Transcript:

BRAD NEWBOLD 0:00
Hello, everybody, and welcome to We Measure the World, a podcast produced by scientists, for scientists…

MASON STAHL 0:07
But what’s really makes this a particularly challenging question is that when you look at the chemistry of the rice, right, it is much, there’s much less clear spatial patterns to it, there are some. And it’s much less, at least at first glance, when you just look at say something like arsenic in the soil versus arsenic that ends up in the rice grain, there is a very weak relationship, and it actually in our data so far, looks like a negative relationship. And so like, one’s intuition might be like high arsenic in the soil, high arsenic in the grain. And so that we actually look it looks like there’s at least, you know, in our, in our, you know, 80 plus fields that we’ve sampled, the relationship is weak, and it actually looks negative. And so that suggests right that there’s, it’s not, it’s not going to just be an easy answer.

BRAD NEWBOLD 0:57
That’s just a small taste of what we have in store for you today. We Measure the World explores interesting environmental research trends, how scientists are solving research issues, and what tools are helping them better understand measurements across the entire soil plant atmosphere continuum. Today’s guest is Mason Stahl. Mason is an associate professor in the environmental science policy and engineering program and the geosciences department at Union College. His research spans the fields of hydrogeology, geochemistry and water resources. And he studies how groundwater and surface water quality respond to natural and anthropogenic changes to the environment, and what this means for the health of people and the environment. And today, he’s here to talk about his research he’s conducting in Cambodia, examining how hydrology and climate affect the uptake of naturally occurring arsenic into rice. So Mason, thanks so much for being here.

MASON STAHL 1:49
Thanks so much for having me. I really look forward to the conversation.

BRAD NEWBOLD 1:53
All right. So first off, we usually like to get a little bit of background information from our guests. So we just like to know how you got into the sciences in general, and especially getting into environmental engineering and environmental sciences?

MASON STAHL 2:07
Oh, perfect. Yeah, so in college and undergrad, I was actually a major in math and an economics. And part of the reason for that was I was interested in both of those subjects. But I really knew I wanted to go into science long term for my career, but I just couldn’t figure out what science I wanted to go into. And so I chose math, because I thought it was, you know, it’s a universal kind of aspect of all sciences. And so I was taking math, really liking it, but knew that wasn’t the long term career I wanted. And sort of by chance, I just lucked into it. In my first term of my senior year, there was a class being offered on groundwater hydrology, and so on a just a kind of total whim, I took it, and immediately fell in love with that class. So I was really lucky at Grant Garvin at Tufts University. And, you know, he’s a kind of one of the, you know, he has a really like, illustrious, you know, history and impact on the field of hydrogeology. And so I was lucky to take a class with him, and that inspired me, and he really took me under his wing. And then I took a bunch of other geology classes throughout my senior year, and then immediately decided, like, Okay, I want to go into water resources. I love this field, I’m really excited about the environment and in particular water resources. And so then went right to grad school after. And then I consider myself really fortunate again, in that I got to study under Charles Harvey at MIT. And he was is a fantastic scientist and advisor. And so when I got there, I didn’t, I just knew I wanted to do something in the field of water resources. And he had a range of different projects. And he’s a hydrologist but really kind of a scientist who covers, you know, all different aspects of the environment. And he had a project on groundwater arsenic in Bangladesh. And that really stuck out to me because I was excited about kind of seeing different parts of the world and blending, you know, bringing together my interest in the environment, but also an interest in kind of science that had a clear and direct impact. And so, for my PhD, I worked on arsenic and groundwater and was really focused in Bangladesh and also in Vietnam. And so I lucked out to and that I got to work with a number of colleagues at other institutions in particular, I got connected with Ben C. Bostick, who was at Columbia University, and we did a lot of work when I was in, he was a faculty member there, and he helped mentor me on a lot of my research in Vietnam, on arsenic and groundwater there. And I’ve continued to work with all of those got those folks since. And so just sort of fortuitous. I took a class on a whim my senior year and undergrad and then have followed through kind of continually.

So how was your first trip? Had you traveled much before you just took on this research trip in Southeast Asia?

