Climate change represents the single most significant policy challenge in the 21st century. It is a multifaceted and global threat to society that affects a range of concerns, from the environment and food security to health and economic inequality. Recognizing the urgency of the threat, Mathematica recently established a climate change practice to help develop innovative and sustainable solutions that meet ambitious reduction targets for carbon dioxide emissions while also helping vulnerable communities become more resilient.
The guest for this episode of Mathematica’s On the Evidence podcast is Tulika Narayan, who leads the company’s climate change practice. Narayan is an agriculture and development economist by training who previously served as senior director for food security and agriculture within the company’s International research unit.
The focus on climate change stems from Mathematica’s long-standing commitment to marshal the best available evidence to confront the most serious problems threatening the collective good and improve people’s lives. To do so, we’re drawing on our technical expertise in research and data analytics and our subject matter expertise in areas like health, education, poverty reduction, clean energy, and sustainable agriculture.
On the episode, Narayan discusses the following topics:
- The importance of using data and evidence to understand and respond to climate change’s impacts on people’s health and overall well-being
- The role that big data and digital technology can play in improving our understanding of climate change’s impacts and how to address them
- The significance of organizations like Mathematica, with deep roots in social science research and data analytics, becoming more involved in informing climate policy
- The challenges and opportunities for communities across the world to share evidence and learn from one another when presented with problems like climate change or pandemics that don’t respect national borders
Listen to the full episode.
With the advent of data analytics, we can better tailor programs for highest impact. And also use the same information to evaluate if the programs are impactful in addressing climate change.
I’m J.B. Wogan from Mathematica, and welcome back to On the Evidence, a show that examines what we know about today’s most urgent challenges and how we can make progress in addressing them.
On today’s episode, we’re going to talk about perhaps the most urgent policy challenge facing humanity in the twenty-first century, which is climate change. And we’re going to talk about the role that data and evidence can or should play in confronting the climate crisis.
My guest is Tulika Narayan, an agriculture and development economist who now leads the climate change practice at Mathematica.
Climate change poses a unique threat to society in that it touches on so many different areas of life. From the environment, to food security, to health, to economic inequality. And it has a global reach. Such a multifaceted problem demands innovative and sustainable solutions that simultaneously meet ambitious reduction targets for carbon dioxide emissions while also helping vulnerable communities become more resilient.
While climate change is a relatively new focus for our organization, it stems from a longstanding commitment to marshal the best available evidence to confront big, thorny problems threatening the collective good and ultimately to improve people’s lives. To do so, we draw on our technical expertise in research and data analytics and our subject matter expertise in areas like health, education, poverty reduction, clean energy, and sustainable agriculture. If we needed any reminder of the urgency of climate change as a problem, Tulika and I spoke in the fall of 2022 on the heels of one of the hottest summers on record in the United States. About a week before our interview, the Associated Press reported that record-breaking floods in Pakistan had killed more than 1,700 people and affected another 33 million while causing $40 billion in damages. Only a year earlier, the World Health Organization declared climate change the single biggest health threat facing humanity.
In our interview, Tulika and I discuss what it means for researchers and evidence-driven decision makers to think about the implications of climate change on people’s health and overall well-being. We discussed the role that big data and digital technology can play in improving our understanding of climate change’s impacts and how to address them.
We touch on the importance of organizations like Mathematica with deep roots in social science research and data analytics becoming more involved in informing climate policy.
And finally we talk about the challenges and opportunities for communities across the world to share evidence and learn more from one another when presented with problems like climate change or pandemics that don’t respect national borders.
A full transcript of the conversation is available in the show notes and on Mathematica.org. I hope you find the conversation useful.
Tulika, there is great consensus that as we find solutions to reduce carbon emissions, we’ll have to come up with solutions to address its knock-on effects as well. The knock-on effects of a changing environment on people, from access to food, to heat-related illnesses, to the contraction or expansion of entire industries. So, for a research organization like Mathematica, which has a long history of studying the effects of programs and policies on human well-being, what does it mean for us and our work that policy leaders are increasingly concerned about climate changes’ effects on people?
