This item is only available as the following downloads:
RIP CURRENT HAZARD AS AFFECTED BY SAND BAR HEIGHT ON FLORIDA AND CUMBERLAND ISLAND BEACHES BY STEPHEN B. LEATHERMAN A thesis Submitted to the D ivision of Natural Sciences New College of Florida In partial fulfillment for the requirement for the degree Bachelor of Arts in Environmental Sciences Under the sponsorship of Dr. Sandra Gilchrist May 2013
ii ACKNOWLEDGEMENTS I would like to thank my family for their encouragement, support, and love. Thanks to my fa ther who sparked my interest in science and has helped guide me along my collegiate career. Thanks to my mother for always being there for me and for her unconditional love. Thanks to my sister for her support as well. I would also like to thank Dr. Sandra Gilchrist, my academic advisor and thesis sponsor, for being extremely helpful and supportive during my four years at New College. I would also like to thank my thesis committee Dr. Elzie McCord and Dr. Suzanne Sherm an for their improvements to my thesis.
iii Table of Contents Chapter Title Page Acknowledgements ii Table of Contents iii List of Figures and Tables iv Abstract vi Chapter 1 Introduction 1 Chapter 2 Rip Current Formation 6 Chapter 3 Wave Tank Exp eriment: Sand Bar Height vs. Rip Presence and Strength 12 Chapter 4 Field Methodology 24 Chapter 5 Results 32 Chapter 6 Discussion and Conclusion 43 References 47
iv LIST OF FIGURES AND TABLES Figures Title Page 1 Annual Rip Current Deaths 1 2 Beach Flag System 4 3 Rip Current Diagram 7 4 North Carolina Rip Current 9 5 Rip Current 10 6 University of Miami Wave Tank S chematic 14 7 Wave Tank Rip Current 14 8 a Rip Speed versus Depth above B ar 15 8b Plunge Point versus Depth above B ar 16 8c Rip Speed versus Bar H eight 21 8d Plunge Point versus Bar H eight 22 9 Beach S lope at Cumberland Island 23 10 Study Area 24 11 Cumberland Island 25 12 Lido Beach 26 13 Miami Beach 27 14 Pensacola Beach 28 15 Surveying Diagram 30 16 Pensacola Survey Data 34 17 Miami Survey Data 35 18 Lido Survey Data 36 1 9 Cumberland Survey Data 37
v Tables Title Page 1 Effect of Wave Energy and Water Level on Rip Strength 18 2 Effect of Bar Height on Rip Current Strength 19 3 Sand Bar Height Relative to Channel D epth for each B each 38 4 Lido Beach Rip Resc ues and Visitation numbers 39 5 Miami Beach Rip Rescues and Visitation numbers 39 6 Pensacola Beach R ip R escues and Visitation N umbers 40 7 Cumberland Beach R ip R escues and Visitation N umbers 40 8 Rip Rescues per One Million Visitors for each B each 40 9 National Climatic Data Center Storm D ata 41 10 Mean Significant Wave Height for each L ocation 42
vi RIP CURRENT HAZARD AS AFFECTED BY SAND BAR HEIGHT ON FLORIDA AND CUMBERLAND ISLAND BEACHES Stephen Leatherman New College of Florida, 2013 ABSTRACT Rip currents are seaward flowing currents usually caused by a break or hole in a study examines the ro le of the height of sand bars on rip current hazard on three Florida beaches (Miami Beach Miami, FL Li do Beach, Sarasota FL and Pensacola Beach, Pensacola FL ) and one Georgian beach (Cumberland Island Beach Cumberland Island, GA ) These areas were chosen for their variety in sand bar heights Sand bar heights on these beaches were measure d using surveying equi pment. When possible, a rip channel was surveyed with at least three survey lines to capture the features of the rip channel. Rip current hazard on each beach was measured with lifeguard rip current rescue statistics, drowning statistics, and wave height statistics. A positive correlation was found between sand bar height and rip current hazard. From this correlation, it is sugg ested that rip current forecasts should include sand bar height in their models to provide better accuracy. __________________________ Dr. Sandra Gilchrist
1 Chapter 1: Introduction Rip currents take approximately 100 human lives in the United States annually and probably over 500 lives globally (Brander and MacMahan, 2011). They are responsible for more deaths than hurricanes, tornadoes, sharks, and lightning strikes (Fi gure 1 ). These seaward flowing curre nts are deceptively dangerous. T hey are often seen as calm water between breaking waves, offering an inviting place to swim. F igure 1 Annual Rip Current Deaths. Weather and marine deaths averaged from 1994 to 2003 by the National Weather Service. Rip currents are responsible for more deaths than floods, tornados, lightnin g, and hurricanes. ( R etrieved f rom ripcurrents.com/watertracer, 2013).
