 So thanks, first of all, for being here at CSE Conference. It's our first time participating in the conference, but I think the idea that drives the conference resonates a lot with our own mission and also seeing a lot of familiar faces and names. So very excited and grateful to be here with you all. And also excited to catch the Kamalama that I did not get to catch yesterday. So an introduction, I'm Krista Hazenkope. I'm with OpenAQ. We are a non-profit based in Washington, DC, and our mission is to fight air inequality with open data and community. And air inequality, if you have not heard of it before, is a term our community has coined to express the severe human rights and public health issue that the lack of clean air to breathe across the world poses to all of us. And so in this talk, I'm gonna cover a few points. I'll talk a little bit about OpenAQ's motivation in our mission. I'll introduce the OpenAQ platform. I'll share a few cool uses of the platform in our community and I'll highlight two particular use cases involving COVID-19 and air quality. And then share what's next for our community and how to get involved if you're interested. I'll also hopefully end a couple of minutes early to take questions. But if I don't, please do post any questions in the conference's Q&A Slack channel and I will definitely answer there. So as I mentioned, air inequality is a huge issue. It's really one of the biggest public health and human rights issues of our time. Air pollution in general, indoor and outdoor, air pollution causes one out of every eight deaths on the planet. It's a huge issue. And one of the things we've noticed in community after community is a basic issue to tackling air pollution at the local level is access, basic access to air quality data. And when that access is enabled, a community can tackle, begin tackling the issue. And so a great example of this that we've seen was in Beijing about 10 years ago. About 10 years ago, the US Embassy in Beijing launched a little air quality monitor on its rooftop. It started sharing out air quality data via Twitter. Soon, third-party apps started scraping that data and suddenly millions of Beijingers for the first time were walking around with air quality information on their phone that they could see in real time. This created a huge stir and actually caused a series of actions from the Chinese government that improved open access to air quality data and also initiated many different policies, air pollution policies, both in Beijing and across the country in terms of monitoring but also mitigation pollution. And air pollution is still clearly a huge issue in China and in Beijing, but it has significantly improved year on year for the past about four or five years. And you really can trace a lot of that progress to the initial stir that was caused by this little bit of open data. And so what we have found in the OpenAQ community across the world are other examples as well where basically when you open up this access to data that communities can affect policy change. And so what we see as the challenges in many different locations is there's a lack of access to air quality data often in the most polluted places and that when data is shared, it's often across many fields besides air pollution but often shared in an inconsistent format from place to place or a format that's temporary. So on a say government website, air quality information shared on an hourly basis and it gets updated and that data is lost through the access after that data has appeared. And these factors prevent civil society from taking action to improve their air quality in their local community. And so that's where our mission comes in. We have a very simple mission which is to aggregate and harmonize existing disparate air quality data largely from government sources currently to put it into one format and make that data available in the ways that enable a larger public audience to use that data in the ways that their expertise best suits. And so essentially extracting more value, more impact from that data. So this can be data uses including app development, education, public health analysis, other scientific use cases, policy work, a wide array of activities. And so the idea behind this is that the platform, the platform harmonizes this disparate air quality data enables the convening diverse sectors to improve the coordination around building solutions. And in general, you're creating a healthier and more efficient data sharing ecosystem which really positions the civil society to better fight air inequality in local regions and across the world. So to briefly introduce the OpenAQ platform. So if you go to openaq.org, you'll see a world map of data from across the world and at various locations from currently 93 countries. This data is also accessible for download on the platform. Everything you see on the website and then our platform is open source and it's been created by an open source community. I link here to our GitHub page with the different repositories of our platform's code. And the data is made available in a few different ways. One is through the website I showed, but also through an open API and there's a link to the docs there. This is the same API that powers the website and anything that's built on top of our platform by ourselves. And then there's also the API and the website provide the past 90 days worth of data. We can also access that data via S3 buckets on AWS. So we see quite a bit of usage of data from the platform. We see on average about 33 million data requests per month to the platform. So this equates to about 400 million data requests per year and these come from all over the world. I will say there is a larger amount of data usage from the places where there is existing data. So where you see data gaps, you see less usage which makes total sense and speaks to the need of generating more data in places with currently without any. And so as I mentioned, our platform