB 6,687 0+00 66+87 6 2 1972 16.5-18.5 20' SOUTH WAVE BARRIER SWB 2,574 66+87 92+61 2 1 1972 21.5' 20' OLD RIVER SOUTH LEVEE ORSL 14,533 133+00 278+33 12 3 1972 16.0'-17' 12' OLD RIVER SOUTH LEVEE NEW ORSLN 4,389 0+00 43+89 - 2 - 16.0'-17' 12' OLD RIVER NORTH LEVEE ORNL 19,128 127+00 318+28 16 4 1973 16.5-17' 16'-20' OLD RIVER NORTH LEVEE BULKHEAD ORNLB 5,374 0+00 53+74 - 2 - 16.5-17' 16'-20' DOW BARGE CANALS SOUTH LEVEE DBCSL 31,828 0+00 318+28 30 8 - 16'-19' 20' DOW BARGE CANALS NORTH LEVEE DBCNL 29,199 0+00 291+99 30 8 - 16'-19' 20' EAST STORM LEVEE ESTML 18,900 11+00 200+00 21 5 1968 22' 20' OYSTER CREEK LEVEE OCL 29,354 200+00 493+54 22 8 1967 15.5'-22' 20' EXTENSION OF OYSTER CREEK LEVEE EOCL 11,217 669+00 781+17 11 3 1965 16.5' 20' Note: Values are approximations based on historical records. Extensive survey P&P sheets are in attached binders.

16

Table 2. East Bank Brazos River Levee (EBBR L1&L2) General Analysis of Design Features

Location Length Type of Protection Embankment Slope Levee Crown Design Crest Elevation

(MSL)

Vert. Horiz. Width (FT) Surface

0+00 80+00 8,000 Levee 1:3 1:3 12 Pave 15.0

80+00 85+00 500 Levee 1:3 1:3-1:6 12 Pave 15.0-16.0

85+00 125+00 4,000 Levee 1:3 1:6 12 Pave 16.0

125+00 130+00 500 Levee 1:3 1:6 12 Pave 16.0-17.0

130+00 164+00 3,400 Levee 1:3 1:6 12 Pave 17.0

164+00 166+00 200 - - - - - 17.0-18.0

166+00 177+00 1,100 - - - - - 18.0

177+00 181+00 400 - - - - - 18.0-17.0

181+00 193+50 1,250 - - - - - 17.0

193+50 195+50 200 - - - - - 17.0-18.0

195+50 201+50 600 - - - - - 18.0

201+50 203+50 200 - - - - - 18.0-17.0

203+50 224+00 2,050 Levee 1:3 1:6 12 Pave 17.0-19.5

224+00 383+16 15,916 Levee 1:3 1:6 20 Pave 19.5

“ – “ denotes data not available

17

Table 3. South Levee (SSTML) General Analysis of Design Features


(MSL)


1+00 12+25 1,125 Levee 1:3 1:6 20 Pave 19.5-17.5

12+25 20+50 825 High grnd - - 20 Pave 17.5

20+50 32+00 1,150 Levee 1:3 1:6 20 Pave 17.5-18.5

32+00 42+50 1,050 Levee 1:3 1:6 20 Pave 18.5

42+50 46+00 350 - - - - - 18.5-22.0

46+00 173+75 12,775 - - - - - 21.25

173+75 176+75 300 - - - - - 21.25-19.75

176+75 183+45 670 - - - - - 19.75

183+45 187+95 450 - - - - - 19.75-15.25

Table 4. Old River and Freeport Harbor South Levee (ORSL & ORSLN) General Analysis of Design Features


(MSL)


0+00 57+50 5,750 Levee 1:3 1:3 12 Turf 17.0

57+50 142+00 8,450 Vert. wall - - 16.5

142+00 206+00 6,400 Levee 1:3 1:6 12 Pave 16.0

206+00 268+40 6,240 Vert. wall - - - - 16.5

268+40 269+40 100 16.5-16.0

269+40 278+53 913 Levee 1:3 1:6 12-20 Pave 16.0

18

Table 5. Old River and Freeport Harbor North Levee (ORNL & ORNLB) General Analysis of Design Features


(MSL)


0+00 29+00 2,900 Levee 1:3 1:3 12 Turf 17.0

29+00 101+00 7,200 Vert. wall - - - - 16.5

101+00 130+50 2,950 Levee 1:3 1:3 12 Pave 17.0

130+50 134+00 350 Vert. wall - - - - 16.5

137+00 150+00 1,300 - - - - - 17.5

150+00 152+00 200 - - - - - 17.5-17.0

152+00 166+00 1,400 - - - - - 17.0

166+00 168+00 200 - - - - - 17.0-16.5

168+00 308+40 14,040 - - - - - 16.5

308+40 310+40 200 - - - - - 16.5-19.0

310+40 311+62 122 - - - - - 19.0

19

Table 6. Dow Barge Canal South Levee (DBCSL) General Analysis of Design Features


(MSL)


40+00 45+00 500 Levee 1:3 1:3-1:6 12 Pave 20.0-21.0

45+00 55+00 1,000 - - - - - 21.0-19.0

55+00 65+00 1,000 - - - - - 19.0-20.5

65+00 80+00 1,500 - - - - - 20.5-19.0

80+00 82+00 200 - - - - - 19.0-20.0

82+00 86+00 400 - - - - - 20.0-19.0

86+00 90+00 400 - - - - - 19.0-20.5

90+00 108+00 1,800 - - - - - 20.0-21.0

108+00 140+00 3,200 - - - - - 21.0-16.0

140+00 191+00 5,100 - - - - - 16.0-17.0

191+00 205+00 1,400 - - - - - 17.0-16.5

205+00 220+00 1,500 - - - - - 16.5-17.0

220+00 225+00 500 - - - - - 17.0-16.0

225+00 289+00 6,400 - - - - - 16.0-17.0

289+00 297+00 800 - - - - - 17.0-19.0

297+00 318+00 2,100 - - - - - 19.0

20

Table 7. Dow Barge Canal North Levee (DBCNL) General Analysis of Design Features


(MSL)


0+00 101+00 10,100 Levee 1:3 1:3-1:6 12 Pave 17.0

50+00 52+00 200 - - - - - 22.0-21.0

52+00 58+00 600 - - - - - 21.0-22.0

58+00 68+00 1,000 - - - - - 22.0-20.0

68+00 71+00 300 - - - - - 20.0

71+00 82+00 1,100 - - - - - 20.0-25.0

82+00 91+00 900 - - - - - 25.0-20.0

91+00 98+00 700 - - - - - 20.0-21.0

98+00 102+00 400 - - - - - 20.0-19.0

102+00 110+00 800 - - - - - 19.0

110+00 122+00 1,200 - - - - - 19.0-16.0

122+00 147+75 2,575 - - - - - 16.0

147+75 148+50 75 - - - - - 16.0-22.0

148+50 151+00 250 - - - - - 22.0-16.0

151+00 225+00 7400 - - - - - 16.0

225+00 229+00 400 - - - - - 16.0-15.0

229+00 262+00 3300 - - - - - 15.0

262+00 264+50 250 - - - - - 15.0-16.0

264+50 290+50 2600 - - - - - 16.0

21

Table 8. Dow Barge TB General Analysis of Design Features


(MSL)


0+00 22+00 2,200 Levee 1:3 1:3 12 Pave 17.0

22+00 24+91 273 Vert. wall - - - - 17.0

24+91 26+50 43 Vert. wall - - - - 17.0

26+50 27+40 115 Vert. wall - - - - 16.0

27+40 30+30 290 Vert. wall - - - - 17.0

30+30 35+30 500 Levee 1:3 1:3 12 Pave 17.0

35+30 46+45 1,115 Vert. wall - - - - 17.0

46+45 52+60 615 Levee 1:3 1:3 12 Turf 17.0

Note: This system is the tie-in U-turn segment for DBCNL to DBCSL. Table 9. East Levee (ESL) General Analysis of Design Features


(MSL)


0+00 25+00 2,500 Levee 1:3 1:6 20 Pave 20.0

25+00 210+00 18,500 Levee 1:3 1:6 20 Pave 21.0

27+00 200+00 17,300 - - - - - 22.0

22

Table 10. Oyster Creek Levee (OCL) General Analysis of Design Features


(MSL)


200+00 210+00 1,000 22.0 210+00 220+00 1,000 Levee 1:3 1:6 20 Pave 21.0-20.5 220+00 287+00 6,700 Levee 1:3 1:6 20 Pave 20.5 287+00 292+00 500 Levee 1:3 1:6 20 Pave 20.5-19.5 292+00 350+00 5,800 Levee 1:3 1:6 20 Pave 19.5 350+00 356+00 600 Levee 1:3 1:6 20 Pave 19.5-17.0 356+00 401+00 4,500 Levee 1:3 1:6 20 Pave 17.0 401+00 406+00 500 Levee 1:3 1:6-1:3 20-12 Pave 17.0-16.2 406+00 410+00 400 Levee 1:3 1:3 12 Pave 16.2 410+00 411+00 100 Exist Vert Wall - - - - 16.2-16.5 411+00 420+00 900 Levee 1:3 1:3 12 Pave 16.5 420+00 421+00 100 - - - - - 16.5-16.0 421+00 425+60 460 - - - - - 16.0-16.5 425+60 426+50 90 - - - - - 16.5-16.0 426+50 432+00 550 - - - - - 16.0-15.5 432+00 436+50 450 - - - - - 15.5-16.0 436+50 442+00 550 - - - - - 16.0 442+00 445+50 350 - - - - - 16.0-15.5 445+50 480+00 3,450 - - - - - 15.5 480+00 490+00 1,000 - - - - - 15.0 490+00 492+00 200 - - - - - 15.5-16.0 492+00 493+53 153 - - - - - 16.0

23

Table 11. Oyster Creek Levee Extension (EOCL) General Analysis of Design Features


(MSL)


669+12 679+51 1,039 Levee 1:3 1:6 20 Pave 16.5 679+51 681+94 243 Vert. Wall - - - - 16.5 681+94 782+33 10,039 Levee 1:3 1:6 20 Pave 16.5 Table 12. Wave Barriers (North and South of GIWW) (NWB & SWB) General Analysis of Design Features


(MSL)


- - 9,900 Levee 1:3 1:6 20 Turf 20.0

Table 13. Waste Water Canal Levees (WWCL) General Analysis of Design Features


(MSL)


- - 899 Levee/Struc - - 20 Pave 20.0

24

Table 14. (Recommendation) Levee Analysis Breakdown vs. Modeled Storm Surges FHP Levee System

Station Limits H&H Risk Assessment Report Data

Levee Name Length Sta. Sta. Average

Elev.

50-Year Storm Surge

(Fig. 2)


(Fig. 3)


(Fig. 4) East Bank Brazos River Levee - 1 13,810 0+00 138+10 16' 7.0'-9.0' 11.0'-13.0' 15.0'-17' Wastewater Canal Levees -1 511 0+00 5+11 20' 7.0'-9.0' 11.0'-13.0' 15.0'-17' Wastewater Canal Levees -2 385 0+00 3+85 20' 7.0'-9.0' 11.0'-13.0' 15.0'-17' East Bank Brazos River Levee - 2 25,269 128+00 380+69 17-19.5' 7.0'-9.0' 11.0'-13.0' 15.0'-17' South Storm Levee 18,795 0+00 187+95 17.5-21.25 7.0'-9.0' 11.0'-13.0' 15.0'-17' North Wave Barrier 6,687 0+00 66+87 17.5-18.5 7.0'-9.0' 11.0'-13.0' 15.0'-17' South Wave Barrier 2,574 66+87 92+61 21.5' 7.0'-9.0' 11.0'-13.0' 15.0'-17' Ole River South Levee 14,533 133+00 278+33 16.0'-17' 7.0'-9.0' 11.0'-13.0' 15.0'-17' Old River South Levee New 4,389 0+00 43+89 16.0'-17' 7.0'-9.0' 11.0'-13.0' 15.0'-17' Ole River North Levee 19,128 127+00 318+28 16.5-17' 7.0'-9.0' 11.0'-13.0' 15.0'-17' Ole River North Levee Bulkhead 5,374 0+00 53+74 16.5-17' 7.0'-9.0' 11.0'-13.0' 15.0'-17' Dow Barge Canals South Levee 31,828 0+00 318+28 16'-19' 7.0'-9.0' 11.0'-13.0' 15.0'-17' Dow Barge Canals North Levee 29,199 0+00 291+99 16'-19' 7.0'-9.0' 11.0'-13.0' 15.0'-17' East Storm Levee 18,900 11+00 200+00 22' 7.0'-9.0' 11.0'-13.0' 15.0'-17' Oyster Creek Levee 29,354 200+00 493+54 15.5'-22.0' 7.0'-9.0' 11.0'-13.0' 15.0'-17' Extension of Oyster Creek Levee 11,217 669+00 781+17 16.5' 7.0'-9.0' 11.0'-13.0' 15.0'-17'

25

Table 15 Example Return-Year and Percent Annual Exceedance Probability Correlation

Return-Year 2

5

10

15

25

50

100

150

200

250

500

Percent Annual Exceedance Probability

50

20

10

6.67

4

2

1

0.67

0.50

0.40

0.20

Table 16 Nodal Stations 2, 14, 26, and 33 Top-of-Levee Percent Annual Exceedance Probability Nodal Station Top of Levee Elevation Percent Annual Exceedance Probability 2 (Freeport Harbor) 4.66-m or 15.3-ft 0.58 (172-yr) 14 (Dow Barge Canal North Levee) 4.47-m or 14.7-ft 0.58 (172-yr) 26 (East Levee) 6.31-m or 20.7-ft 0.30 (333-yr)* 33 (South Levee) 6.44-m or 21.1-ft 0.30 (333-yr)* * Estimated by extending stage-frequency rating. Table 17 Conditional Non-Exceedance Probability Using a 1% Exceedance Probability Event (100-yr) Nodal Station Conditional Non-Exceedance Probability in Percent 2 (Freeport Harbor) 84 14 (Dow Barge Canal North Levee) 84 26 (East Levee) Greater than 95 33 (South Levee) Greater than 95

