Table 1. Rip current risk after Verbeek
.
By the information derived from literature, an overview of the literature concerning forecast systems for rip currents is presented.
The amount of literature addressing swimmer's safety in relation to the presence of rip currents is very limited, even though various authors estimate that approximately 70 - 95% of drownings and surf rescues are rip current related. The available papers focus mainly on the Australian and US coast. Rip current issues have received more attention in these countries than elsewhere, probably because of the large number of beach visitors in combination with a high energetic coastline. Furthermore, some documented research on this topic was found for the UK, the Balearic Islands, Korea and New Zealand.
Worldwide, forecasting of rip currents has received most attention in the US. This all started with the research undertaken by Lushine (1991) in Florida. In East Florida there are an average of 21 drownings per year, which is larger than the number of deaths due to tornados, thunderstorms, lightning, and hurricanes combined. Lushine developed a predictive index for rip currents; the Lushine Rip Current Scale (LURCS), which was founded on statistical analysis of lifeguard logs, newspapers and medical records of rip current drownings. The correlation between the drownings and wind and sea conditions was investigated and correlations captured in a forecast system. The empirical forecast model placed the greatest weight on moderate to strong onshore winds and also included swell height and timing of low water.
Lascody (1998) continued the work of Lushine by reanalyzing 10 years of rescue data. This showed that 75% of the rescues could be related to long period swells, and that long period swell (>12s) were always associated with a greater risk of rip currents even for low wave heights. Therefore, he adapted the LURCS model to account for the greater impact of swells. The adapted LURCS checklist to calculate the daily rip current threat includes the following factors:
Based on the adapted LURCS scale, rip current statements were issued, stating the rip current threat to be normal, greater or much greater than normal. As more than 50% of drownings occur on weekends or holidays, the thresholds for issuing a warning was set lower at these days to account for the increased number of visitors.
Engle et al. (2002) took the LURCS scale even further and included new predictive factors to improve the system, based on analysis of rescue data from Daytona Beach, Florida. They found that the frequency of rip current rescues increased significantly with shore-normal wave incidence, mid to low tidal stage, deep water wave height between 0.5-1m, wave period between 8-10s.
Rip current intensity will go up with increasing wave height and period, but as the surf conditions become heavier, the number of swimmers is likely to reduce. This explains the peak of rip current rescues at a certain wave height/period.
Based on their findings, Engle et al. included an improved tidal factor and a wave direction factor in the LURCS model and eliminated the wind factor.
Gensini (2009) argued that although the LURCS system is a useful tool to indicate the threat of rip current formation, it does not take into account certain physical variables (e.g. cloud cover or precipitation) and societal factors (e.g. weekends and holidays). To investigate the role of these factors, he constructed a national database of rip current fatalities throughout the US. Next, he performed a geographic analysis of rip current fatalities and tried to establish the relationship between meteorological and demographic factors and number of drownings.
It appears that NOAA / NWS is the only national governmental institute that issues rip current warnings at present. Read more...
Australia has a high energetic shoreline and knows about 10,000 rescues per year of which 50 drownings (figures for 1990s). Short and Hogan (1994) stated that at the time of writing, identification and quantification of beach hazards received very limited attention and public and media awareness was very poor. They identified physical beach characteristics that constitute hazards to people, which consist of:
An analysis of the public safety of beaches in NSW was made by establishing a beach safety rating, which is primarily based on classification of the beach environment on the six beach states defined by Wright and Short (1983). This classification is based on the dimensionless number to differs from a dissipative, intermediate and reflective beach state. Furthermore, wave conditions are included in the beach safety rating. The morphodynamic factors considered as potential hazards and thus contribute to the beach safety rating are overall water depth and variability; size of breaking waves; prevalence and intensity of rips; existence of longshore trough and currents; occurrence of wave setup and set down. Extra factors which are not explicitely included in the rating but should be considered are non-uniform factors, such as wave height variation, headlands, stage of tide and wind. It should be noted that the dissipative and longshore bar-trough beach states are not present in NSW, so that the safety rating of these two classes seems not to be based directly on Wright and Short's data analysis.
Short and Hogan state that furthermore, the safety of beachgoers depends strongly on the swim ability of the individual bather (after Power, 1992). A person's swim velocity ranges from 0.2 to 1.5 m/s depending on its fitness. Mapping the variation in swim ability on surf zone conditions shows a large variability in safety zones for different swimmers on a single stretch of beach.