was really created and is still in large part contributed by an open source community who also use the data. So for example, an open source contributor in our community from Mongolia, a software developer, Dolgun, wrote a code to access data in Bosnia because there was a small group in two of the Bosnia who were recording data by hand from a website. Dolgun heard about this. He wrote the adapter that got added to our system. And then this group can now access that data much, much more easily than they were before. Or another example is a Swedish app developer wanted to ingest Swedish air quality information into his app. Contacted the Swedish environmental administration to adjust their API to provide all of the data and metadata that our system requires so that the data could be ingested. And as I mentioned, and I'll give a few examples of other community uses of the data, but there's been one popular example is Smokey the Air Quality bought Al on Twitter and Facebook and on airpollution.io created by a developer who was based in Delhi, India at the time. And it really takes the information that's shared on our platform, the raw information and helps present it in a very human way in a way that connects with the public in Mongolia. And so we use really the gravity of this massive data set to convene local communities across sectors and we've done this in cities across the world. In fact, David, the first speaker in this session was a participant in the workshop we held in Accra, Ghana. And the idea of these workshops is to get folks from across different sectors in a local area where there's an air inequality issue and find the low hanging fruit that everyone can get behind to take an action. So for instance, in Sarajevo Bosnia, the group, the participants there decided that while there was a lot of open air quality data, what was lacking was an understanding of how much pollution was coming from which sources. So the group, which comprised folks from a think tank, some scientists who could do the research, some air quality agency staff, they came together and decided to launch a field campaign on figuring out just that, how much pollution was coming from which sources in Sarajevo for the first time. And they're just wrapping up the work of that project now. So I'll mention a few community uses of the data that we have seen over the past two years, and also briefly mentioned that who our community is. So we conducted a survey a couple of months ago to figure out who is in the community. And you can see it's quite diverse, several different countries, but you do still see the domination of the community being focused on where there is a lot of access to air quality data already. And again, points to this data access gap where there's simply no data to really ingest publicly. So one use case that we've seen from the research community is using the air quality data to inform air quality forecasts. So a team at NASA, for example, they have an atmospheric chemistry model of air quality at the city level for various places around the world. And they can test how well their model works by taking the observation data they access from the OpenAQ platform in near real time to assess their model. Another example is the media. So data-driven journalists can access the information and make data visualizations that are relevant to their region or the world in this case. So Bloomberg News recently launched Bloomberg Green. And one of the pieces of that has been this air pollution visualization that shows air quality across the world. And they ingest data from the OpenAQ API to do that. Another example, this is not so much a data use case, but an example of how governments can be incentivized to open up data. Once you have a massive set of harmonized data in one place, is we were recently contacted by the Icelandic Environmental Agency to help them build their API in a way that would let their data be shared onto OpenAQ, which was awesome. So now that data appears on OpenAQ. And again, I think it speaks to the fact that if you do have data in one place in a harmonized fashion, it can get more data. So data can get more data and incentivize governments to open their data in a fully harmonizable fashion. So I did wanna highlight a couple of examples we've seen from our community recently around COVID-19 and air quality. You may have seen some articles in the news around air quality and how it's been changing with COVID-19 lockdowns and what this can mean. So one example we've seen from our community comes from India, from Kudakunda at Urban Emissions. There's been a fantastic series of analyses. I have a link there to the article. I can also post that in the Slack channel that shows an analysis of what air pollution, specifically in this case shown here, nitrogen dioxide was doing before and then during the onset of the lockdown period in Delhi. Nitrogen dioxide is a pollutant that is associated with transportation cars. And you can see a pretty marked decrease after the lockdown period. And so the reason Sharath's able to do this sort of analysis is because the data's available in a near real-time fashion and also in one harmonized place. And he's done this analysis for various pollutants in various cities across India. A really interesting preliminary analysis that's been done globally comes from this Norwegian group where they took satellite data and they also accessed data from the OpenAQ platform for 27 countries and did a more global analysis of the impact of lockdowns across the world. So they looked at the two-week period after a lockdown and compared that to air quality. Previous to that period, they also controlled for meteorology and other conditions that could be changing during this time. And they found over all pollutants and over the 27 countries, they looked at a 20% decline on average of pollution. And then they went a step further and linked that to some human health impacts as well. And what we can glean from this is