26

Figure 1 – Freeport and Vicinity

27

Figure 2 - Refer to Table 14 for general analysis comparison

28


29


Photo 1. General levee reach depicting slope and crown condition

Photo 2. General levee reach depicting slope and relief near inner coastal reach

31

Photo 3. General protection reach depicting location of panels (storm) along dock

Photo 4. General protection system barrier and removable panel site tie in to earthen levee

32

Photo 5. General levee protection system wave barrier depicts service road (crown) condition

Photo 6. General levee protection system wave barrier depicts access and grade along perimeter

33

Photo 7. General levee protection system depicts slope vegetation growth

Photo 8. Levee reach depicts gravel service road inside of plant facility and grade/ relief

34

Photo 9. Levee feature (WWCL structure) ties into embankment at elevation 20’ MSL

APPENDIX B

Risk Assessment for Exposure to Hurricane Conditions at Freeport, Texas

Risk Assessment for Exposure to Hurricane Conditions at Freeport, Texas

Prepared for

Velasco Drainage District

June 6, 2005

Page2

Billy L. Edge, Manish Aggarwal, Oscar Cruz-Castro, Robert E. Randall, Patrick J. Lynett

Texas Engineering Experiment Station

Department of Civil Engineering Ocean Engineering Program

Texas A&M University

TABLE OF CONTENTS

INTRODUCTION .....................................................................................................................1

THE COMPUTATIONAL GRID..............................................................................................2

THE ADCIRC MODEL ..........................................................................................................10

Tidal Propagation................................................................................................................ 13

Wind Forcing ...................................................................................................................... 14

APPLICATION OF THE ADCIRC MODEL .........................................................................15

Tidal Circulation ................................................................................................................. 15

Tropical Storm Surge.......................................................................................................... 19

THE EMPIRICAL SIMULATION TECHNIQUE .................................................................28

Storm Consistency with Past Events .................................................................................. 31

Storm Event Frequency ...................................................................................................... 32

Risk-Based Frequency Analysis ......................................................................................... 33

Frequency-of Occurrence Relationships............................................................................. 33

WAVE RUNUP ANALYSIS ..................................................................................................40

The STWAVE Model ......................................................................................................... 40

Computational grid ............................................................................................................. 42

STWAVE Results and Analysis ......................................................................................... 47

The COULWAVE Model................................................................................................... 52

COULWAVE Results......................................................................................................... 53

Frequency Indexed Wave Runup........................................................................................ 61

CONCLUSIONS......................................................................................................................67

ACKNOWLEDGEMENTS.....................................................................................................68

REFERENCES ........................................................................................................................69

Pageii

TABLE OF FIGURES Figure 1 Study Area .................................................................................................................. 3 Figure 2 Computational Domain ............................................................................................. 4 Figure 3 Bathymetry in the Gulf of Mexico ............................................................................ 5 Figure 4 Grid Resolution in the area of interest....................................................................... 6 Figure 5 Bathymetry in the area of interest ............................................................................. 7 Figure 6 Comparison of tides at Freeport Harbor.................................................................. 17 Figure 7 Comparison of tides at Pleasure Pier....................................................................... 18 Figure 8 Large and small scale plots of Hurricane Claudette’s track .................................... 21 Figure 9 Raw surface elevation data for Hurricane Claudette at Freeport Harbor ................ 22 Figure 10 Surge only surface elevation data for Hurricane Claudette................................... 23 Figure 11 Simulated-observed data for Hurricane Carla (1961) for pleasure Pier ................ 25 Figure 12 Simulated-observed data for Hurricane Alicia (1983) for pleasure Pier............... 26 Figure 13 Maximum surge at the approximate time of peak surge for hurricane Claudette

along the area of Freeport coast ...................................................................................... 27 Figure 14 Location of points used for input to EST model ................................................... 30 Figure 15 Frequency relationships and the mean value for the Freeport Harbor .................. 35 Figure 16 Mean Value Frequency with standard deviation bounds for Freeport Harbor...... 36 Figure 17 200 year storm surge values around the Freeport Levee System .......................... 37 Figure 18 100 year storm surge values around the Freeport Levee System .......................... 38 Figure 19 50 year storm surge values around the Freeport Levee System ............................ 39 Figure 20 Computational Domain .......................................................................................... 44 Figure 21 Example of storm surge time series from Hurricane Celia, surge is taken to

coincide with the worst incoming waves arriving at Freeport........................................ 45 Figure 22 Example of wind field generated by hurricane Celia, worst incoming waves at time

195000 seconds............................................................................................................... 46 Figure 23 Arrangement of transects used to measure wave parameters................................. 49 Figure 24 Typical output of STWAVE, vectors indicate wave direction............................... 50 Figure 25 Typical output of STWAVE, colors indicate wave height..................................... 51 Figure 26 Results of two dimensional COULWAVE simulation showing wave height at

entrance to Freeport Harbor for Hurricane Carla (-1.0 long). Note the diffraction of waves as they enter channel............................................................................................ 54

Figure 27 Maximum Runup Frequency of Occurrence at TR1 .............................................. 62 Figure 28 Maximum Runup Frequency of Occurrence at TR2 .............................................. 63 Figure 29 Maximum Runup Frequency of Occurrence at TR3 .............................................. 64 Figure 30 Maximum Runup Frequency of Occurrence at TR5 .............................................. 65 Figure 31 Maximum Runup Frequency of Occurrence at TR6 .............................................. 66

Pageiii

LIST OF TABLES

Table 1: List of stations used for tidal verification ................................................................... 8 Table 2: Tidal verification of ADCIRC along open coast ...................................................... 17 Table 3: Tropical storms set.................................................................................................... 20 Table 4: Comparison of Storm surge computations with observed data measured from MSL

......................................................................................................................................... 24 Table 5: Hypothetical storm events ........................................................................................ 28 Table 6: Elevations of tidal datums referred to Mean Lower Low Water in meters .............. 43 Table 7: Tidal elevations referred to Mean Sea Level in meters ............................................ 43 Table 8: STWAVE deepwater boundary conditions .............................................................. 48 Table 9: Results from COULWAVE program for the MHHW condition.............................. 56 Table 10: Results from COULWAVE program for the MSL condition ................................ 58 Table 11: Results from COULWAVE program for the MLLW condition ............................ 59

Pageiv

EXECUTIVE SUMMARY

An analysis of potential flooding by storm surge and wave runup and overtopping was performed for the area of the Texas coast centered about Freeport for Velasco Drainage District. Results of the analyses can be used to evaluate storm protection afforded by the existing levee configuration or used to evaluate various alternatives for providing hurricane flood protection.

The finite element numerical hydrodynamic model Advanced Circulation (ADCIRC), was used to numerically simulate the propagation of tides and storms across the Gulf of Mexico to the study area. The model was verified by comparing model-generated tide time series with the corresponding time series reconstructed from existing harmonic analyses and on-site measurements of surface elevation. Storm event simulations are verified by comparing simulated results of water surface elevation with archived storm measurements. The study used the model to reproduce historic events; all storms that significantly impacted the study area since 1886 were simulated. In order to insure that the most severe events have been included for all open coast stations, simulations also included ten hypothetical events that could likely occur.

The storm surges were compiled with tide conditions to represent high, slack and low water for neap, spring and mid range tides to use with the statistical procedure known as the Empirical Simulations Technique (EST). The EST used the historic and hypothetical events to generate a large population of life-cycle databases were used to compute mean value maximum storm surge elevation frequency relationships. Frequency computations were made at 35 levee locations in the Freeport area. The maximum storm surge plus tide value for the 200-year storm was found to be 5.47 m and occurred in Oyster Creek. For the 100-year storm the maximum storm surge plus tide value of 3.98 m also occurred in Oyster Creek as did the maximum for the 50-year storm. However, none of the storm surge plus tide values were greater than the current height of the levee system. Wave conditions at the toe of the levee included predicting the maximum waves generated by the storm at an appropriate offshore position and determining the transformation of those waves as they came close to the shoreline. The initial wave transformation was determined using STWAVE. Results from STWAVE were used to compute the boundary condition for the nonlinear model COULWAVE which modeled the transformation of the waves when the surge was high enough to overtop the barrier islands; this condition allowed the waves from offshore to attack the levees. The COULWAVE is a version of the Boussinesq wave model. This model also computed the wave runup on the levees. The frequency-of-occurrence relationship was determined for five locations along the levee. The highest expected value for the maximum wave elevation was computed to be 6.6 m above NAVD 1988 for a 200-year storm and 4.9 m for a 100-year event. The highest levels of wave runup occurred west of the Freeport area at transects 1 through 3. The areas to the east of Freeport had approximately 0.5 m less runup elevation. The maximum wave elevation includes the storm surge, tides, and wave setup and runup. Although some storms did have a maximum runup

Pageii

in excess of 8 meters, the return period of those events would be much more than 200 years. The contribution to wave overtopping of the levees by wind is not included in this analysis.

Page1

INTRODUCTION

Freeport is an important industrial center and deepwater port located on the Texas

coast. The community has a diversified source of income, but is predominantly dependent on the petro-chemical industry. The principal sources of income are derived from processing petroleum and petroleum by-products. Brazoria County houses one of the world’s largest chemical complexes with the Dow Chemical being the principal employer. Since this area is exposed to storm surges resulting from tropical and extra tropical events, a levee system was constructed at Freeport, TX, in 1982. The levee was constructed for providing flood damage protection to the area. The levee system consists of a series of levees and pumping stations that protect an area of about 68 square kilometers. The project was completed in 1982. The levee system is vital to protection against flooding of the nation’s most vital petro-chemical industry worth almost $500 billion.

A comprehensive analysis of storm damage potential was performed along the Texas

coast centered about Freeport for Velasco Drainage District. The study deals with numerical modeling of tides/storms, run-up on levees and frequency analyses of storm surge elevations associated with historical tropical and hypothetical storms to obtain extreme event storm surges and run-up values. Results of the analyses can then be used to evaluate storm protection afforded by the existing levee configuration or used to evaluate various alternatives for providing hurricane flood protection.

The main computational resource for this project is the finite element numerical hydrodynamic model Advanced Circulation (ADCIRC), (Leutich, R.A., Westerink, J.J., and Scheffner, N.W., 1992) to numerically simulate the propagation of tides and storms at the study area. This task first requires development of a computational grid for the study area. The ADCIRC model then uses the computational grid to simulate tidal circulation and storm events. The model grid is verified by comparing model-generated tide time series with the corresponding time series reconstructed from existing harmonic analyses and on-site measurements of surface elevation. Storm event simulations are verified by comparing simulated results of water surface elevation with archived storm measurements. Once the model has been shown to be capable of reproducing historic events, all storms that significantly impacted the study area since 1886 are simulated. In order to insure that the most severe events have been included for all open coast stations, simulations include hypothetical events that could likely occur.

Following the numerical simulations for all the selected storms, the database of computed surges and tides are used as input for a statistical procedure known as the Empirical Simulations Technique (EST). This procedure uses historic events to generate a large population of life-cycle databases that are post-processed to compute mean value

Page2

maximum storm surge elevation frequency relationships with standard deviation error estimates. Frequency computations are made at 35 locations in the Freeport area. These stations are located at points of interest within the domain and help in establishing the extreme event storm surges along the levee system. This study requires completion of the following five sequential modeling tasks: 1) grid development, 2) application and calibration of the ADCIRC model, 3) verification of model results for tides and storms, 4) application of STWAVE and run-up calculation, 5) application of the EST model and calculation of frequency relationships. Detailed descriptions of each task are provided in the following sections. THE COMPUTATIONAL GRID

The region of general interest within the Gulf of Mexico for this application consists of the Freeport area as shown in the Figure 1. A problem often encountered in the modeling of near-shore regions in the Gulf of Mexico is that the computational boundaries of the model are not well removed from the area of interest. The Gulf of Mexico is a semi-enclosed basin that contains numerous amphidromes that affect the amplitude and phase of a storm surge or tide propagating from open water onto the shelf. Circulation within the Gulf of Mexico is primarily a function of wind and pressure; therefore, a basin-scale domain is required to capture tropical and extra tropical events. If the model boundary conditions are only specified near shore on the shelf, then boundary condition errors are introduced into the solution because the assumed boundary conditions are posed in a dynamic flow region, i.e., the transformation of the flow field over changing bathymetry enclosed within an irregular boundary. Flow features such as resonant shelf edge waves, hurricane forerunners, and/or complex wind patterns associated with a hurricane driving the flow onto the shelf, make it desirable to define larger computational domains, including regions well beyond the continental shelf adjacent to the area of interest. The modeling strategy has been to define the entire Gulf of Mexico as the computational domain and to refine the region of interest using the significant grid flexibility offered by the finite elements and the ADCIRC codes. Using the entire Gulf as the pertinent domain is quite convenient from a variety of perspectives. Most important, two well-defined open ocean boundaries of limited extent can be used to specify the boundary forcing functions that define the interaction between the Atlantic Ocean and the Caribbean Sea with the Gulf. The open boundary across the Strait of Florida was selected to run from Cape Sable in Florida to Havana, Cuba. The second open boundary stretches just south of the Yucatan Strait. The resulting finite element grid consists of 28,266 nodes, 52,624 elements and is shown in Figure 2. Minimum node-to-node spacing in the study is approximately 50 m. The bathymetry for the computational domain is shown in Figure 3. The increased resolution of the study area is shown in Figure 4. Bathymetry in the study area is shown in Figure 5. This large domain approach to specification of boundary conditions virtually eliminates contamination of model results from poorly defined boundary conditions.

