Loading...
THORNTON ROAD TREE PROTECTION REVETMENT STUDY '! - ( I ........ ~ >(- Thornton Road Tree Protection Revetment Study Study performed for Kyle Associates, Arborist Consultants, Tampa, Florida Steven D. Hunt, EI Project Engineer, Investigator, Co-author August 4,1994 -~, I. ..... ........ ~...~...... ~.. .... ... . - W. ........~.. ~.... ~ - . !I\~:\\~....'IJI ..~ '; I..' .,. '\'..~ , <J"i<<"'''!i0(~~,I\' -~~~~~~:-o_,i"'~~~7it~~~i:{~~- challenger Enterprises POBox 285 Inc. safety Harbor. Florida 34695-0285 phone (813) 726-7612 voyage of HMS challenger. 11372-11376 Science and Engineering in the Traditions of the Challenger Expedition (( -Jr,- h , , (1/; ... '"':. ~-. I I !" ~'" Thorton Road Tree Protection Revetment Proposal Study Task: Detail the characteristics of the protection measures necessary to preserve specimen trees currently endangered due to shoreline erosion adjacent to Thorton Road at Tampa Bay. Evaluate the protection material characteristics necessary using different material types for cosUbenefit ratio calculations by others. The proposed protection for this site is a rubble revetment placed along the waterline area. Backfill, as necessary, may be placed behind the structure to provide protection and stability to the root structure of the trees currently threatened by erosion. This investigation estimates the size of the protection material necessary to withstand the expected wave forces. These estimates are provided for a range of storm forces corresponding to different design storm frequencies. Structure damage estimates are provided for potential excessive forces impacting the structure. Step 1. Determine design storm wave and water level criteria for estimating the forces on the proposed structure. The design wave and water level calculations needed for designing a protection structure utilize estimated deep water storm wave characteristics and local bathemetry. For the purposes of this estimate, several different design storm frequencies are used. Wind speed and direction data from the National Hurricane Center were used with a wave prediction computer program to estimate the storm wave characteristics at the site. Tables 1 and 2 show the results of the computer program. In both tables, the wind velocity Table 1 Maximum Predicted Wave Characteristics Under Average Conditions WIND STORM AVERAGE WAVE WAVE FETCH VELOCITY FREQUENCY DEPTH HEIGHT PERIOD ANGLE Knots Years Feet Feet Seconds Degrees 5 <1 12.0 0.41 1.10 110 10 <1 12.0 0.96 1.61 110 15 <1 12.5 1.47 2.13 110 20 1 12.5 1.94 2.53 110 25 1 13.0 2.38 2,81 110 30 2 13.0 2.79 3.03 110 35 4.5 13.5 3.17 3.21 110 40 5.2 13.75 3.53 3.36 110 45 6.5 14.0 3.88 3.50 110 represents the sustained rather than peak value over the area. The depth value is an average for the fetch with a factor added for wind setup or surge (nominal average depth is 12.0 feet for all fetch angles). The results of the computer program are the significant wave height, Hs, and corresponding wave period as Challenger Enterprises, Inc. Thorton Road Bank ErosionlTree Protedion Page 1 ..., ,-, '"':~ J I ~ indicated in the Tables. The fetch angle is the angle, in degrees from North (Azimuth), at which the wave height is maximum for all exposure angles at the site. This angle nominally represents the longest fetch length. Note that adjustments were not made for structures within Tampa Bay, such as bridges. For the project site, these structures would tend to reduce the wave height and therefore incident wave energy. Table 2 Maximum Predicted Wave Characteristics Under Storm Conditions WIND STORM AVERAGE WAVE WAVE FETCH VELOCITY FREQUENCY DEPTH HEIGHT PERIOD ANGLE Knots Years Feet Feet Seconds Degrees 50 8 14 4.18 3.60 110 60 14 15 4.91 3.88 110 70 23 16 5.63 4.14 110 80 48 17 6.35 4.38 110 90 65 18 7.07 4.61 110 100 100 19 7.80 4.84 110 110 125 20 8.53 5.05 110 120 200 21 9.27 5.26 110 130 300 22 10.01 5.46 110 Table 1 data represents wave characteristics which are expected from non-tropical storm events. Winter and early spring storms can account for sustained wind speeds at the top end of