2 Although these currents are more deadly than many natural disasters, they receive far less attention. This is because rip currents do not result in massive economic losses. Hurrican es cost the United States billions of dollars annually (National Science Board, 2007). Rip currents are localized to small sections of beach, are almost unobservabl e to the average beachgoer, and do not result in economic costs. Unfortunately, the result is that the average person knows very little about the dangers of rip currents. The nature o f beaches as local and tourist recreation sites further adds to the hazards that rip currents pose. Every year, hundreds of millions of people attend beaches all over the world. Two hundred and fifty two million people attended beaches in t he United States alone in 2012, according to the Unite d States Lifesaving Association This e normous amount of beach traffic includes not only locals, but many to urists from o ther states and countries. Many times, beach tourists live in inland areas with no be aches an d have a lack of experience and knowledge about the ocean. Also, many to urists from other countries do not speak English, and are not a ble to read beach warning signs or may be unfamiliar with the beach hazard flag system. This lack of public knowledge contributes to the high annual death toll of rip currents. Currently, the United States educates the public about rip currents through the beach hazards flag syst em (Figure 2 ) as well as online and through the weather channel. The beach hazard flag system is the method in which beach lifeguards pos t current beach conditions, using different colored flags to represent danger These flags
3 are green, yellow, red, an d purple, and are posted on top of the lifeguard stands at most U.S. beaches. Red flags are the serious hazard conditions. One red flag mean s dangerous current or high surf. Two red flags are the most serious warning, and they represent that the beach is closed for swimming. A yellow fla g means caution. T his flag allows swimming but with caution due to possible rip currents or high surf. A green flag means no danger present. Purple flags represent marine wildl ife hazard. Typically, purple flags are used to warn of jellyfish or man o wars being present ; however, swimming is still allowed. The beach hazard flag system is useful for warning beachgoers of hazards, but is not without problems. The problem with a colored flag system is t hat it is not standardized throughout the world. Other countries can and have d ifferent meanings for flag colors and tourists visiting the beach can become confused
4 Figure 2 Beach Flag System. A green flag denoted by an arrow, is di splayed at Pens acola Beach, FL, representing calm conditions on the beach warning flag system. ( Photo by Stephen B. Leatherman, 2012 ). Another way for beachgoers to learn about current beach condit ions and hazards is through local TV news and weather channels. The National Weather Service provides rip current forecasts for regional areas. In Florida, these regions are Miami, Melbourne, Jacksonville, Tallahassee and Tampa. Their forecasts are based on predictive models of rip curren ts devel oped by Lushine (1991), later modified by Lascody (1998). These models account for wind speed and direction, wave height, and tide level. Lascody (1998) modified the original model to include long period swell waves. These long period waves are the best type of waves to set up dangerous rip
5 currents, and are caused by large offshore weather events. The National Weather region. In this study, the role of sand bar h eight on rip current hazard was studied on three Florida beaches and one in Georgia. Brander and MacMahan, 2011, Bruneau et al., 2009, Kennedy and Thomas, 2004 have shown that lower tides typically result in stronger rip currents. This is because as the water level drops during low tide, the sand bar becomes higher relative to the water level. The hole (or gap) in the sand bar acts as a relief for the wat er built up landward of the bar and funnels the water seaward in the form of a strong rip current. T his effect of sand bar height on rip cu rrent presence and strength is shown in a wave tank experiment in Chapter 3. This relationship between sand bar height and rip current presence and strength bar heights can vary greatly even in the same geographic region. This is a problem because wave energies can be similar in a region, but the varying sand bar heights wil l cause the rip current hazard to be different between region s Currently, the National p forecast model do es not account for sand bar height The goal of this study is to show that sand bar height on beaches is an important variable in determining rip hazard on beach es and that it should be included in rip forecast models to give a more accurate rip forecast by beach instead of by region
6 Chapter 2: Rip Current Formation The driving force behind rip currents is longshore variation in wave height (Haller et al, 2002) This is typically caused by the refraction and diffraction of waves due to variable sand bar sizes. Other bathymetric features such as pits can cause wave refraction and diffraction as well. This causes the waves to have varying heights alongshore. When the waves break and water is pushed up on the beach, the differences in wave height will cause a differential wave set up on the beach. The backwash is concentrated in some areas due to the differential wave set up, and can develop topographic depressions, such as a hole or gap in the sand bar. After bar holes are formed, the rip current s persist in this area as the hole s in the bar s allow for relief of the wave set up. The resulting seawar d flowing current is called a bar gap rip current also known as a fixed rip (Figure 3) 0.3 to 0.8 meters per second (Brander and McMahan, 2011).
7 Figure 3 Rip Current Diagram The bar gap rip current is shown piercing through a hole in the sand bar, flowing seaward. The arrows represent the direction of water flow. ( Taken from Leatherman, 2003 ). Wave height is the most important factor in the strength of rip current s as the example, a three meter high wave is nine times as powerful as a one meter high wave. H igher, stronger waves will cause the rip currents to flow much faster. Offshore storms will usually produce the s trongest rip currents due to increased wave heights. However, these offshore storms also have longer period waves, which also increase the strength of a rip current. Long period, shore normal waves are perfect for genera ting rip
8 currents. As waves break over the sand bars perpendicular to shore, they will create large rip currents where holes in the sand bars exist. The long period of the waves will prevent allo wing the rip current to be faster. While the previously mentioned bar gap rips are the most common, there are also other types of rip currents: permanent rips, fixed rips, flash rips, traveling rips, and more. A permanent rip is a rip caused by a structu re such as a groin, jetty, pier, or storm drain. The backwash from the breaking waves is concentrated along the groin or jetty, flowing offshore as a strong rip. A storm drain will have fast flowing water shooting directly onto the beach, which will even tually create a hole in the sand bar and cause conditions for a rip current to form. A flash rip can last for a few minutes; they are created when a big set of waves break on the beach, pushing more water than normal up the beach and creating a sudden rip current. These currents do not typically result in a hole or gap in the sand bar, and they only last for a few minutes at most. A traveling rip is a rip current that can move up and down the beach, depending on wave action and direction of approach. Ev en more types of rip currents such as mega rips and swash rips exist. In addition to being the most common type of rip current, bar gap rips are also the most deceptive. Whereas a permanent rip current will have a groin or jetty to identify its location, the bar gap rips can be found anywhere along the beach with no v isual clues. In fact, as the rip current flows seaward past the sand bars, it actually
9 knocks down the height of the incoming waves to make that area appear to have calmer water compared to a djacent areas This is detrimental for the uneducated beachgoer who will typically choose the area with calmer water, unfortunately choosing the exact area of the rip curren t (Figure 4 ). Figure 4 North Carolina Rip Current. A f amily in North Caroli n a enters the water where a bar gap rip is present, due to the seemingly calm waves. A black arrow indicates the gap in the sand bars and the location of the rip current. (Photo by Stephen P. Leatherman). Bar gap rips can stay in the same location for lo ng periods of tim e. Due to the gap in the bar, these rips can persist for days or months, until the gap is filled in slowly
10 over time or by a coastal storm. Bar gap rips are strongest when waves approach perpendicular to shore (called normally incident waves). The typical rip current has a characteristic mushroom cloud shape as seen from the air (Figure 5 ), but is not as easy to see from the beach. This mushroom cl oud is darker than t he surrounding water and occurs when the fast flowing rip current causes sand or other sediment to be suspended in the current. The mushroom shape is the typical shape kwash water after traveling past the bar. Figure 5 Rip Current. The classic mushro om cloud shape of a rip current with the arrow repre senting the neck of the cu rrent and the circle showing the scale of a surfer. ( Photo from Los Angeles County Co astal Monitoring Network, 2002).