Page3

Figure 1 Study Area

Page4

Freeport

Cuba

Yucatan

Figure 2 Computational Domain

Page5

Bathymetry

10.0283.0556.0829.01102.01375.01648.01921.02194.02467.02740.03013.03286.03559.03832.04105.04378.0

Cuba

Yucatan

Freeport

Figure 3 Bathymetry in the Gulf of Mexico

Page6

Figure 4 Grid Resolution in the area of interest

Brazos River

BRAZOS RIVER FREEPORT

Page7

Figure 5 Bathymetry in the area of interest

The bathymetry in the Gulf of Mexico varies dramatically, as is illustrated in the Figure 3. Bathymetric data in most of the Gulf was obtained from the grid developed by Scheffner et al. (2003), GeoDas (a database developed by National Oceanic and Atmospheric Administration, NOAA), USACE surveys and surveys conducted by Texas A&M University in the area of interest. The grid was generated as a combination of finite element grid developed by Scheffner et al. (2003) and modified in the area of interest with details added. The grid in the area of interest was added using SMS (Surface Modeling System). To get a mesh/grid with density radiating from the center of the Freeport channel, size functions in SMS were used along with celerity and wavelength functions so that smaller elements are obtained closer to the shore to correctly model the area of interest.

Page8

The model was “spun up” (started with progressively greater forcing such as tides or winds) from homogeneous initial conditions using a time ramp to avoid problems with short period gravity modes and vortex modes in the sub internal frequency range. A very smooth hyperbolic tangent ramp function, which acts over approximately one day, was applied to both boundary conditions and direct forcing functions. A 6-day spin-up was determined to be more than adequate for all conditions of interest. A time step of 6 sec was used for tidal propagation and a time step of 2 seconds was used for storm simulation in order to accommodate the strong gradients associated with strong winds in case of a storm. Using lower time steps resulted in oscillations and long term instabilities. The Courant number (The Courant Number constraint requires that the distance traveled by advection during one time step is not larger than one spatial increment ) based on wave celerity ranged from 0.0025 to 0.82. This is related to explicit treatment of the non-linear terms, since these problems do not arise for fully linear simulations. Time weighing factors of 0.35, 0.30 and 0.35 were used in the GWCE (Generalized Wave Continuity Equations). The parameter τ0 was set equal to –0.005 as this signals ADCIRC to use 0.005 in deep water and 0.02 in shallow water, so that a balance is set between the primitive continuity and wave equation portions of the GWCE equation. Tidal water surface elevation data computed with the ADCIRC model were recorded at 15 locations for verification purposes. These locations are listed in Table1. Storm surge water surface elevations were archived for 12 locations within the area of interest for subsequent computation of frequency-of-occurrence relationships.

Table 1: List of stations used for tidal verification

S. No Location Longitude ° Latitude °

1 Corpus Christi -97.38928 27.08113 2 Freeport Harbor -95.34277 28.95019 3 Sabine Pass -93.83873 29.68882 4 Galveston bay entrance south jetty -94.69849 29.32304 5 Round Point Galveston Bay -94.78059 29.31827 6 Galveston bay entrance -94.70587 29.34739 7 Bolivar roads -94.78388 29.34029 8 Galveston (channel) (2) -94.78774 29.31305 9 Galveston Pleasure Pier -94.78747 29.28495 10 Galveston channel -94.80136 29.31203 11 Jamaica beach -95.00899 29.19919 12 Morgan point -94.9766 29.6756 13 Clear lake -95.06118 29.55583 14 Lynchburg landing -95.09607 29.77917

Page9

15 San Luis Pass -95.12016 29.07968

Page10

THE ADCIRC MODEL

Water-surface elevations and currents for both tides and storm events are obtained from the large-domain long wave hydrodynamic model ADCIRC (Advanced Circulation model; Luettich, Westerink, and Scheffner 1992). ADCIRC is a finite element (FEM) code that makes use of the Generalized Wave Continuity Equation (GWCE) for improved stability and efficiency over other FEM hydrodynamic codes. Included within the code are features that allow the user to include tidal and atmospheric forcing in the computations. Wind can be input in a variety of different formats and could be derived from any source that the user has available. The model was developed as a family of two- and three- dimensional codes with the capability of:

a. Simulating tidal circulation and storm surge propagation over large computational domains while simultaneously providing high resolution in areas of complex shoreline and bathymetry. The targeted areas of interest include continental shelves, near shore areas, and estuaries.

b. Representing all pertinent physics of the three-dimensional equations of motion.

These include tidal potential, Coriolis, baroclinic and all nonlinear terms of the governing equations.

c. Providing accurate and efficient computations over time periods ranging from

months to years. The 2-dimensional, Depth Integrated (2DDI) model formulation begins with the depth-averaged shallow-water equations for conservation of mass and momentum subject to incompressibility and hydrostatic pressure approximations. The Boussinesq approximation, where density is considered constant in all terms but the gravity term of the momentum equation, is also incorporated in the model. Using the standard quadratic parameterization for bottom stress and omitting baroclinic terms and lateral diffusion and dispersion, the following set of conservation statements in primitive, non-conservative form and expressed in a spherical coordinate system are incorporated in the model (Flather 1988; Kolar et al. 1994):

Page11

( cos )1 [ ] 0 (1)cos

tan1 1 [ U ]cos

1 [ ( )] * (2)cos

tan1 1 [cos

UVUHt R

U U UU V f VR Rt R

Pg UHR

V V VU V UR Rt R

s so o

φζφ λ φ

φφ λ φ

τ λζ η τρ ρφ λ

φφ λ φ

∂∂ ∂+ + =∂ ∂ ∂

∂ ∂ ∂+ + − + =∂ ∂ ∂

∂− + − + −∂

∂ ∂ ∂+ + −∂ ∂ ∂ ]

1 [ ( )] * (3)cos

f V

Pg VHR

s so o

τ φζ η τρ ρφ λ

+ =

∂− + − + −∂

where : ζ = free surface elevation relative to the geoid, U, V = depth-averaged horizontal velocities, H = ζ + h = total depth of water column, h = bathymetric depth relative to the geoid, f = 2 Ω sinφ = Coriolis force, Ω = angular speed of the Earth, φ = latitude in degrees, λ = longitude in degrees, Ps = atmospheric pressure at the free surface, g = acceleration due to gravity, η = effective Newtonian equilibrium tide potential,

oρ = reference density of water, α = effective Earth elasticity factor,

,s sλ φτ τ = applied free-surface stress,

*τ = Cf (U2 + V2)1/2/H, bottom shear stress,

fC = bottom friction coefficient, R = radius of Earth, t = time. In order to overcome general stability problems encountered when finite element models depend upon the direct solution of these primitive forms of the governing equations (Gray 1982), the ADCIRC code was developed around the generalized wave

Page12

continuity equation (GWCE). Combining a time-differentiated form of the momentum equations yields this form of the primitive equations. With the inclusion of a simple eddy viscosity model for closure (Kolar and Gray 1990), the GWCE in Spherical coordinates takes the form:

0

0 00

2( ) ( cos ) tan1 1[ ( ) ]2 cos cos

[ 2 sin ( ( ) ) ]cos

( ) ( cos ) tan1 1[ ( ) 2 sin ]cos

1 [ ( ( ) ) ( *

HUV HUVUVH Rt R R

t

sHHV g HU HUR

HVV HVVUUH HUR RR

H gR R

Pso

Pso

ζ φζ φτ φ λ φ λ φ

τ λω φ ζ αη τ τφ λ ρ

φ φω φφ φ λ φ

ζ αη τφ λ

ρ

ρ

∂ ∂∂∂ ∂+ − + −∂ ∂ ∂ ∂∂

∂− + − + + − −∂

∂ ∂∂− + + +∂ ∂ ∂

∂ ∂− − + + −∂ ∂ 00

0

) ]

[ tan ] [ tan ] 0 (4)

sHV

VH VHR Rt

τ φτ

ρ

φ τ φ

−

∂− − =∂

The ADCIRC-2DDI model solves the GWCE (Equation 4) in conjunction with the primitive momentum equations given in Equations 2 and 3.The equations are solved using a FEM grid, made up of linear triangular elements (only three nodes per element). The model domain can be as extensive as an entire ocean basin, or more localized, as in the case of a small bay or estuary. The numerical solution of the governing equations presented above follows a two-step procedure in ADCIRC code. First, the GWCE (Eq. (4)) is solved. The linear terms in the GWCE are discretized using a Galerkin weighted residual, three time level, and implicit scheme. The non-linear terms, along with Coriolis forcing, atmospheric forcing and tidal potential are solved explicitly (Westerink et al. 1994a). The explicit formulation of these terms has the advantage that the solution depends only upon the previous time step. On the other hand, the implicit terms depend upon the solution of a system of equations, arranged in a banded matrix. The second step in the solution of the governing equations, after solving the GWCE, is to solve the momentum equations (Eq. 2 and 3). Most of the terms of the momentum equations are handled in a Crank-Nicholson, two-time level, and implicit discretisation scheme. The explicit terms in the momentum equations are the *τ terms, the convective terms and the eddy viscosity terms.

Page13

A feature of ADCIRC that makes its application particularly useful in storm surge modeling of bays along Texas Gulf coast, is the capability of wetting and drying in the computational cells. Most of the coastal basins that make up the estuaries along the Texas Gulf coast are very shallow, with depths that are often no more than a meter. In addition to the shallow bay depths, the topography of coastal lands is that of flat coastal plains, with very gentle slopes. The barrier islands in Freeport are just a few meters above mean water level. During extreme meteorological events like hurricanes, it is possible that shallow areas may become dry from “blow down” (due to water being driven from the area by storm winds). On the other hand, the surge during a storm can cause extensive inland coastal flooding as is evident historically in all large storms affecting the Freeport area over the time period considered in this study.

An element based technique for wetting/drying was developed for implementation in ADCIRC. Conceptually, the algorithm assumes removable barriers exist along the sides of all triangular elements of the grid. Nodes of the elements are designated as “dry” nodes, “interface” nodes, and “wet” nodes. All elements connected to a dry node are assumed to have barriers in place such that there is no flow through the element, i.e. a dry element. An element connected to all wet nodes is a wet element and is included in the full flow domain. Interface nodes connect wet and dry elements. Boundaries connecting interface nodes are considered as standard land boundary nodes at which the water level rises and falls against the element barrier.

Tidal Propagation

Tidal potential forcing, which causes the normal observed periodic water level changes in large bodies of water, is included in ADCIRC. Other popular large scale hydrodynamic models, like SLOSH and RMA2, do not include tidal potential forcing. ADCIRC determines the magnitude of the tidal potential η in equation (4) at each grid node and each model time step by the relationship:

,

000

( , , ) ( ) ( ) co s2 ( ) ( )

j n

t t Lt tB j v tjn jnj T jn

η λ φ φπ λ=

− + +

∑

as given by Reid (1900), where: j = tidal species

0 = declinational 1 = diurnal 2 = semidiurnal

jnB = amplitude constant of the nth tidal constituent of species j,

jnF = time dependent nodal factor,

Page14

jnv = time dependent astronomical argument,

jL = function for species j

20 3sin 1L φ= − ,

1 sin(2 )L φ= , 2

2 cosL φ=

0t = a reference time, usually the beginning time of simulation jnT = period of constituent n of species j. The values of f and ν for the constituents used for the tidal potential computations are determined for the specific time that a model run begins using LeProvost database (Westerink, J.J., Leutich, R.A., and Scheffner, N.W., (1993)). Note that tidal potential was not used during the simulation of tropical storms. Tides were combined after the simulations during the frequency analysis.

Wind Forcing

In addition to the capability for tidal forcing within ADCIRC, there are provisions to input atmospheric and wind forcing information into the simulations. Several formats for the wind data are supported, including a fleet numeric and National Weather Service (NWS) wind file format. For this study, the Planetary Boundary Layer (PBL) model (Cardone et al. 1992) supplies the atmospheric forcing information. This model was developed to simulate hurricane generated wind fields using basic characteristics about a particular storm that can be easily retrieved from sources such as NWS archives of past hurricanes, as well as forecast data for a currently active storm. This model simulates hurricane-generated wind and atmospheric pressure fields by solving the equations of horizontal motion that have been vertically averaged through the depth of the planetary boundary layer. The PBL model requires input defining both the hourly location of the eye of the storm and a set of meteorological parameters defining the storm at various stages of development. These parameters include latitude and longitude of the eye of the storm, track direction and forward speed measured at the eye, radius to maximum wind, central and peripheral atmospheric pressures, and an estimate of the geostrophic wind speed and direction. A two-step process is used to generate wind fields for use by ADCIRC from the storm data. First, a program for the PBL model is used to determine the track of the storm as one our ‘snapshots’. These snapshot data include the radius of maximum wind, which is approximated using a nomograph that incorporates the maximum wind speed and atmospheric pressure anomaly (Jelesnianski and Taylor, 1973). In the second step, the PBL model computes the wind field and pressure field of the hurricane.

Page15

APPLICATION OF THE ADCIRC MODEL

Application of the ADCIRC model requires verification to ensure that grid resolution, bathymetry, and boundary conditions were acceptable to properly simulate conditions in the defined domain. For comparison of tidal simulations with observed tides, verification was accomplished using 8-constituents (M2, S2, N2, N1, K1, O1, Q1, and P1), as these constituents comprise most of the tidal energy, with tidal elevations calculated using software XTIDE which in turn uses published harmonic series. Verification for storm events is achieved by comparing computed surface elevation time series to surge data available from the National Ocean Survey (NOS) for the Pleasure Pier recording station. The following two sections describe the tidal circulation and storm surge verification effort.