this Table. Table 2 data principally relates to tropical cyclone events. This data is utilized as the design parameter even though the winter storms typically are longer in duration with magnitudes nearly that of a hurricane. The reason is the wind direction typical of the winter storm events. Winter storms on Florida's west coast produce, at their peak, west to northwesterly winds. The project site is geographically shielded from these events. It is typical that in the interval leading to actual landfall of a winter storm front, that southerly winds of significance are experienced. The impact of these pre-storm conditions is less than a tropical storm or hurricane. Hurricanes typically approach from the south to southwest. With counter-clockwise winds, the primary wind direction at storm impact runs along the longest fetch area intersecting the site. For the structural design estimates, 10,25, 50 and 100 year design storm frequency data is used. For the proposed rubble mound structure, significant wave height is suitable for design calculations, so no estimate of the highest 10% or 1 % wave characteristics is necessary. In estimating the forces on a structure produced by wave action, it must first be determined if the wave breaks prior to impact. With the nearshore depths and offshore slope in the project area quite shallow (a 1 :200 average slope is indicated in the survey data), it is expected that breaking wave conditions will be encountered. The Shore Protection Manual (SPM, 1973) provides a methodology for estimating the depth of breaking for an advancing deepwater wave (section 7.121). Utilizing this method and including any wind setup or storm surge elevations, it was verified that the structure is subject to breaking wave forces throughout the range of wave data in this investigation. Challenger Enterprises, Inc. Thorton Road Bank ErosionlTree Protection Page 2 .-. . I I .. ",. Given that the structure is subject to breaking waves, the SPM provides an equa- tion (7-4) to estimate the design breaker height for a given deep water wave cli- mate. Table 3 shows the results of applying this equation to the storm return periods being investigated. These values can be used to determine the overall height of the structure nec- essary to prevent overtop- ping (neglecting wave run- up). For example, the de- sign wave height for the 100 year design storm is 6.63 feet with an estimated surge elevation of 7 feet. To prevent overtopping, the absolute structure height must exceed 7 (surge) + 6.6 (wave) or 13.6 feet. As it is not believed that overtopping is a concern, wave run-up calculations were not performed. In this location, the trees requiring protection are located at an elevation of approximately 11 feet, except for one specimen located at an elevation of 7 feet. Since placing material on top of the root structure of an existing tree can often harm the tree, the maximum design structure height is limited to 11 feet. In the area of the lower elevation specimen, the structure height must stop before this design goal. The impact ofthis is discussed in the recommended configuration discussion. A rubble structure depends on backfill for support, so placing rubble above the elevation of the backfill is not recommended. As the primary purpose of this structure is to prevent erosion from toppling the trees, overtopping is acceptable. Examining the design criteria, overtopping (including wave run-up) can be expected to occur with design storms of 50 year return period and greater. .. Table 3 Design Wave Heights DESIGN STORM FREQUENCY DEEPWATER WAVE HEIGHT DESIGN WAVE HEIGHT SURGE or SETUP Years Feet Feet Feet 10 2.5 4.42 3.76 25 4.0 5.69 4.84 50 5.2 6.43 5.47 100 7.0 7.80 6.63 Step 2. Using the wave and water level data, estimate the characteristics of the armor stone layer necessary to withstand the predicted forces. Determine the optimum cross section of the revetment and cross sectional areas of the armor layer, underlay- er and bedding material. The SPM provides an equation (7-105) for predicting the weight in pounds of an individual armor unit in the primary cover layer necessary to protect against a given impact wave condition. For this investigation, a two stone thickness with individual stone weights ranging from 0.75 to 1.25 of the estimated design weight is assumed. The estimates utilize the design wave heights (from step 1), the unit weight of the armor material, the angle of the structure in degrees from horizontal and a stability coefficient which varies with the shape and material making up the armor layer. For material unit weight, two values are used: 1 . 