11 New studies show that rip currents flow like circulatory eddies and only 20% o f the water exits the surf zone with the other 80% circling back to the sand bars (Brander and MacMahan, 2011). Brander and MacMahan also state that this new findin g contradicts the popular swim parallel to the beach escape method for being stuck in a rip current, since 80% of the time if the swimmer does nothing he will be brought back to the safety of the sand bar. Also, if a swimmer chooses to swim parallel to the beach he may end up swimm ing against a longshore current which can distress the swimmer and pull him into the rip current Another aspect of rip currents to consider is the seasonal variation in beach berm height and sand bar height. The berm of a beach is the area of deposition of beach material caused by wave action on the backshore of the beach. The berm of the beach is a lmost horizontal. During the summer, the berm is low and wide, as the small summer waves build the berm up. During winter, strong waves from offshore storms are present. These strong waves will cut into the berm and cause severe erosion. The eroded ber m sand is moved offshore to form sand bars. Due to these waves and erosion, the berm is narrow and high during the winter and the sand bars are more prevalent.
12 Chapter 3: Wave Tank Experiment: Bar Height versus Rip Presence and Strength The height of a sand bar plays a major role in rip presence and strength. This experiment was designed to test the role of the height of the sand bar with respect to the presence and strength of a rip. Field observations have indicated that a larger bar results in stronger rip currents (Dean and Thieke, 2011), and these authors have called Rip currents have been observed to be more prevalent and stronger within a few hours of the minimum low tide based on field observations (Brander and Short, 2001). This suggests that it is not only the bar height that is important, but the height of the bar relative to water level. At low tide the relative height of the sand bar to the water le vel is increased, which thereby exerts a stronger influence on rips, everything else being equal. Variations in set up gradients created by alongshore variable wave breaking drive topographically controlled rip cu rrent circulations in nature (Bo nneton et al., 2010), but these conditions cannot be replicated in a 2 D wave tank. The role of varying bar height, tidal range, and wave energy was investigated through this simple wave tank experimentation. In one case, bar height was hel d constant while the water level was varied to simulate a changing tide. In the other case, the water level remained constant, and bar height was varied. In both sets of experiments, three wave energies were utilized to simulate higher, moderate, and low er wave conditions.
13 The University of Miami wind wave tank was utilized for the rip experiments; it measured 15m x 1m x 1m, and had a 1:12 beach slope (Figure 6 ) The model bars were lead weights with a width of 8.5 cm and a length of 39 cm; their heigh t was varied throughout the experiment by adding additional flat lead weights. A bar gap in the center of the tank simulated the case of a fixed rip channel; the gap between the bars was 22 cm (the width of the tank being one meter). The beach slope and bars were immobile, and sand was not used in order to reduce the number of variables in the experiment. Monochromatic 0.5 Hz waves were generated with a hydraulic ram which was set at varying amplitudes. Digital line scan cameras were used to record the surface elevation at three locations at a sampling rate of 500 Hz. To measure rip speed, the movement of neutrally buoyant pieces of eel grass through the bar gap were recorded with a video camera. In each test, three to six pieces of eel grass were trac ked as they moved 30 cm relative to the tape on the bottom of the tank. An average rip speed was determined from these data. Yellow fluorescent dye was used as a marker in some cases (Figure 7 ).
14 Figure 6 Univeristy of Miami Wave Tank S chematic. This schematic shows a cross section of the wave tank. Inside the wave tank is an artificial beach made out of wood, with metal sand bars attached to it. The height of the beach at the plunge point is 40 cm, and the distance from the plunge point to the bar i s 1.2m ( Taken from Leatherman et al., 2013) Figure 7 Wave Tank Rip Current Yellow fluorescent dye shows rip current formation through a break in the artificial bars with retu rn flow toward the bar (Taken from Leatherman et al., 2013 ).
15 In the first se t of experiments the bar height was held constant at 5 cm, and the water level was varied (Table 1; Figures 8a & 8 b). Five runs with differing water levels were conducted: 8.5 cm, 6.5 cm, 4.5 cm, 2.5 cm, and 0.5 cm above the bar. The plunge point position relative to the bar was also recorded. All runs had three different incident significant wave heights 8.4, 8.6 and 8.8 cm recorded at the toe of the beach slope at a single frequency (0.5 Hz), representing lower, moderate, and higher wave energies. Significant wave height is defined as the mean wave height of the highest third of the waves (abbreviated as Hs). While this range of wave heights seems small relative to total water depth of 40 cm, it resulted in significant changes in breaker posi tion ( order of 1 m; see Figure 8 b) and rip speeds. Figure 8 a Rip Speed versus Depth above B ar. This figure shows the rip speed versus depth ab ove bar for high (8.6 cm Hs) medium (8.4 cm Hs) and low (8.2 cm Hs) wave energies denoted by a triangle, square, and circle, respectively.
16 Figure 8 b Plunge Point versus Depth above B ar. This figure shows the plu n ge point position versus the dep th above the bar for changing water level experiment The high (8.6 cm Hs) medium (8.4 cm Hs) and low (8.2 cm Hs) wave energies are denoted b y a triangle, square and circle, respectively. In the second set of experiments, the water level was held constant at 8 cm above the bottom at the bar position, and the bar height was varied (Table 2). There were nine runs with the bar heights at 8 cm (0 cm above bar so that the water level was the same as the bar height), 7 cm, 6 cm, 5 cm, 4 cm, 3 cm, 2 cm, 1 cm and 0 cm (no bar). There is a clear increasing trend in the onset of occurrence of rips and their speeds as water level decreases and relative height of bar to water depth increases (Table 1, F igure 8 a). Run 1 (8.5 cm above the bar) showed no rip currents for any energy level. As the water level was decreased, rip currents formed and became stronger with
17 lower wate r levels. This supports the field observations of rip currents being strongest when the water level was near the low tide elevation (Lushine, 1991; Brander and Short, 2001). Comprehensive laboratory experiments of rip current circulation using drifters i n a wave basin (Kennedy and Thomas, 2004) also indicated that stronger velocities occurred at lower water levels.