Tidal Circulation

Tidal circulation is simulated within ADCIRC by specifying a surface elevation time series at the Florida Strait and just south of the Yucatan Strait as shown in the computational grid of Figure 2. This boundary condition specification is accomplished by reconstructing an 8-constituent tidal elevation time series at each open water boundary node of the grid based on amplitudes and Greenwich epoch values obtained from a database incorporated in the SMS software also known as LeProvost database. Additionally, tidal potential terms are specified at each node of the computational grid. The ADCIRC model has an internal harmonic analysis option in which individual constituent amplitudes and epochs are computed at user specified locations during the tidal simulation. Verification of tidal circulation was made by comparing both ADCIRC computed harmonic constituents and ADCIRC computed time series with existing constituent data and reconstructed time series at each of the 15 verification locations listed in Table 1. Comparisons of ADCIRC versus published Harmonic Analysis (HA) computed constituent amplitudes and Greenwich epochs (G) are shown in Table 2 for two locations. Because the Gulf of Mexico is a semi-enclosed body of water, approximately 10 to 15 days of spinup time are required for the tide to come to a dynamic equilibrium, i.e. when the tides are acceptably reproduced. The harmonic analysis used for the comparisons in Table 2 were based on a 43-day simulation of tides and during this time the harmonic analysis was computed for the 29-day (one lunar month) period of days 15 through 43. A period of comparative less wind activity was chosen to effectively compare the real-time data and ADCIRC simulated time-series. In order to demonstrate a degree of acceptability for the constituent comparisons shown in Table 2, a tidal elevation time series for days 15 through 43 is shown in Figures 6 and 7 at different recording stations.

Page16

As shown, the comparisons are quite acceptable and fully adequate for the statistical generation of stage-frequency relationships.

Page17

Table 2: Tidal verification of ADCIRC along open coast

Constituents Galveston Pleasure Pier Freeport Harbor Amp – m

Mod / HA G – deg Mod / HA

Amp – m Mod / HA

G – deg Mod / HA

K1 .162/. 197 33.2/42.6 .175/.188 18.7/26.6 O1 .162/. 187 20.8/32.1 .167/.178 13.2/15.9 P1 .051/. 056 14.2/18.5 .059/.065 17.1/21.6 Q1 .034/. 043 11.6/16.2 .037/.046 2.2/4.8 N2 .025/. 037 290.2/285.1 .024/.035 261.1/258.4 M2 .102/. 138 317.6/295.7 .092/.101 273.5/265.2 S2 .024/. 040 278.9/282.6 .033/.035 284.3/178.9 K2 .009/. 011 271.3/282.1 .014/.015 266.1/272.6

0 100 200 300 400 500 600 -1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

Hours

Wat

er S

urfa

ce E

leva

tion

(m)

Comparison of ADCIRC and Xtide for Freeport Harbor -95.34277 28.95019

ADCIRC XTIDE

Figure 6 Comparison of tides at Freeport Harbor

Page18

0 100 200 300 400 500 600 -1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

Hours

Wat

er S

urfa

ce E

leva

tion

(m)

Comparison of ADCIRC and XTIDE for Galveston Pleasure Pier -94.78747 29.28495

ADCIRC XTIDE

Figure 7 Comparison of tides at Pleasure Pier

It is only necessary to show that the model acceptably reproduced water levels for all gauges to demonstrate approximate verification in a more global sense with verification efforts concentrated on the study area. However, in order to demonstrate that the overall verification was acceptable, all model to reconstructed prototype tidal time series are presented in Appendix A. As shown, the comparisons are acceptable and demonstrate that the ADCIRC model is properly reproducing tides throughout the general area of interest as well as in the specific study area.

Page19

Tropical Storm Surge

The PBL model was used during this study and was coupled with ADCIRC in the form of a wind file, which can be input to the ADCIRC model to simulate wind affects of the storms for the area of interest. Peripheral atmospheric pressures were assumed equal to the standard atmospheric pressure of 1013 millibars (mb) and the geostrophic wind speeds were specified as 6 knots in the same direction as the moving eye of the storm. All additional data are computed from data contained in the National Oceanic and Atmospheric Administration’s (NOAA) Hurricane DATabase (HURDAT) of tropical storm events (Jarvinen, Neumann, and Davis 1988). This database is updated yearly and now contains descriptions of all hurricane, tropical storm, and severe tropical depressions that have impacted the east coast of the United States, Gulf of Mexico, and Caribbean Sea from 1886 through 2002. The database contains latitude and longitude locations of the eye of the hurricane with the corresponding central pressure and maximum wind speeds at 6-hour intervals. The recent hurricane Claudette (July 2003) was simulated by using track data from weather databases, which contain track information: latitude, longitude, time along with minimum pressure and maximum wind. The goal of this component of the study is to compute frequency-of-occurrence relationships for storm surge plus tide in the Freeport area. In order to develop these relationships, it is necessary to identify tropical storms that have historically impacted the study area. This was accomplished by making use of the tropical storm database (Scheffner, et al, 1994) that was generated through simulation of 134 historically based storm events along the east coast, Gulf of Mexico, and Caribbean Sea. The database uses the HURDAT database described above as input. For 486 discrete locations along the U.S. coast, peak storm surge values corresponding to storm events, which produced a surge of at least 0.305 m, were archived and indexed according to event, location, and surge magnitude. The DRP database was used to select 26 storm events for the present study beginning with the hurricane of 1886 and extending through Hurricane Claudette (2003). These events, shown in Table 3, represent the historical training set of storms. An example plot of the storm track and location every 6 hours of Hurricane Claudette is shown in Figure 8. The track for each storm event of the historical training set is shown in Appendix B.

Page20

Table 3: Tropical storms set

HURDAT No./Name Date of Storm HURDAT No./Name Date of Storm 1. #5 8/12/1886 14. #602 – Carla 9/3/1961 2. #117 8/27/1900 15. #690 – Celia 7/31/1970 3. #183 7/13/1909 16. #703 – Edith 9/5/1971 4. #211 8/5/1915 17. #704 – Fern 9/3/1971 5. #232 8/1/1918 18. #722 – Delia 9/1/1973 6. #295 6/27/1929 19. #809 – Chris 9/9/1982 7. #310 8/12/1932 20. #812 – Alicia 8/15/1983 8. #324 7/25/1933 21. #841 – Bonnie 6/23/1986 9. #397 8/2/1940 22. #867 – Chantal 7/30/1989 10. #405 9/16/1941 23. #874 – Jerry 10/12/1989 11. #445 8/24/1945 24. #923 – Dean 7/28/1995 12. #565 – Audrey 6/25/1957 25. #965 – Frances 9/8/1998 13. #586 – Debra 7/23/1959 26. #1001 – Allison 6/5/2001 27. #1016 – Claudette 7/5/2003

Page21

-100 -99 -98 -97 -96 -95 -94 -93 -92 -91 -90 23

24

25

26

27

28

29

30

31

32

33

Freeport

Claudette, 09-15 JUL 2003

12-9 12-15

12-21

13-313-9

13-1513-21

14-3

14-9

14-15

14-21

15-315-9

15-1515-21Latitude °

Longitude °

-96.2 -96 -95.8 -95.6 -95.4 -95.2

28.2

28.4

28.6

28.8

29

29.2

Freeport

Claudette, 09-15 JUL

15-15

Latitude °

Longitude ° Figure 8 Large and small scale plots of Hurricane Claudette’s track

Page22

This set of tropical storms is verified by comparing National Ocean Survey (NOS) measured tide gage records taken at Pleasure Pier and Freeport Harbor. These data are ideal for storm event verification effort as it allows calibration of the radius of maximum wind that is an important input to the PBL model to optimize the model for comparison of the set of storms for the area of interest. Due to spin-up time of 10-15 days required for tidal simulations, the decision was taken to compare storm surge elevations without tidal forcing. Therefore, surge only storm surge time-series were constructed by removing the astronomical tide from the raw NOS tide gage records, and the ADCIRC surge was computed without tidal forcing. For example, Figure 9 shows a time series of NOS data for Freeport for Hurricane Claudette (2003). As is evident from Figure 9, the storm surge is accurately captured; however, the tides are not accurately simulated due to spin-up time required for tidal simulation in the Gulf. In Figure 10, surge-only data is shown, which is computed by subtracting tide and pre-storm datum from the raw signal.

Claudette (7/9/2003)

-1

-0.5

0

0.5

1

1.5

2

0 1 2 3 4 5 6 7 8Time, days from 7/9/2003

Surg

e - m

eter

s

Observed Data Adcirc Freeport

Figure 9 Raw surface elevation data for Hurricane Claudette at Freeport Harbor

Page23

Claudette (7/9/2003)

-0.5

0

0.5

1

1.5

2

0 1 2 3 4 5 6 7 8

Time, days from 7/9/2003

Surg

e - m

eter

s

Observed Data Adcirc Freeport

Figure 10 Surge only surface elevation data for Hurricane Claudette NOS Data are available for the Pleasure Pier for the period starting August 1957 through August 2003. Data for Freeport Harbor is available only for period starting March 1995 through August 2003. This period represents almost a 50-year period and encompasses majority of the storms at the Pleasure Pier on Galveston Island. All storms shown in Table 3 were simulated using ADCIRC model without tide. Storm surge peak values for events before August 1957 were compared to anticdotal data from National Hurricane Center’s historical archives by removing tides. Storm event simulated hydrographs for events after August 1957 were compared to the NOS hydrographs at Pleasure pier and Freeport Harbor Gage stations. Results and comparisons of peak surge

Page24

values in meters relative to mean sea level for each event for which data are available are shown in Table 4.

Table 4: Comparison of Storm surge computations with observed data measured from MSL

Storm No. Pleasure Pier (m,msl) Freeport Harbor (m,msl)

ADCIRC NOS ADCIRC NOS

1. #5 0.796 0.880 2. #117 2.141 0.788 3. #183 1.493 0.931 4. #211 3.059 1.372 5. #232 0.780 0.576 6. #295 1.875 0.917 7. #310 3.040 0.612 8. #324 0.689 0.687 9. #397 0.264 0.270 10. #405 2.155 2.159 11. #445 1.002 1.351 12. #565 – Audrey 0.516 0.449 13. #586 – Debra 1.105 0.954 0.562 14. #602 – Carla 2.314 2.46 3.295 15. #690 – Celia 0.952 0.91 1.525 16. #703 – Edith 1.158 1.25 0.927 17. #704 – Fern 1.477 1.32 1.446 18. #722 – Delia 1.227 1.14 1.227 19. #809 – Chris 0.434 0.401 0.409 20. #812 – Alicia 2.710 2.334 1.023 21. #841 – Bonnie 0.431 0.69 0.327 22. #867 – Chantal 0.593 0.682 0.359 23. #874 – Jerry 1.390 0.997 0.967 24. #923 – Dean 0.859 0.84 0.498 25. #965 – Frances 0.791 0.84 0.877 0.865 26. #1001 – Allison 0.605 0.792 0.345 0.585 27. #1016 – Claudette 1.481 1.523 1.707 1.772 Figure 11 shows a comparison of simulated-adjusted raw data for Hurricane Carla (1961) for the Freeport Harbor. A comparison of simulated-observed data for Hurricane Alicia (1983) is also provided in Figure 12 for pleasure Pier. Hurricane Alicia represents

Page25

the most intense storm for which the observed data are available. As shown, the model generated surge matches the observed data very closely with respect to maximum surge as well as shape. The maximum observed surge at Pleasure Pier for hurricane Alicia was 2.710 m and the maximum surge computed from ADCIRC was 2.334 giving an error of approximately 14 cm. Model generated storm surge hydrographs and comparisons wherever available are included in Appendix B along with hurricane track plots.

Carla (9/3/1961)

-0.5

0

0.5

1

1.5

2

2.5

0 2 4 6 8 10 12 14 16


Surg

e - m

eter

s

Adcirc Pleasure Pier Observed Data

Figure 11 Simulated-observed data for Hurricane Carla (1961) for pleasure Pier

Page26

Alicia (8/15/1983)

-0.5

0

0.5

1

1.5

2

2.5

3

0 1 2 3 4 5 6


Surg

e - m

eter

s

Observed Data Adcirc Pleasure Pier

Figure 12 Simulated-observed data for Hurricane Alicia (1983) for pleasure Pier A spatial distribution snapshot plot of the maximum surge at the approximate time of peak surge for hurricane Claudette (2003) along the open coast area of Freeport is shown in Figure 13. Model results agree very well with observed data reported from NOA database where storm surge of 1.3 – 2 meters above normal tides was observed and both the barrier islands were inundated. Model simulations for tidal elevations are considered to be acceptable for use in the frequency analysis. Additionally, model results compare well to available storm surge data for a variety of storm events and qualitatively compare well to post-storm visual

Page27

surveys. The conclusion of the tidal and storm surge verification effort is that the model reasonably reproduces known historical events such as tide and tropical storms.

Figure 13 Maximum surge at the approximate time of peak surge for hurricane Claudette along the area of Freeport coast

Although this study is based on a historical storm set, it is important to consider not only historical storm events but also potential storm events that could reasonably be expected to occur. For example it is reasonable to assume that the storm could easily have tracked within + one degree of the actual track. Therefore, in order to insure that the most severe events have been included for the area of interest, simulations include hypothetical events that could occur. For example, the tracks of the most intense events in the historical storm events are shifted along the coast such that all maximum storm surge events that could occur are taken into consideration. Ten events were defined as perturbations of HURDAT storms 117 (August 1900), 211 (August 1915), 310 (August 1932), Carla (September 1961) and Alicia (August 1983) to augment the historical storm set. Each perturbation is represented by a shift of the storm track reported in HURDAT database as indicated in Table 5. These events have been shown to be among the most severe storms to have impacted the study area between 1886 and 2003. Use of these 10 hypothetical events increases the total to 37 storms that are used as a “training set” for the study. Thus the training set is comprised of 27 historical storm events plus the 10 perturbations.