150 Ibs/ft3 - average for bulk limerock which is commercially available. 2. 115 Ibs/ft3 - average density of concrete rubble currently proposed for the project. Estimates of the stability coefficient vary with the material composition and configuration. For quarried Iimerock, this value is estimated between 3.0 and 4.0. For concrete rubble, the value ranges from 2.0 to 3.0. In the calculations, both ends of the stated range of values are used to establish lower and upper limits of individual stone size. In addition to a variation in stability coefficient, two slopes angles are investigated, 1 :1.5 and 1 :3. The 1 :1.5 slope is the minimum slope which can be recommended for Challenger Enterprises, Inc. Thorton Road Bank ErosionlTree Protectioll Page 3 I I ~ protection structures subject to breaking wave forces. This minimum is investigated since one of the overall project design guidelines is to minimize impact of the project on the intertidal area of the shoreline. Applying the parameters outlined to equation 7-105 results in a range of average stone size for the top armor layer of the rubble structure. The calculated significant weight of the armor stone for the given conditions) within the range of design values and wave impact forces are found in Table 4. Table 4 Unit Weight (Pounds) of Primary Armor Layer Stones UMEROCK CONCRETE RUBBLE STABlUTY COEFFICIENT ----------------r-----------------lr-------------------r----------------- 3.0 I 4.0 2.0 I 3.0 ; STRUCTURE SLOPE T I I ------- ------- I --------- ------- --------- -------- I --------- ------- Stonn 1 on 1 on 1 on 1 on 1 on 1 on 1 on 1 on Frequency 1.5 3.0 1.5 3.0 1.5 3.0 1.5 3.0 10 730 365 550 275 4,025 2,015 2,690 1,345 25 1,560 780 1,170 585 8,590 4,295 5,725 2,860 50 2,250 1,125 1,685 845 12,400 6,200 8,265 4,135 100 4,000 2,000 3,000 1,500 22,080 11,040 14,720 7,360 Reviewing Table 4, note that in addition to the predicted wave impact force, the density of the material has a dramatic affect on the unit weight of the armor stone required for protection. One reason for this sensitivity is that the density is found in both the numerator and denominator of the equation used to formulate the results presented in this table. Another reason the prediction is sensitive to density is that density is applied in relation to the density of seawater (64 Ibs/fe) in the denominator of the equation (specific gravity of the material). As the density decreases, the resistance to the wave impact forces is decreased proportional to the specific gravity of the material. Another significant result demonstrated in Table 4 is the effect of the structure slope on the unit weight. The slope is found in the denominator of the prediction equation. To estimate the thickness of the primary armor layer, it is assumed that an average individual stone is approximately cylindrical with a length roughly twice the radius. Using this approximation, a total individual unit volume and shape can be estimated. Corresponding to the unit weights found in Table 4, the volume of an individuallimerock unit ranges from 1.89 fe (0.66 ft. radius, 1.32 ft. height) to 26.7 ft3 (1.61 ft. radius, 2.32 ft. height). Concrete rubble units range from 11.7 ft3 (1.23 ft. radius, 2.46 ft. height) to 192 ft3 (3.13 ft. radius, 6.26 ft. height). Table 5 provides a complete listing for all design criteria. The recommended cross section for the protection structure consists of three layers of rubble material. Figure 7-116 ofthe SPM provides an excellent drawing ofthis configuration. The top or primary armor