That's why Robert Cieri, a PhD student at the University of Utah and Blue Waters Graduate Fellow, is studying the airflow patterns in monitor lizards.
"We're trying to understand the evolution of the respiratory systems in higher vertebrates", stated Robert Cieri. "On one side there's mammals, like cows and sheep, and the other side is birds and reptiles. We have a good understanding of how the bird lung works - there's a unidirectional flow pattern throughout much of the lung, so when the animal is breathing in or breathing out air takes the same pathway from the back of the lung to the front of the lung, which is really fascinating and different from how the human lung works. For a long time, biologists thought unidirectional flow was really about efficiency, it provided enough energy for birds to be endothermic - create your own heat like humans do - and to fly.
"But about 10 years ago, my advisor - Professor CG Farmer at the University of Utah - discovered unidirectional flow patterns also in alligator lungs. So now we have the same trait in a cold-blooded reptile that is semi-aquatic and not flying, and not doing a lot of high-level aerobic activity. So we need to go into other reptiles and figure out what's going in their lungs to get at the root of where these traits came from Do they have lungs that are functionally more similar to birds or similar to ours?"
Robert Cieri said the monitor lizards, which are found in warm climates - Africa, Asia, Australia, and a lot of Pacific islands between Asia and Australia - greaty vary in body size, metabolic rate, and habitat. He's intrigued lizards with the same general body design, and the same general lung design, has lung traits that might vary with habitat use and body size. They're also a very active lizards, he noted, "so they're a good group to study because they might represent the lizard lung that's been selected and evolved most for high activity."
One reason airflow patterns are so important, said Robert Cieri, is that it is "just plain interesting that there are fundamentally different lung designs in vertebrates". By contrast, if you look at the development of hearts in reptiles and amphibians, they have a very clear progression, he explained, from a simpler fish hearts to birds and mammals with similar designs.
"It's interesting", he stated, "that lung evolution took a different path, and made a switch in the vertebrate family tree, going one way in birds and reptiles and another in mammals. By understanding why that switch was made and what those choices mean in an evolutionary sense, you can get a sense of the shared evolutionary history of different animals, understand more about the selective pressures that led to each lineage."
He also pointed out that from a functional standpoint the airflow could be really important. As the air goes into the lungs, it exchanges oxygen and carbon dioxide with blood. Robert Cieri said this process is a more efficient exchange if you have countercurrent exchange. For example, if you have the blood going from left to right and the airflow going from right to left, you can get more of the oxygen out of each parcel of air. He was quick to point out that biologists don't think that birds are actually doing that type of exchange in their lungs. But he said there's "a lot of functional implications of how lung design could influence gas exchange under different conditions that depend on which way the air is going".
The Blue Waters graduate fellowship provides enough money to fund his education, letting Robert Cieri focus solely on his research. More importantly, said Robert Cieri, the fellowship provided access to the leadership-class Blue Waters supercomputer.
"This access has been transformative to my research, as it has allowed me to run more models than I could do with just campus resources, and increase the number of hypotheses I can test in my research. We're trying to understand what direction the air is flowing through these lungs and how the structure determines the airflow patterns. The best way we've found to do that,is to use computational fluid dynamics (CFD) simulations. We create computational mesh based on the lung geometry of that species, then using the software that engineers have developed to predict the airflow around cars or through air ducts , simulate how air would flow through the lung and then validate those simulations on real lungs."
"This is better than trying to measure the airflow direction on a real lung, because many reptile real lungs are really small, complicated and very fragile structures. Once I have one of these models built, I can go in and change some stuff - what if we move this wall to that wall, what if we close off the main chamber, what if we make this hole three times bigger - and that's allowing us to have a lot of experimental opportunities to ask questions about how these lungs are working. These models are very computationally expensive, so I need to have a large computing allocation to do them. As a Blue Waters fellow, I've had the opportunity to ask more questions. As a biologist, this makes me happy."
Robert Cieri said he usually runs on 96 Blue Waters compute nodes at once, as he's discovered that's the most efficient size for his models. The simulations anywhere from 24-100 hours.
"Nature didn't make the lungs simple, the models are very complex", explained Robert Cieri. "Each model has roughly one million elements. What makes them complicated is that were making the lungs breathe, in a lot of CFD simulations you just have air flowing through a solid structure or a static structure, it's not moving. But since the animals are breathing, we're trying to have the meshes expand and contract and then that expansion and contraction causes the flow and is part of the reason the reason they're so computationally expensive."
The work Robert Cieri is doing is fundamental research, although he says it could one day potentially be applied to develop better artificial lungs.
"Biologists have wanted to know how bird reptile lungs work for years. But we didn't have a good methodology to answer the question because lungs are fragile, you try to measure things they break, if you try to put some type of recording device in there you can't always trust your measurements because you're bending the lung around and moving it, and many tests taking the lung out of an animal and running an intervention on a delicate structure.
HPC has made this research possible, because without the simulation techniques that are accurately able to replicate the flow pattern from CT scan data or imaging data, my PhD wouldn't be possible. My work represents how research can pull together techniques from multiple fields, and how interesting melding biology and engineering and physics can be."