18 Table 1. Effect of Wave Energy and Water Level on Rip Strength High Energy (8.8 cm Hs) Bar Height (cm) 5 5 5 5 5 Depth of water above bar (cm) 8.5 6.5 4.5 2.5 0.5 Percent of bar height to total water depth (%) 62.9 56.5 47. 4 33.3 9.1 Plunge Point from bar (m) 0.95 0.35 0.1 0.1 0.15 Rip Speed (cm/s) 0 7 15 19 18 Moderate Energy (8.6 cm Hs) Bar Height (cm) 5 5 5 5 5 Depth of water above bar (cm) 8.5 6.5 4 .5 2.5 0.5 Percent of bar height to total water depth (%) 62.9 56.5 47. 4 33.3 9.1 Plunge Point from bar (m) 1.45 1.1 0.35 0.1 0 Rip Speed (cm/s) 0 0 3 9 21 Low Energy (8.4 cm Hs) Bar Height (cm) 5 cm 5 5 5 5 Depth of water above bar (cm) 8.5 6.5 4.5 2.5 0.5 Percent of bar height to total water depth (%) 62.9 56.5 47. 4 33.3 9.1 Plunge Point from bar (m) 1.5 1.3 1.05 0.1 0.2 Rip Speed (cm/s) 0 0 0 9 13
19 Table 2. Effect of Bar Height on Rip Current Strength Higher Energy (8.8 cm Hs) Bar Height (cm) 8 7 6 5 4 3 2 1 0 Depth of water above bar (cm) 0 1 2 3 4 5 6 7 8 Percentage of water above bar to total water depth (%) 0 12.5 25 37.5 50 62.5 75 87.5 100 Plunge point from bar (m) 0.1 m 0.15 0.3 0.15 0.17 0.23 0.2 0.18 0.25 Rip Speed (cm/s) 34 20 19 16 9 0 0 0 0 Moderate Energy (8.6 cm Hs) Bar Height (cm) 8 7 6 5 4 3 2 1 0 Depth of water above bar (cm) 0 1 2 3 4 5 6 7 8 Percentage of water above bar to total water depth (%) 0 12.5 25 37.5 50 62.5 75 87.5 100 Plunge point from bar (m) 0.1 0.18 0.25 0.2 0.25 0.27 0.18 0.25 0.35 Rip Speed (cm/s) 23 23 19 12 11 0 0 0 0 Lower Energy (8.4 cm Hs) Bar Height (cm) 8 7 6 5 4 3 2 1 0 Depth of water above bar (cm) 0 1 2 3 4 5 6 7 8 Percentage of water above bar to total water depth (%) 0 12.5 25 37.5 50 62.5 75 87.5 100 Plunge point from bar (m) 0.15 0.21 0.11 no data 0.28 0.37 0.2 0.3 0.32 Rip Speed (cm/s) 26 15 18 10 9 0 0 0 0
20 There is a consistent spread in rip speeds as wave en ergy varies (Table 1). Figure 8 a indicates that higher wave energy relates to increased rip presence and speeds. This result is hardly surprising and agrees with field observations, including those of Bruneau et al (2009). Higher energy waves created rip currents at run 2 (6.5 cm bar), moderate energy resulted in rips on run 3 (4.5 cm bar), and rips were only generated for two cases at the lowest ener gy. Therefore, rip currents were generally stronger with higher energy waves; this relationship would have likely been more pronounced if a larger range of laboratory wave heights had been utilized. Haller and colleagues (2002) showed in their experiment al study of nearshore dynamics that higher wave heights at the bar resulted in higher rip velocities measured at the rip neck, which correlates with our findings. The results of the second set of experiments clearly show that independently varying the bar height greatly affects rip current strength (Table 2, Figure 8 c). Run 1 (8 cm bar) had strong rips for all energy levels; the rips weakened as the bar height was decreased. No rips were observed with 3 cm and lower bars. While there was a clear trend in wave energy vs. rip presence and strength in this series of experiments, the higher wave energies only resulted in an average increase of 3 cm/s in rip speed. This trend was likely related to changing breaker type plunging waves grading to spilling typ e waves with different runs as the plunge position did no t change significantly (Figure 8 d). Haller and colleagues (2002) found that the ratio of wave height at the onset of breaking to water depth was larger for plunging breakers than spilling breakers, w hich was also observed in our laboratory experimentation. When the water depth was too
21 large relative to the bar height, then the waves would move over the bar and break further landward near the shoreline. In these cases, the bar would no longer affect the nearshore dynami cs, and rip currents were not present (e.g., see cases where rip speed is zero in Table 2). Figure 8 c Rip Speed versus Bar H eight. This figure shows rip speed versus bar height for high (8.6 cm Hs) medium (8.4 cm Hs) and low (8.2 cm Hs) wave energies denoted by t he triangle, square, and circle, respectively.
22 Figure 8 d Plunge Point versus Bar H eight. This figure shows the plunge point versus bar height for high (8.6 cm Hs) medium (8.4 cm Hs) and low (8.2 cm Hs) wave energies denoted by t he triangle, square, and circle, respectively. Sand bars on beaches have greatly varying heights. For example, beaches in Palm Beach, Florida have a sand bar height of about two meters (Dean and Thieke, 2011) whereas beaches in Cumbe rland Island, Georgia have subtle sand bars of 20 30 cm heights based on personal observation (Figure 9 ). These laboratory experiments showed that rip currents are more likely to occur and be stronger on beaches with higher sand bars. Therefore rip curr ents are likely prevalent at Palm Beach, Pensacola Beach and other beaches with large bars during the appropriate wave and water level
23 conditions. The wave tank tests also suggest that beaches with minimal to no bars wou ld not be susceptible to rips. Figure 9 Beach S lope at Cumberland Island. Cumberland Island, Georgia is characterized by very fine sand, a gently sloping beach profile, and low wave activity (Photo by S. P. Leatherman). This experiment demonstrates that the ratio of bar hei ght to total depth is important for the development of rip currents, and the threshold decreases as wave energy increases. Because of the inherent limitations of the quasi two dimensional wave tank, it is likely that the bar to total depth ratio at which r ip currents form in the laboratory will not necessarily scale with wave height in the same way for field investigations.
