Skip To Main Content
Dr. Ali Mostafavi headshot


Two years ago, Hurricane Harvey made landfall in Texas, leaving devastation and destruction in its wake. Dr. Ali Mostafavi, assistant professor in the Department of Civil and Environmental Engineering, shares how his expertise and work in infrastructure and disaster resilience has unearthed ways to improve how people and cities respond in the wake of disaster – looking through the lens of Harvey’s aftermath.

Episode Transcript

Steve: Howdy, and welcome to Texas A&M Engineering presents SoundBytes, the podcast where faculty, students and staff share their passion, experience and expertise. On this episode of Engineer This!, we have Dr. Ali Mostafavi, assistant professor in the Zachry Department of Civil and Environmental Engineering. I'm Steve and with me, as always, is my co-host, Hannah.

Hannah: Hi there, and welcome to hurricane season. Today, we're going to talk a little bit about how communities, from the individual level all the way up to municipalities and government, can help boost resilience and really bounce back from natural disasters, as well as talk a little bit about how social media plays a part in all of that.

Steve: I really love this conversation.

Hannah: It's very new age.

Steve: Yeah. Awesome. Well, we hope you enjoy the episode. Welcome, Dr. Mostafavi, thank you for joining us.

Dr. Mostafavi: My pleasure. Thanks for having me.

Steve: Tell us about what inspired you to get into this line of research.

Dr. Mostafavi: Okay, so, you know, there are a number of reasons, you know, I'm a civil engineer, by training, I, by training from undergraduate, we learn how to design things, right, design buildings, bridges, roads. And in the design, we are given, you know, in our homework in our undergraduate studies, that, hey, design a bridge for this load due to an earthquake, for example, or to do to wind. And all, that's, that's what we'll learn. But, you know, another aspect of it is, you know, when we think broadly, we see that, you know, these built environment that we design is saving lives, you know, for people, in the case of a disaster. So, because of that, I became interested in understanding how we can improve our built environment, when facing with disasters, it becomes highly interdisciplinary, that, you know, if you want to be a good, you know, researcher in the field of disasters, you have to understand, you know, fundamentals from social sciences and other disciplines. And that was a phenomenal, you know, and very attractive feature for me to get interested in this line of research, because not only I was able to apply my skills and knowledge of civil engineering, but also I was able to work with and learn other disciplines in assessing disasters.

Hannah: I think it's really cool that you bring up saving lives. Having a real impact on people even kind of in a more background setting of looking at infrastructure really does kind of impact a very broad audience. And just kind of breaking it down, what, what is an infrastructure system?

Dr. Mostafavi: Sure. So infrastructure systems are the built environment that all of us interact in on a daily basis, so you can think about everything that you interact with, from the time that you wake up, like the water that you use for taking a shower, it’s infrastructure systems, the electricity, you know, that you use to brew coffee, it’s infrastructure systems, you go out, you know, to work, you know, roadways, bridges that you pass through, you know, infrastructure systems, you call your loved ones, telecommunication, infrastructure systems, you know, you go to grocery store, you know, to buy, you know, grocery infrastructure systems, or call them critical facilities, you know, you go to pharmacy to pick up some vitamins. So these are, you know, the fundamental components of communities that help us, you know, live through our lives in a day by day basis.

Steve: So it's all that stuff that I mean, really, is essential to, like you said, our lives, and probably we take for granted on a daily basis.

Dr. Mostafavi: Absolutely, but not when, when we have an event like a disaster, and when they get disrupted, you know. If we, because we are so dependent on them, you know, we feel, you know, the disruption, and it causes hardship in our daily lives. And it's interesting to know that when these, these outages are extensive, they could be they could be very devastating in terms of the impacts that they have on well-being of people. Just imagine a person who is who is on oxygen, or some family with a small baby, you know, they cannot go to three days without electricity.

Steve: Right.