I had travelled a little bit, but never never to Asia. And so that was a really new and exciting experience. And Bangladesh in particular was a totally new world. And so it was really eye opening just from from a number of different perspectives. But also, one key thing was just seeing how water resources and how I mean water supply differs across the world. And so, in Bangladesh, it’s really common, and a large fraction of the population is on what are called tubewells. So just the shallow kind of drilled well, you know, two or three inch diameter well, and it might have a hand pump on it, and they collect water that way. And so that differs a lot from what we tend to see in the US and in Europe, you know, a lot of people in the US are on domestic wells, but it’s, it’s a bit different of a situation and many more people are on, you know, public supply, which is treated. And so it’s really interesting to see that and just to see the environment, and how people are getting the water and the issues that are associated with water supplies that are a little more decentralized, that we don’t always face in this country. Because we have often regulation that is, especially for public supply. And so yeah, that was a really eye opening experience, to kind of see the different environments and how people are supplied with kind of one of the most critical things in our day to day lives, which is our water.

BRAD NEWBOLD 6:36
So before we get too much into the weeds, and into the interesting research topics, you are our first guest that I can remember who is involved directly in Environmental Engineering. And so we were wondering if you could just give us a little background into that field, how it fits in with environmental sciences, whatever you’d like there.

MASON STAHL 6:55
Yeah, sure. So there’s definitely a lot of overlap. And you know, so in Environmental Engineering, there’s kind of a wide range of different specialties and disciplines and kind of different hats that people wear. So the field that I’m in hydrology is often a really cross cutting field. So you’ll find hydrologists in environmental science departments, and environment, in geoscience departments, Environmental Engineering. And I think what really kind of ties that together is a few things. So much of my research, and a lot of hydrologists are looking at how human activity right impacts the environment. And in particular, trying to understand, right, you know, how to mitigate negative impacts to the environment from human activity, or how to remediate problems that we’ve had, you know, so think about something like legacy contamination from mining or industrial activity, there’s plenty of people in hydrology and environmental chemistry are really concerned about that. And trying to think about, well, how do we bring, you know, restore the environmental quality? And so Environmental Engineering, often, I think, while there’s, you know, it bleeds into many different fields, I think, a key aspect is thinking about, you know, how can we develop better technologies, techniques, approaches to ensuring that we can supply clean water, clean air, clean food, you know, clean soils to people. And so it brings together a lot of different sciences, chemistry, biology, you know, physics. But there’s often that kind of applied or practical approach to it too, which is, you know, about supply, ensuring that we mitigate environmental exposures and minimize the exposures to harmful chemicals in the environment to people, and then also how we remediate and improve. And then there’s a whole range of other things that an environmental engineer might do. But that’s kind of the I would say, the big picture of, you know, the connections of environmental engineering and environmental science and geosciences.

BRAD NEWBOLD 8:57
Do you get a lot into? I mean, as you’ve talked about, dealing with interdisciplinary groups and teams. Also, along with that, I would assume I mean, you’ve talked to you just kind of mentioned in passing about about issues with regulation, which may or may not be available. So then I would assume that this might get into implications on public policy and health safety from a a, you know, local to, to national government level. Is that right?