24 Chapter 4: Field Methodology Sand bar height was experimentally shown in the previous chapter to have a great effect on rip current presence and strength The goal of this study is to show this effect through field studies which have never been previously conducted. Three beaches in Florida and one beach in Georgia were chosen as the study area (Figure 10 ) due to their variability in sand bar heights. T hese beaches are Miami Beach, Miami, FL, Li do Beach, Sarasota, FL, Pensacola Beach, Pensacola, FL, and Cumberland Island Beac h, Cumberland Island, Georgia. Figure 10 Study Area The study area of the beach surveys. Miami Beach is represented by the white triangle, Lido Beach is represented by the yellow square, Pensacola Beach is represented by the orange circle, and Cumberland Island Beach is represented by the
25 Cumberland Island Beach (Figure 11 ) is a remote beach on a barrier island off of Georgia. This beach is special in that its sand bar is very small. The beach also has a very shallow slope, and low wave energy. Due to the lack of sand bars and wave energy this beach does not experience rip currents and has no recorded rip current deaths (Dean, 2010) Cumberland Island is a recognized national seashore and is only accessible by boat, and is limited to three hundred visitors per day. There are no lifeguards present at Cumberland Island, Georgia. The surveys for Cumberland Island Beach were conducted on August 14, 2012 by Stephen B. Leatherman and Stephen P. Leatherman. Figure 11 Cumberland Island. Stephen P. Leatherman rests after surveying at Cumberland Island Beach, Georgia. ( Phot o by Stephen B. Leatherman ).
26 Lido Beach (Fi gure 12 ) is located on a barrier island off of Sarasota, FL. This beach is a very popular tourist destination, and has many visitors annually. This beach is characterized by low wave energy and small sand bars. Since Lido Beach is located on the Gulf Coast of Florida, it does not typically experience strong waves such as those found on the east coast of Florida Lido Beach experiences rip currents and rescues, and has lifeguards. The surveys for Lido Beach we re conducted on November 9, 2012 by Stephen B. Leatherman and Stephen P. Leatherman. Figure 12 Lido Beach. Stephen B. Leatherman sets up the surveying equipment in Lido Beach, Sarasota. (Photo by Stephen P. Leatherman).
27 Miami Beach (Figure 13 ) Miami, FL, is the most popular beach in Florida. Tens of millions of visitors arrive at Miami Beach annually. Miami Beach is characterized by medium sized sand bars and moderate wave energy. The surveys for Miami Beach were conducted on May 7, 2012 by S tephen B. Leatherman and Stephen P. Leatherman. Figure 13 Miami Beach. (Photo by Stephen B. Leatherman). Pensacola Beach (Figure 14 ) in Pensacola, FL, is located on the panhandle of Florida. This area is known to experience the most rip currents and s trongest wave energy (Houser et al. 2011). Rip currents at this beach are morphologically controlled by transverse ridges which refract the incident wave field to create alongshore variations in wave height the driving force behind rips. This configurat ion sets up a
28 rhythmic transverse bar and rip state, with rip currents present between the ridges (Houser, et al. 2011). Pensacola Beach also has the largest sand bars of all the study areas. The surveys for Pensacola Beach were conducted on July 17, 20 12 by Stephen B. Leatherman, Stephen P. Leatherman, and Klaus Meyer Arendt. Figure 14 Pensacola Beach. Stephen B. Leatherman takes survey measurements with Dr. Meyer Arendt positioning the meter rod. (Photo by Stephen P. Leatherman). Surveying equipment was used to measure the profile of each beach (Figure 15 ) The surveying equipment consists of a transit attached to a tripod, and a meter rod. The transit is a surveying instrument used to measure horizontal and vertical angles. The
29 transit m ust be completely level o n top of the tripod. T o do this, the transit has a hemispherical mounting plate with three levels on it that allo ws for quick leveling of the transit. The rod is a measuring device that can expand up to 5 meters tall. The transi t contains a scope that the user looks through at the rod. The scope shows three horizontal lines called the upper crosshair, center crosshair, and lower crosshair. When the transit is perfectly level, the surveyor second surveyor who is holding the rod. It is necessary to keep the scope from moving during the process. T he person with the rod must be in line with the scope for the scope to take measurements on the rod This is done by looking through the scope and using hand signals to have surveyor two come in line with the scope. Surveyor one will then take three measurements from the rod: the value of the upper, center, and lower crosshair. Using this, a distance can be calculated by subtracting the value of the lower crosshair from the value of the upper crosshair, and multiplying by one hundred. The distances for all of the surveys conducted were in meters. The location of the transit is positioned on the high water line, and the surveys are conducted in lines going offshore, perpendicular to the beach. At least three survey lines were conducted for each beach, with each line going as far offshore as possible until the deep water made it impossible to hold the rod straight. The goal of the survey lines was to capture the height of the sand bar as well as the depth of a rip current channel. This was done by conducting one survey line in the middle of a rip channel, and two survey lines on the sand bar adjacent to the rip channel. This allowed for a general mapping of the rip current channel. By surveying inside of the rip current channel, we were able to measure the depth of the
30 channel relative to the adjacent sand bar heights. This can then be compared wit h the other beaches to determine the sand bar heights of the beaches. Figure 15 Surveying Diagram (a) The upper, center, and lower cross hair are displayed as the surveyor looks through the transit. (b) The instrument is shown looking at the meter rod (called the level rod in this diagram), and the cross hairs from the transit are displayed intersecting the meter rod at the point seen through the scope of the transit. ( Taken from Cruz, 1983). Rip current hazard s were primarily assessed through the U nited States Lifesaving cs. These statistics are available online at www.usla.org, and show the number of rescues from rip currents at each beach with USLA lifeguards present. According to the USLA, 80% of all lifeg uard rescues are caused by rip currents R ip current rescue statistics for each beach were then normalized by annual visitors to be able to compare them.
31 Rip current hazard s were also assessed through the use of wave hindcast data. These data were retrieved from the Wave Information Study conducted by the US Corps of Engineers (Jensen, 2010). This study generated a wave cli matology for each area by using a computer model of the observed wind fields over an area This study provides an average of significant wave height for eac h beach and is used in the analysis of rip current hazard.