Dr. Mostafavi: You know, that's, that's the real impact that you will see that there are different segments in our communities that use and, you know, rely on these, you know, systems. I did a study in in Nepal after the earthquake in 2015, and I started doing the research with the mindset that, you know, the people's attitude and perception about this infrastructure system is similar to the countries that have these services like 24/7, and and they cannot live without them. But when I went to Nepal, I observed an interesting phenomena. I realized that in Nepal, in Katmandu, the capital, with the people, a large portion didn't have, you know, tap water, okay. And, and for those who had electricity in house, they had extensive load shedding schedules, and they had like, couple of hours a day, and they have like, the rooftop storage for water, things like that, and basically, their perception and and their expectation from infrastructure systems was very different from what, you know, we tend to believe, you know, in in communities that have access to the services 24/7, and when I, when I started collecting information about how people were impacted, it was interesting to realize that some people even didn't recognize that the power was out because they didn't have power, you know, most of the time anyway, so they didn't recognize that the substation or reservoir close by was damaged because of the earthquake. And and those people had more capacity to deal with that, because they had all those of those sorts of substitutes, like, you know, the rooftop, you know, storage, or the power generator, or means to or just ability to use without, to live without electricity. So that's basically the difference in terms of, you know, not only the role of those services, but also how we appreciate them and how we interact with them is another important aspect of, you know, studying infrastructure systems and impact on communities in disasters.

Steve: You know, that's incredible to compare to how it is for us where, you know, we have a bad thunderstorm, and the power goes out for 15 minutes, and we all get so anxious, about you know, what are we going to do? You know, we've got to get out the flashlights, and we're acting like it's a major emergency

Hannah: Got to get the Wi-Fi back up.

Steve: Right. Yes, it's incredible the difference.

Dr. Mostafavi: Yes, yes. It's incredible the difference and, and, you know, one of one of the things that my my research team is investigating is that how we can improve this, the the capacities of this society for dealing with these disruptions. So we let's let's get get this straight, we cannot make our infrastructure systems fail-safe, meaning that they will fail anyway.

Steve: Right.

Dr. Mostafavi: In disasters, if you want to make them fail-safe, it requires billions of trillions of dollars of investment, and still, still with that level of investment, something like Harvey, which is a one in in how many years event that some people say that it’s a 1,000 year event, some people say it's a new trend, Anyway, something like that will impact that. There is no way we can make the infrastructure system fail-safe. So we have to make them safe to fail.

Hannah: What is resilience?

Dr. Mostafavi: That's a great question. So resilience, by definition, is the ability of a system to, to cope and bounce back. So the best, you know, short definition is, to me, is bouncing back. So, you know, resilience at the individual level is somebody has a tragedy, you know, you you get impacted, but you bounce back to your normal life faster, the faster you bounce back, you're more resilient.

Steve: So you mentioned Hurricane Harvey. When you look at the resilience that we saw there, I imagine your team is probably taken a look at that. What have you found as you've gone through and kind of evaluated how that was handled?

Dr. Mostafavi: Absolutely. So Harvey had different, different facets in terms of, you know, understanding resilience, which were really unique. So we saw the really important resilience, you know, characteristics in Harvey in Texas communities. So we saw, you know, significant participation of volunteers and community members in the relief effort that significantly helped the, the entire community to cope and bounce back from Harvey. That was a resilience characteristic, people were selfless, you know, and, you know, contributing, volunteering, donating in different capacities, in terms of, you know. We are, you know, looking into data from social media to see how people share and receive information about disaster.

Steve: Oh wow.