layer consists of a section two units thick. This double thickness is a factor in the selection of the stability coefficient discussed previously. Since the toe elevation (Le. depth) of the structure is less than one and one half times the design wave height, the primary armor layer extends all the way to the toe. The primary Challenger Enterprises, Inc. Thorton Road Bank ErosionlTree ProtectioII Page 4 I I Table 5 Average Volume and Radius of Primary Armor Layer Stones UMEROCK CONCRETE RUBBLE STABlUTY COEFFICIENT ~-------3~~-------r-------~~~-------lr--------;~---------r-------;~~------- ; STRUCTURE SLOPE ; I I ------- ------- I --------- ------- --------- -------- I --------- ------- Stonn 1 on 1 on 1 on 1 on 1 on 1 on 1 on 1 on Frequency 1.5 3.0 1.5 3.0 1.5 3.0 1.5 3.0 10 4.87 ft' 2.43tt' 3.87 tt' 1.83ft' 35.0ft' 17.5ft' 23.3ft' 11.7 ft' 0.92' 0.73' 0.84' 0.66' 1.7S' 1.41' 1.55' 1.23' 25 10.4 t' 5.20ft' 7.80ft' 190ft' 74.7ft' 37.3ft' 49.Sft' 24.9ft' 1.lS' 0.94' 1.07' O.SS' 2.28' 1.Sl' 1.99' 1.58' 50 15.0 ft' 7.5Oft' 11.2tt' 5.63ft' 107.Sft' 53.9ft' 71.9ft' 36.0ft' 1.33' 1.06' 1.21' 0.96' 2.58' 2.05' 2.25' 1.79' 100 28.7 ft' 13.3 ft' 20.0ft' 10.0ft' 192ft' 96.0ft' 128ft' 64.0ft' 1.62' 1.28' 1.47' 1.17' 3.13' 2.48' 2.73' 2.17' layer rests on two additional layers, an underlayer and a bedding layer. The thickness of the primary layer is estimated at 1.5 times the long dimension of the armor unit to account for random placement. Using these factors, the thickness of the primary layer ranges from 3.5 ft for the largest unit size and a 1: 1.5 slope to 2.0 ft for the smallest unit size and a 1:3 slope for limerock. For concrete, the same factors yield 9.4 ft for the largest unit size and a 1 :1.5 slope to 3.7 ft for the smallest unit size and a 1:3 slope. The underlayer consists of two thicknesses of stone with an average unit weight nominally 10% of the primary layer. The underlayer only extends through a portion of the overall structure height. This layer provides protection in two ways. It provides a buffer between the primary layer and the bedding material and likewise prevents the bedding material from washing through the larger voids in the primary layer. For estimation, a fixed one foot thick underlayer of 50 - 100 Ib stone is used for either choice, quarried limerock or concrete rubble, of primary armor material. The bedding layer prevents the larger armor layer material from settling into the sand foundation due to wave action. It also prevents irregularities in the armor stones from piercing a fabric soil retainer, if used. For both the Iimerock or concrete primary armor layers, the bedding consists of a 0.75 foot thick layer of 10 - 50 Ibs. material. The bedding material extends from the top of the structure down to the toe. Step 3. Using the cross section data developed in step 2 and the plan view survey of the area to be protected, estimate the total volume of each type material required for the project. Utilizing the cross section data from step 2, the volume of material required for the various combinations of slope and material characteristics can be estimated. These estimates assume 80 linear feet of shoreline are protected to the elevation of the structure. The unit weights (density of 150 Ibs./ft3 assumed) of the underlayer and bedding material are fixed and therefore subject only to the variation provided by varying the slope of the structure itself. The portion of the structure which must be varied to account for the lower elevation is not considered in these estimates. The estimates are only performed for the 50 year design Challenger Enterprises, Inc. Thorton Road Bank ErosionlTree Protection Page 5 I I storm. Calculations are proportional to the size difference of the armor layer material for other design criteria. For a 1 on 1.5 sloped structure, it is estimated that 76 tons of bedding material and 89 tons of underlayer material are necessary. These values are the same regardless of the type of material. 