Data for Global Food Security Insights with Dr. Michael Humber, University of Maryland

(soft music) Today we're talking with Dr. Michael Humber from the University of Maryland about the NASA Harvest Program. Dr. Humber provides background on the challenging food insecurity situation around the world, taking a deeper dive into some of the unique challenges facing Ukraine, which we will explore during our upcoming event in April. We hope you enjoy it. Hi, and welcome to "Through the Human Geography Lens," a podcast brought to you by the Worldwide Human Geography Data Working Group, or WWHGD. I'm Terry Ryan, and I'm Gwyneth Holtz. And today we are here in the studio with Dr. Michael Humber, an associate research professor in the department, a geographical sciences at the University of Maryland, and the data lead on the NASA Harvest Portal. Mike, it's great to have you on the show. Thank you so much for having me here. I'm really excited. Yeah, we are too. So just to dig right in, what is NASA Harvest? And the Harvest Portal. Can you tell us about that? Sure thing. So NASA Harvest is a consortium operating out of the University of Maryland. And we're really focused on food security and agriculture on the world. And part of our key mission is that we're using satellite data and Earth observations to understand how propl and extent productivity and impacts of weather, climate, socioeconomic factors, and political drivers impact food availability and to a somewhat lesser extent price. And right now, that consortium consists of about 50 different organizations. We're into our six-year operations this year, so it's an exciting time for us. We get to refocus a lot of our activities and decide what we're doing over the next five years. And the hub, as we call it, based out of University of Maryland is led by our director, Dr. Molde-Mecher O'Sheff. But we have dozens of partners from different universities, NGOs, humanitarian organizations, private sector, farmers, co-ops, all over the world. And I'm sure we'll talk a lot more about this later, but a lot of the data and projects within harvest are really just so spread out that truth that I can't take credit for very much of it at all myself. It's there are dozens and dozens of fantastic researchers and technicians that are working on it. So as you can imagine, we do generate a lot of data and data sets and analyses and those sorts of things. And harvest portal is really where we just store all those data, make them accessible to the public, and make them usable from both within and outside of the consortium. So I know you're also really involved in the group on Earth Observations Global Agricultural Monitoring, GeoGlam, crop monitor for early wearing. Did I get that right? It's a mouthful, but yes, here you go. OK, it is a mouthful. So what data is included in that? And is that part of the NASA harvest portal? And what is that organization? Yes, so I guess maybe I'll do a little bit of storytelling first here, because we need to understand what geo and geoGlam are before we can start talking about the current environment for early warning. So when we say geo, we're talking about the group on Earth Observations, which is-- it's an international coordinating by that's kind of designed around linking together all the different Earth-observing systems, whether they're based on satellite data or ground measurements or sea level buoys, things like that. And that's an organization that dates back to the early 2000s. GeoGlam is the Geo Global Agricultural Monitoring Initiative. And that's sort of a subsidiary of geo and in a sense. And NASA harvest is actually part of NASA's contribution to geoGlam as an initiative. And geoGlam itself actually has kind of an interesting genesis story behind it. It was conceived by the G20 in response to food price spikes that happened in 2008 and 2011. And those were brought on by droughts and major wheat-producing regions. And those spikes actually more than doubled the cost of wheat at the time. And that leads to what we call price volatility. And when you have price spikes and volatility, those impact everyone around the world. It's not just the people that are growing the crop, but it's also the vulnerable populations that are ultimately importing, consuming that crop. So with that sort of backdrop, the G20 in 2011 issued a declaration that endorsed GeoGlam as well as what's called the Agricultural Market Information System or AMIS. And the goal there was to have these two groups work together so that we can increase transparency in agricultural markets where GeoGlam is really providing data on production using satellite data. And then AMIS provides some context for the market. So I know that's a lot of background, but getting into the crop monitor part. A couple of years after that, we had this sort of meeting where we decided that what we'll plan to do is gather together all these experts from different agricultural ministries, private companies, NGOs, other organizations in one place, and just ask them, how are the crops doing in your country? And most of the time, everyone sort of agrees that they'll look at their country and look at the other countries. And we more or less have a consensus on what the conditions are. But in some cases, they don't, and then we'll work those out. And then we compile all of that into report. And that gets issued in what's called the AMIS Market Monitor and part of that is the GeoGlam crop monitor, which has the information on the crop conditions. So once that gets pulled into a market monitor, they'll add more information on things like market conditions, fertilizers, things like that. And that's really how we try to get the information from the hands of the ag ministries and remote sensing people into the hands of the economists and traders and try to bridge that gap. So originally, we were working just AMIS countries. So it was G20 plus eight other countries that most impact global grains markets. And these are kind of places where if you have a bad year because of a drought or something like that, it could impact global foods of supplies and global food prices. But that's really only half the equation. What the Girling Warning Crop Monitor does is effectively looks at the rest of the countries, ones that if you have a bad season, it could actually impact food security and food availability in that country. So in that