Dr. Mostafavi: We see, we see significant, you know, resilient behavior in terms of people's, you know, participation in disseminating important information about, “Hey, this bayou is overflowing. If you are in this area, evacuate.” That's a self-organizing resilience behavior, that it doesn't come from the emergency office, it's coming from somebody in that community, posting on social media, Twitter or Facebook, requesting the fellow community members to pay attention to that. These are phenomenal resilience behaviors that we see. On the other hand, we see areas that require improvement. So we see that, you know, in, in planning for development of the urban areas, and I'm talking specifically here about, you know, Houston, which is one of our study test beds, you know, we see that, you know, these urban areas are, are composed of highly connected systems, you know, transportation system, in a flood control system. And, you know, if we don't understand the relationship between these systems, and and understand how, you know, changes in one system influence the other, we get, you know, unintended consequences. And an example of that could be in a case of, you know, Highway 99, which was developed for relieving some of the traffic in Houston. And it's great, but what happened was, you know, in development of the transportation plan that included the development of Highway 99, the flood impacts of that development was not fully, you know, considered and, and plans were not in place to mitigate the impacts. So that development led to, you know, housing, you know, development in areas around 99 that increased, reduce the green land, and increased the pressure on Addicks and Barker reservoirs. Therefore, you know, when Harvey happened, at that amount of rainfall, you know, green land was gone, so there was no capacity to absorb the water, so all the water went to the pool of the reservoir. And guess what happened, you know, the pressure on those reservoirs was significant, and a decision was made to release water into downstream neighborhoods that most of them were outside the floodplain, and most of them didn't have flood insurance. Okay, you're talking about these complex dependencies that are difficult to capture, but this is a new norm. Urban systems are, are composed of complex, interconnected, you know, systems and networks. And we cannot design and develop one system in isolation from other systems. So what we are learning is that first, we have to recognize and incorporate the relationship between different systems, for example, how transportation development affects for risk, how, you know, development of flood risk affects the access to critical facilities and hospitals, right, where we should build new hospitals, so that if flooding happens, you know, the most vulnerable segments of our population can access those facilities. So these are the questions that right now, with the current, you know, practices, are usually not fully considered. Okay. And and this is where our research comes to the picture. And we're trying to understand these relationships and build models. And these models are, you can imagine as computer games that you can simulate different scenarios of flooding and hurricanes.

Steve: Oh interesting.

Hannah: Very cool.

Dr. Mostafavi: And see that, you know, with the existing system, how your your community will, will be impacted as in what roads would be closed, what percentage of the community will lose access to hospitals and critical facilities, And what are the characteristics of those people who lose access? Are they from part of town that are considered to be an older population or from a part of town that’s considered to be a lower income that have less resources? And then based on that, say, “Okay, how we can optimize the resource allocation, a prioritization of our investment to minimize the impact.” And the good thing about simulations is that, in the real world, if you build the reservoir or a road somewhere, it's there forever, literally, but in a simulation, we’ll build it, remove it, build it somewhere else. And then with that we can, you know, test different scenarios to see what is the best combination of solutions that will mitigate the impacts under most of the scenarios. As I said, you know, we cannot make, you know, fail-safe systems, but we say, “We've simulated a thousand scenarios and say that this combination survives under 990. The 10 additional scenarios that this system will fail is too expensive to implement.” So we kind of stress test different solutions before we implement them. And and it's not prediction, it's, it's a, it's a evaluation, it's a new practice in how we design systems, as opposed to designing individual buildings or individual roads and bridges.

Jenn: Jenn the producer here. So this interview is really interesting, but unfortunately some things had to be cut for time. But I wanted to butt in to include something that stood out to me. When Dr. Mostafavi was talking about his research after the earthquake in Nepal, he went on to tell us that he was at the site of ground zero. It's also the trip that sparked his interest in disaster recovery research. So I guess it's kind of like his origin story in this research, which I think is pretty cool. All right. I know I've interrupted the interview, y'all go on and get back. See you later.

Hannah: When Harvey was happening, I was living in Houston at the time. And I was very fortunate that I had a family in College Station that I could go to while it was storming and everything was happening there. So I was really only affected in that I had to watch it happen from afar. And when they started releasing the levees, which I found it really interesting that you, you mentioned, my area was one of the areas that was affected by flooding. And so they released Buffalo Bayou in the middle of the night, and it flooded right up to the road that my apartment complex was on. And so there's a mandatory evacuation between the highway all the way up to that street, and then my apartment complex was on the other side of the road. So it was a little bit close to home. Would your model be able to kind of look at that map and say, “If we release this, what would the effect be?”

Dr. Mostafavi: Yes. So that's, that's exactly what we are trying to do in one of our projects, that is, you know, we are very excited to receive support from the National Science Foundation. If you're trying to build the models that we can dynamically evaluate the propagation of flood risk, based on different scenarios, and one of this is exactly what you just said, because when we talked to, you know, emergency managers and, you know, personnel, after Harvey in Houston, Harris County, you know, we realized that they, they didn't have means of understanding where the water will go next. You know, they were told, you know, that, you know, these neighborhoods will be flooded, and they were, you know, they had to be just reactive to the situation. But, you know, using the models, we can have this predictive capacity to, in near real-time, not, you know, fully real-time. And by real-time, I mean, that, you know, models that, you know, run as things unfold. So, but we can have models that can tell us, you know, maybe an hour or two ahead of time that, based on the current propagation, you know, pattern of flood, the next neighborhood that will be flooded is this one. Plus, the data that we get from recent events like Harvey will help us to have robust techniques to supplement existing hydraulic and hydrological models that has been around for many years for planning, but this time with more computational capacity, so that we can, you know, make decisions right on the spot, not just based on the judgment, but also based on some, you know, outputs that the model give us about the where the flood will go, which neighborhoods could be impacted, so that if we need to evacuate them, we evacuate them some hours before they get impacted.