32 Chapter 5: Results The results of the beach surveys found Pensacola Beach to have the largest sand bar heights relative to chann e l depth at 0.9 meters (Figure 16 ), followed by Miam i Beach at 0.6 meters (Figure 17 ), Lido Beach at 0.35 mete rs (Figure 18 ) and Cumberland Island Beach at 0.25 meters (Figure 19 ) (Table 3 ) These results are in line with field observations at each location. At Cumberland Isla nd Beach, no rip current channel was present. Therefore, the height of the sand bar for Cumberland Island is not relative to a rip current channel, but is determined based on a projected profile. The beach profile for Lido Beach, Sarasota, appears errati c. However, this can be explained by the axis of the graph as well as the timing that the profile was conducted. The distance axis for the beach profiles range from 90 meters on Lido Beach to 150 meters on Cumberland Island. These distance axis are hori zontally squished to fit to the page, which exaggerates the vertical components of the profiles. On November 9, 2012, the beach profile survey was conducted for Lido Beach. At this time, Lido Beach was experiencing strong waves due to a tropical storm an d a recovery sand bar migrating onshore (Stephen P. Leatherman, personal observations). A recovery bar is a phenomenon that occurs after a tropical storm or hurricane rapidly erodes the sand from the beach and deposits the sand offshore as a sand bar. Af ter this sand bar is in place offshore, it will begin to migrate onshore to rejoin the berm of the beach, and is called a recovery bar The tropical storm waves can erode a beach in a day, giving it a winter profile overnight, and the summer waves can pus h the recovery bar to rejoin the
33 berm in less than a week This is very similar to the seasonal changes that occur to the sand bar and berm of a beach, except in a much smaller timeframe. Due to this, the sand bar height relative to channel depth for Lido Beach may not be well represented in this study, but the height obtained does match expectations.
34 Figure 16 Pensacola Survey Data. Pensacola Beach shows a large sand bar at around 40 meters distance with a sand bar height of 0.9 meters relative to t he channel depth The sand bar height is denoted by the vertical arrow. 2.8 2.7 2.6 2.5 2.4 2.3 2.2 2.1 2 1.9 1.8 1.7 1.6 1.5 1.4 1.3 1.2 1.1 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 20 40 60 80 100 120 140 Elevation (meters) Distance (meters) Pensacola Beach Survey Lines and Bar Height Sand Bar Rip Channel Rip Channel Sand Bar 0.9 m
35 Figure 17 Miami Survey Data. Miami Beach shows a medium sized bar at around 30 meters distance with a height of 0.6 meters relative to channel depth The sand bar height is denoted by the vertical arrow. 2.6 2.5 2.4 2.3 2.2 2.1 2 1.9 1.8 1.7 1.6 1.5 1.4 1.3 1.2 1.1 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 10 20 30 40 50 60 70 80 90 100 Elevation (meters) Distance (meters) Miami Beach Survey Lines and Bar Height Sand Bar Rip Channel Rip Channel Sand Bar 0.6 m
36 Figure 18 Lido Survey Data. Lido Beach shows a small bar located at around 39 meters distance with a height of 0.35 m compar ed to the channel depth The sand bar height is denoted by the vertical arrow. 2 1.9 1.8 1.7 1.6 1.5 1.4 1.3 1.2 1.1 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 10 20 30 40 50 60 70 80 90 Elevation (meters) Distance (meters) Lido Beach Survey Lines and Bar Height Sand Bar Rip Channel Sand Bar 0.35 m
37 Figure 19 Cumberland Survey Data. Cumberland Island Beach shows a very small and flat bar at around 105 me ters distance with a height of 0.25m with no rip channel present. The sand bar height i s denoted by the vertical arrow. 2.1 2 1.9 1.8 1.7 1.6 1.5 1.4 1.3 1.2 1.1 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 20 40 60 80 100 120 140 160 Elevation (meters) Distance (meters) Cumberland Island Beach Survey Lines and Bar Height Sand Bar Sand Bar Sand Bar 0.25 m
38 Table 3 Sand Bar Height Relative to C ha nnel Depth for each Beach Beach Sand Bar Height Relative to Channel (m) Pensacola Beach 0.9 Miami Beach 0.6 Lido Beach 0.35 Cumberland Island Beach 0.25 The rip current hazard at each beach is determined by the number of rip current rescues p er beach, per million visitors. The statistics for Miami Beach and Lido Beach were obtained through the United States Lifesaving Association (Table s 4 and 5 ). The USLA statistics for Lido Beach were obtained from the years 2004, 2005, 2007, 2008 and 2009, whereas the statistics for Miami Beach were o btained in 2011. Lido beach experienced an average of 52.8 rip current rescues in those years, with an averag e of 4,309,939 visitors (Table 4 ). Miami Beach experienced 234 rip current rescues, and had an attendance of 13,268,841 visitors (Table 5 ). Pen Between 2004 and 2009, 759 swimmers were rescued at Pensacola Beach according to Houser et al (2010). This averages out to 152 rescues per year during that time period. According to approximately 80% of all rescues are caused by rip currents. T he number of rip rescues per year can be averaged to approximately 121.6. According to th e Pensacola Beach Convention
39 The Pensacola Beach rip rescues and at tendance is displayed in table 6 Cumberland Island Beach has no recorded rip current res cues or d eaths. This beach has an annual attendance of only 45,000 visitors per year, according to the National Parks Conservation Association (2013) Cumberland Island Beach is part of Cumberland Island National Seashore, which is a national park and is only acc essible by boat, with a limit of 300 visitors per day. This leads to the low visitation number on Cumberland Island Beach Table 7 displays the rip rescues and annual attendance for Cumberland Island Beach. Table 4 Lido Be ach Rip Rescues and Visitation N umbers ( obtained from www.usla.org ). Year Attendance Total Rip Surf 2004 4,149,915 63 42 21 2005 4,321,023 49 31 18 2007 4,555,485 119 92 27 2008 4,073,065 783 57 721 2009 4,450,207 957 42 910 Average 4,309,939 394.2 52.8 339.4 Table 5 Miami Be ach Rip Rescues and Visitation N umbers ( obtained from www.usla.org ). Year Attendance Total Rip Rescues Surf 2011 13,268,841 261 234 27
40 Table 6 Pensacola Beach Rip R esc ues (Houser et al., 2010) and V isit ation N umbers (www.visitpensacola.com). Year Attendance Rip Rescues Averaged from 2004 2009 3,843,766 121.6 Ta ble 7 Cumberland Island Beach Rip Rescues (Dean, 2010) and Visitation N umbers (www.npca.org) Year Attendance Rip Rescues Each Year 45,000 (from NPCA) 0 Table 8 presents the rip rescue statistics for all beaches normalized by population. The table displays the average number of rip current rescues per year, per million visitors. Table 8 Rip Rescues per One Million Visitors for each B each Location Rip Rescues per 1,000,000 visitors Pensacola Beach 31.64 Miami Beach 17.64 Lido Beach 12.25 Cumberland Island Beach 0
