Cornell University researchers have launched an online dashboard that estimates mortality from wildfire smoke in real time. The tool uses data from several federal sources including the U.S. Environmental Protection Agency’s AirNow website, which displays air quality sensor data. The innovation comes after smoke from Canadian wildfires colored the skies and negatively impacted air quality in the Northeast last summer. WAMC spoke with Alistair Hayden, a professor in Cornell’s Department of Public and Ecosystem Health and a co-principal investigator of the dashboard.
Hayden: So I want to start out with just why we built this dashboard, which is that smoke is bad for your health, and we really want to communicate that it can be bad for your health while a smoke wave is happening. I get a lot of questions during smoke waves about, you know, what are the impacts in my community? How bad is it for me? And we built this dashboard to really try and help communicate some of those impacts. So it's really a communications-focused tool, and we hope it'll also be useful for policymakers to understand what the potential impacts of wildfire smoke is in their jurisdictions.
Levulis: And you mentioned a smoke wave there, certainly upstate New York experienced that with the Canadian wildfires last summer, correct?
Yeah, so I actually, prior to moving here in New York about two years ago, I was a division chief in the California Governor's Office of Emergency Services, so I saw a lot of smoke out there. Lived it myself, and when I moved out here, I thought I was perhaps going to be free of that wildfire smoke. But then, sure enough, we started to experience smoke here on the east coast as well. So this dashboard is built both for you know, our colleagues and friends on the west coast as well as our friends and neighbors here on the east coast as well.
So what data goes into this mortality risk calculation on this dashboard that we're speaking about?
So the way we built this dashboard was we were really just trying to make available much faster estimates of the mortality that might be attributable to wildfire smoke. So in the past, there have been many different ways of getting at the number of deaths due to wildfire smoke. The most detailed and lengthy one is you actually go in and look at death certificates, and you kind of correlate any spikes you see in the number of death certificates with the amount of smoke in the air. That usually takes months to even years, to make that really detailed estimate. And so another method is you take work that other people have done of that detailed type, and they have developed equations where they see, OK, when smoke goes up this amount, we expect deaths to go up that amount. And even those estimates can take months to become available because of some of the delays in getting the data about how much smoke there is. So really, our innovation here was to take those existing methods and optimize them for speed, so that the estimates are available within 24 hours of the smoky day. So we're estimating the concentration of wildfire smoke in the air, and then using those detailed epidemiology equations in order to then estimate the mortality attributable to the smoke.
And now, due to wildfires in California this summer, the dashboard has already gotten some real-world experience. How is it performed so far?
Yeah, so we're still very much in the validation phase. Actually, even some data from earlier this year, there were some wildfires in Oklahoma and Texas, and we can see some of those smoke impacts show up on Texas actually. And then, yes, there have been some wildfires on the west coast that seem to be related to impacts we're seeing on our dashboard here. So, so far, We're very happy that it's able to generate these estimates in real time. And then now we're going to be doing the validation phase, where we're talking to practitioners on the ground. You know, are you seeing levels of smoke like what our model is showing? Talk to medical folks on the ground. Are you actually seeing health impacts like our model is predicting? And then to really validate it again, we're going to have to wait for those months to years, for those very detailed analyzes to come out.
And now in launching this dashboard and announcing it, Cornell notes that wildfire smoke and heat waves have never been declared as federal natural disasters that qualify for financial resources. Why is that? And is the hope that a dashboard like this could change that?
So it might actually surprise many folks to hear that there has never been a disaster declared by the federal government for heat or smoke. And there's a number of reasons for this. It's actually been in the news lately. There have been a number of organizations, environmental organizations, worker rights organizations, who petitioned FEMA to try and change this. So it's a very hot topic right now for both heat waves and smoke waves. And talking to my emergency management colleagues, some of the reasons why they think one has never been declared are because we actually just don't have the real time estimates of the impact. So the way that we define a disaster in this country is, when the impacts exceed the resources of the community who's being impacted. So you know, it can be a hot day, but that's not necessarily a disaster until the community is overwhelmed. It can be a smoky day, but it's not a disaster until they're overwhelmed. And it's the same definition that we apply to hurricanes and tornadoes. You know, if a hurricane touches down and doesn't harm anyone and doesn't damage any property, is that a disaster? Well, probably not. So that is definitely a major reason why we developed this dashboard, is to try and help understand what are the numerical impacts of wildfire smoke and understand that at the community level in real time, so that we can start thinking about what is the level of impact that we might need to declare a disaster. Now that's not going to be the only thing we need for a federal disaster to be declared for heat or smoke. We need to make sure that FEMA is sufficiently funded in order to respond to these and we also need to ensure that the role of emergency managers in these events is really, like clearly spelled out, because there haven't been a lot of responses to this type of disaster, and so we need to build that knowledge base of how our emergency managers can help.
