Would you drive into floodwater?

New research is delving into why people enter floodwater and what can be done to change this risky behaviour. 

By Radhiya Fanham from the Bushfire and Natural Hazards CRC. This article was first published in Issue Four 2020 of Fire Australia.

It is widely agreed that entering floodwater, whether by foot or by vehicle, is highly risky. There is usually little to no visibility of what is underneath the surface of a flooded road, and even seemingly shallow, moderate flowing waters can sweep a person off their feet or cause a vehicle to lose control. 

Globally, floods are the highest cause of fatalities from natural hazards, and the second highest in Australia, so it makes sense that plenty of flood risk communication campaigns have been instituted over the years. The question is: are they effective?  

Associate Professor Melanie Taylor, Occupational Psychologist at Macquarie University and lead researcher of the Bushfire and Natural Hazards CRC Flood risk communication project, said while previous research has identified the reasons behind flood fatalities, not enough research has investigated the behavioural aspect of entering floodwater.

The project commenced in July 2017 to develop an understanding of the motivations, beliefs, decision-making processes and information needs of at-risk groups for flood fatalities. State Emergency Services (SES) from around Australia have worked closely with the research team from the onset of the project, taking particular interest in the behaviours of their own personnel when it comes to risk taking and driving into floodwater, as well as their attitudes to the public entering into floodwater.

“There has been a lot of work generally around drowning, and some work on driving into floodwater, but it’s mostly been qualitative,” A/Prof Taylor said.

“We wanted to do something a bit more quantitative to provide some more compelling statistics in terms of public behaviour and also SES behaviour.”

Community Capability Coordinator at the NSW SES, Mr Joshua McLaren, said the changing climate requires new and improved communication strategies.

“What we’ve started to understand in a changing climate, and what we’re seeing now going into a La Niña, is that we really need to shift gears to think about how we change the paradigm around how we operate with communities," Mr McLaren said.

"If we have really effective communication with members of our community, we can potentially save thousands of people through one action."

Defining the message

Having a clear definition of ‘floodwater’ is essential when conducting research on behaviours around floodwater and potential responses to flood risk communication. In the early stages of the project, the research team worked in consultation with the NSW SES to develop a working definition of floodwater on roads to provide a clear idea of what is being referred to in driving into floodwater messaging. This was defined as an environment where:

• water is across the road surface
• there is little to no visibility of the road surface marking under the water (i.e. uncertainty of the road quality/integrity and possibly depth)
• there is water on normally dry land - either flowing or still.

"It's not rocket science," A/Prof Taylor said. “You might think you know what it’s like underneath the water, but it could have been eroded away or potholes may have formed, and you don’t know the integrity of the road surface. Of course, the really dangerous aspect is that you could just get swept off.” 

As a psychologist, A/Prof Taylor considered human nature in the reasons why one might enter floodwater despite the clear warnings and communications against it. One of the possible hypotheses, she said, was people are just not engaging with the messaging and that is why they are continuing to enter floodwater.

“We have looked at how both SES personnel and the public conceptualise floods to try to gain insights into why people might be rejecting or dismissing the messaging,” A/Prof Taylor explained.

“If someone sees an advert that shows deep, flowing water that’s clearly dangerous, that’s the image they have of what floodwater is in the context of not to drive through it. However, when they come across water that looks quite benign, they might not see that as ‘floodwater’ and, therefore, feel the message doesn’t apply.”

Mr McLaren said communicating the risk of driving into floodwater is a particular challenge.

“Disasters aren’t something that people deal with everyday,” he said.

“The challenge for emergency services, and this is where Mel and the team’s work is so important, is how do we effectively engage people when they are so busy with their own lives, and how do we utilise communication and messaging to engage people who have never experienced the risk?”

Behaviour

To look at the extent to which entering floodwater is influenced by behaviour, the research team consisting of A/Prof Taylor, Dr Katharine Haynes and Dr Matalena Tofa conducted extensive survey and quantitative research to investigate the conditions in which people drive through floodwater and the decision-making around what was being thought at the time and why they chose to go through instead of turning around.

A public survey was distributed between December 2018 and January 2019 and was constructed to be proportionally representative of the adult Australian general population by state and balanced for age and gender. The survey consisted of eight main sections: driving details; demographics; experiences of entering floodwater, either on land or in flooded rivers; willingness to drive through water on roads; experience of driving into floodwater; experience of turning around in floodwater; general attitude to risks; and flood risk messages.

Over half of the 2,184 respondents who undertook the survey had driven through or been driven through floodwater, and 41% had been through floodwater on more than one occasion in the past five years.

The central question is why is it that, despite advice and warnings, people are still entering floodwater? What are they doing at the time, and why are they disregarding risk messages? By understanding what people are doing when entering floodwater, we can find additional levers to influence their behaviour.