context, we started working with groups like FuseNet, the Family and Early Warning System Network, and other contributors to take that same approach and apply it to more or less the rest of the world. And that's really how the early warning crop monitor was born. It's a bit of a different emphasis because we're looking more at impacts and human impacts than just production itself. But the advantage of this type of system is that we can really bring in information from all over the world in one place. And one of our key contributors who was essential in getting the early warning crop monitor started, I learned, she's a big contributor in food security arena. He always reminds us that with the technology we have today, there really should be another family. There should never be another family that we don't see coming. There's no excuse for it. We have the technology to see it coming months in advance. And if we see it coming early, we can do something about it. And that's really kind of a spirit of the crop monitor is to get that information earlier so that we can do something about it. Wow. I'm glad you took us on that journey and provided all that history because I definitely didn't have that kind of context. So thank you for that. Yeah, I agree. So, Mike, could you tell us about some of the recent insights that you're seeing using these tools that you've been speaking about looking at global food insecurity, particularly surrounding the conflict in Ukraine? Sure. So Ukraine is actually-- it's been a lot of action from us on that. We've done quite a lot of work in Ukraine since the beginning of the war. And we've had a lot of our partner step up to provide data, resources, analysis. It's really been all hands on deck. And when you think about Ukraine, it's important to remember that Eastern Europe is one of the bread baskets of the world. It's where you have a lot of green production. You have wheat, barley, corn coming out of that region, sunflowers, particularly important in Russia and in Ukraine. And a lot of those grains that come out of there ultimately end up in Africa, the Middle East, and other places. And that's why when you think about the disruptions that are going on in the Middle East, you have to think about where the regions that are actually going to be impacted by that in terms of consumption, not just the production. A lot of it has to do with the fact that Ukraine is one of the main suppliers to the world food program. And then on top of that, sunflower oil is actually a major staple for food security. Sunflower is considered a food security crop because the oils used for cooking in many places. And in that sense, it's really the world's bread basket is one, not just the Eastern European bread basket. And it sort of took a while for mainstream media to catch on. But that was really the major story that once we start putting the pieces together, you can see that the impacts are much more far reaching than just Eastern Europe. But really, that's not everything that was kind of at risk. You also have to think about this times with the political context. So you also have sanctions going into Russia and Belarus, which is slowing the flow of grains out of those regions. You have fertilizers or a major output of Belarus and Russia. So Potash, for example, very important. And at the same time, you have the Canadian railways are starting to strike. So between those three countries, you have about 70% of the Potash market is tied up in some way. So that all feeds into production down the road. There might be fertilizer on the field now. But next year, the situation becomes a lot more murky. And some of this is starting to resolve now. But what we think about what we can do and what we can see with the data that we have, it was pretty clear early on that we needed to work at a higher resolution. And that's where Planet, for example, really stepped up and helped us get some high resolution imagery with very quick revisit times over all of Ukraine. And we start to look at that data alongside some of the more human-oriented data. So not just the satellite imagery, but things like information on supply chains, import export data, who do these countries normally export to? Naval blockades were something that were a really big deal, damaged ports, grain prices kind of went all over the place early on, and information on oil. So the cost of diesel was spiking, causing prices to increase all over the world as well. And these are all things that contribute to two-food security issues. It's access and price that are main drivers. So when we're looking at all this data, we're actually able to remove some of the uncertainties. So even with satellite data about having to be in the country, we're able to understand where different crops are planted, how much the crops are planted, and ultimately how much of that was harvested. So we're able to monitor those conditions throughout the season within the season, which is really quite difficult to do. And by being able to do that, we're able to get a better idea of how much of the crop that was planted was actually ultimately harvested. And there are reasons why you would not be harvested. The obvious one would be if there's a conflict that actively happening in your field, you're not going to go out and harvest. We're actually doing some detection of, we've been calling them shadows of things like unexploded organs in fields. So we can actually see the individual landmines and things like that that are in the fields that are preventing farmers from harvesting that. And there is still a lot of uncertainty in the numbers of the total amount that's been exported. But even in the season, we're able to see that the actual conditions of the crops were really not that bad. It was always just a question of how much was going to be available for export at the end of the year. With all of these data points, is there anything that surprised you? I mean, a few of the things that you just mentioned surprised me, but you're the expert here. Anything that you were surprised to see? I think when you start to look at some of these data, that's what we don't traditionally work with. You start to see a lot of interesting patterns. So for example, when you think about agriculture, especially during this season, you don't really think about fires. But in Ukraine, fire is actually a very important part of