Steve: So you say that you can use these models to look a couple hours ahead of time. But could you also use these models say, when you were working on new developments. You know, like in Houston, they're constantly changing, building. Would, say, city planners be able to use something like this to run these models and find out, okay, if we put this highway right here, then that's going to flood this whole area, and try and use that to find a better place?

Dr. Mostafavi: Absolutely. Absolutely. There, you know, there, I've talked about models, there are different types of models, you can think about it, as I mentioned about different types of video games, right? So these are simulation models in which we simulate, you know, if, if, have have you played Sim City.

Steve: Yeah.

Dr. Mostafavi: Something, something like that, you know, essentially what we build is something like that, that we build the elements of a city, but we defined physical properties, and physical measures and engineering formula into that, so that we can simulate the behavior of the buildings and roads and bridges and the propagation of flood, etc. So, you know, these models can be used in different ways. As I mentioned, for planning purpose, it could be used to test if we build, let's say, the third reservoir here, and then simulate, like, thousand scenarios of hurricanes and flooding, and see that, you know, how, how would be the risk in that case? So let's move it. Let's go with another plan, another type of infrastructure solution. What if we add, you know, these green infrastructure that, you know, it's a local retention ponds of biosoils? How would that reduce the flood rates? That's exactly what we are doing. So that's one element of that. And the time for those simulations would be you just have the model, you simulate the scenario, and then you will say that in a period of like, 10 years, 20 years, what would be the risk impact? So we are looking at the time horizons, which are longer so. But another, you know, scale to evaluate is this, you know, during the event and response that, you know, our timeframe would be, even in minutes, to look into how, you know, events unfold, how disruptions happened, and based on these time steps, we can help with emergency response and, you know, a community's capacity to cope with the disruptions.

Steve: Absolutely.

Hannah: Talking about infrastructures and resilience, is this something that individuals can help with? Is it just more like municipalities?

Dr. Mostafavi: That's a great question. I think the more we do research in this area, we get to this belief that resilience is everybody's responsibility, and everybody can contribute to resiliency. So let's say from individual households, what we found in disasters is people who are not impacted have a significant role in, you know, providing relief to their fellow community members. And as I mentioned, Harvey and Texas was a wonderful example of such resilience behavior. You know, people who are impacted, you know, if you are educated and you have the right expectations, right, for example, myself, if I know that, you know, in for the next event, I have to be prepared for a week of power outage, for a week of not being able to go to grocery grocery store, I would prepare, you know, ahead of time, so I won't wait until like, it's the 48 hours from the hurricane landing, right? So this is the capacities that we can build individual, I can check, you know, my my street drainage and see that whether the drainage is is, you know, has any debris to clean up or something like that. So that's at the individual household level. Another interesting phenomena that we observed, is, I talked about social media and people's activities on social media, which is increasingly, you know, popular, becoming popular in, in disasters is the, you know, it's the participation of volunteers on social media, people who by their expertise know things, and they try to help others, right? So a civil engineer, or somebody with, you know, expertise in, you know, weather patterns or hurricanes, you know, doing activities, finding, you know, reliable sources of information on social media and trying to retweet them, and, you know, mention their neighbors or community organization, so that this information is propagated, you know. Having good, having a good situational awareness in disasters is a fundamental element of being resilient. So, these activities, somebody might then be affected by, “Hey, I have this expertise, and I know that, you know, some other, you know, fellow residents might not know that we are living close by one of the most, you know, risky reservoirs. So let me just tell them that, ‘Hey, folks, if Harvey, you know, unfolds as we expect, you know, there is a chance that, you know, this happens.’” And if, if people know that three days before, compared to just one hour before that, “Hey, water is being released,” that's, that's different, you know, in terms of resilient behavior. So we see that, and we see, we call them influential users, right? And we borrow that from marketing that we have, you know, social media influencers to try to sell you things, you know, through, you know…

Hannah: Selling safety.

