If you need to get from point A to point B, an app can show you the fastest route, or the shortest route — or even the route with the fewest tolls, if you’re in a place with them.
But, until now, you’ve not really been able to see the shadiest route — as in, the coolest one, in which you’ll spend the least amount of time in the sun.
Hua Wei is an assistant professor in the School of Computing and Augmented Intelligence at Arizona State University and has been working on ways to incorporate artificial intelligence into making information about shade more available and accessible. The result: two projects that come at the issue from different angles.
As part of The Show’s occasional series called Throwing Shade, Wei talked more about his work.
Full conversation
MARK BRODIE: Hua, how big of a role do you think AI could play in increasing shade and access to it?
HAU WEI: I would say a lot, because right now think about shade or think about a simple map that we have in the map providers in Google or Apple. They don’t provide any like shade map. And I think that is really something that we need in Phoenix. Not only in Phoenix, but I feel like for the worldwide cities —like in Paris, in South Africa and in China, in Japan, we need that. A lot of cities are undergoing the heat wave.
BRODIE: Yeah, so you mentioned something like Google Maps or Apple Maps, and it sounds as though one of your projects, which is called Shaded Route Planning, it almost sounds like it’s sort of Google Maps, but instead of just having options for like if you want to drive, if you want to take public transit, if you want to walk, that kind of thing, it also adds an option for what is the shadiest route. Is that a fair analogy?
WEI: Yeah. Yeah, you’re right. So that kind of like Google Map, but instead of telling you the fastest route, it can also tell you the coolest one or the combination like 50% of fastest and 50% of coolest. So we use AI models to detect the shade directly from set of the images and lines that up with each sidewalk and bike passes.
BRODIE: Do you find that it matters what kind of shade it is, be it natural or man made in terms of the kind of impact it might have on somebody traveling through a city?
WEI: A city, I think it matters when you have different, when you are in different cities. I feel like in Phoenix, as long as you are in the shade, no matter it’s structural shade from the buildings or the tree, the shade from the tree it helps. But for those cities who have like wet heat, not dry heat, I think the structural shade matters more.
BRODIE: So let me ask you about the other project you’re doing, Deep Shade. And this is one where it almost sounds like a planning tool where you can. You’re sort of using AI to figure out where shade is going to be. Am I getting that about right?
WEI: Yeah, you’re right. You’re right. So that’s sort of like a movie. So it actually predicts how the shadows will move throughout the day. It’s a generative AI model. So you can ask, what does downtown Phoenix look like at 4 p.m. tomorrow or 4 p.m. in July 4? So they will literally draw the shadows where they should be.
BRODIE: And is this the kind of tool that you can use to sort of figure out, you know, if you want to have shade in a particular place, like what kind of structure or what kind of tree might need to be in that place to make it shady at a particular time, maybe even on a particular day?
WEI: Yes. Yeah, you’re right. I think this one going to be very helpful for the urban planning. So if you want to have a shade over there at a certain time of the day, we can design how high the building or how high the tree should be and in order to cover that area.
BRODIE: So you mentioned, you know, this might be helpful in city planning and in trying to think about, you know, ways that this might be useful. You know, things like where to put a park or maybe where to put a bus stop or how to shade. Those kinds of things are ones that come to mind.
But I’m guessing that there are instances that maybe aren’t as readily obvious where this could be useful.
WEI: Yeah, I think like one direct thing the city planners can do is use the shade map that we developed to figure out which part of the city or which part of the route, like do not have any shade. Then they want to connect some part of it to make sure there is a certain route that is always under the shade.
Preferably they would have pedestrians going from one direction to a bus stop and they want to make sure that people commute to the bus stop and that route specific route is under the shade. That is one direct thing that we can think of.
The other ones I can think of is the schools and universities, when they are having students running through different buildings and how they can plan for the trees and for the buildings to make sure they have the shade covered for the routes.
BRODIE: So obviously Phoenix has a goal of having a certain percentage of the city, you know, with shade, canopy. Have you had conversations with anyone within the city about how to maybe work with this tool to help make that happen?
WEI: Yeah, actually we are. I think I’m. We’re really lucky to be in Arizona because the local governments here get how serious the heat problem is. And right now we are working with city planning and sustainability teams to bring our shade data into the workflows.
BRODIE: How much do you think that this tool can help in that effort?
WEI: For example, they can use tools to spot areas that are super exposed, like neighborhood without tree cover or bus stops with no shade, and prioritize those for improvement. That is just a simple idea. But having actual numbers or visualizations help them make data driven decisions, not just rely on iteration.
BRODIE: Well, I guess when you talk about data driven decisions and you’re talking about cities, that often also includes money. And I would think that cities would be maybe a little reluctant to spend money on shade or maybe more apt to spend money on shade if they had, you know, a model like this that said, OK, this is, we’re pretty sure this is going to work as opposed to, we hope this is going to work.
WEI: Yes, you’re right. So this is going to be when they have a budget and where they want to plant or plan first. And this data and this quantitative evaluation going to be the thing that can be helpful for them.
BRODIE: How else do you think that you might be able to use AI to solve the problem of shade?
WEI: I think AI is really well positioned to solve this problem because previously before AI, people are using the real simulations. They have to run, set up their simulators and they have to know exactly where the plants height and what is the building’s height. They build 3D simulations and they have to place where the sun could be in certain day of the year.
But with the simulations or the AI’s generative AI, we can just input text and AI with well trained data and then we can basically generate shade very easily. We do not rely on the expert simulators anymore.
KJZZ’s The Show transcripts are created on deadline. This text is edited for length and clarity, and may not be in its final form. The authoritative record of KJZZ’s programming is the audio record.