Page28

Table 5: Hypothetical storm events

Storm Event Perturbation shift in degrees longitude #117 August 1900 - 2 events +/- 1.0 degree longitude #211 August 1915 - 2 events +/- 1.0 degree longitude #310 August 1932 - 2 events +/- 1.0 degree longitude Carla September 1961 - 2 events +/- 1.0 degree longitude Alicia August 1983 - 2 events +/- 1.0 degree longitude

The ADCIRC model has been shown to satisfactorily reproduce tidal circulation and tropical storm surge elevations within the project area. The model simulations are used to generate a database of historical and historically based storm events for use in generating maximum surface elevation frequency-of-occurrence relationships at the sites as shown in Figure 14. The following section describes the statistical approach used to generate multiple life-cycle simulations of storm-activity for the study area and the subsequent post-processing of results to generate surge versus frequency-of-occurrence relationships. THE EMPIRICAL SIMULATION TECHNIQUE

The Empirical Simulation Technique (EST) is a procedure for simulating multiple life-cycle sequences of non-deterministic multi-parameter systems such as storm events and their corresponding environmental impacts. Essentially, it is a “Bootstrap” resampling-with-replacement, interpolation, and subsequent smoothing technique in which random sampling of a finite length database is used to generate a larger database (Borgman, 1992). The only assumption is that future events will be statistically similar in magnitude and frequency to past events. As stated above, EST is a generalized procedure applicable to any cyclic or frequency-related phenomena (Scheffner et al. 1999). For example, if one can parameterize storm events as well as obtain or simulate corresponding historical impacts for these events, EST could be used to investigate life-cycle scenarios of storm conditions. The EST begins with an analysis of historical events that have impacted a specific locale. The selected database of events (the training set) is then parameterized to define the characteristics of the event. The interdependence of parameters is computed directly from the respective parameter interdependencies contained in the historic data. In this manner, probabilities are site-specific, do not depend on fixed parametric relationships or assumed joint probability distributions. The impacts of events may be known or may be simulated by other models (e.g., hurricane events can be characterized by parameters such as central pressure, forward speed, etc. and their impact may be simulated with appropriate hydrodynamic and storm wind models). Parameters that describe an event, i.e., a storm in this discussion, are referred to as Input Parameters or Input Vectors. Response parameters or response vectors define event-related impacts

Page29

such as storm surge elevation, inundation, shoreline and dune erosion, etc. These Input Parameters and Response Parameters are then used as a basis for generating life-cycle simulations of hurricane activity with corresponding impacts. The descriptive characteristics of the storm event with respect to the specific location of interest are determined by the input parameters or input vectors. For tropical storms these input parameters are studied at the point when the eye of the hurricane is closest to the station of interest. These vectors are defined as:

1) tidal phase during the event, with 1.0 corresponding to high water slack, 0.0

MSL at maximum ebb, -1.0 low water slack, these represent relative values that are defined for each station

2) radius of maximum wind for the hurricane when the eye is closest to the

hurricane in nautical miles. 3) minimum distance from the eye of the storm to the location of interest in

nautical miles.

4) pressure at the hurricane eye in millibars (mb)

5) wind speed in the hurricane at the instant of eye hitting the coast, measured in knots.

6) direction of forward propagation of the eye of the hurricane 7) tidal range during the event: with spring, neap or mid tide conditions. The maximum storm surge elevation reached at specified gauge locations is defined as the response vector of the storm at that location. The specified response vector for this study was determined by simulating the specific storm event via the ADCIRC hydrodynamic model using the computational domain shown in Figure 2. The output vector(s) represents the environmental response to the storm. This response is defined at location X and is a direct consequence of the storm via the storm parameter values defined at the point of nearest proximity of the storm eye to point X. For the case of stage-frequency analyses, maximum surge is assumed to occur when the eye of the storm is nearest to location X. In order to establish the training set of storms 27 historical events and 10 storm perturbations were used to produce a total of 37 events. Each of the 37 storms was simulated without tide to produce a set of surge-only responses at the stations. The storms are then assumed to have taken place at different phases of the astronomical tides: 1) high tide, 2) mean (MSL = 0.0) tide, and 3) low tide. It is further assumed that the storm could occur during the lunar cycles of: 1) spring tide, 2) between spring and neap, and 3) neap tide. Input vectors representing these phases of the tide are described above.

Page30

This combination of tide and lunar cycle produces 9 surface elevations for each of the 37 storm events of Table 3 at the station locations shown in Figure 14. This procedure produces a total input/response vector training set of 333 (37*9) tide plus surge events for each station location. It is also considered that the mid-tide level would have twice the probability of occurrence on the MHHW or the MLLW. Similarly the mean tidal range would have twice the probability of occurrence as spring or neap tide.

Figure 14 Location of points used for input to EST model The water surface elevation that is one of the response vectors for the EST analysis was calculated as follows. Analysis of tidal values for the tides at Freeport harbor show that the approximate peak tidal elevation at spring, mid, and neap cycle is 0.35, 0.268, and 0.20 m. The four primary model-generated diurnal and semi-diurnal tidal constituents for the Freeport Harbor study area are the K1, O1, M2, and S2 with the amplitudes of 0.137, 0.125, 0.07, and 0.02 m respectively. The values of spring and neap tides are calculated using these constituents as follows assuming that most of the tidal energy is contained in these constituents, using the relationship:

GULF OF MEXICO

BRAZOS RIVER

FREEPORT

Page31

spring high tide = K1 + O1 + M2 + S2 mid spring/neap high tide = K1 + O1 neap high tide = K1 – O1 + M2 + S2

This relationship generates an acceptable approximation of 0.352, 0.261, and 0.191 m versus 0.35, 0.268, and 0.2 m for the 9 combinations of astronomical and lunar tidal effects. Therefore this relationship was used for all locations.

Storm Consistency with Past Events

The first major requirement for the use of EST is that future events will be statistically similar to past events. This criterion is maintained by insuring that the input vectors for simulated events are similar to those of past events and the input vectors have similar joint probabilities to those historical events of the training set. For example, a hurricane with a large central pressure deficit and low maximum winds is not a realistic event – the two parameters are not independent although their precise dependency is unknown. The simulation of realistic events is accounted for in the nearest-neighbor interpolation-bootstrap-resampling technique developed by Borgman (Scheffner, et al. 1999 and Borgman, et al. 1992). By using the training set as a basis of for defining future events, unrealistic events are not included in the life cycle of events generated by the EST. Events that are output by EST are similar to those in the training set with some degree of variability from the historic/historically based events. This variability is a function of the nearest neighbor: therefore the deviation from historic conditions is limited to natural variability of the system. The basic technique can be described as follows. Let X1, X2, X3, . . . Xa be n independent, identically distributed random vectors ( historic storm events) each having two components [Xi=xi(1),xi(2); I =1,n]. If there are no hypothetical events, each event Xi has a probability pi of l/n. If one storm event is used to generate two hypothetical events, then the original storm and each of the two perturbations are assigned a probability of one-third of l/n. A cumulative probability relationship can be developed in which each storm event of the total training set of 333 surge plus tide events is assigned a segment of the total probability of 0.0 to 1.0. Therefore each event occupies a fixed portion of the 0.0 to 1.0 cumulative probability space according to the total number of events in the training set. A random number from 0 to 1 is then used to identify a storm event from the total storm training set population. The procedure is equivalent to drawing and replacing a random sample from the full storm event population. The EST is not simply a resampling of historical events technique, but rather an approach intended to simulate the vector distribution contained in the training set data base population. The EST approach is to select a sample storm based on a random number selection from 0 to 1 and then perform a random walk from the event Xi with n

Page32

number of response vectors to the nearest neighbor vectors. The walk is based on independent uniform random numbers on (-1,1) and has the effect of simulating responses that are not identical to the historical events but are similar to events, which have historically occurred. However it is important to point out that it is possible is that the response value of water surface elevation (i.e. tide plus surge) may be greater than the greatest value in the total training set or it could be smaller than the smallest of the training set. The process can be summarized as follows. Select a specific storm event from the training set and proceeds to the location in multidimensional input vector space corresponding to that event. From that location, perform a nearest neighbor random walk to define a new set of input vectors. This new input vector defines a new storm, similar to the original storm but with some variability in parameters.

Storm Event Frequency

The second criteria to be satisfied is that the number of storm events selected per year must be statistically similar to the number of historical events that have occurred at the area of concern. Given the mean frequency of storm events for a particular region, a Poisson distribution is used to determine the average number of expected events in a given year. For example, a Poisson distribution can be written in the following form:

!);Pr( ses s λλλ −= (5)

for s=0,1,2,3,…. The probability Pr(s;λ) defines the probability of having s events per

year where λ is the historically based number of events per year.

In the present study, historical data were used to define λ as: λ = 0.2307 (27 historical events/117 years or one event every 4.33 years)

Output from the EST program is N repetitions of T years of simulated storm event responses. For this study, N = 500 repetitions of a T = 200 year sequence of storm activity are used. It is from the responses of those 500 life cycle simulations that frequency-of-occurrence relationships are computed. Because EST output is of the form of multiple time –series simulations, post processing of output yields mean value frequency relationships with definable error estimates. The computational procedure followed is based on the generation of a probability distribution function corresponding

Page33

to each of the T-year of simulated data. In the following section, the approach adopted for using these storms to develop frequency-of-occurrence relationships is given.

Risk-Based Frequency Analysis

The primary justification for applying the EST to a specific project is to generate risk-based frequency information relating to effectiveness and cost of the project with the level of protection provided. The multiple life-cycle simulations produced by EST can be used for developing design criteria in two approaches. In the first, the actual time series are input to an economics based model that computes couple storm inundation, structure response, and associated economics. The model internally computes variability associated with the risk-based design. The other application is the post processing of multiple time-series to generate single-response frequency relationships and associated variability.

Frequency-of Occurrence Relationships

Estimates of frequency-of-occurrence begin with the calculation of a probability distribution function (pdf) for the response vector of interest. Let X1, X2, X3, . . . , Xn be n independent, identically distributed, random response variables with a cumulative pdf given by Fx (x) = Pr [ X < x ] (6) where Pr[X<x] represents the probability that the random variable X is less than or equal to some value x, and Fx(x) is the cumulative probability density function ranging from 0.0 to 1.0. The problem is to estimate the value of Fx without introducing some parametric relationship for probability. The following procedure is adopted because it makes use of the probability laws defined by the data and does not incorporate any prior assumptions concerning the probability relationship. Assuming a set of n observations of data, the n values of x are first ranked in order of increasing size. In the following analysis, the parentheses surrounding the subscript indicate that the data have been rank-ordered. The value x(1) is the smallest in the series and x(n) represents the largest value. Let r denote the rank of the value xI such that rank r = 1 is the smallest and rank r = n is the largest.

Page34

An empirical estimate of Fx(xI), denoted by Fx(xI), is given by Gumbel (1954) (See also Borgman and Scheffner (1991) and Scheffner and Borgman (1992)) as:

)1()( )( +

=n

rxF rχ (7)

for xI, r = 1, 2, 3, . . ., n. This form of estimate allows for future values of x to be less than the smallest observation x(1) with a cumulative pdf of 1/(n+1), and to be larger than the largest values with cumulative pdf of n/(n+1). The cumulative pdf as defined by Equation 7 is applied to develop stage-frequency relationships as follows. Consider that the cumulative probability for an n-year return period storm can be written as

nnF 11)( −= (8)

where F(n) is the simulated cumulative pdf for an event with a return period of n years. Frequency-of-occurrence relationships are obtained by linearly interpolating a stage from Equation 7 corresponding to the pdf associated with the return period calculated by Equation 8. Equations 7 and 8 are applied to each of the N-repetitions of T-years of storm events simulated via the EST. Therefore, there are N frequency-of-occurrence relationships generated. From these results, the standard deviation is determined to provide an estimate of the variability of the result. The standard deviation is computed for each return period as:

2

1(1 / ) ( ) ]

N

nn

N x xσ−

=

= − ∑ 9)

where x is the mean value of x. An example set of 500 frequency relationships and the mean value for the Freeport Harbor are shown in Figure 15. Figure 16 shows the mean value with the +/- one standard deviation error bounds. Similar plots for each of the stations for which frequency results are reported are given in Appendix B. The extreme event storm surge for locations around the Freeport levee for 200, 100 and 50 year storms are shown in Figures 17-19. All frequency tables corresponding to stations of Figure 14 are given in Appendix 3. Surge elevations are affected by many variables such as offshore bathymetry, storm/shoreline orientation, location with respect to the Gulf of Mexico, and the local topography. Therefore, the frequency-indexed surge distribution varies from one end of the project to the other.