You ask a very important question, and as – and as you know, Mathematica has been working with agencies in the U.S., and indeed across the world, that are in the business of providing key services to communities. And that is all guided by our mission to have an impact on well-being of people worldwide and our focus particularly on driving an equitable and just world where evidence drives this decision. And the impact that climate change is having and is expected to have on people globally, with large inequities in how the impacts will be felt and vast disparities. And in the abilities of communities and countries to adapt to it. Our role studying the effects of these programs and policies on human beings becomes immeasurably relevant and insightful.
For example, you know, our longstanding work in supporting the health and human services, including the Centers of Medicare and Medicaid in the United States means that we have in-depth understanding of the populations served by these agencies in the program, their access to safety net services and their vulnerabilities. And because we manage a lot of these datasets, we have the ability to take the information from these datasets and translate that in understanding how climates might impact them. For example, we developed the ClimaWATCH tool that combines data on temperature and dew point data on social vulnerability and racial composition on the Medicaid beneficiaries. And this tool helps to understand how heat waves would impact communities, who is most susceptible, by what types of diseases and the eventual impact of those illnesses.
So, these are, this is just an example of the way in which the work that we do can really bring important information to bear that policymakers can use in their work.
And that – that is speaking to the implications in a domestic context, but I know you’ve done a lot of work on international policy research, so I’m curious, what does it maybe mean for evidence on a more global scale or in countries outside of the U.S.?
Yes, we are. We are – I have also been doing a lot of work in the international space, and our work has been with many donor agencies working on the impact of their programs on communities of interest. And, you know, in the international space, climate change’s impact on communities was complicated by the fact that there are many market imperfections. So, let’s take the case of the ClimaWATCH tool, which has been developed in the context of the U.S. domestic space.
But if you are trying to understand the impact of climate on health, in the developing country context, access to health care is poor, implying that this is not just a lack of data that limits our ability to measure health impact. It is that the same stressors have a complex area of impact on people. In developing countries, for example, access to insurance is poor, the large amounts of labor force are in the informal sector with low job security and reliance on daily wages. Which means that any sick days can translate to food insecurity with compounding impacts on health. And rural populations in developing countries depend on agriculture as a key source of income. So, which means that the climate with the stressors such as heat are not only impacting your health and well-being, it’s also possibly affecting your yields, and that can plunge them into food insecurity, make them more susceptible to diseases, and even mortality because there is poor access to health services.
You have already touched on so many different – I mean, it seems like this would be a really exciting area of research to be looking at, the knock-on effects of climate change and to be looking beyond just, you know, does something make carbon emissions go up or down? But I imagine that also might present challenges. Are there – are there challenges with trying to measure the social and health effects of climate change or policies designed to address climate change?
When you actually mentioned and asked me a question and talked about reduction in greenhouse gas emissions, and that is speaking to the work that has to be done in addressing mitigation, and the piece that we were opening and having this conversation about was around, what is the knock-on effect on people, which is essentially this issue of adaptation. How can we make people more resilient to climate and climate stressors? And on, well, there are two things I would say. One, when we are looking at overall addressing climate change and we kind of simultaneously have to work on both mitigation and adaptation interventions, a place where sort of we come in is to really help policymakers understand what tradeoffs there might be when they are looking at interventions. What is the impact of their interventions in, say, improving adaptation that could also influence its contribution on greenhouse-gas emissions?
For example, with the USAID Bureau of Africa, with the sustainable agriculture team, we were developing a decision support tool that not only took a look at the climate forecast to sort of look at what is the predictive impact on yield of the new technologies that by looking at – now, these are technologies that are meant to make the farmers more resilient, they are meant to be crops that are more drought tolerant such as millets.
We were looking at what that would look like but also simultaneously understanding what will be the influence of those interventions on mitigation itself because these crops were also meant to provide better fodder for livestock and livestock, as you may know, is associated with greenhouse-gas emissions. Now, this is just a small example of the tradeoff that can exist between technologies that are providing more resilience but at the same time may be associated with better emissions.
Another example would be where are you coming in with investments in irrigation, and if investments in irrigation leads to crop choices such as better cultivation of rice on larger – larger tracts of land which are also associated with greenhouse-gas emissions.
