As has been shown in the past few weeks, flash floods can develop quickly in rural and urban areas with mild to catastrophic effects. That’s why flash floods are so important research, according to Andrew Kruczkiewicz, a meteorologist and senior researcher at the National Disaster Preparation Center at the Columbia Climate Institute. “Some parts of the world will never see potential disasters like wildfires or tropical cyclones, but there are few areas with zero risk of flash floods,” Kruczkiewicz said. And, because the potential impact varies greatly, “it requires additional sensitivity because of the way you communicate risk, because the mountain flooding in the Mill is very different from the 30-foot water wall.”
Kruczkiewicz’s current research focuses on extreme weather events such as flash flooding, application of climate and weather data, and forecasting to reduce disaster risks and promote humanitarian action. In the discussion below, Kruczkiewicz talks about why the floods in Texas are so devastating, how the warning system considers very different populations (EG, entertainment – seekers and locals) and how we integrate technology and local knowledge to avoid such a rampant future.
What causes these floods and why are they so disastrous?
When we talk about flash floods, there are many factors: meteorology, climate-related, geophysical, and flash floods related to the built environment. Meteorologically, it interacts with the mesoscale convective vortex (MCV) in central Texas, and the moisture interactions associated with the residues of tropical storm Barry. MCV interactions allow the formation of new, intense and rainy thunderstorms. Furthermore, these storms are very slow. When it is determined that the impact of flooding on life and livelihoods is so great that the degree of dryness on the ground is also a contributing factor.
One of the fingerprints of climate change is that the warm atmosphere causes the tropical system to be wet. But there have been a long time of mountain flooding in this area. Will climate change worsen? We should ask this question. Other factors include socio-economic transformation, land cover changes and land use changes. Broadly speaking, we must consider all of this to develop revised policy and disaster preparation, warnings and anticipated approaches.
Some news reports say this is a “smoke alley”. The flood-affected areas are indeed part of the Flash Torrent Lane, and in this area of central Texas, we see higher basic risk levels. But these factors are not only in the country of Texas, but also into northern Texas, even Oklahoma, and then into New Mexico. “Floor torrents” is not a well-defined term and more work should be done to define and map various types of Floor torrents as they are areas experiencing higher risks and therefore a more specific approach is needed depending on their location. For example, cities can also have mountain torrents in cities, which is mainly due to poor urban planning? Yes, if there is a higher level of risk than other areas of the city.
Will better communication around the risk of flash torrents help save lives?
Few flash torrents in the world have zero risk, so we need different types of risk communication. What does the “floor torrent risk” mean to people who already live in mountain torrent alleys? How many flash flood warnings have been issued in the past five years, and what are the flash floods that have caused? There must be some false alarms. But what are their impacts for those mountain torrents that do occur? Of course, nothing is close to what we see.
In areas where flash flooding is common, things locals can handle may be considered “ordinary” flash flooding – which may wash off your car or cattle, which can cause economic stress, which can have a disproportionate impact on low-income people, which is very different from a 30-foot wall of water. I think what we are learning is that the binary of “no flash flood warning” and “frost flood warning” is not enough. “Flash torrent emergency” is a relatively new mechanism. Its purpose is to solve the problem of different intensities of mountain flooding. So I think we need more sensitivity to send these different types of messages.
In addition, many people who died do not live in this area. They were there for entertainment purposes. Who is responsible for communicating risks and raising awareness – telling them that they are entering an area where the baseline risk is low but definitely not zero? What seasons or months will increase baseline risk? What does an increased risk of flash floods mean in areas where the baseline risk level is already higher than that in the surrounding areas? With the flooding of intensity, we are at critical moments to review and modify methods to reduce flash flood risks, early warnings and anticipated actions that will support the development towards a more comprehensive system.
What tools or technologies may help warn people to evacuate?
For us, this is an opportunity to better combine new technologies with “traditional” technologies such as siren systems or NOAA weather radios using VHF frequencies. People no longer use these radios because they think the phone will solve the problem. They believe that if the risk suddenly increases, they will get automatic push notifications. But sometimes these notifications don’t come. VHF frequency is much more reliable – no need to charge the temperature radio. You put some batteries and they last for a long time.
I haven’t heard a lot of people talking about how to use local knowledge (or we can even say indigenous knowledge) to perceive and manage flash torrent risks before building the field. There are other ways even before the siren and the radio. What is that? How do they inform the next generation of risk awareness and warning systems of the mountain flood?
Can new technologies such as AI modeling help here?
I think there is an interesting lens here to do with our trust in AI, climate technology and weather applications. In many places, we are learning that AI-driven forecasts are better than numerical weather forecasts when we talk about average or moderate rainfall. That’s because we have a lot of training data representing these types of events. But when we talk about events that are far beyond the norm, we don’t have historical data to train the AI modeling infrastructure, so we have no evidence that AI-driven products do better in predicting disasters in a given geographical area.
How can we better prepare for such extreme events?
At the very least, everyone should have the potential for extreme weather to occur in a given night. When we have periods of no disaster, social science shows that we are becoming comfortable. In our minds, we know there is a risk of mountain flooding, but we don’t want to think about that, especially when we try to enjoy the holidays. This is a common phenomenon of disaster response and risk reduction, and we need to do better to address complacency and underestimation risks.
Who should be responsible for the communication risks?
The main question to be asked is, what types of communication drive people to take meaningful action at the right time? Until recently, one approach was based on frequent and consistent messaging from trusted and authoritative sources. But the “trustworthy” element is key here. To what extent does everyone in a community or county trust the source of trust? Does the person at the highest risk trust the same source as the person at the medium risk? For example, if you are traveling in a new area, such as working in a job or tourism industry, do you think of the “source of trust” the same as those living in that area?
If there is a policy response to Texas’s flash flooding and more focus on improving risk communication and warning messages for people who are out (such as those on vacation or camping), it may mask equally important factors: Some people Living In these areas, data suggest that, especially for mountain floods, those at the highest risk are more likely to be underserved financially. Even if they receive a warning, they may not be able to evacuate or can only evacuate areas at higher risk. How many people received warnings during this event that they could not leave for various reasons? It is a great opportunity to ask these questions and see if they can inform the policies of people living at various risks that face flash torrent risks, in addition to developing ways to communicate risks and warning messages for people to enter such areas for work, leave or other reasons.