159 tons (average for the discussed stability coefficients) of Iimerock or 231 tons of concrete are required to protect the desired shoreline area. For a 1 on 3 sloped structure, it is estimated that 133 tons of bedding material and 157 tons of underlayer material are required. 222 tons (average for the discussed stability coefficients) of limerock or 318 tons of concrete are required to protect the desired shoreline area. Recommended Configuration: To meet the minimum purposes for which this estimate was generated, the 1 on 1.5 slope would provide adequate stability and protection for the area. and allow space for backfill which may be required to properly protect the trees in question. The 1 on 3 slope, while requiring more material due simply to the increased cross sectional area, may be less expensive overall due to the lower primary stone total weight requirement. This structure slope may not be acceptable with the restriction of no structure below the mean high water line. With these criteria, the 1 on 1.5 slope is acceptable for the protection. Use of the 50 year design storm frequency is recommended as providing economically feasible protection. Reviewing Table 6 in the next section, the impact of a 100 year storm event is only expected to cause up to 20% damage of the structure. With less maintenance after construction and minimal repair cost due to impact of a more severe storm, use of the 50 year design storm is deemed acceptable. As far as material, the individual unit size necessary for concrete rubble protection increases very rapidly with increasing design storm frequency. This size reaches a point which makes it difficult to manage as well as being aesthetically unpleasing. Use of limerock is recommended, but it is understood that economic conditions may force the use of concrete. Challenger Enterprises, Inc. Thorton Road Bank ErosionlTree Protection Page 6 I I ~ Additional Discussions 1. Estimate the damage percentage when the structure is exposed to a storm event exceeding the design parameters. The Shore Protection Manual provides methodology to estimate the percentage of a structure damaged when impacted by forces greater than those used in the structure design. Damage is defined as armor units sufficiently dislocated as to require replacement actions to restore the structure to design service levels. The methodology is based upon data collected on various rubble structures exposed to sub-design, design and super-design wave loading. Application ofthis methodology provides findings set forth in Table 6. As an example of the use of Table 6, a Iimerock structure designed to withstand a 25 year design storm impacted by a 50 year storm can expect between eight and thirteen percent damage. These findings indicate the proposed structure parameters providing economically sound protection for the risk factors of the site. Table 6 Damage Estimates in Percent Following Impact of Design Storm Waves. Top numbers are for concrete, bottom for limerock. Actual Storm Frequency Design Storm 10 25 50 100 Frequency 10 0% 17-22% 45-50% >50% 0% 15-20% 35-40% >50% 25 10-15% 30-35% 8-13% 25-30% 50 15-20% 10-15% 100 0% 0% 2. Items of special concern during construction. Construction must be undertaken with full recognition that the overall project objective is to protect the trees of the site. Particular attention must be given to the presence of the root system and the impact of construction thereon, either immediately or over the long term. All work must be planned with the arborist and is subject to his inspection, guidance and approval. Two particular soil requirements may be in conflict and such conflict must be resolved to meet tree survival criteria. These two conditions are (1) adequate backfill compaction to support the rubble structure and (2) minimal compaction of soils around the tree roots, as required, for their function. From the engineering Challenger Enterprises, Inc. Thorton Road Bank ErosionlTree ProtedioII Page 7 I I perspective (condition 1), all soils within two feet of the bedding layer (and geotextile underlayment) are to be compacted by a vibratory compactor consistent with backfill of buried pipes. Soils beyond this