MASON STAHL 9:23
Yes. Yeah, that’s right. So a lot of our work. A key motivation is we see okay, for instance, I’ll go back to the arsenic in groundwater and I continue to work on that. And actually, we have a project right now, where we’re looking in the US northern plains, so South Dakota and Nebraska, and looking at arsenic and and uranium and surface water and groundwater there. And kind of the key trigger when we think about what considering exposures, but we look at what regulatory guidelines are and so for thinking about what percentage of the population is exposed, how significant of a problem is this. And so like the EPA for arsenic for instance, that’s a guideline of 10 parts per billion or micrograms of arsenic per liter of water. And if you’re on public supply water, and it exceeds that, right, then, because of the Safe Drinking Water Act, right, you need that municipality needs to come into compliance and would have to, you know, treat the water to try to bring it into compliance to bring it into compliance with EPA guidelines. And the World Health Organization similar, similarly sets guidelines for, like contaminants in drinking water. And they have the same guideline for arsenic that the EPA does. And so when we find places where there’s, you know, an exceedance, you know, we start to ask a couple questions, right? So we say, Well, why is this happening? So, is this a naturally occurring contaminant? Or is this something that’s related to human activity like say, you know, industry or mining, some kind of contamination like that, and sometimes it’s a, it’s a mix, it might not be that the contaminant is introduced by people, but practices that we do, for instance, maybe agricultural practices, or earrings like, or how we irrigate or pump ground or pump water might impact the movement of contaminants, or the mobilization of even naturally occurring occurring contaminants. And so, we often, you know, we’ll try to say, Well, why is this happening and get to the kind of fundamental mechanisms and science but then a really crucial thing, and we’re doing this right now. And in the Dakotas. And in Nebraska, we’re trying to say, Well, can we make predictions based on say, environmental variables to say, here’s an area that we predict would be high risk? Because we can’t, of course, have a well for sampling water in every single location on Earth, right? That’s not practical. But we might want to go and install new wells when we’re thinking about developing areas, right? And we’d like to know beforehand, if we can, this is a high risk area. Or maybe we have places where we know people are currently drinking groundwater, but we’ve never tested it. And so we’d like to know well, is this an area where we should really focus our our next step stages and testing and kind of put up a red flag and say, hey, this might be an area that we’ve we’ve never measured it before the water quality, but based on, you know, our predictive models, this is an area where we might expect really high concentrations. And so we try to understand the underlying mechanisms so that we can make better predictions and tease out areas that are higher risk and lower risk. And then also you can make estimates of exposure to by bringing in, you know, populations to that. So that’s, um, I yeah, sorry. So hopefully that kind of gets to the question you’re asking there?

BRAD NEWBOLD 12:38
Definitely, definitely. Let’s get started with your project in Cambodia, you talked about looking at arsenic in the water as it is then. Let’s say what’s, the word, uptaken? But, but you’re looking at at arsenic in the groundwater as it gets into rice, and how that affects human health and the environment, other things like that. So could you give us a quick, you know, abstract of that research project? What are what are you exploring? You know, where, what hypotheses are you testing there?

CHRIS CHAMBERS 13:07
And also like, like, why arsenic? You know, you think about all of the things that are in our environment these days that are problematic, or, you know, and we can we can circle back to this later, too. But what, what are the big consequences in the environment and human health?