Page35

0

1

2

3

4

5

6

7

8

9

10

11

10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200

Surfa

ce E

leva

tion,

m

Return Period (years)

Frequency Spread

Figure 15 Frequency relationships and the mean value for the Freeport Harbor

Page36

-1

0

1

2

3

4

5

6

7

10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200

Surfa

ce E

leva

tion,

m

Return Period ( )

Frequency Curve

80% Confidence +/- 1 S.D.Mean

Figure 16 Mean Value Frequency with standard deviation bounds for Freeport Harbor

Page37


Figure 17 200 year storm surge values around the Freeport Levee System

Page38



Page39



Page40

WAVE RUNUP ANALYSIS

Large waves are often associated with the arrival of hurricane storm surge and hurricane force winds. The analysis of these waves and their runup is an important part of the assessment of risk for exposure to hurricane conditions at the levees maintained by the Velasco Drainage District. Wind waves generated offshore and even nearshore by wind from strong storms and hurricanes become very important since the area of interest can be greatly affected by these factors. Since the barrier islands may become submerged during large hurricane events, the levees which are located from one to eight miles behind the normal beaches may be affected by the waves as well as the surge. As the wind waves leave the area of the winds where they are generated they will travel outwards from the center of the storm. Some of these waves will travel towards Freeport. As they travel in this manner, the waves will be affected by the changes in the bottom depths. This will result in changes in the wave height, direction and wave length and thus the resulting steepness. The hydrodynamics of the transformation of waves as they enter shallow water from the deep ocean and then continue to travel across barrier islands and wetlands is quite complicated. The present report explains the use of two nested models employed to analyze the transformation of the waves as they travel from the storm center to Freeport, for different storms. Estimation of the wave heights and periods generated by a given storm in the time of its passage through the northwestern Gulf of Mexico was used to provide input to the model STWAVE. The nearshore output from STWAVE was used as a boundary condition for a second model called COULWAVE which also calculated the wave transformation over the barrier island and across the wetlands and lastly the wave runup onto the levees.

The STWAVE Model

In order to determine wave conditions at the Freeport area, a model called STWAVE (Steady-state spectral WAVE model) (Smith, Resio, and Zundel 2001) is used to estimate wind-wave growth, transformation, and propagation in the offshore-nearshore area, where changes in wave height, period, direction, and spectral shape, are influenced by variations in bathymetry, water level (surge induced water level), and current. STWAVE is considered a steady-state finite difference model based on the wave action balance equation, which requires the following model assumptions:

• Mild bottom slope and negligible wave reflection • Spatially homogeneous offshore wave conditions • Steady-state waves, currents, and winds • Linear refraction and shoaling • Depth-uniform current • Bottom friction is neglected

Page41

• Linear radiation stress Therefore, STWAVE is capable of simulating diffraction, depth-induced wave refraction and shoaling, current induced refraction and shoaling, depth- and steepness-induced wave breaking, wind-wave growth, wave-wave interaction and white capping that redistribute and dissipate energy in a growing wave field. Refraction and shoaling are calculated by applying the conservation of wave action along backward traced wave rays. Only ray directions propagating toward the shore with angles between –87.5 and +87.5 degrees are included. Energy propagating toward the offshore is not considered. Diffraction is evaluated in STWAVE through smoothing of wave energy in a given frequency and direction band using the following form:

( ) ( ) ( ) ( )( )θϖθϖθϖθϖ ,,255.0,55.0, 11 ajajajaj EEEE −+ ++= (10) where E is the energy density in a given frequency and direction band, and the subscript j indicates the grid row index. Strong wave height gradients that occur in shelter regions are smoothed by equation 10, while providing no turning on the waves. Laboratory measurements performed by Smith, Resio, and Vincent (1997), with irregular wave breaking on ebb currents demonstrated that a breaking relationship in the form of the Miche criterion was accurate:

( )kdLH mo tanh1.0max

= (11) As a result equation 11 is applied as a maximum limit on the zero-moment wave height. Nonlinear transfers of wave energy to higher frequencies that occur during breaking are not represented in the model. The transfer of momentum from the wind field to the wave field makes waves grow. Therefore the flux of energy, Fin, into the wave field in STWAVE is given by:

guCF m

w

ain

2*85.0

ρρλ= (12)

where λ = partitioning coefficient that represents the percentage of total atmosphere to water momentum transfer that goes directly into the wave field (0.75) ρa = density of air Cm = mean wave celerity u* = friction velocity (equal to the product of the wind speed, U, and the square root of the drag coefficient, CD = .0012+.000025U)

Page42

Due to STWAVE being a half-plane model, only winds blowing toward the shore (+x direction) are included in the simulations. Wave damping by offshore winds and growth of offshore-traveling waves are neglected. As energy is fed into the waves from the wind, it is redistributed through nonlinear wave-wave interaction. Energy is transferred from the peak of the spectrum to lower frequencies and to higher frequencies. In STWAVE, the frequency of the spectral peak is allowed to increase with fetch (or equivalently propagation time across a fetch). The equation for this rate of change of fp is given by:

( ) ( )7

33

4

*37

1 59

−

+

∆

−= t

guff

ipip ζ (13)

where the I and I+1 subscripts refer to the grid column indices within STWAVE and ζ is a dimensionless constant as indicated by Resio and Perrie (1989). The energy flux to high frequencies is represented in STWAVE as:

( )dk

kEg

p

pE

43

2932

1

tanh

ε=Γ (14)

where Γ E = energy flux ε = coefficient equal to 30 Etot = total energy in the spectrum divided by (ρw g) kp = wave number associated with the peak of the spectrum

Computational grid

Wave conditions in the Freeport area were estimated by using the STWAVE program, which requires a computational grid. An SMS interface was used for developing the grid shown in Figure 20; this arrangement covers the Freeport area and its offshore zone. The grid has a total of 72,600 elements, and each element is 100 x 100 meters. Extension of the grid along the shoreline is approximately 33 kilometers, and it is 22 kilometers wide. Data related to bathymetry, tidal elevations, storm surge, wave setup, wave height, and period are required as inputs for the model. The bathymetry was obtained mostly from surveys conducted by Texas A&M University (in some areas and canals), GEODAS (data base from NOAA), USACE

Page43

surveys, and some points from the ADCIRC grid prepared by Scheffner et al. (2003); all of these bathymetries are the same as those used for the ADCIRC program and are adequate for this work. Tidal elevations were determined relying on information obtained from station 8772440 at Freeport, Dow barge canal, which is monitored by NOAA. Table 1 shows elevations of tidal datums referred to Mean Lower Low Water (MLLW), in meters. Therefore from Table 6 were established tide elevations for three conditions considered during the model runs. It was assumed that a combination of storm surge and wave setup could happen during any of the following tide elevations MHHW, MSL and MLLW. This gives three different runs for each one of the storms selected. In order to use these elevations, and make them compatible with the bathymetry, they were referred to the MSL (NAVD 1988). Table 7 shows the tidal elevations used for the runs referred to MSL.

Table 6: Elevations of tidal datums referred to Mean Lower Low Water in meters

Highest Observed Water Level (09/11/1998) = 1.800 Mean Higher High Water (MHHW) = 0.536 Mean High Water (MHW) = 0.494 Mean Sea Level (MSL) = 0.291 Mean Tide Level (MTL) = 0.289 Mean Low Water (MLW) = 0.083 Mean Lower Low Water (MLLW) = 0.000 Lowest Observed Water Level (02/25/1965) = -1.295

Table 7: Tidal elevations referred to Mean Sea Level in meters

Mean Higher High Water (MHHW) = 0.245 Mean Sea Level (MSL) = 0.000 Mean Lower Low Water (MLLW) = -0.291

A thorough analysis was made to determine storm surges, which were obtained from ADCIRC, for each one of the 37 storms considered for this study. Several plots were used to identify the maximum surge and wind field associated with the worst wave conditions arriving at Freeport. This was done for each storm. To illustrate the procedure Figure 21 shows a plot of the surge for a given storm, and Figure 22 shows the wind field related to the worst condition of both surge and incoming waves for the same storm at Freeport.

Page44

Figure 20 Computational Domain

Page45

Figure 21 Example of storm surge time series from Hurricane Celia; the surge is taken to coincide with the worst incoming waves arriving at Freeport

Surge selected

TIME SERIES Freeport, TX

Page46

Figure 22 Example of wind field generated by hurricane Celia; the worst incoming waves are at time 195000 seconds

Page47

Once this procedure was completed for all of the storms, it was possible to get main angle of incidence (θ) of the waves and surges. Table 8 presents these results. A method indicated by the Shore Protection Manual (USACE 1984) was employed to estimate offshore wave height and period at the point of maximum wind speed. The information related to pressure at the eye of the depression, maximum wind speed, position and forward speed of the hurricane, was obtained from data available at “http://weather.unisys.com/hurricane/atlantic/” website and from data estimated by the Planetary Boundary Layer program for the radius of maximum wind speed. By making use of the formulation mentioned above wave height (Ho) and period (Ts) were estimated and propagated to the offshore part of the STWAVE grid, these results are shown also in Table 8. Once the information was gathered a series of pre runs for each storm were made in order to know the wave heights near the shoreline. This information was employed to estimate wave setup, which was considered as 15% of the mean wave height arriving to the shoreline. The wave setup for each storm is shown in Table 8. With tide elevations, storm surge, and wave setup known, it was possible to determine the tide level, still water elevation that is used by STWAVE, which in this case is equal to the sum of tide elevation plus storm surge plus wave setup. As a result there are 3 different conditions for the still water level, since the only parameter that changes is astronomical tide elevation (see Table 7).

STWAVE Results and Analysis

Thirty-seven storms were simulated for three different tidal conditions (MLLW, MSL, MHHW), for a total of 111 runs. The simulations provided angle of incidence, significant wave height, peak period, and wave spectra for the whole domain. In order to know the parameters mentioned above at specific points of interest, six transects were established perpendicular to the coast, along them were measured wave height and wave period. Figure 23 shows the arrangement of these transects and Figure 24 and 25 show typical STWAVE outputs. The results from the simulations showed that waves were only able to reach the levees for storms that flooded areas above the average topography elevation on the order of 1.00 meter. There are a total of 12 storms for the MHHW condition, and 7 storms for each one of the remaining conditions MSL and MLLW, which allowed waves to overwash the barrier islands. Appendix D shows the plots of wave height and period measured for these storms, as well as a cross section of each transect used in the simulations. For the rest of the storms, there were no waves reaching the levees from offshore. A problem was noted with the results of STWAVE as the waves approached the levees. The wave and runup values were much too large for most storms. This situation was corroborated when the wave runup calculations were made for hurricane Carla, for the

Page48

condition of MHHW where wave run up was very large and not supported by any of the photographic evidence from that storm.

Table 8: STWAVE deepwater boundary conditions

Hurricane Storm Surge Wave setup Ho Tp θ Name (m) (m) (m) (sec) Alicia 1.011 0.240 7.025 10.240 84.000

Alicia, (+1.0 long) 1.395 0.300 4.913 8.563 -5.128 Alicia, (-1.0 long) 0.340 0.045 0.699 3.230 78.780

Allison 0.230 0.230 7.533 10.600 -26.022 Audrey 0.395 0.128 1.960 5.410 76.667 Bonnie 0.302 0.180 3.104 6.810 81.300 Carla 3.107 0.450 8.542 11.290 33.638

Carla, (+1.0 long) 2.000 0.390 7.130 10.316 3.780 Carla, (-1.0 long) 4.304 0.585 11.605 13.161 72.923

Celia 1.500 0.353 9.473 11.891 23.298 Chantal 0.260 0.140 2.462 6.062 82.374

Chris 0.360 0.225 6.480 9.834 70.388 Claudette 1.996 0.363 7.300 10.430 29.570

Dean 0.330 0.240 6.582 9.911 -55.091 Debra 0.524 0.255 6.284 9.680 17.988 Delia 1.210 0.315 8.791 11.454 6.745 Edith 0.820 0.330 8.539 11.289 70.807 Fern 1.574 0.255 8.765 11.437 17.893

Frances 0.930 0.225 7.857 10.830 40.840 Jerry 0.370 0.045 4.392 8.096 83.364

Storm 117 0.837 0.225 11.730 13.787 -68.040 Storm 117, (+1.0 long) 2.917 0.480 9.606 11.973 -47.790 Storm 117, (-1.0 long) 0.150 0.255 8.230 11.080 -66.495

Storm 183 1.020 0.240 8.066 10.972 -56.058 Storm 211 1.403 0.330 13.871 14.388 -62.272

Storm 211, (+1.0 long) 3.303 0.300 11.451 13.073 -28.940 Storm 211, (-1.0 long) 0.398 0.270 11.321 13.000 -66.740

Storm 232 0.260 0.045 1.062 3.980 64.530 Storm 295 0.920 0.180 6.196 9.616 32.847 Storm 310 0.285 0.210 5.356 8.940 -58.388

Storm 310,(+1.0 long) 3.005 0.300 7.623 10.670 -4.551 Storm 310,(-1.0 long) 0.130 0.188 5.522 9.078 -69.013

Storm 324 0.510 0.173 3.065 6.764 75.756 Storm 397 0.055 0.203 5.436 9.007 -83.831 Storm 405 2.050 0.345 9.547 11.936 -1.505 Storm 445 1.350 0.300 5.366 8.949 -1.703

Storm 5 0.957 0.150 2.962 6.650 39.080 Note: Wave incidence angle is referenced to the STWAVE local coordinate system.

Page49

Figure 23 Arrangement of transects used to measure wave parameters

Page50

Figure 24 Typical output of STWAVE, vectors indicate wave direction

Page51

Figure 25 Typical output of STWAVE; the colors indicate wave height

The situation mentioned above was of serious consideration; therefore it was decided to use a different program that could simulate properly the wave evolution from relatively deep water to the breaking point and from there across the barrier island and marsh to the levees. However the results obtained from STWAVE for the nearshore area were employed as the boundary conditions for the model COULWAVE that is described in the next section.