41 Another source of data for assessing rip current hazard was the National Climatic Data datasets of climatic information. The NCDC has a storm dataset that show s the reported injuries and deaths caused by storm phenomena. Included in these data are the reported deaths and injuries from rip currents. This data set is not complete and is known to underreport casualties an d damage estimates (Ashley and M ote, 2005) However, it has been used by others and proven useful in determining rip current fatalities (Genisi and Ashley, 2010). occurred at Pensacola Beach, 8 at Miami B each, and none at Lido Beach or Cumberland Island (Table 9 ) Table 9 National Climatic Data Center Storm D ata This table shows national rip current deaths from 2000 2012 ( retrieved from www.ncdc.noaa.gov. ). Location Year Deaths by Rip Currents Pensacola Beach 2000 2012 21 Miami Beach 2000 2012 8 Lido Beach 2000 2012 0 Cumberland Island Beach 2000 2012 0 Additionally, average significant wave height was determined fo r all areas using the U.S. Corps of Engineers Wave Information Study This study uses previous wind speed and direction data to hindcast surf conditions. The dataset is available from 1980 to 1999, and has hundreds
42 of data stations in the Gulf of Mexico and Atlantic coast. The data displays a mean significant wave height for each station. The mean significant wave height was averaged from 1980 to 1999 for each study area using the closest data station. The result s are displayed below in table 10 Table 10 Mean Significant Wave Height for each Location. A veraged from 1980 to 1999 (Jensen, 2010). Location Average Significant wave height from 1980 1999 (meters) Cumberland Island Beach 1.02 Pensacola Beach 0.919 Miami Beach 0.893 Lido Beach 0.713
43 Chapter 6: Discussion and Conclusion The mean significant wave heights from the U.S. Corps of Engineers Wave Information Study (Jensen, 2010) showed Cumberland Island Beach to have the highest wave height at 1.02 meters, followed by Pensacola Beach at 0.91m, Miami Beach at 0.89m, and Lido Beach at 0.71m (Table 10 ) The stations used to record these data are located offshore at varying depths. This creates a problem for accurate comparison of wave heights for these locations. Rip currents are a nearshore coastal process which typically extends only hundreds of mete rs offshore at the most. However, the data stations from the US Corps of Engineers are located thousands of meters offshore. By using wind speed, direction, and fetch distance, the US Corps of Engineers can simulate w ave heights at each station. However as a wave enough water to begin to be affected by the bottom bathymetry and the wave will steepen Cumberland Island Beach is locat ed in the Georgia bight, which has the longest continental shelf on the east coast, extending almost 400 miles offshore. This propagate towards land, they st art to break due to the shallow slope of the continental shelf. This results in significantly lower wave heights nearshore than offshore, which can explain the large wave heights measured at the data station offshore of Cumberland Island Beach. Georgia also has very fine sand, and the lowest sand bar heights of all the study areas. Fine grain sand and small sand bars are indicative of lower wave energy. In fact, Cumberland
44 Island Beach was found to have a average annual wave height of only 0.6 m eters at a buoy located 15 km offshore (Dean, 2010) The bathymetry of the ocean affects this wave height data in all locations, not just in the Georgia bight. Miami has the narrowest continental shelf of all the study areas, which only exceeds a few kilomete rs offshore. This would normally allow large ocean waves to propagate to shore without being slowed down or broken, but this is not the case. Miami Beach is shielded from the large ocean waves by the Bahamas, which explains the relatively small waves on Miami Beach compared to much larger waves in Palm Beach north of Miami. Pensacola Beach and Lido Beach are both located in the Gulf of Mexico, which is known for smaller waves than the Atlantic coast. Lido Beach and Pensacola Beach both are on the very w ide continental shelf bordering the west coast of Florida. However, it is much narrower off of Pensacola Beach in the panhandle of Florida. Furthermore, Pensacola typically experiences stronger waves due to storm events. Tropical storms formed in the Ca ribbean usually head northward bringing stronger swell type waves towards the panhandle of Florida. Similarly, large frontal system rarely extend far enough south to create large waves for Lido Beach, but this is much more common at Pensacola Beach. De termining rip current hazard s for all locations depends on many factors. The most important factor in determining the strength of a rip current is the wave heights over the bars, as e nergy of the wave is pro portional to the square of the wave height (Bowe n 1969) Based on the Wave Information Study (Jensen, 2010) we can determine the wave heights for all beaches at the locations of the data stations. However, the wave heights at the data stations do not
45 reflect the actual wave heights that break over the sand bars. From field observations, it was seen that Pensacola Beach and Miami Beach had similar wave heights, with Lido Beach having significantly lower waves, and Cumberland Island Beach having the low est waves. Accounting for the width of the continental shelf at these locations, this was to be expected. The wave height data at the data stations almost reflected this observation, with the exception of Cumberland Island, GA. However, data from a buoy closer to shore measured annual wave heights of only 0.6 meters which does match expectations. High waves alone are not enough to form a rip current, as sand bars must be present for the formation of a rip. Factors that this field study analyzed in dete rmining rip current hazard was sand bar height. By traveling to each location and taking survey lines, the profiles of all the beaches were recorded. Comparing the sand bar heights by beach, it was found that Pensacola Beach had the largest bars, followe d by Miami Beach, then Lido Beach, and finally Cumberland Island Beach. The results of the data show a positive correlation between sand bar height and sto rm data shows that the number of deaths due to rip currents is highest in Pensacola Beach, accurate enough to be used in rip current fatality comparisons (Genisi a nd Ash ley, 2010). These result s also show a positive correlation between rip current deaths and sand bar height current hazard s were found to be greater on beaches w ith higher sand bars. Currently, the
46 models. This f ield study has shown that sand bar heights are i mportant factor s in determining rip current hazard on beaches, and rip current forecasts should include this factor in their models. This is very important in the case of two beaches with similar wave heights but different sand bar sizes. In these cases, rip current hazard could increase greatly on the beach with the h igher sand bars, but will not be forecasted by curre nt prediction models. Furthermore, including sand bar height in rip current models will make the forecasts specific to each beach instead of a large region. This will help in the accuracy of the forecasts, and can eventually include other beach specific factors in the models t o further increase the accuracy of the forecasts. Rip currents take hundreds of lives each year in the United States alone and account for more deaths than hurricanes, tornadoes or lightning strikes (Brander and McMahan, 2011) T here is clearly a need fo r better or more accurate rip current forecasts.