Of the 1,167 people who reported driving into floodwater in the survey, a fifth were returning home from work, and a further 17% were either on holiday, sightseeing or on a leisure drive (Figure 1). The vast majority of participants  (90.7%) reported that they drove through floodwater without any negative consequences.

“It again speaks to behaviour that people enter floodwater fairly often,” A/Prof Taylor said.

“If you do it and get away with it, there’s this idea that you don’t have to be cautious the second and third time, especially if it’s the same location. However, it can be quite a different situation every time and there could be a lot of danger underneath in terms of obstructions. You wouldn’t drive into a bushfire, so why drive into floodwater?”

The majority of contexts in which people drove into floodwater related to common or mundane situations or activities and were therefore not about ‘urgent’ or high-stakes situations, and many people indicated they had carefully considered the situation. This suggests decision-making is not sudden or impulsive and means there is an opportunity to influence the decision-making process.

Training and rescue

"The results of this survey research are also important to guide things like training for the SES," A/Prof Taylor said. 

“It’s better to give SES personnel a scenario that’s actually realistic so they can begin to understand the situation in which entering floodwater happens. It’s not only about decisions being made on duty in life or death situations, it’s also those more mundane moments when maybe there is a lack of consideration to risks, or reversion to habitual driving behaviours.”

NSW SES Flood Rescue Coordinator, Mr Carl Manning agreed and explained how this research is aiding frontline SES personnel in their decision-making during flood operations.

“I’m currently involved in the flood rescue site commander project for frontline flood rescue personnel, and it’s really helpful to be able to expect outcomes based on likely behaviours of the public that this research has informed,” Mr Manning said.

“We will definitely be using this behavioural science to consider when undertaking future flood rescue operations.”

Challenging the behaviours

Despite quite a number of public risk messaging campaigns in recent years, the study found there was generally fairly poor levels of awareness of campaigns and recollection of messages. Surprisingly, there also did not appear to be a positive link between campaign awareness and lower levels of risk taking.

“The theory tells us that to get people to take protective behaviour, in this case, not to drive through floodwaters, the first thing is that more needs to be done to make it clear that there is a risk, especially with more benign situations,” A/Prof Taylor said. 

Risk is often difficult to assess and a poor ability to perceive risk is likely to result in increased engagement in risky driving behaviours. To better understand the decision-making processes involving drivers’ decisions to enter floodwaters, the research team explored the ability to recognise floodwater hazard and adequately assess the level of associated risk, in a newly developed online tool, EXPERTise 2.0 (Expert Intensive Skills Evaluation).

“The second pillar of the equation is actually to try and give people solutions to the issue, which, in this case, is not driving into floodwater,” A/Prof Taylor said. 

“Past behaviour often predicts future behaviours, so we need to modify risk communication and influence risk perception. People have got to know that it’s a risk to start with if we want them to engage with the messaging and then we need to provide them with ways to manage the risk; this is a fundamental aspect.”

As the coordinator for flood rescue operations, Mr Manning said the most common poor decision he has seen people make is underestimating the effect of floodwater. 

“Floodwater is always more powerful than people expect, and the sheer force of the water alone is something people don’t anticipate when entering it,” he said.

The NSW SES is currently reviewing their messaging and flood risk management procedures to look at how they can better structure their risk messaging.

“Over the last five to ten years, we really didn’t think much into the complexities of driving into floodwater,” Mr McLaren said.

“Mel and the team’s work has been a trigger for us to understand that it is really complex and it’s probably time for us as an agency to go back to the table and look at how we can better structure our messaging to use behavioural insights and psychology.”

EXPERTise 2.0 assesses the user’s ability to interact with task-related cues and form diagnoses, and the research aims to develop and validate a measure of cue utilisation in the context of driving into floodwater.

“It doesn’t tell us what specific cues people are using but it gives us information about the extent to which they are using cues,” A/Prof Taylor explained. 

“We can see whether people using cues are more risk aware and, while we are still refining the tool, we have already found that people who use cues have higher expertise in judging risk and are therefore more likely to make safer decisions.”

Mr McLaren said the team’s research and the EXPERTise 2.0 program is helping emergency services begin to understand people’s decision-making processes.

“The research shows that there’s often quite complex behavioural issues that are driving people’s actions,” he said.

“If we are going to effectively design communication and messaging, we need to understand what is driving people’s decision-making so that we can actually target that when a) we are working with people on the ground, and b) how we are messaging during an operation or when it is flooding.

“We live in a fast-paced and challenging world, and to have an asset like the Bushfire and Natural Hazards CRC to work alongside us is so important to make sure that the programs and messaging that we’re using with communities is evidence-based and grounded in research."

In conjunction with AFAC and SES research end users, the results of this research are now informing the co-development of a set of public communication guidelines and the establishment of a set of national community safety announcements for use by the ABC in emergency broadcasting.

To find out more about this research visit the Flood risk communication project page.