sort of a silver lining to the COVID-19 lockdowns for air quality policy going forward. And so there's a link there to the preprint. I know it's small. Again, I can share it on the Q&A Slack channel if anyone is interested. But again, the reason this preliminary analysis could be done in such a timely manner is because the data's available programmatically and in a harmonized fashion. So that the people who wanna do such an analysis can get to doing what they do as quickly as possible and apply their expertise instead of wrangle data. So if you'd like to hear briefly what's next for the OpenAQ community and how to get involved, that's how we'll end this out. So what's next for us? We are really excited to be expanding our platform and really launching a pilot platform that will include low cost air quality sensing in our system. Right now, we primarily share government and some research grade air quality data. And one of the biggest demands we've gotten from our community is the ability to host low cost air quality sensing data. So we are not launching low cost sensors ourselves, but rather saying for low cost sensing initiatives out there that wanna share their data in a harmonized way alongside both other low cost sensing initiatives as well as these other government sources that we're building a pilot platform to do that. And one of the hopes is that this helps fill in those data gaps in regions where there's not government monitoring or there's not open sharing of that data by the government yet. And we're always working on various tools and light visualizations that help the community do more. We ourselves aren't building an app or sort of an end product, but rather building the tooling infrastructure that helps others do more with the data. So for example, we're working on a data averaging tool that averages spatially and temporally in our system so that you can get back the air quality average daily value in Beijing as opposed to the hourly level at the station level, which is what our Roth system provides now. So with that, I'll close out. I'll say if you would like to get involved in our community or want to hear more, we would love for you to reach out to us. I've had my email in the bottom right hand corner at chris.openacq.org for many of the slides. So please feel free to reach out that way or at info at openacq.org. We also have a public Slack channel that I'd invite you all to join as well where we have a bunch of folks in different sectors who share their interests around air inequality or share what they're building on top of the platform or ask questions about how to use the data. So with that, I'll say thank you again for everyone's attention and for the ability to present today. Yeah, thank you. It's really great. A lot of people in the chat saying wow and clapping along the way. So it's amazing work. So just real quick wanted to remind everybody who's listening there is a, the way the day works is there is another session, session seven that's starting. So another crowdcast URL that's starting in a couple of minutes. And so check your schedules, check the links and make sure that you attend different sessions throughout the day. But I do think we have time for one question. I mean, I was just thinking about myself. I work at the University of California and I know there are researchers that use this kind of data to like crunch and analyze against social science and humanities data to like kind of say like, what are the effects of air quality on ethnographic research or on different things? There's a project out of UC Irvine called the asthma files. It's like about trying to bring air quality into the discussion around social science. And so I was just thinking, talking a lot about public health, but then also like how the access to this information affects other types of research and like opens up so much more cross disciplinary research. And I just thought like, you know, your emphasis is on public health, but I just thought like, what are your thoughts around kind of how it affects and how it can affect the social sciences and the humanities as well? Yeah, I think that's a fascinating question. You know, I'm a physical scientist by training, but I think some of the aspects that have captivated me most about seeing how people are using the data or its impact have been around more the social science realm. I, you know, early on, we did a really light analysis of just seeing how, with access to these real-time air quality levels. And then if you look at Google searches that correspond in various regions during apocalypses, how interest changes or shifts from people Googling air pollution versus climate change actually. So I think it's fascinating to see how folks are motivated to change. The other thing that I'd point out is, you know, with the example from Beijing or several other examples we've seen, it's not that the science changed. It's not that what really changed was the way the information was presented to them, to people, to the public. And then the public getting activated on the issue was the social issue. It wasn't really this sort of technology issue. The data and the technology that lets that data be shared was, I guess, the key thing that needed to be in place, but it's really getting that data in a way that you can let the social impacts happen more so than I think anything else. Yeah, I love this trigger event too. That's like the Beijing came from, you know, a little bit of either intentional or accidental advocacy by the US Embassy there that then sparked this information, you know, ripple effect that created social change. And I think like the potential that you're creating with creating that same kind of ripple effect in other communities is amazing. So really great work. That's awesome. Thank you. Yeah, so I think that's it for time. So we have the questions for Krista that will be moving over to Slack and also slides are up on Zenodo and we'll see you in the next session. Thank you.