the way that they manage the agricultural lands. And if you look at something like the satellite-based fire data, you'll see fires spread out all over the country. But what we saw last year was that those fires were almost exclusively concentrated right along the areas surrounding the front lines of the conflict. And that's something that we haven't really -- there aren't many examples that you can look to to see something that clear and well-defined in the satellite data records. So that was one that was quite interesting. I think sort of having to work with this higher resolution things like the unexplored ordinance, those are things that we -- there's no existing pipeline for monitoring that for us. That was something that we had to sort of build. And then it also pushed us into working a lot more with economic and price data. So we've really started working in this sort of almost economics and trade framework that we weren't really too much into. We had started, but understanding the relationship between things like natural gas prices and fertilizer production and how existing trade agreements and sanctions impact the flow of goods from country to country, those were all sort of new things that, you know, the results are quite surprising in a lot of those cases. Another thing -- so you've spent a lot of time discussing all of these ripple effects or some of these ripple effects from Ukraine and the crisis there. And we're actually planning a W.W.A.C.D. event that's going to focus on the ripple effects of Ukraine. And what we're finding in our limited research -- I know this is something you're looking very closely at, but in our limited research, we're seeing that it's -- as time has gone on, it has been a little bit more difficult to determine what is happening downstream that is actually a direct effect of impact related to food security. Considering COVID-19 and climate change, all of these sort of systems problems seem to be connected. So can you talk about how your organization is thinking about these three different impacts the crisis in Ukraine, COVID-19 impacts and climate change? Do you think there's any value in decoupling them as, you know, different kinds of impacts or just considering them as a larger systems problem as we're looking to the future? That's a great question. So I think when you sort of look at these individually and decouple them to use the same term there, it simplifies the problem and makes it more solvable in a way. But I think that, you know, when you start to think of these systems and how they're related to each other, you really need to kind of think about them as connected and dynamic and ever changing it, especially now. When you look at what's going on in Ukraine and the sort of economic policy that's been spurred on by COVID-19 where you have, you know, the Fed is trying to get inflation under control and that's not just in the US, it's also your central bank doing the same thing ever around the world. You know, inflation is cropping up as an issue. You know, I think that those things combined, the war plus, you know, the economic impacts from COVID, it's really created a new political and, I think, to a greater extent, a new macroeconomic reality that we really can't ignore. These things are not happening in a vacuum. So, you know, we're looking at the Fed doing its thing with the rates. We're in person, we haven't seen in two decades. We haven't inflation that hasn't been seen in much longer than that. European Central Bank following the same pattern. And there's no real clear indication of when those rates are going to go back down if they're going to go back down anytime soon. And that means that we're entering these sort of decades long cycles. And when you contextualize that against climate change, you know, if we're looking at a 10 year, 20 year rate cycle, then really that's much longer than the time that we have to take massive action on climate change. You know, you can't be planning 20 years out for climate change because the reality is that action needs to be taken now. And, you know, I think that when you add in the backdrop of the war in Ukraine, which is, you know, directly affecting supply chains. And you also has these sort of longer term effects where you have these sanctions that are, you know, potentially the crippled economies for decades. You have new trade alliances being redrawn all over the place. We're seeing, you know, actions and changes in trade partnerships between country like China, India, Turkey, Syria, others. You know, these countries, their role in the global economy is changing. And, you know, what we really need to do is, you know, again, while it doesn't make sense to simplify these problems in order to be able to solve them, I think it would be a mistake to continue with a sort of business as usual model where we have the paradigms currently shifting and we're applying models from a previous, you know, reality. You know, a lot of what's happening out, it's really complex. It's really unprecedented. And especially if we just think about, you know, from the solid perspective, we don't have enough time series going back to find examples that we can look to and see what happens. So I think, you know, there are, there are opportunities to be flexible and sort of aware of how complex this system is, even if even if you are, you know, breaking into smaller parts, you have to be aware that what's going to happen in the next 10 years might be very different than what's happened in the last 10 years. Things really are, you know, different now. Yeah, that seems like it's just going to be so complicated to begin to think about what those models really look like. And speaking of some of these models and what the what the future holds, what are the biggest data gaps that you're seeing? Yeah, so there are two that come to mind when I think about data gaps. In particular, the one that is sort of immediate and people are working on now is what are called in-situ measurements or field data measurements. And these are ground observations that we use to calibrate and invalidate the models. And, you know, I think what's important to note about agriculture and crop on a particular is that it's very tricky because crop one changes a lot from year to year. It changes within the year. And you have a plant that wasn't there, then it grows and then it's in S's, you harvest