So you have to sort of balance these things, and I am hoping that we are able to look at, more broadly, the dual impact of interventions and how they – and how they will affect each other so that we can find win-win strategies for addressing climate change.
Does it demand more data, more data sources, and a more sophisticated ability to synthesize or interpret the data from these different sources? Because it seems like you’ve got – there’s the emissions data, but then there may be other things you’re looking at like economic productivity of the agricultural sector, which seems like that would require a different skillset or subject matter expertise or – or familiarity with the different dataset than just emissions. Is that – is that going to be a part of the challenge going forward?
That’s an important question, and – and you are basically putting your finger on this general idea and notion that as we go from supporting work with our domestic clients, where data are rich, there is huge, large administrative data, the kind of data that we are able to access in developing the ClimaWATCH tool and the other work that we do. So – so that is accurate. But the way that I come to it is not what is the data that I need to do the analysis, but here’s all this data available, and what can we do best to utilize and leverage what we have?
But overall, where we are headed is actually there is so much more rigor than we ever had before. And as donors are engaging and over time there has been a lot of focus on that, health programs, other sector programs, are increasingly gathering data on service delivery. And at the same time, data innovations are ongoing that combine big data and also new analytical techniques, such as machine learning to study that data.
You know, just a few weeks back I was at the SatSummit that brings together leaders in satellite industries and experts in global development. Do you know that there are more than 2,000 satellites circling the earth, and there are ever more numbers of those planned in future years. And as you know, they have also been commercialized and there have been pathbreaking innovations. You know, some satellites are as small as the size of a shoebox.
So, the costs of the data generated by these satellites is decreasing and they are providing more information at higher and higher resolution. And these satellites have the power to bring data to data-poor regions, regions that are hard to reach, regions that are insecure. And satellites have been supported to use early warning systems in plant formations to measure crop yield pixel by pixel. And Mathematica, as you know, is using the, kind of, remote sensing, you know, data from these satellites, just to estimate crop types and crop yields in our projects, including in Niger and Burkina.
So, this is just an example. Overall there is a sea change in the data available, and with advanced analytics applied to it, it can begin to map the earth along many socioeconomic factors, agroecological factors, factors that influence disease prevalence, factors that influence adaptive capacity of communities. And what that – what that means is that with the advent of these analytics we can better tailor programs for highest impact and also use the same information to evaluate if the programs are impactful in addressing climate change.
I was thinking about the timing of this interview, and I feel really fortunate that we’re talking in 2022 and that we’re – we’re talking as you’ve come into this new position overseeing Mathematica’s climate change work because I think it’s a really interesting moment for the – the country and maybe the world in terms of how it – it deals with climate change. And I’m thinking about a couple of things. Like in the past couple years Congress passed the Inflation Reduction Act and the Infrastructure Investment and Jobs Act, which both have provisions that are intended to address climate change. And then for the first time ever the White House has a national climate task force and an Office of Domestic Climate Policy. And then there also are major donor agencies such as USAID, the Millennium Challenge Corporation, and the Rockefeller Foundation, just to name a few of those agencies, they’ve all committed themselves to addressing climate change.
And then internally, even here at Mathematica, our staff have identified climate change as the most pressing problem in the next decade. There was a staff survey that surfaced that insight. And that has led us to focus more on how we contribute our evidence-based expertise to combatting climate change.
But for all of the activity occurring now, when I was prepping for our interview, I was looking at your resumé, and your resumé is a reminder that it’s not as if social science research about climate change is new. I mean, I saw, just to pull up your resumé, I saw some fun early publications that I would love to read that were about like the environmental impact of wood preservatives, or the costs and benefits of a container recycling rule. And those are, you know, early – early naughts research. And then even just like 10 years ago, research on mitigating greenhouse gas emissions from the rice and livestock sector in Vietnam. Which is all to say that what seems to me that is new is the technology, as you were referring to, like those satellites, the – the increasing number of satellites proliferating around our – our – our planet. And the analytical capacity to do something with the data that comes from that technology.