The COULWAVE Model

The model used for all the simulations related to wave propagation in shallow water and runup on the levees is called COULWAVE, for Cornell University Long and Intermediate Wave Modeling Package. The COULWAVE program developed by Dr. Patrick J. Lynett∗ makes use of Boussinesq and Boussinesq-type equations, which have been used extensively in coastal regions, including interaction with porous coastal structures (Lynett et al., 2000), and internal wave propagation and evolution (Lynett and Liu, 2002 (a), (b)). The governing equations employed in this numerical model allow for the evolution of fully nonlinear (wave amplitude to water depth ratio = O(1) ) and dispersive waves over variable bathymetry. This model has the ability to simulate a wide range of wave processes such as shoaling, diffraction, refraction, wave-wave and wave-current interactions, and nonlinear transformation. The model is not appropriate in general for simulations in deep water – beyond where the bottom affects the wave height and direction. So, the STWAVE model was used for the deepwater transformation from the near the storm center to the nearshore and COULWAVE was used to take the waves from the nearshore over the barrier islands, marsh and onto the levees. The numerical model utilizes a moving boundary algorithm, which was developed for use with depth integrated equations. The model is used here in conjunction with a fixed grid finite difference scheme. Founded around the restrictions of the high-order numerical wave propagation model, the moving boundary scheme employs linear extrapolation of free surface and velocity through the wet–dry boundary, into the dry region. The technique is numerically stable, does not require any sort of additional dissipative mechanisms or filtering, and conserves mass. The moving boundary has been tested for accuracy using one- and two-dimensional analytical solutions and experimental data sets. Nonbreaking and breaking solitary wave runup is accurately predicted, yielding a validation of both the eddy viscosity breaking parameterization and the runup model (Lynett et al., 2002). A better explanation of the governing equations can be found in the COULWAVE Code Manual (Lynett et al., 2004), which is a summation of the author’s analytical and numerical work concerning depth-integrated wave theory, submarine landslide modeling, and runup modeling.

∗ Assistant Professor, Department of Civil Engineering, Texas A&M University, College Station, TX, USA

Page53

The inputs for COULWAVE were obtained from STWAVE and they are presented in Appendix D. The program was run as one dimensional, and the results are described in the next section.

COULWAVE Results

A total of six transects were analyzed for each one of the 12 storms that were considered for the MHHW condition. The results are summarized in Table 9 that shows transect number, tide level, significant wave height (Hs), peak period (Tp), maximum runup and maximum wave runup elevation referred to the MSL (NAVD 1988). Similar results are given for the MSL and MLLW conditions in Tables 10 and 11 respectively. It should be noticed that runup values given in Tables 9, 10 and 11 are for the maximum runup out of 400 waves, which is approximately a little over an hour’s duration. It can be seen in those tables that the worst runup conditions for all of the storms are for transect number 4. Hurricane Carla (-1.0 long) is the storm which presents the largest maximum runup levels for each transect, except transect 4. The worst case scenario for maximum runup is actually Storm 211 (+1.0 long), transect 4. Even though the tide is higher for Carla (-1.0 long), the larger wave height and increased shoaling in the slightly shallower water leads to greater wave heights near the front levee for Storm 211 (+1.0 long). Appendices E, F and G, show the graphical outputs of the COULWAVE model for MHHW, MSL and MLLW conditions respectively. The runup values and elevations for transect 4 for the large storms seem to be out of line with the other transects. Looking in detail at the location of the transect, it is seen that it is drawn directly through the jetties and the harbor to the revetment, all through water of approximately 12 m. As a result, the waves arrive without much loss of energy due to diffraction, breaking on the jetty structures or spreading away to the sides of the channel. This is a problem using just a “simple” one-dimensional model. Therefore, it was decided that a two-dimensional simulation would be done for one of the largest storms to determine the actual effect of the simplification to a 1-D model. Figure 26 shows the results of wave height for a two-dimensional run for hurricane Carla (-1.0 long). It can be seen that wave heights along transect 4 are not as large as given by the 1-D model. This simulation demonstrates that the 1-D results from transect 4 overestimate the likely wave runup and runup elevation. The two-dimensional run shows the spread of energy quickly away from transect 4 as the waves move through the inlet; this allows the wave height and runup to be diminished well below that predicted with the 1-D method. It is thus assumed that the effects of transect 4 can be represented as an average of transect 3 or transect 5.

Page54

Figure 26 Results of two-dimensional COULWAVE simulation showing wave height at entrance to Freeport Harbor for Hurricane Carla (-1.0 long). Note the diffraction of waves

as they enter the channel. Also note that North is downward in this figure.

Page55

Page56

Table 9: Results from COULWAVE program for the MHHW condition

Hurricane Transect Tide Level Hs Tp Max Runup

Max Runup Elev

Name (m) (m) (sec) (m) MSL (m) TR1 1.940 0.11690 197.69530 0.27854 2.21854 TR2 1.940 0.09890 197.69530 0.17064 2.11064 TR3 1.940 0.15023 197.81770 0.36916 2.30916 TR4 1.940 1.45670 10.51570 2.89060 4.83060 TR5 1.940 0.12721 98.90880 0.29783 2.23783

Alicia, (+1.0 long)

TR6 1.940 0.00000 0.11340 0.00000 1.94000 TR1 5.134 1.13630 197.50720 2.43150 7.56550 TR2 5.134 1.40120 39.57840 2.96670 8.10070 TR3 5.134 1.70410 199.53100 3.49580 8.62980 TR4 5.134 2.98200 9.99470 5.20000 10.33400 TR5 5.134 1.19940 33.25520 2.43510 7.56910

Carla, (-1.0 long)

TR6 5.134 0.75338 199.53100 1.51830 6.65230 TR1 3.802 0.54959 195.06640 1.04170 4.84370 TR2 3.802 0.72013 195.06640 1.09780 4.89980 TR3 3.802 0.87549 197.35900 2.12160 5.92360 TR4 3.802 3.67680 13.33510 6.30450 10.10650 TR5 3.802 0.60135 197.35900 1.39360 5.19560

Carla

TR6 3.802 0.34078 197.35900 0.92015 4.72215 TR1 2.635 0.21180 197.89210 0.65289 3.28789 TR2 2.635 0.32680 197.80430 0.75145 3.38645 TR3 2.635 0.40368 197.80430 0.89178 3.52678 TR4 2.635 2.99120 11.11260 5.71800 8.35300 TR5 2.635 0.27971 98.90210 0.60192 3.23692

Carla, (+1.0 long)

TR6 2.635 0.09646 197.80430 0.31844 2.95344 TR1 2.098 0.33143 197.35900 0.63073 2.72873 TR2 2.098 0.19142 197.35900 0.61656 2.71456 TR3 2.098 0.36919 197.35900 1.31870 3.41670 TR4 2.098 3.51990 13.36070 5.82530 7.92330 TR5 2.098 0.27967 98.67950 0.69673 2.79473

Celia

TR6 2.098 0.00000 0.16305 0.00000 2.09800 TR1 2.604 0.21180 197.89210 0.65289 3.25689 TR2 2.604 0.32680 197.80430 0.75145 3.35545 TR3 2.604 0.39453 197.80430 0.84678 3.45078 TR4 2.604 2.92860 11.11260 5.66730 8.27130 TR5 2.604 0.26510 98.90210 0.62738 3.23138

Claudette

TR6 2.604 0.09646 197.80430 0.31844 2.92244 TR1 2.074 0.30855 195.06640 0.64289 2.71689 TR2 2.074 0.17470 195.06640 0.63047 2.70447 TR3 2.074 0.35964 197.35900 1.30200 3.37600 TR4 2.074 3.51990 13.36070 5.82530 7.89930

Fern

TR5 2.074 0.27624 98.67950 0.68773 2.76173

Page57

TR6 2.074 0.00000 0.16063 0.00000 2.07400

Table Continued…


Max Runup Elev


Storm 117, (+1.0 long)

TR6 3.642 0.28566 199.53100 0.82806 4.47006 TR1 1.978 0.31309 197.51480 0.82338 2.80138 TR2 1.978 0.20343 197.51480 0.79400 2.77200 TR3 1.978 0.23988 197.46650 1.45040 3.42840 TR4 1.978 4.00790 197.51480 6.02740 8.00540 TR5 1.978 0.25080 98.73320 0.76145 2.73945

Storm 211

TR6 1.978 0.00000 0.16063 0.00000 1.97800 TR1 3.848 0.72613 197.50720 1.38510 5.23310 TR2 3.848 0.95281 197.50720 1.73470 5.58270 TR3 3.848 1.01500 199.53100 2.43520 6.28320 TR4 3.848 4.67540 13.30210 7.93110 11.77910 TR5 3.848 0.65643 199.53100 1.67200 5.52000


TR6 3.848 0.33502 199.53100 1.00440 4.85240 TR1 3.55 0.43329 197.89210 1.18890 4.73890 TR2 3.55 0.50488 197.89210 0.89703 4.44703 TR3 3.55 0.69254 197.80430 1.78260 5.33260 TR4 3.55 3.10180 11.11260 5.91990 9.46990 TR5 3.55 0.52265 98.90210 1.17510 4.72510

Storm 310,(+1.0 long)

TR6 3.55 0.29045 197.80430 0.71215 4.26215 TR1 2.64 0.11447 65.02210 0.26453 2.90453 TR2 2.64 0.10953 14.34310 0.26977 2.90977 TR3 2.64 0.58609 197.35900 1.29830 3.93830 TR4 2.64 3.73520 13.33510 6.56590 9.20590 TR5 2.64 0.37330 98.67950 0.87235 3.51235

Storm 405

TR6 2.64 0.11271 197.35900 0.37214 3.01214

Page58

Table 10: Results from COULWAVE program for the MSL condition

Hurricane Transect Tide Level Hs Tp Max

Runup Max Runup

Elev Name (m) (m) (sec) (m) MSL (m)

TR1 4.889 1.04070 197.50720 1.96990 6.85890 TR2 4.889 1.11850 197.50720 1.98050 6.86950 TR3 4.889 0.76932 15.34850 1.03550 5.92450 TR4 4.889 1.75660 15.34850 4.08100 8.97000 TR5 4.889 0.53200 33.25520 0.73938 5.62838

Carla, (-1.0 long)

TR6 4.889 0.37283 33.25520 0.56586 5.45486 TR1 3.557 0.48900 195.06640 0.94274 4.49974 TR2 3.557 0.69724 195.06640 0.99713 4.55413 TR3 3.557 0.38658 12.49110 0.68651 4.24351 TR4 3.557 1.28160 11.74760 2.80710 6.36410 TR5 3.557 0.28220 65.78630 0.41508 3.97208

Carla

TR6 3.557 0.23241 0.54370 0.28489 3.84189 TR1 2.390 0.20768 197.89210 0.54643 2.93643 TR2 2.390 0.17807 197.89210 0.48477 2.87477 TR3 2.390 0.13804 197.80430 0.21928 2.60928 TR4 2.390 2.29000 11.11750 4.62440 7.01440 TR5 2.390 0.10617 98.90210 0.16655 2.55655

Carla, (+1.0 long)

TR6 2.390 0.00000 0.01977 0.00000 2.39000 TR1 2.359 0.19710 197.89210 0.53663 2.89563 TR2 2.359 0.17041 197.89210 0.48931 2.84831 TR3 2.359 0.13396 197.80430 0.21195 2.57095 TR4 2.359 2.25490 11.11750 4.58540 6.94440 TR5 2.359 0.10056 98.90210 0.15747 2.51647

Claudette

TR6 2.359 0.00000 0.16789 0.00000 2.35900 TR1 3.397 0.47076 197.50720 1.08130 4.47830 TR2 3.397 0.74869 197.50720 1.10030 4.49730 TR3 3.397 0.35532 39.90620 0.65454 4.05154 TR4 3.397 1.73210 15.34850 3.60200 6.99900 TR5 3.397 0.23509 66.51030 0.43927 3.83627


TR6 3.397 0.16638 0.14789 0.31437 3.71137 TR1 3.603 0.59590 197.50720 1.46870 5.07170 TR2 3.603 0.87123 197.50720 1.29020 4.89320 TR3 3.603 0.39389 39.90620 0.76322 4.36622 TR4 3.603 1.80750 15.34850 3.83860 7.44160 TR5 3.603 0.26672 66.51030 0.53710 4.14010


TR6 3.603 0.19361 14.07770 0.36863 3.97163 TR1 3.305 0.36993 197.89210 0.92546 4.23046 TR2 3.305 0.46559 197.89210 0.80712 4.11212 TR3 3.305 0.30025 49.45110 0.66627 3.97127

Storm 310,(+1.0 long)

TR4 3.305 1.10410 11.11260 2.41480 5.71980

Page59

TR5 3.305 0.28307 16.76310 0.39246 3.69746 TR6 3.305 0.10346 11.11260 0.23061 3.53561

Table 11: Results from COULWAVE program for the MLLW condition


Max Runup Elev


Carla, (-1.0 long)

TR6 4.598 0.34800 18.13920 0.49743 5.09543 TR1 3.266 0.39426 195.06640 0.87306 4.13906 TR2 3.266 0.62715 195.06640 0.92139 4.18739 TR3 3.266 0.31292 12.49110 0.58515 3.85115 TR4 3.266 1.26060 11.74760 2.72340 5.98940 TR5 3.266 0.23793 14.30140 0.38567 3.65167

Carla

TR6 3.266 0.00000 0.16063 0.00000 3.26600 TR1 2.099 0.15576 197.89210 0.40848 2.50748 TR2 2.099 0.14443 197.89210 0.38900 2.48800 TR3 2.099 0.09097 197.80430 0.15515 2.25415 TR4 2.099 2.28560 11.11750 4.52530 6.62430 TR5 2.099 0.08320 98.90210 0.12858 2.22758

Carla, (+1.0 long)

TR6 2.099 0.00000 0.15073 0.00000 2.09900 TR1 2.068 0.15125 197.89210 0.32147 2.38947 TR2 2.068 0.14919 197.89210 0.31190 2.37990 TR3 2.068 0.08684 197.80430 0.14772 2.21572 TR4 2.068 2.24960 11.11750 4.35740 6.42540 TR5 2.068 0.08060 39.56090 0.12504 2.19304