MASON STAHL 13:26
Oh, yeah. So great, great question. Okay, so I’ll start with why arsenic, and then I’ll jump into discussing our work ongoing in Cambodia. So the kind of the question why arsenic is a great one? And the answer the the reason is, and I’ll give a little history here. So exposure, when we think about human exposure to environmental contaminants, arsenic is one of these environmental contaminants where we’ve actually had a mass exposure globally, and in particular, in South and Southeast Asia, to naturally occurring arsenic, right. And so, as you as you brought up in your question, there’s this wide range of naturally occurring chemicals, industrial chemicals, and that we can find in the environment, but in the case of arsenic, so going back about now, well, almost 30 years, there started to be cases that people presenting with what looked like arsenic poisoning in West Bengal, India, and in Bangladesh. And so, just to kind of give a little context here, historically, India, Bangladesh and many other countries, right, relied on surface water for their drinking water. And so that’s, you know, often it’ll be low in arsenic, naturally occurring arsenic. But the problem is with surface water, especially when it’s untreated is that they can often be very high in pathogens. So you might be drinking from a river where wastewater is discharged. People are bathing animals are going into it. And so you consume that water and you can get serious gastrointestinal Disease and exposures like that. And those can be really deadly. Right? So people, you know, diarrheal disease and can can really increase his mortality in this room. So it’s a it’s a huge health burden. And so there was a big shift this to groundwater. And a lot of those countries because they were abundant in groundwater, typically, it was actually really easy to put in a well, so for the case of Bangladesh, the Bengal basin, that Bengal Delta, right, it’s these three large, some of the largest rivers in the world coming. And this is a nice flat floodplain, and it’s sands and silts and clays that are deposited. And you can actually just drill in really easily with this super cheap, actually human powered rig where you’re just kind of like raising and lowering a pipe and putting a well at 50 feet, and pump out groundwater. And so the impetus was, well, people are dying from diarrheal disease, and it was a really serious disease burden there. And so they switched over to groundwater. And they have this cheap, abundant, easy to access groundwater. And groundwater tends to be much lower in pathogen so when water flows through the soil, and sediment tends to filter out these bacteria and viruses. And so you’re often like you can drink it and without treating it pretty safely. Now the problem is right, you might have what we call geogenic are naturally occurring contaminants can be present. And so these are elements, right contaminants that can get released from weathering of the soils and the sediment sedimentary material. And so at the time when they were doing this going back more than 30 years now, but it was kind of came to light that arsenic poisoning was happening about 30 years ago. But going back many decades, they’ve made this switch over to to groundwater, and mortality from diarrheal disease went down. And now the problem is, when you drink water with pathogens in it, you quickly find out that you’ve had that pathogen within a few days, right? You know, they’re sick. If you’re drinking water with arsenic in it, there is there’s no taste to arsenic in water, there’s no smell, you can’t see it. And so it wasn’t always super common to test for arsenic and water. And people weren’t really thinking about this being you know, contain a naturally occurring contaminant. And so you drink it. And you don’t notice, after a day, you don’t notice after a week, you’d probably don’t notice after a few years. But what’s happening is this chronic exposure leads to a whole host of diseases, so cancers, increased risk of diabetes, heart disease, cognitive impairment, and then ultimately, you’ll start to see if you have really long term chronic exposure sores actually developing, you get this condition called keratosis. And you get these like kind of black sores on your hands, soles of your feet, your chest, and then you know, and so that’s what started to manifest in West Bengal, India, and Bangladesh. And then people started saying, Oh, this looks like arsenic poisoning. Why is this happening? And people started testing the groundwater and it came out that okay, there’s a really kind of unfortunate confluence of sort of events here is that the geo, the geochemical conditions in the sediments there. In the aquifers, are what we call reducing, so they’re really low in oxygen. There’s a lot of organic material. And when that happens, you can break down iron oxides, naturally occurring iron oxides. So microbial processes will do this. And you dissolve iron into the water, iron is not toxic. But oftentimes the iron has associated with it arsenic and in the solid phase. And so when you dissolve that iron into water, then you also release the arsenic. And that’s a toxic element. And so this, this came about, right, and people started saying, Okay, we have this huge issue with arsenic poisoning. And in a country like Bangladesh, where there’s, you know, I think something on the order of like 140 million people, so really populous country, and a large fraction of them are on groundwater, we have this mass, essentially, mass poisoning event of a whole country. And the problem was similar in parts of India, in Cambodia, in Vietnam. And so that’s kind of what spurred, now going back a few decades, a real big push to understand well why is this happening. So even that mechanism that I told you about iron oxide reduction, right? At first people were saying, where’s this arsenic coming from? isn’t natural, is it not natural? How is it getting released? You know, so there was a lot of there’s been a lot of research into that. And that’s was part of what I did in grad school and have continued to do is look at that kind of mechanistic understanding. And then we started asking questions, okay, it’s naturally occurring and it gets mobilized under these conditions. But how do things like geomorphology affect it? So like, well, you know, how a river deposits and erodes in an area. How does groundwater pumping impact the movement of arsenic? or the mobile the rate of its release or mobilization? So, you know, I and many other scientists have been asking these questions for years. And so there’s a real recognition of arsenic being a widespread, naturally occurring contaminant and it’s not just limited to South and Southeast Asia. The population there is heavily exposed. But we actually have a problem here in the United States. So, you know, several million people are uh, so there was a USGS study about six, six or seven years ago, estimating the number of Americans who are on private well supply. And I think it was several million who are estimated to be drinking groundwater that exceeds EPA guidelines in the US alone. So it’s a problem here too. And in a lot of cases, it is naturally occurring. Arsenic is, you know, somewhat abundant in the Earth’s crust and in the soils and kind of found, you know, in many soils and sediments around the world. And the conditions can be such that in some places, it’s it’s mobilized. And the question that we’ve really tried to tackle too, is, it’s not always the there’s a lot of variation, even at the very small country scale or county scale, you’ll see like, one well might be really high, and you go kilometer away, and a nearby Well, in a very similar environment is quite low. And so a lot of our research, too, is trying to understand what what’s driving this variability, you know, and, and how can we make better predictions about risk. So that’s kind of the why arsenic. And then when we get to Cambodia, the issue is, well, we were looking actually in Cambodia, we’re not on this project, I’ve done some recent work on groundwater arsenic in Cambodia, and trying to predict that and so I have been working on that. But in this project that I’m using some of some of your sensors, we’ve act, we’re actually really interested in arsenic in the rice crop, and in particular, how it gets into the grain from because if you think about human exposures, right, there’s a couple routes of exposure to arsenic, right. So first off, it has to be in the prerequisite for the environmental exposures, it has to be present in the environment. But that’s not sufficient, right. It has to get then, if you’re going to ingest it, it has to get released from whatever solid phase right like the soils or sediments, and then either into the water, which you then drink, or released into the water, and then take it up by a plant, and then incorporate it into the grain that you then eat. And so we’ve looked a lot and the work I was just talking about a minute ago, we’ve looked a lot at how it gets released into the water, and then you might ingest it just through drinking the water. But then the next question, the step is, well, what about our food supply? And so we’re looking at arsenic in the rice grain. And there’s additional complexity here, because now it’s not just that it got released from the soils and into the water, but then it needs to get taken up into the plant, and then ultimately into the grain, which we’re gonna eat. And so that’s kind of that’s the question we’re looking at now. Well, what are the environmental conditions that drive arsenic uptake into the rice grain? And in particular, you know, so I’m working with a number of colleagues on this. And so I just like to mention a few. So I’m working with Ben Bostick, so he’s a professor at Columbia University. And then the other Co-PI with us is Dan Susa, who’s at San Diego State. Then we also have a colleague Conky Afan at International University in Cambodia. And then we have our students working with us as well. So we have a nice team, a really great team of colleagues who are all working together on this, but um they’re thinking about arsenic in rice. We care a lot about this, because that’s another primary round of exposure, right. People drink a lot of water, right. That’s a necessity for life. And we eat a lot of food and other necessity for life. And so…

