Marine experts challenge reliability of vessel strike prediction models

A new study has raised questions about current approaches to predicting the risk of vessel strikes on whales.  

Led by researchers at Heriot-Watt University in Edinburgh, Scotland, an international team of scientists compared eight models currently in place to assess and forecast the risk of ships colliding with whales. The study reveals notable inconsistencies in the results across the different models. 

Most reports of vessel collisions involve large whales but all species can be affected. Globally, strikes are under-reported and can often go undetected, particularly when they involve large vessels. Collisions can result in animals being injured or killed, and vessels may also sustain substantial damage. 

Approaches for identifying areas of high collision risk have been used by marine managers and policy makers for more than 20 years to try to understand where vessels are more likely to encounter large marine life and explore where it might be most effective to deploy management measures to reduce the risk of fatal collisions. 

Once an area is identified as being of high collision risk, steps such as slowing down vessels and moving them away from areas where there is a high likelihood of them encountering whales are just some of the ways that have been shown to be effective at reducing incidents.

For the first time, this study has considered how using different approaches for evaluating strike risk has consequences for areas identified as high risk. The significance of these findings could hold global implications for the management of vessel traffic.

Emily Hague is the lead author and a researcher from the Institute of Life and Earth Sciences at Heriot-Watt University. 

She said: “In this paper, we looked at the different methods that exist to map collision risk and found that they came up with similar, but fundamentally different, predictions. Some approaches identified the same places as being ‘high risk’, whereas other models identified quite different locations or sizes of areas.

“These methods have previously been seen as interchangeable, which we show is not necessarily the case.”

She continues: “We hope to improve transparency surrounding whale-collision risk mapping, so managers and policymakers know the limitations and strengths of each approach and can consider the best approach and type of data they need for their particular area and requirements.”

“We hope this work can contribute to the body of work that is working towards reducing ship strike to whales, and ultimately aiding their conservation.” 

The paper compares existing methods for predicting vessel strikes including the most-up-to-date techniques.

The team used the same datasets to compare the outputs of each of the different approaches. These included data on the distribution of the whales in the study area and data on the vessels that were present over a comparable period.  

Dr Lauren McWhinnie who is an Assistant Professor at Heriot-Watt University and the project lead, explains why there is a need for accurate modelling. 

“The reality is we’ve got lucky and now have a second chance with these animals after commercial hunting pushed many species to the brink of extinction,” she said.  

“Fortunately, there are many species now showing signs of recovery but our maritime sector has largely evolved without the presence of large whales in any great numbers. This has meant that we are only now starting to understand how these animals might be impacted by human activities such as shipping” 

“In the future, if we want to try and ensure the further recovery of these species, then we need to put in place measures to mitigate against modern-day risks that pose a threat to whale survival. We know that for some populations of whales, vessel strikes are one of the primary threats to their recovery, so we need to be proactive and take the steps needed to protect them.”

The authors highlight the importance of careful consideration when deciding upon which approaches are most appropriate to use for predicting vessel collision risk. These choices, the paper explains, have substantial implications on the areas identified by risk maps and could have subsequent implications for management or policy decisions arising from the results.    

Emily concludes: “It's important that we use the best available data and models so that we can better inform shipping companies how they can minimise the likelihood of colliding with whales.”

Vessel strikes are one of the greatest threats to a whale’s life. With the number of ships rapidly increasing to keep pace with global demand for goods, the issue is only set to get worse. However, there is a growing political willingness to take action to protect the oceans and its habitats, with the UN Ocean Decade underway, and the International Maritime Organisation (IMO) increasingly working with the shipping industry to ensure heavily trafficked areas do not affect large whales. 

The paper, titled: ‘Not all maps are equal: Evaluating approaches for mapping vessel collision risk to large baleen whales’, has been published today (Oct 17) in the Journal of Applied Ecology and is part of a wider four-year project being funded by the European Union Horizon 2020 Research and Innovation Programme.

It involved contributions from international researchers, scientists and organisations including the Wildlife Conservation Society in Canada, the School of Earth and Ocean Sciences at the University of Victoria, the Department of Geography, Environment, and Geomatics at the University of Ottawa, Fisheries and Oceans Canada, the Greenland Institute of Natural Resources, Transport Canada, the School of Biological Sciences and School of Geosciences at the University of Aberdeen, the National Decommissioning Centre, and the Department of Geography at the University of Victoria.

---

---