two foot dimension are to be hand tamped to compaction criteria established by the arborist. Given the site conditions, the recommended construction sequence is to place the geotextile over the planimetric footprint of the structure from the seaward extent to the toe of the slope and draped up over the face of the slope to the base of the trees. The bedding layer at the bottom of the structure is then to be placed over the geotextile only from the water's edge to the toe of slope. The next step is to increase the structure two feet vertically by additions of armor stone, underlayer and bedding stone leaving a gap at the upland end for placement of backfill. The geotextile material is then carried from the slope face out over the stone layer just placed in order to expose the backfill space for work. Backfill is now placed and compacted to the height of the stone. The geotextile is then transferred back to the slope face to expose the stone for placement of another two foot layer from the seaward edge (per the grade line) to the geotextile line. This process is then repeated. Once the structure reaches full height, no more vibratory compaction of the backfill is required. The remainder of the backfill is placed, tamped and landscaped under the direction of the arborist. Challenger Enterprises, Inc. Thorton Road Bank ErosionlTree Protedion Page 8 I U.S. DEPARTME~T OF COMMERCE National Oceanic and Atmospheric Administration NATIONAL WEATHER SERVICE National Hurricane Center 1320 S. Dixie Highway, Room 631 Coral Gables, Florida 33146 August 15, 1989 ITilriJ AUG 1 7 1989 Andrew M. Nicholson, PE-PLS Director of Public Works/City Engineer City of Safety Harbor Florida 750 Main Street Safety Harbor, FL 34695 Dear Mr. Nicholson: In response to your request of 1 August 1989, enclosed are 17 plots generated by the National Hurricane Center Risk Analysis Program. Although the input site location for these plots is Tampa, Florida, the statistics generated by the program are valid for your location 15 miles to the west. As we have recently acquired the equipment necessary to produce these plots in color, we have chosen to provide them to you instead of the microfilm version. Also enclosed is NOAA Technical Memorandum NWS NHC 38, by Charles J. Neumann which describes the graphical output, its derivation and correct interpretation. Sincerely, ~At~'~ Colin J. McAdie Research Meteorologist Enclosures (~~ ~~1 < ~ ~~1>. ;;~ '"\,f'''Hll.I.'l1'" .......~ TROPICAL CYCLONES (1886-1988) PASSING WITHIN 75 N.MI. OF TAMPA FL STORM MAXIMUM WIND CLOSEST STORM STORM STORM NUMBER (KTS) NEAR STORM POINT OF HEADING FORWARD INDEX DAY FOR CNTR AT CLOSEST APPROACH (DEGS) SPEED AT NUMBER STORM NAME YEAR MONTH (GMT) YEAR PT. OF APPROACH (CPA) N.Mi. AT CPA CP A (KTS) 1 9 4 4 ~4: i B 6 ~ 4 4 4 4 I ~ ij~ ! ~ 4 4 i ll:! - 3 4 4 ~ 4 5 ~ I 4 4 ~ ~8 4 4 ~ ~ 4 In 1 10 ~ ~ 4 ~~ 4 1 ~':I ~ ~ 4 IE 4 4 1 . A 1 4 1 1 : 4 .4 ~ .0 ~ 4 I :~ ~ 9 ~ 4 I 4 II!I 4 4 -, 4 A 4 ~ ~ 4 4 ~ IE !~ 1 :i AM 4 IE 4 O~ II-! 4 ~1 1 ~1 4 N A 1 A ~ 4 II i~ ,I ~ 4 ~~: t I 4 2~ ~ ~~ 14:~ CHART 1 i05W 40N i00W 95W 90W B5W BOW 75W 70W 65W 60W 55W 40N 25N 11 1/ b ~'~ ~I~{~~/- " . ~ 1/ V/' 'fh- ~~ 4 " /. / '" 'I, ---A~~. ~ u::_ d--~~~-X. I \1\ )<J~~' ~ y~~-' V-~~ 35N ~~\ J ~/J '7 ' 'U/ "" _I~ (\ Vf}/ ~J:' ~/~ ____-~~/ I / \. '. ~ ~ ~ r{~ 'v, If/ //1\ /1 ....,,)_./ ~f .__ ",,;;::"- ~ ~~,-' ~ ~f..-7 -- // 30N .Y -...... ~ ., - _____ / ___/ .. ~ _.L---- Y ---z ~~ ~ / I ~~~ ---------- '- ~ ~ I ~'l' l{; I :,r~~ L~- ~ ,/ (\. ~~ ,-~~ "I--- 20 N \ { --------A ~ I ~ 7"<>-""" "'" ~ ~ 20 N ~ ~ --./ j r\1~~~~ ~~~~ -____~ ~ \\f\~~,\~~"'~~,-~ ____ ............ ~ ...... -.......... . ~ . ~~ "~I---. '-... ~~ ~ "- }, __ ~ --=~________ 15N I\~ ( , (~~~ ~ - 35N 1- 30N 25N -, i5N N=55 1 iON i05W i00W 95W 90W B5W BOW 75W 70W 65W TROPICAL CYCLONES PASSING WITHIN 75 N.MI. OF TAMPA, FL iON 60W 55W 1886-1988 CHART 2 (SITE LOCATION MOVED TO 28.0N, 82.9W) 105W 100W 95W 40N 1 I I + I B5W BOW 90W 75W 70W 65W 60W t 35N' 1 + ~ 1 1 30N: I , + i T t ---- /~ / 55W 40N 35N / / c 30N ~ // 25N; t i' 1 i i 20N' i5N: 't ! r---.