Dr. Mostafavi: Exactly. So that's, that's wonderful. Actually, you know, I’ll make sure that I cite you for that because those influential users, actually they sell safety, and it's phenomenal how they emerge. You know, they're consistent, they rely on their expertise, and they work day and night to just filter through the noise, you know. Guess how many tweets were posted, just in Houston over 40 days for Harvey, just Harvey related. Can you make a guess?

Steve: It has to be in the millions.

Dr. Mostafavi: More than 30 million, and I would say most of it is noise, okay. And to calm down the noise and get to the essence of information that community needs to respond right away is important. And those people play essential role in that. And another phenomena that we observed was fake news, right? Probably you saw some images of, you know, the airport, Intercontinental being flooded, which was fake news, but those, those people are fundamental in debunking that and say that hey, that's not true. Because some people try to take advantage of the situation and spread panic, right, but those people sell safety, as you mentioned, and play an important role. That's another role that they play. And, and, you know, I don't want to talk about the roles of engineers and city planners, etc. It's obvious, but if you go to different levels, from individual to household to neighborhood to, to community level, I think everybody has a role. And I think we have to acknowledge that, and understand that and we hope that our research can help with, with that, you know, understanding, so that everybody knows that all of us have a role to play.

Hannah: We got really lucky at my apartment complex, because we have kind of have like a social media, when I was still living there, specifically for the apartment. And we had people doing just that where anybody who was trapped outside would be sending messages in asking, “How's the apartment? Is everything okay?” “There's a giant tree outside of mine. Has it been blown over?” And everybody who was there would take pictures and send out information and tell us when parts of the storm came through, if there was any parts of the apartment complex that was particularly hit, and how much it was actually affected, which was incredibly calming for people who were stuck outside the city.

Dr. Mostafavi: Absolutely, absolutely. And, and that's the power of, I would say, you know, networks that we are, we are so blessed to use. Maybe when we, you know, started using social media, we never, you know, recognized that they will help us to be more resilient, but now, they are becoming a fundamental component of a community resilience. And that's because of their network effect. And, you know, and we see, we see the increasing usage and also, you know, the capabilities that they provide. So we tend to believe that they increase the collective intelligence of communities in responding to these disasters.

Steve: It's great to have that perspective, because a lot of what we're hearing, especially right now, is there's a focus on the negative aspects of social media, and rightfully so. There are quite a few. But keeping in mind that there are areas like this, that it really does make an impact that, especially in situations like this, can be life or death, sometimes.

Dr. Mostafavi: Absolutely, and, and actually, that's, that's why it's important to study the social media, you know, impact during disaster in a situation, so if you use that right, in a right way, it can help build capacity for improving the collective intelligence, but the key is to understand, you know, how people use it, how people share information, how people receive information, and then try to improve it, right? So by studying the behaviors of people on social media, when different built environment disruptions happen, qw can understand what are the good activities in terms of tweeting, retweeting, posting, etc., with the content, etc., and what are the not so good activities that lead to spread of rumor, etc., and we have to learn these things. If we just use them and don't understand them, then, then they can become, you know, sometimes dangerous, you know, tools, but if you fully understand how people interact with them, and use them and try to use the learning to educate and improve the usage of these in future events, we see great potential and, and honestly, I think, you know, one factor that contributed significantly to people's ability to respond in Harvey, without, without a doubt was the, the extensive use of social media.

Hannah: So, where do you go from here?

Dr. Mostafavi: We are trying to understand the, the variations and disparities and risks for different, you know, sub-populations, and use that to see how we can minimize the societal risk and, you know, build societal capacity. And, and what makes this, you know, research, very exciting for us is the partnerships that we have with the stakeholders, you know, in, in the region, and at the national level. We work very closely with different organizations, city of Houston, you know, has very exciting initiative on resilience, acceleration, that we are actively participating in, working closely with the Harris County Flood Control District, U.S. Army Corps of Engineers, and, and try to learn and try to just communicate what we are learning so that you know, all of us can, can improve and also help to disseminate research findings, to, to them so that they can account for that as they make new plans, and, and and decisions. So we tend to say that after disaster, we have a window of opportunity to be more resilient, and we try to maximize that. And by working closely, and if we work together, I think we can, we can be very prepared for future events.