it. So it's a very dynamic land surface to be mapping. But then you also have the year to year crop rotations, you have differences in the crops that are planted, so on and so forth. But then you also have the sort of slower evolving changes in seed technology. And that results in different ways that the genes are expressed in the field. So we need to constantly be feeding these models new new field data in order to make sure that they're keeping up with changes in how the cropping system is actually appearing from a satellite perspective. So that's a big one. Right now I can say that geoglam has taken sort of a leading role in starting to coordinate some of those field data collection efforts internationally and making sure that the data is being collected is having its value really maximized. And there's a lot of work being done to try to make that data collection process faster, easier, more robust, that we get more data faster with less costs. But I think the other thing that's sort of a big gap is when it comes to understanding supply chains. It has been getting a lot of attention lately. And to be honest, I think supply chain really is kind of the wrong term in a lot of senses, because what we need to understand is the food system has a whole, not the individual supply chains. And it's really more complex than you can imagine it anywhere from the seed and fertilizer production all the way through consumer actually using the goods and the waste management and everything in between it. It's a very, you can make it as complex as you want really. And I think that within food systems, the biggest gap that I see is data from the grain trade itself. So it can be quite hard to come by. It's difficult to manage because there's so much data that's generated. And if you do get your hands on it, it tends to be very expensive. And part of the reason for that is that the vast majority, something like 70 to 90% of global grain trade is actually controlled by four or five companies, depending if you want to go or not. And yes, it's really incredible how much of the grain trade is consolidated into those companies. And they have subsidiaries and things like that. But it's very difficult to get information out of those companies. And you need to have some that gets out the firm, feels of lading at the ports and things like that. That's a big blind spot. No one really has cheap, easy affordable access to that type of information. And that leads to a very big hole in the understanding of supply chain and food systems. Well, thanks for pointing those two out. What about maybe a new data set that you've recently run across that you just got really excited about that you want to share with a group that maybe others don't know about? I think. I know. I think there are lots of new exciting, not just data coming online, but methods in processing tools that are out there. So I guess to more directly answer the question, when we think about new data sets, and I'm a satellite guy, so I'm going to go that route. But we have a lot of new instruments that are coming out. You have private sector, public sector, both launching new instruments in unprecedented rates. And the past year, these national things like Landsat mode is very well calibrated, well built instruments that were sort of work courses designed to work for a number of different fields. Moving forward, starting to see more purpose built satellites because they can fit into that smaller form factor. So we're starting to see missions that are going up that are focused on just wildfires. We're seeing focus focuses on just visual things. We're starting to see things like light hour instruments, so I particularly shout out to Jenna, the Global Ecosystems Dynamics Investigation, which is a light hour sensor that was put on the International Space Station. As being used to very accurately measure global biomass, it really sort of an unprecedented precision scale. And I think those are all very exciting because they're new sources of data, new ways of looking at the services, but they're also because some of them are very specific and in their purposes. They are, they're also very good and very tailored to what it is they're doing. So I think there's a lot of new opportunity there. And then on the processing side of things, I'm sure a lot of people are from Google or Fengen and systems like that, where you have all the data in one place and the computer structure built alongside of it. So there are those advantages, but we also are looking at things like one of our partners at IFP is working with the University of Sheffield and in building a natural language processing workflow that allows them to automate the definition of different supply chain components based on scraping articles from the internet, whether they're academic or or blog type of articles. So I think those are a really neat ways of handling these sort of bigger data problems that are just as important as the data itself, because if you can get the data, but you can't process it, then it's not as useful. So I think those are really important aspects to it. That's fascinating. Thanks for sharing that. It is fascinating. Mike, thank you so much for joining us today. It was a great discussion. Please join us next time for another conversation on human geography and human security and through the human geography lens. Our human geography library has over 5,000 data sources and references, including data applicable on food insecurity from the Famine Early Warning Systems Network or FuseNet. The Crop Explorer from the US Department of Agriculture, Foreign Agricultural Service, and Hunger Map Life from the World Food Program. If you're interested in learning more about human geography and the WWCD, check us out at www.hd.org, where you can find the human geography library and access presentations and recordings from more than 50 data driven events. If you have interesting, openly available geospatial data to share, feel free to contribute your data at www.hd.org. I'm Gwyneth Holt and I'm Terry Ryan. Thanks for joining and we hope to see you again next time. We really appreciate your support. If you enjoy this show, please take a moment to leave a survey review in a rating on Spotify, Apple Podcasts, or your favorite podcast platform. And we hope you'll share the podcast with your friends on social media. Thanks again for listening. [Music]