-.. · I -')3 I I ,--r: iON i05W 100W 95W 90W B5W BOW 75W 70W HURRICANES PASSING WITHIN 75 N.MI. OF TAMPA, 25N 20N i5N -, r---'..~ .;wc I - 10N 65W 60W 55W FL 1886-1988 (SITE LOCATION MOVED TO 28.0N, 82.9Wl CfJtRT 3 SUMMARY FOR HURRICANES: NUMBER OF YEARS: 103 NUMBER OF HURRICANES: 20 MEAN NUMBER OF OCCURRENCES PER YEA~ .194 MEAN RECURRENCE INTERVA~ 5.2 YEARS SUMMARY FOR HURRICANES AND TROPICAL STORMS: NUMBER OF YEARS: 103 NUMBER OF HURRICANES AND TROPICAL STORMS: 55 MEAN NUMBER OF OCCURRENCES PER YE~ .534 MEAN RECURRENCE INTERV A~ 1.9 YEAR I = HURRICANE I"l ;1 Ii = TROPICAL STOR~l u -- 3 3 2 ~ n 2 II II I, 1 ~ ~~ I ~ ~I'G rr n~ 1 II I'!. I II .il' I III _. I 1II1 I ill I II I I'! ", I I I! I! 0 0 C> In C> In C> In C> In C> In C> In C> In C> In C> In C> In en en C> C> .... .... N N C") C") "<:t "<:t In In lO lO r- r- eo eo eo eo en en en en en en en en en en en en en en en en en en .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... TROPICAL CYCLONES PASSING WITHIN 75 N.MI. OF TAMPA, FL 1886-1988 (Note: Storm intensity is determined at time of closest approach to site) CHART 4 TROPICAL CYCLONES PASSING WITHIN 75 N. MI. OF TAMPA, FL TOTAL NUMBER OF STORMS: 55 NUMBER OF HURRICANES: 20 NUMBER OF TROPICAL STORMS: 35 YEARS INCLUDED: 1B86 - 1988 MEDIAN OCCURRENCE DATE (SEP 9) en 6 LiJ U GJ a: a: 5 :::l U U Cl lL. Cl 4 a: LiJ co ::E :::l :z: 3 2 N ~ 0 ,... .... 'q' CD 'q' CD .... It"l en CO> ,... 0 'q' CD N It"l en N Ie 0 'q' re .... I!J en CO> N .... .... .... N .... N .... N N .... .... CO> .... .... N ~ Z ~ ! c: ;"i c: c: c: >- >- z z ~ ~ (!) !5 0.. ~ En I-- I-- > > u U <C <C ~ ~ ~ <C <C ~ ~ 1i w u u !2 0 ~ w 7 7 :E =f :E =f <C en en 'i' ~ 1" ? , I I .l, I clJ I , c!. I I ...., It"l g] Ie N Ie en ,... .... 'q' CD Ie 0 CO> ,... 'q' CD It"l en ,... .... .... N .... N N N .... .... CO> .... N .... N N .... .... ~ ~ ~ ~ ~ ~ ;"i c: c: >- >- ~ z ~ ~ ~ t!) (!) ~ En I-- I-- > > u U 0.. 0.. <C <C => => => u u 0 0 w ~ J J J :E <C <C :E :E J <C <C en 0 0 z z 0 DATE WHEN STORM NEAREST TO SITE CHART 5 6 .... 5 4 3 2 1 _I o 40 35 30 I NNE NE I ENE E ESE 5E S~- ~~ en w 25 en <( u LL 0 20 ~ :z w u c: 15 w a.. 10 5 0 " DIRECTION DISTRIBUTION FOR STORMS PASSING WITHIN 75 N.MI. OF TAMPA, FL NUMBER OF CASES. 1886-19Ba ALL STORMS/HURRICANES/TROPICAL STORMS: 55/20/35 RESUlTANT DIRECTIO~ ALL STORMS!HURRICANES/TROPICAL STORMS: 002/360/004 RESULTANT SPEED (KTS): ALL STORMS/HURRICANES/TROPICAL STORMS: 8,0/9.1/7.3 MEAN SPEED (KTS): ALL STORMS/HURRICANES/TROPICAL STORMS: 11.8/11.8/11.7 ~ m TROPICAL STORMS ~ = HURRICANES - (DIReCTION rRQt-() iI' 5 55li I 5h' ft'SIi I k' Ii~~'ri Nk' NS'r: N t . '-"" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" '" C> .... N l"l ... Ln le ,... !e Rl C> .... 7 l"l ... Ln C> .... N l"l .... Ln '" ,... a:> '" C> .... N l"l ... Ln '" ,... a:> '" N N N N N N N l"l l"l l"l .... 7 'T 'T C> 'T C> C> 'T C> C> 'T .... .... .... .... .... .... .... .... .... .... I I , I , , , , I I , , I I , I I I , I I , I , , , I , I C> C> C> C> ." C> C> C> C> C> C> 2 C> C> C> C> C> C> C> C> C> C> C> C> C> C> C> 2 C> C> C> C> C> C> C> C> C> .... N ~ ... Ln '" ,... a:> gj C> N .... ... Ln '" .... N l"l .... Ln '" ,... a:> '" C> N .... .... Ln '" ,... a:> '" N N N N N N N N l"l .... l"l .... .... .... l"l C> C> C> C> C> C> C> C> C> .... .... .... .... .... .... .... .... .... .... DIRECTION (DEGS) TOWARD WHICH STORMS ARE MOVING CHART 6 NUMBER OF STORMS (1886-1988) PASSING WITHIN SPECIFIED DISTANCES FROM TAMPA, FL 60 Y Q A + ax + CX2 ./" A c .1687 50 a Q .4504 C .. .0047 - C 40 en :::E a: . 0 . l- . en 30 . . . . lJ.. . 0 . a: AYEl\ASE UJ co Rl<\i FOfl :::E :=J SITE :z: 20 I I ~I . . . r. . . . . 10 ..... . --"" ",' . . . . . R = .995 'I:( 0 0 10 20 30 40 50 60 70 80 DISTANCE IN NAUTICAL MILES (X) CHART 7 40 35 30 E en 25 lJJ en <( u u. 20 0 t- :z: lJJ u 15 a: lJJ a. 10 5 '. WEIBULL DISTRIBUTION FITTED TO HISTOGRAM OF OBSERVED MAXIMUM WINDS FOR TROPICAL CYCLONES PASSING WITHIN 75 N.MI. OF TAMPA, FL -- y w .. (CA/a) (X-34)A-i EXP [- (X-3<)A /s1 A .. 1.07167 B" 35.09151 C .. 1000.00 N .. 55 MEAN = 61. 0 X Sx = 22. 3 o .