47 References Ashley, W. S., and Mote, T. L. (2005). Derecho hazards in the United States. Bull etin of the Am erican Meteorological Society 86:1577 1592. Bonneton, P., Bruneau, N., Ca stelle, B. and Marche, F. (2010). Large scale vorticity generation due to dissipating waves in the surf zone. Discrete and Continuous Dynamical Systems Series B, 13, 729 738. Bowen, A. J. (1969). Rip Currents theoretical investigations. Journal of Geophysical Research 74:5467 5478. Brander RW, MacMahan, J (2011) Future challenges for rip cur rent research and outreach. Leatherman SP, Fletemeyer J (eds) Rip currents: Beach Safety, Physical Oceanography and Wave Modeling 1 29. Brander, R.W. and Short, A.D. ( 2001 ) Flow kinematics of low energy rip current systems. Journal of Coastal Research 17, 468 481. Bruneau, N., Castelle, B., Bonneton, P., Pedreros, R., Almar, R., Bonneton, N., Bretel, P., Parisot, J., and Senechal, N. (2009) Field observations of an evolving rip current on a meso macrotidal well developed in ner bar and rip morphology. Continental Shelf Research 29: 1650 1652. Cruz, C. R. (1983). Fishpond Engineering: A technical manual for small and medium scale coastal fis h farms in Southeast Asia. Food and Agriculture Organization of the United Nations Photo retrieved from http://www.fao.org/docrep/field/003/E7171E/E7171E03.htm Dean, R. G. (2010). Conditions at Sea Camp Beach on Cumberland Island, GA on June 24, 2006 an d The Likelihood of The Presence of Rip Currents. A consulting report for the U.S. Department of Justice. Dean, R.G. and Thieke, R.J., ( 2011 ) Surf zone hazards: rip currents and waves. In: Leatherman, S.P. and Fletemeyer, J. (ed.), Rip Currents: Beach S afety, Physical Oceanography, and Wave Modeling CRC Press International, Boca Raton, Florida, p. 107 123. Genisi, V. A., and Ashley, W. S. (2010). An examination of rip current fatalities in the United States. Nat ural H azards 54:159 175. Haller, M., Dalrymple, R. and Svendsen I. (2002). Experimental study of nearshore dynamics on a barred beach with rip channels. Journal of Geophysical Research 107(C6): doi: 10.1029/2001JC000955. Issn: 0148 0227 Houser, C., Barrett, G., and Labude, D. (2011). Alongs hore variation in the rip current hazard at Pensacola Beach, Florida. Natural Hazards, 57:501 523
48 Houser, C., Caldwell, N., and Meyer Arendt, K. (2011) Rip Current Hazards a t Pensacola Beach, Florida. Rip Currents: Beach Safety, Physical Oceanography and Wave Modeling 175 197. Jensen, R. ( 2010 ) Wave information Study: US army corps of engineers. Retrieved from: http://wis.usace.army.mil/WIS_Documentation.shtml#po Kennedy, A. and Thomas, D. (2004) Drifter measurements in a laborat ory rip current. Journal of Geophysical Research 109 C08005, doi:10.1029/2003JC001927 Leatherman, S. B., Leatherman, S. P., and Haus, B. K. (2013). Bar height and rip current presence and strength. Shore and Beach 81: 19 22. Leatherman, S. P. (2003). Survival Guide: What You Need to Know about Sharks, Rip Currents and More Before Going in the Water. Yale University Press, New Haven, CT. Leatherman, S. P. ( 2011) Rip Currents terminology and pro active beach safety. Leatherman, S.P. and Fletemeyer, J (ed.), Rip Currents: Beach Safety, Physical Oceanography, and Wave Modeling CRC Press International, Boca Raton, Florida, p. 259 271. Lascody, R. L. (1998) East Central Florida Rip Current Program. National Weather Digest, 22: 25 30. Lushine J. B. (1991). A study of rip current drownings and related weather factors. National Weather Digest 16:13 19. National Science Board. ( 2007 ). Hurricane Warning: The critical need for a national hurricane initiative, NS 06 115 36 National Climatic Data Center storm data (2012). Retrieved from: http://www.ncdc.noaa.gov/stormevents/listevents.jsp?beginDate_mm=01&beginDate_dd=01 &beginDate_yyyy=2000&endDate_mm=12&endDate_dd=31&endDate_yyyy=2012&eventType =%28Z%29+Rip+Current&county=ALL&zone=ALL&submitbutton= Search&statefips=12%2CFLOR IDA# National Parks Conservation Association (2013). State of the Parks, Cumberland Island National Seashore. Retrieved from: http://www.npca.org/about us/center for park research/stateoftheparks/cumberland/CUIS web.pdf Pen sacola Beach Visitation Numbers. (2009). Pensacola Professional Reports. Retrieved from: http://www.visitpensacola.com/professional/reports United States Lifesaving Association ( 20 13 ) National Lifesaving Statistics. Retrieved from: http://www.usla.org/statist ics/public.asp.