CHRIS CHAMBERS 23:31
That’s parallel, right, if you don’t mind, because, yeah, drinking water is one thing, it’s generally a different water source, then, you know, then what you’re thinking about for your food, because it’s Yep, rice isn’t accessing the groundwater necessarily. It’s mostly So is this just the same problem cascading through different compartments of the ecosystem? Or is this a completely separate problem?

MASON STAHL 23:55
So that’s a great question. So you’re right. Rice is usually not like tapping into groundwater, right. And so if we’re not irrigating the rice, it’s taking up soil moisture, and kind of in, you’ve seen, I’m sure most people have seen pictures of rice fields, their pond and they’re flooded, right, or at least part of the year, they’re flooded. And so those create those reducing conditions that are favorable to the release of arsenic from the soil, but it still needs to get uptaken and stuff plant. And so in Cambodia, actually, irrigation is pretty minimal compared to a lot of other countries in the region. There. It’s a much, much less industrialized, a form of agriculture there. And so a lot of it is just rain fed. And so it’s flooded, it floods from riverine flooded or flooding or rain, and so they’re not actually pumping high arsenic groundwater onto the fields. And so you’re right that the arsenic is coming from the soil zone and getting uptake and now there’s a little bit of a wrinkle in some parts of the world, say like Bangladesh and Vietnam. Rice, irrigation of rice is more common with groundwater. And so the bang So this is an issue. There’s widespread groundwater irrigation. And they’re often irrigating with really high arsenic water. And so it can then become kind of this to two routes of arsenic source for irrigated groundwater, both from the soil zone, and then also from irrigating and just pumping on top of the fields, high arsenic on groundwater, and that can get incorporated into the plant as well. But yeah, so a lot of to, one other thing to think about too, so right, just like we’re exposed through our water, we can get exposed through our food. And rice is the important crop for a number of reasons, right. So it’s 20% of the direct calorie intake for people on earth comes from rice, so that one crop accounts for 1/5 of all direct calorie consumption for people on Earth. And so it’s a huge like, supply of just you know, nutrition and energy for people. But that is, that’s the case, on average globally. But then you go to a country like Cambodia, and it’s nearly 70% of calories, they’re come from rice alone, so that one food source provides seven nearly 70% of calories. So that if you’re thinking about relative exposure there, it’s really going to be water and your food. And in particular, the rice crop and rice is is also a unique crop and that it has a propensity to uptake arsenic. And that’s for a number of reasons. First, just the environmental conditions in a rice paddy are really conducive to these reducing conditions that favor the release of arsenic into the soil and soil water and then into the plant. But But additionally, the rice plants the uptake pathway for a number of nutrients, so for Salicylic acid, so plants need silica, the rice plants for structural integrity, so they take up silica, they take up phosphate, that’s a key nutrient for plants. And arsenic, dissolved arsenic, chemically behaves very similar to phosphate and solicit acid. And so you can have the plant trying to take up its nutrients that it needs. And it’s incidentally uptaking arsenic. And so then it gets incorporated into the grain because the right the plant is trying to draw up its nutrients and water. And in doing so, right can can incidentally take up this, you know harmful chemical that then gets into the grain. And so so that’s also it’s the environmental conditions are favorable and kind of the the biological, physiological conditions of the rice plant itself are favorable to this. So it’s a kind of confluence of things there that lead to high arsenic in the grain itself.