Claudette

TR6 2.068 0.00000 0.16083 0.00000 2.06800 TR1 3.106 0.38998 197.50720 1.22270 4.32870 TR2 3.106 0.56723 197.50720 0.97201 4.07801 TR3 3.106 0.30198 39.90620 0.63626 3.74226 TR4 3.106 1.71790 15.34850 3.49530 6.60130 TR5 3.106 0.19795 99.76550 0.33597 3.44197


TR6 3.106 0.07080 15.34000 0.20579 3.31179 TR1 3.312 0.48622 197.50720 1.38880 4.70080 TR2 3.312 0.69541 197.50720 1.16560 4.47760 TR3 3.312 0.33912 39.90620 0.67155 3.98355 TR4 3.312 1.79200 15.34850 3.70860 7.02060 TR5 3.312 0.21511 16.62760 0.42443 3.73643


TR6 3.312 0.08930 14.43000 0.21000 3.52200 Storm 310,(+1.0 long) TR1 3.014 0.29227 197.89210 0.78580 3.79980

Page60

TR2 3.014 0.34583 197.89210 0.72106 3.73506 TR3 3.014 0.25070 49.45110 0.54478 3.55878 TR4 3.014 1.09930 11.11260 2.37650 5.39050 TR5 3.014 0.18813 39.56090 0.31661 3.33061

TR6 3.014 0.00000 0.01107 0.00000 3.01400

Page61

Frequency Indexed Wave Runup

Similar to the analysis for the storm surge elevations, a frequency index for the maximum wave runup elevation was performed using the empirical simulation technique (EST). As before, 37 different storms were considered. For each storm it was assumed that the occurrence at MHHW, MSL and MLLW was equally likely to occur. However, since the range between spring and neap tide is not that great in this area, only the mean tide conditions were considered in the EST analysis. In other all words, all 27 historical storms, hurricane and large extra tropical, that have had an effect on this coast were used along with ten hypothetical storms for each of the above noted tidal conditions. The results of the EST give a frequency curve for each transect. The results given in Figures 27-31 show the return period for a given maximum runup elevation. Note that these runup elevations are above NAVD 1988. For example Figure 27 shows that the maximum expected elevation of the water, including tide, storm surge, wave setup and wave runup to be about 4.9 m for a once in a one-hundred year event. The one factor that this does not include is the effect of the wind on the runup and potential overtopping.

Page62

1

2

3

4

5

6

7

8

9

10

10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200

Max

Run

Up

Elev

atio

n (m

)


Frequency Curve

+/- 1 S.D.Mean

Figure 27 Maximum Runup Frequency of Occurrence at TR1

Page63

1

2

3

4

5

6

7

8

9

10

11

10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200

Max

Run

Up

Elev

atio

n (m

)


Frequency Curve

+/- 1 S.D. Mean


Page64

1

2

3

4

5

6

7

8

9

10

11

10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200

Max

Run

Up

Elev

atio

n (m

)


Frequency Curve

+/- 1 S.D. Mean


Page65

1

2

3

4

5

6

7

8

9

10

10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200

Max

Run

Up

Elev

atio

n (m

)


Frequency Curve

+/- 1 S.D. Mean


Page66

1

2

3

4

5

6

7

8

9

10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200

Max

Run

Up

Elev

atio

n (m

)


Frequency Curve

+/- 1 S.D. Mean


Page67

CONCLUSIONS

This report describes an investigation of tropical storms occurring along the Freeport, Texas area. The goal of the study was to determine the degree of risk of flooding for the Freeport area that is protected by the levee system. This was accomplished by generating storm surge plus tide frequency-of-occurrence relationships. Results are based on numerical simulations of tidal elevations and storm surges for tropical storm events that have historically impacted the study area. Simulations were made using the long-wave hydrodynamic model ADCIRC. The model results were input to the EST statistical simulation model to produce multiple life-cycle simulations of tide and storm surge activity along the coast to generate surface elevation versus frequency-of-occurrence relationships for 35 selected locations within the study area. As a result of the study the maximum storm surge plus tide value for the 200-year storm was found to be 5.47 m and occurred in Oyster Creek. For the 100-year storm the maximum storm surge plus tide value of 3.98 m also occurred in Oyster Creek as did the maximum for the 50-year storm. However, none of the storm surge plus tide values were greater than the current height of the levee system. In addition to investigating the frequency–of-occurrence for storm surges along the levee system, an analysis was also made to determine the frequency-of-occurrence for runup of wind-generated waves. The procedures to determine the wave conditions at the toe of the levee included predicting the maximum waves generated by the storm at some offshore position and determining the transformation of those waves as they came close to the shoreline. The wave transformation was determined using STWAVE. The results from STWAVE were used to compute the boundary condition for the nonlinear model COULWAVE which was used to model the transformation of the waves when the surge was high enough to overtop the barrier islands allowing the waves from offshore to attack the levees. The COULWAVE model also computed the wave runup on the levees as well. The frequency-of-occurrence relationship was determined for five locations along the levee. The highest expected value for the maximum wave elevation was computed to be 6.6 m above NAVD 1988 for a 200-year storm and 4.9 m for a 100-year event. The highest levels of wave runup occurred west of the Freeport area at transects 1 through 3. The areas to the east of Freeport had approximately 0.5 m less runup elevation. The maximum wave elevation includes the storm surge, tides, and wave setup and runup. Although some storms did have a maximum runup in excess of 8 meters, the return period of those events would be much more than 200 years. The contribution to wave overtopping of the levees by wind is not included in this analysis.

Page68

ACKNOWLEDGEMENTS

The work presented here would not have been possible without the contributions made by many individuals. In addition to the students listed as authors, Laura Lynn Robinson and Murtaza Bakrawala assisted with the early work. We would also like to thank Dr. Norman Scheffner and Dr. David J. Mark of the Corps of Engineers ERDC (Engineer Research and Development Center) for their help with some of the problems that arose during the application of ADCIRC. Dr. Jeff Waters and his colleagues at the Galveston District of Corps of Engineers were very helpful in providing bathymetric data and understanding as the wave transformation modeling built upon the storm surge results. Lastly, Mr. George Kidwell and Mr. Mel McKey of the Velasco Drainage District have made substantial contributions to the completion of this modeling study. They have both shown a thorough understanding of the complexity and the processes by which this type of risk based analysis proceeds. Their patience and support is sincerely appreciated. Hopefully, this report will provide necessary information for them as they continue to fulfill their duties to the Velasco Drainage District.

Page69

REFERENCES

1. Borgman, L.E., Miller, M.C., Butler, H.L., and Reinhard, R.D., Empirical simulation of future hurricane storm histories as a tool in engineering and Economic Analysis. ASCE Proc. Civil Engineering in the Oceans V. College Station, TX. 2-5 Nov. 1992.

2. Borgman, L.E. and Scheffner, N.W., 1991, “The simulation of time sequences of wave height, period, and direction.” TR-DRP-91-2. USACE, WES, Vicksburg, MS.

3. Cardone, V.J., Greenwood, C.V., and Greenwood, J.A., 1992, “Unified Program for the Specification of Hurricane Boundary Layer Winds Over Surfaces of specified Roughness,” contract report CERC-92-1, USACE, Waterways Experiment Station, Vicksburg, MS.

4. Flather, R.A. 1988, “A Numerical model investigation of tides and diurnal-period continental shelf waves along Vancouver Island.” Journal of Physical Oceanography 18, 115-139.

5. Gracia, Andrew W. and Flor, Thomas H., 1984, “Hurricane Alicia Storm Surge and Wave Data,” Technical report CERC-84-6, USACE, Waterways Experiment Station, Vicksburg, MS.

6. Gumbel, E.J., 1954,“Statistical theory of extreme value and some practical application.” National Bureau of Standards Applied Math. Series 33. U.S. Gov. Pub. Washington, DC.

7. Harris, D.L., 1963, “Characteristics of the Hurricane Storm Surge,” Technical Paper No. 48, U.S. Dept. of Commerce, U.S. Weather Bureau, Washington, DC.

8. Jarvinen, B.R., Neumann, C.J., and Davis, M.A.S., 1988, “A tropical Cyclone Data Tape for the North Atlantic Basin, 1886-1983: Contents, Limitation and Uses,” NOAA Technical Memorandum NWS NHC 22.

9. Jelesnianski, C.P. and Taylor, A.D., 1973, “A Preliminary View of Storm Surges Before and After Storm Modifications.” NOAA Technical Memorandum ERL WMPO-3, Weather Modification Program Office, Bounder CO

10. Kolar, R.L., Gray, W.G., Westerink, J.J., and Leutich, R.A. 1994, “Shallow water modeling in spherical coordinates:Equation formulation, numerical implementation, and application,” Journal of Hydraulic Research 32(1), 3-24.

11. Leutich, R.A., Westerink, J.J., and Scheffner, N.W., 1992, “ADCIRC: An Advanced Three-Dimensional Circulation Model for Shelves, Coasts, and Estuaries Report 1:Theory and Methodology of ADCIRC-2DDI and ADCIRC-3DL,” Technical report DRP-92-6, Nov. 1992, U.S. Army Engineer Waterways Experiment Station, Vicksburg, MS.

12. Lynett, P., Liu, P.L.-F. (2002a), “A numerical study of submarine landslide generated Waves and runup”. Proc. Royal Society of London A. v. (458), p. 2885-2910.

Page70

13. Lynett, P., Liu, P.L.-F. (2002b), “A two-dimensional, depth-integrated model for internal wave propagation”, Wave Motion, v. (36), p. 221-240.

14. Lynett, P., Liu, P. L.-F., Losada, I., and Vidal, C. 2000, “Solitary wave interaction with porous breakwaters”, Journal of Waterway, Port, Coastal, and Ocean Engineering (ASCE), v. 126, (6), p. 314-322.

15. Lynett, P. et al. 2002. “Modeling Wave Runup with Depth-Integrated Equations,” Coastal Engineering, v. 46(2), p. 89-107.

16. Lynett, P. et al. 2004. “COULWAVE Code Manual”, Available online at: http://ceprofs.tamu.edu/plynett/

17. Resio, D. T., and Perrie, W. (1989). “Implications of an f—4 equilibrium range for wind-generated waves,” J. Phys. Oceanography 19, 193-204.

18. Scheffner, N.W., Mark D.J., Blain, C.A., Westerink, J.J., and Leutich, R.A., 1994, “ADCIRC: An Advanced Three-Dimensional Circulation Model for Shelves, Coasts, and Estuaries Report 5:A Tropical Storm Data Base for the East and Gulf of Mexico coasts of the United States,” Technical Report DRP-92-6, August 1994, U.S. Army Engineer Waterways Experiment Station, Vicksburg, MS.

19. Scheffner, Norman W., Clausner, James E., Militello, Adele, Borgman, Leon E., Edge, Billy L., and Grace, Peter E, 1999, “Use and application of the Empirical Simulation Technique: Users Guide,” Technical Report CHL-99-21, December 1999, U.S. Army Engineer Waterways Experiment Station, Vicksburg, MS.

20. Scheffner, N.W., Carson, F.C., Rhee, J.P., Mark, D.J., 2003, “Coastal Erosion study For the Open Coast from Sabine Pass to San Luis Pass, Texas: Tropical Storm Surge Frequency Analysis,” USACE, Waterways Experiment Station, Vicksburg, MS.

21. Smith, J. M., Resio, D. T. and Zundel, A. K. (2001). “STWAVE: Steady-State Spectral Wave Model; Report 1: User’s manual for STWAVE version 3.0,” Instructional Report CHL-99-1, U.S. Army Engineer Research and Development Center, Vicksburg, MS.

22. Smith, J. M., Resio, D. T. and Vincent, C. L. (1997). “Current-induced breaking at an idealized inlet.” Proc. Coastal Dynamics ’97, ASCE, 993-1002.

23. USACE. (1984). Shore Protection Manual, 4th Ed., Coast. Engrg. Res. Ctr., vol. 1., pp. 3-84 – 3-87.

24. Westerink, J.J., Leutich, R.A., and Scheffner, N.W., 1993, “ADCIRC: An Advanced Three-Dimensional Circulation Model for Shelves, Coasts, and Estuaries Report 3: Development of a Tidal Constituent Database for the Western North Atlantic and Gulf of Mexico,” Technical Report DRP-92-6, June 1993, USACE, Waterways Experiment Station, Vicksburg, MS.

APPENDIX C

Non-Federal Sponsor Request for Termination

APPENDIX D

Cost Share Allocation

Freeport & Vicinity Hurricane/Flood Protection February 24, 2005

Cost Thru 31 January 2005

Federal Non-Federal TotalTitle/Feature Cost Cost Cost

Engineering Analysis & Design (22P0) $106,189.53 $78,831.56 $185,021.09Damian/Gallion Surveying & Mapping $137,721.80 $0.00 $137,721.80USAED, Waterways Experiment Station $5,707.64 $31,092.36 $36,800.00

Plan Formulation & Evaluation (22R0) $53,475.40 $0.00 $53,475.40

Feasibility Programs & Project Mgmt (22T0) $99,654.04 $31,691.98 $131,346.02

Additional Labor Cost-ESTIMATED $12,500.00 $0.00 $12,500.00

Cost Share Adjustment -$45,136.48 $45,136.48 $0.00

Sub-Total Feasibility Cost $370,111.93 66.46% $186,752.38 33.54% $556,864.31

Estimated Work In Kind $0.00 $183,359.55 $183,359.55

Total Feasibility Cost $370,111.93 50.00% $370,111.93 50.00% $740,223.86

Date post:	13-May-2020
Category:	Documents
Upload:	others
View:	1 times
Download:	0 times

Freeport and Vicinity, Texas Hurricane-Flood Protection ... · Freeport, Texas, Brazoria County,...

Documents