Short and Hogan conclude that beach hazards related to hydrodynamic conditions can be accurately identified in time and space, as well as their impact on a range of bather types. This allows proactive beach safety management, in contrast to reactive (lifeguards) measures.
Surf Life Saving Australia (SLSA) released an iPhone app in 2010, Beachsafe, which provides real-time information about the weather, surf conditions and potential hazards, which the public should be aware of. The app provides information to keep the public informed and stay safe while visiting any of the 11726 beaches on the Australian coast and includes a rip current warning. Read more...
The Balearic Islands have intense tourism and in many beaches, beach user safety is an issue. Beaches are patrolled by lifeguards approximatly 5 months per year. To assist lifeguards, a pilot operational forecasting system (FS) was developed for Cala Millor beach (Alvarez et al., 2010). The system involved numerical forecasting of waves and rip currents with a 36hr forecasting horizon. Maps with beach condition information (waves and currents) were updated automatically to an ftp-site and were used to manage the risk at the beach. Bathymetric surveys had to be carried out every 2 months to update the model. Based on visual validation, the model accuracy was found to be high. However, the system was computationally expensive and the lack of bathymetrical information hampered the transfer of the system to other beaches. Therefore, Alvarez et al (2009) developed an alternative methodology; the Hazard Alert System (HAS).
The HAS forecasting system is based on a nearshore wave database. Authority is used as offshore data source, which issues deep water forecast twice daily. Nearshore wave conditions at 10m depth for the next 36 hours are derived by applying the transformation matrix.
The local beach hazard level is subsequently determined based on the nearshore wave conditions. Hazardous wave directions are defined based on the incident wave angle around the normal of the beach. The hazardous angles of approach differ per beach, and to determine these, the morphodynamic beach state (based on lifeguards descriptions + aerial photographs + sediment size) is taken into account. Furthermore, the hazard level is based on wave height.
A study by Scott et al. (2007) focuses on the coast of Devon and Cornwall, which receive 10 million visitors per year and has 62 beaches guarded by the Royal National Lifeguarding Institute (RNLI). The region is known for a very large tidal range and ocean swell conditions.
Scott et al. constructed a database containing RNLI logged incidents, daily weather, sea conditions and beach population for the 2005 summer season. Analysis of the incidents shows that in 71% of the rescued individuals, rip currents played a role. Other causes for rescues included offshore winds, sandbars, bed return flows, large waves, strong winds, tidal cut-off and plunging waves. These factors all contributed for less than 15% to the total number of rescues (multiple causes possible).
From the analysis, it became clear that the type and level of hazards varied significantly with location and geology of the beach and the hydrodynamic conditions.
The risks were categorized per beach type (following Wright and Short's classification):
As a conclusion of the study, rip currents form the greatest threat to beach user, but several hazards may work simultaneously. Significant variation in beach hazards and severity exists depending on the nature of the hydrodynamic conditions and beach type. The large tidal range introduces additional hazards and enhances rip current velocities on the ebbing tide.
In 2010, the Coastal Processes group at the University of Plymouth has been awarded a research grant to bring further the understanding of macrotidal rip current dynamics and the implications of these surf zone currents for beach safety. The Dynamics of Rip Currents and Implications for Beach Safety (DRIBS) project is supported by the Royal National Lifeboat Institution, and the work in progress aims at:
In the Netherlands, rip currents have received limited attention although per year a handful of people drown after being caught in a rip. One local initiative is known to predict rip current threats at Egmond (W. Verbeek; published on his website www.muien.nl).
In the project Mui Monitoring Project MMP-02, Verbeek (2006) tries to establish a relationship between rip current threat and environmental conditions at Egmond, making use of lifeguard records, wave data and local Argus video images of the surf zone. He found a relation between rip current risk and wave height, that he categorized as shown in Table 1. No other significant relationships could be found, although it should be noted that his analysis was based on a very limited number of observations only.
Table 1. Rip current risk after Verbeek
A rip current measurement and prediction system was set up recently (Kim, 2011). The system is based on the SaDEM model (Sand Deposition and Erosion Model), including the computation of wave deformation, nearshore wave-induced circulations and morphological change. The first model results are presented in Kim (2011); however, the presented model verification is still insufficient to draw conclusions on the system's performance.