So, when thinking about the impact of technology, what are social scientists able to do today that would not have been possible, or at least would have been more challenging, for a researcher like you, a couple of decades ago?
Yeah, and I think that I talked to you about, of course, new data sources that are coming in. And, you know, one of the things that have – has also happened recently is the increased access and use of mobile data. These are all examples of ways in which technologies are advancing in a way that it’s creating passive data. Data where, you know, you’re not doing anything but data is getting generated. So, as people are talking on their phones, you know, their mobile – mobile towers are capturing that activity. And using all of that data, you can track economic activity. You can understand how certain areas and locations are being used. And there’s a lot of innovation going on in accessing that kind of data to do what is called gap filling and understanding economic indicators.
So, for example, recognizing the fact that it is actually more data available than we think, one of the innovations that we’ve been working on is what’s called a resilience platform which recognizes that there’s information from multiple sectors that influences our ability to be adaptive to the climate, our ability to provide sources that can make you more resilient to climate. So, what we are doing with the resilience. platform is we are bringing all the data that are available in a certain context that will be important to program for climate adaptation. And we are also making sure that this information is made available to the communities, to the local governments and agencies that will be planning these interventions so that we can support locally that adaptation.
These, the resilience platform, also includes climate forecasts and the expected climate stressors that are – that are likely to prevail.
And when you take all of this data and you start doing analytics, you can also start looking at some outcomes using predictive analytics. What is likely to be the yields now – impact of the climate stressors? And what influence could that have on food insecurity?
And added to that, if we can include ways of rapidly and with agility measure resilience itself, we can better target and tailor the program into climate adaptation and resilience.
As donors are working to improve climate adaptation, it is vital that we can measure rigorously and accurately if these investments are having an impact. And this is where organizations like Mathematica have a role to play to be able to generate these very valuable and rigorous measures of impact. And this is how resilience platform can support such investment and support such donors in making sure that they are getting the return of investment that they are expecting and simultaneously impacting the communities.
Okay. And you were talking about the resilience platform earlier. I was just curious, how are we defining resilience in this conversation? Is there – is it – is it – it sounds like it crosses multiple domains like health and environment and maybe economic well-being, but are there – yeah, could you give me a couple of examples of the sorts of things that make up resilience and then how that would relate to the sorts of data that would be informing the resilience platform?
I would use your question to say for listeners who are listening to this podcast today and who are really curious, to come to our webinar that we are hosting on December 6 on exactly this topic.
On how to measure resilience. But I do want to say that this is an important theme because resilience has to be defined as resilience of systems, energy systems, resilience of energy systems to be able to provide power, and a resilience of food systems. So, resilience is a concept that describes systems in general that support community well-being. But it’s also a concept that defines communities and individuals. And there is actually several approaches to measuring it. And in this webinar we are hoping to talk about those different approaches and talk about how you can use the right approach and have a framework to use the right approach depending on the need and context. So, I hope that many of your listeners can join for that webinar.
Yeah. And we’ll include a link about the webinar in our show notes.
So, I have a silly question about data and how we can act upon that information and listeners a little bit – learn a little bit more about me. I have one of those smart watches, like a lot of people, and it gives me daily readouts about my sleep from the night before. And the app I use mostly tells me two – two types of information. How much I slept and what the quality of my sleep was. And occasionally I’ll get, you know, a tip about how to sleep better, like, hey, J.B., don’t exercise so close to bed – to your bedtime. But I really would like more of that latter type of action-oriented information especially when I’m, you know, having a stretch of bad sleep. And I’m wondering if there is an analogous situation with climate change data. Like, we have lots of information quantifying the problem with hotter average temperatures or worsening air quality. But at least, you know, in my – my intake and digestive news about science and social science, I don’t see as much about the steps we can take to remedy the problem. So, for example, you know, which evidence-based interventions are most effective at reducing carbon dioxide emissions. Do you – I don’t know if you agree with that take or not, but do you see a knowledge gap around solutions to climate change that evidence-driven organizations like Mathematica might fill?