Steve: You know, I have a lot of friends and family who live in Houston. So, you know, what would you want people like them that live in these kind of high risk areas. What are the key takeaways that you would want them to know, for situations like this to be a part of this preparedness?

Dr. Mostafavi: I would say, you know, be involved with your, your community, you know, you know, currently the city is implementing a hazard mitigation planning. And, you know, everybody can get, you know, let your voice be heard, you know, and, and I think, you know, we cannot be passive citizens of our community and expect good things to happen. We have to proactively and actively participate. And also, we have to also, at the same time, recognize the limitations, you know. It's easy to sit down and say, “Hey, why don't we build this in this way?” without knowing the constraints from engineering perspective, from the policy perspective, but also, you know, recognize the limitations and try to work together. I think collaboration and coordination is, is the key. And, and it's not just for, for the residents to, you know, be engaged and work with these decision makers, but for decision makers to have more community engagement and also collaborate better with each other. As I mentioned, this, this planning for resiliency, hazard mitigation is a collective action process. It’s not a decision that is made by a single organization or a single person, and it's impacting everybody. So if we work together, collaborate more effectively, and, you know, communicate better, I think that, that's the key for, for having really resilient solutions into the future.

Jenn: Hey, it's your producer, Jenn, and I'm here to introduce our next segment where our non-engineering co-hosts try their best to answer engineering questions. It's Ask An Engineer. Steve and Hannah will have 30 seconds to search online to answer the question, and then our guest will tell them how they're wrong. We'll fast forward through the searching part.

Hannah: So Jennifer, what's the question?

Jenn: Your question this week is what influences community disaster resilience? Alright, time’s up. You both now have 15 seconds to answer.

Steve: So I learned that with disaster community resilience, that can mean a lot, especially for different areas.

Jenn: Seven seconds.

Steve: That could mean preparing for fires out in California, for climate change.

Jenn: That’s all you got. Good job.

Steve: Didn't get very far.

Jenn: No. It’s alright. Alright, so Hannah, it's your turn now.

Hannah: Well, I learned that there are many aspects of community that need to be built up to help with resilience, such as education and wellness. 

Jenn: Seven seconds.

Hannah: Transportation, and all these things are kind of interconnected with one another. And you need all of them.

Jenn: And that’s your time. Alright.

Hannah: It took so long for one page to load.

Jenn: So how did they do?

Dr. Mostafavi: I think they do fantastic. So I think they captured two important aspects of community resilience. One is that we are dealing with multiple types of hazards. So multi-hazard aspect of it that is resilience to what; hurricanes, floods, fires, earthquakes, or all of the above. So that's an important aspect, which is very challenging for us researchers to, to account for and, and the second important aspect is the multiple components of our communities that play a significant role in resiliency and their interactions and relationships. So I think they did a great job of capturing two important aspects of community resilience.

Steve: Hey, we did it. 

Hannah: We did a thing. Well, thank you for coming and talking with us. I know that I learned a lot from this.

Steve: Oh, this was a wonderful conversation.

Hannah: It was a lot of fun. I, I was really interested to see how social media kind of came into at all. So that was a really cool thing that I didn't expect. 

Dr. Mostafavi: Yeah, I had fun too. Thanks for having me.

Hannah: Yeah, of course.

Dr. Mostafavi: Thank you. 

Hannah: Thanks so much for tuning in to Texas A&M Engineering presents: SoundBytes. What did you think? Do you have any burning questions you want to ask? Hit us up and let us know at That's bytes with a ‘y’. And keep an eye out for us in ZACH. We like to wander the building from time to time, and we love hearing from you. So come on over and say hello or lend us your voice for a future episode. Finally, just so you know, the views and opinions expressed in this podcast are those of the hosts and guests and do not necessarily reflect the official policy or position of the Texas A&M University System. Make sure to tune in next week. Until then, from everyone on the Pod Squad sounding off. Thanks and gig ’em.