~ MEAN ! PERIOD OF RECORD: 1886 - 1988 34 44 54 64 74 84 94 104 114 124 134 144 154 164 174 184 194 204 MAXIMUM WIND (KNOTS) NEAR STORM CENTER (X) CHART 8 ,. SITE: TAMPA. FL 500 400 300 200 150 -, en 100 a: 80 <t W >- 60 50 Cl 40 0 H a: 30 w n.. 25 ::z:: 20 a: :::J 15 I- W a: 10 ::z:: 8 <t W ::::E 6 5 4 --. 3 RETURN PERIOD Of TROPICAL CYCLONES HAVING AT lEAST SPECIFIED INTENSm (NEAR STORM CENTER. NOT NECESSARILY AT SITE) PI.SSING wrrnIN SPECIFIED 2 DISTANCE (SLOPING SOLID LINES IN N."'I.l FilO'" SITE. OASfED LINE GIVES ESTI"'" TED RETURN PERIOD Of SPECIFIED WINOS AT SITE nSELF. 1 195 (KTS) 35 55 75 95 115 135 155 175 MAXIMUM SUSTAINED WIND (i-Minute Average) CHART 9 .. PROBABILITY OF AT LEAST X TROPICAL CYCLONES (~4 KNOTS) PASSING WITHIN 75 N.MI. OF SPECIFIED SITE OVER N CONSECUTIVE YEARS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 100 0 90 10 ..... en UJ 80 20 u :z UJ en a: UJ a: u ~ 70 30 :z u UJ u a: e a: :::l x u 60 40 u l- e en < e UJ :z ...J 50 50 u. l- e < u. E e E 40 60 >- l- I-< ...J >- I-< I- 30 .534 SlY CD I-< M . 70 < ...J CD H TAMPA. FL e CD a: < 0- CD 20 e 80 -, a: LE6END 0- 1~ x= 2~ 10 3~ 90 4~ 5~ 0 100 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 NUMBER OF CONSECUTIVE YEARS (N) CHART 10 100 90 en UJ 80 u z UJ a: a: ::l 70 u u 0 X I- 60 en -< UJ -J I- 50 -< u.. 0 E LO >- I- 30 I-< -J I-< CD -< CD 20 0 a: a.. 10 0 PROBABILm OF AT LEAST X TROPICAL CYCLONES (~4 KNOTS) PASSING WITHIN 75 N.MI. OF SPECIFIED SITE OVER N CONSECUTIVE YEARS 1 2 3 5 7 9 10 11 12 13 14 15 16 17 18 19 20 4 6 8 . . . . , . , . M. .223 H/Y TAMPA, FL LEGEND -' x= 2 1~~1 3~ 54~ ;::.. ~ ~ - J . ~mi j L If 1[ . -. , , 1 2 3 5 7 9 10 11 12 13 14 15 16 17 18 19 20 4 6 8 NUMBER OF CONSECUTIVE YEARS (N) CHART 11 0 10 - 20 en UJ u 30 z UJ a: a: ::l u 40 u 0 0 z 50 u.. 0 E 60 >- l- I-< -J I-< 70 CD -< CD 0 a: a.. 80 - 90 100 E en UJ en < u u. o ~ UJ u ffi c.. :z .... >- ~ .... -l .... lIJ < lIJ o a: c.. -. ..., .. ':1 .. 15 GAMMA DISTRIBUTION OF STORM TRANSLATIONAL SPEEDS (STORM INTENSITY ~34 KTS) SITE: TAMPA, FL - 12 A = 4.2883 B = 2.2230 C = .3913.10-01 MEANx = 11.76 Sx= 5.11 MODE = 9.53 N = 55 9 - 6 Y=CXAEXP (-X/B) 3 o o 50 60 30 40 10 20 STORM FORWARD SPEED IN KNOTS (X) CHART 12 12 E en UJ en -< to) LL. 9 0 ~ UJ u ffi ~ z 6 .... >- ~ .... -I .... CD -< CD 0 c: ~ 3 " '.0-, .. 15 GAMMA DISTRIBUTION OF STORM TRANSLATIONAL SPEEDS (STORM INTENSITY ~64 KTS) SITE: TAMPA. FL - A = 5.0264 B = 1. 9887 C = .1264 '10-01 MEAN X = 11. 98 Sx = 4.88 MODE = 10.00 N = 23 y=cx AEXP (-X/B) _1 o o 10 20 30 40 50 60 STORM FORWARD SPEED IN KNOTS (Xl CHART 13 ., III '1 .. GEOGRAPHICAL DISTRIBUTION OF STORM FREQUENCY IN VICINITY OF TAMPA, FL S5W sow 25N .... 30N 30N 25N .... S5W sow NUMBER OF TROPICAL STORMS AND HURRICANES PER 100-YEARS EXPECTED TO PASS WITHIN 75 NAUTICAL MILES OF ANY POINT. ANALYSIS IS BASED ON PERIOD OF RECORD 1886 - 1988. ISOLINES ARE AT INTERVALS OF 2 STORMS. CHART 14-01 " .. .,1 . GEOGRAPHICAL DISTRIBUTION OF STORM FREQUENCY IN VICINITY OF TAMPA, FL BOW '-1" j--r---'---- -- --" 30 N I r- i i r I i fl f- I I I I 25N r-' I I i I I i r 2na. 26 L-L-- J_____L__J B5W ~ 20 i . . .___. . --I ~.~J__:L~~J BOW - NUMBER OF HURRICANES PER 100 YEARS EXPECTED TO PASS WITHIN 75 NAUTICAL MILES OF ANY POINT. ANALYSIS IS BASED ON PERIOD OF RECORD 1886 - 1988. ISOLINES ARE AT INTERVALS OF 2 STORMS. eHART 15-01 .. l I " Wi "'l' f. TROPICAL STORM & HURRICANE MOTION (1886-1988) NEAR TAMPA. FL ARROWS DEPICT AVERAGE VECTOR HEADING OF STORMS B5W BOW 1?oJ1 1~1 12.d 11/1 1~1 111' 11-'/ 1c;d Fvf 14:1 7'15 /'15 /,15 114 7'15 )"16 7'16 7'16 J 16 7'Hi 1"~ 10 1~1 1v1 ad 1zA 1M 1~ ad 12/1 1~" 7'12 7'14 ('15 115 7'.15 7'"16 J16 17 /,15 J15 ~'15 30N 25N - .... NUMBERS ADJACENT TO DIRECTION ARROWS GIVE AVERAGE VECTOR AND SCALAR STORM SPEEDS IN KNOTS. VECTOR SPEEDS ARE <= SCALAR SPEEDS. CHART 16-01 30N 25N ~ . I, ,'" " ,- '... HURRICANE MOTION (1886-1988) NEAR TAMPA. FL ARROWS DEPICT AVERAGE VECTOR HEADING OF STORMS B5W BOW ... i ; i -j I ,",,10 r C~ - B5W BOW NUMBERS ADJACENT TO DIRECTION ARROWS GIVE AVERAGE VECTOR AND SCALAR STORM SPEEDS IN KNOTS. VECTOR SPEEDS ARE <= SCALAR SPEEDS. ASTERISK (*) INDICATES < 5 STORMS. eHAAT 17-01