Yeah. That’s an interesting question, but before I take your question about does it have an implication or direct application on interventions that are reducing carbon dioxide emissions or mitigation options. I do want to say that even in that space that you’re talking about, you know, when should I, you know, how did I sleep and so on and so forth, there is – there is, as you know, there are early warning systems around heat and – which are really important to address adverse impact of climate change. Even there, while I would not say there is a knowledge gap, but there is still a lot more refinement that analysis can help with. For example, some of the analysis that we can do and the ClimaWATCH tool can help assess that when there was a heat stress, were there adverse impacts of those heat stress more potentially felt on certain types of communities and certain types of demography? And from there you sort of can uncover in some cases that the communities that were not English speaking were more – more adversely affected. So, does that mean that we need to communicate these early warning systems particularly in those target areas with different languages? This is just a very fine-tuned example of how analytics can really have an impact in your getting to the point of intervention, getting to the point of last-mile delivery.
Okay. So, even – even just in terms of defining the problem, there is still work to be done so that data can still help.
Data can still help. And I – I definitely come from the standpoint where overall we don’t have dearth of technologies, we don’t have dearth of solutions. I think the biggest impact that analytics can do is in sharpening our ability to make those technology and solutions more accessible to where people live. The last-mile gap still remains the biggest challenge. And even so when you come to the mitigation options. Is there a knowledge gap that we can fill up with the most effective ways of reducing carbon dioxide emissions? I would – I would say that overall, you know, the biggest challenge is the politics of it. The most impactful solutions are, in fact, in front of us. But, again, over there also what we can do is shed more light. Analysis can shed more light and understanding that, for those solutions, what are the behavioral questions that we might face? What are the distribution impacts of intervention? Are there pure winners and losers? What is the impact on women and youth? And if you can lay out those – those details more clearly, then we can already find where are the places where there is going to be a socio political impediment to adopting it.
We’re talking in the United States. We do – we do research in more than 50 countries. And occasionally we encounter situations like the COVID-19 pandemic where people across the globe, across different countries, share common challenges and need similar solutions. And climate change strikes me as another problem, or maybe the problem, where communities in different countries can learn from each other. So, how are you thinking about the potential for climate research to inform decision makers here in the U.S. and outside – outside the U.S. in other countries?
That is actually a topic that is very central in my mind, the idea of sharing information, learning from each other, identifying ideas that are scalable. And, in fact, one of the ideas that we are working on right now is actually creating a global social good and the starting point for this that we are – we are talking about is to simply say that let’s bring in all the data that are available for a specific outcome. Say, for example, you want to reduce the impact of climate on health outcomes. This is – to really assess that, this does not require data only that related to climate and health outcomes, but it actually requires data on all sectors that might mediate the impact of climate on health. Such as, as I explained earlier in the talk, more of a dependence of the communities on agriculture. What might be the effect of climate on – on yields? And – and really be able to create a map of communities at the lowest resolution possible on the key factors that might influence the impact of climate on health. And with that foundational understanding, then we can start to target the communities that look similar to each other to create and tailor interventions that are a good fit for communities that have the same kind of mapping or fingerprint. And – and to then be able to also use that to eventually evaluate the impact.
And if that becomes the foundational basis of sort of programming that we can begin to, then scale up the programs that work well to other areas that have the similar kind of attributes. So, I’m very interested with this idea and working to see if we can bring it to fruition in – in some regions and then apply it more broadly.
Tulika, thank you so much for taking the time to talk with me today. I’m really interested to see how the climate change work at Mathematica progresses and evolves. But also I’m hopeful that in general we start to see more progress nationally and globally around addressing climate change.
Okay, thank you.
Thanks to my guest, Tulika Narayan. And thank you for listening to another episode of On the Evidence, the Mathematica podcast.
In the show notes, I’ll include more information about Mathematica’s recent work on climate change. A full transcript of the conversation is available on the blog associated with this episode at Mathematica.org.
To keep up with future episodes of the show, subscribe wherever you listen to podcasts. Or follow us on Twitter.
I’m @JBWogan. Mathematica is @Mathematicanow.
Learn more about Mathematica’s interdisciplinary climate change practice.
Learn more about a webinar hosted by Mathematica on measuring climate resilience.