Sharks have developed quite a notorious reputation in the eyes of the public for being dangerous predators that attack humans, even if unprovoked. LSU Department of Oceanography and Coastal Sciences Assistant Professor Stephen Midway hopes to change this view through his new study by showing that shark attack rates are low and highly variable.
While living in South Carolina in 2015, Midway watched a news report about a high number of recent shark attacks in the area. The report piqued his interest and made him wonder just how unprecedented these events were.
“It caught the attention of the news and was being sold as a very rare occurrence,” Midway said. “I realized that with the right data, we can look at changes over time and look at the shark attacks we might expect in any given area in any given year.”
Midway’s study involved conducting a statistical analysis of shark attacks in various regions of the globe using data collected from the International Shark Attack File at the University of Florida. The data spanned a 55-year period, from 1960 to 2015. Using data from such a long period of time strengthened Midway’s conclusions about the changes of shark attack rates over time.
Midway used time series models to analyze the number of shark attacks in certain areas. Unlike a standard regression model, time series models can detect several trends over an entire set of data. He also divided the globe into different regions that were relevant to certain species of sharks.
“A political jurisdiction is not always the best way to think about an ecological phenomenon,” Midway said. “We split things up to look at attacks in places populated by certain types of sharks that are responsible for most of
After finding the number of attacks that occurred in different regions, Midway adjusted the data based on the regional population to find the rate of shark attacks for each area. Midway said the raw number of shark attacks can be informative but finding the rate of shark attacks based on an estimate of the number of people in the water more accurately reflects the risk of an attack.
Midway concluded that the average rate of attack is very low, but rates vary by location. Regions with higher rates of shark attacks, including the eastern U.S., Hawaii and southern Australia, had larger regional populations and more people spending time in the water than areas with lower rates.
“Southern Australia, Hawaii and the eastern U.S. either have a lot of people living there, so a lot of people in the water, or a lot of tourists,” Midway said. “If you don’t have people going in the water, you’re not going to have shark attacks. Typically, the places where we saw the most attacks are the places with the most people in the water.”
Midway also found that as average shark attack rates increase, the average variability of that rate increases. If a remote area only has a few attacks per year, that number will vary over time, but by very small amounts. In areas with higher rates of attacks, that rate will differ by much larger numbers.
Midway believes the public perception of sharks is largely influenced by media coverage that reports on shark attacks but fails to consider the millions who aren’t attacked. While shark attacks can have serious consequences, Midway’s study proves the public has no reason to live in constant fear of an attack.
“We hope that our findings lead to an increased awareness of how rare shark attacks actually are,” Midway said. “Shark attacks are dynamic in nature and need to be taken into the context of time and place. We need to be conscientious that we’re not always at risk when we get in the water.”