CHRIS CHAMBERS 27:35
You know, it’s easy to think of like China and India as the biggest rice producers. What kind of cultivation are, are tell us a little bit about the cultivation practices? Are we looking at subsistence agriculture? Or is it still an export? And export cash crop for like Cambodia and Vietnam and some of the other countries?

MASON STAHL 27:58
Yeah, it’s a great question. So for, Cambodia, a lot of the farmers are so Cambodia does export rice, I don’t know the exact numbers on that. But I, a lot of the farmers so we’ve been sampling a ton of farms across Cambodia. So we’ve we’ve sampled over 80 different farms, in almost all of the farms actually are pretty small. So we’re talking about, you know, like, you know, an acre or so or several acres, you know, not not very big kind of family owned farms. And essentially, subsistence, you know, they’re selling some of their their crop, but they’re also eating their crop too. And so, it’s very small scale. And it’s generally like minimal mechanization. This is for Cambodia. So like, oftentimes, there’s, you know, little farm machinery that you know, so you don’t see much machinery in the fields, sometimes you do, but it’s not super common, relatively low application of fertilizer, aside from say, like animal dung and things like that. There’s not a ton of chemical fertilizers used, and very little use of herbicides, pesticides. So it’s a fairly, I’d say, not a heavily kind of mechanized type or chemically intensive farming. And so they’ll that’s the case for Cambodia. And they also use very little irrigation, if you will. So Cambodia borders, Vietnam, and if you go and you cross the border from Cambodia, into southern Vietnam, in the Mecca in the Mekong Delta there, you’ll see a distinct shift in agricultural practices. And so it’s much more industrial, like mechanized, larger scale farming in Vietnam, and so have much heavier use of chemical fertilizers, herbicides, pesticides, much heavier use of irrigation. And so there is quite a distinction and a shift between countries so I know I know less about the practices in China, but my impression is that, again, it’s tense, there’s a lot more large scale farming and more mechanized and kind of industrialized system in inside China, as compared to Cambodia. That’s actually one of the advantages for from the scientific standpoint of, of our research, of working in Cambodia is that we’re really interested in how hydrology and climate impact the arsenic uptake into the rice grain. And so it’s, it’s nice in that we actually get the real signal the environmental signal, from the hydrology in the climate, as opposed to say, if we were working in a field that’s always irrigated, well, we can’t really see how variation in rainfall affects arsenic uptake, because the farmers are just controlling the water levels in the field, right. And same thing, if we’re thinking about how soil texture and soil chemistry or quality affects arsenic uptake, if they’re using heavy application of for chemical fertilizers and herbicides and pesticides, that really would mask the kind of natural signal. And so it would make it harder for us to see the underlying environmental and kind of mechanism. So we’re we’re at an advantage, and that’s one of the drivers of working in Cambodia is that we can better tease out the true underlying kind of processes because there’s less environmental control, from the human perspective. There’s less, you know, imposing and environmental control on the field, and more the ultimate outcome that we see is a reflection of the natural environmental conditions. So that’s another advantage of working, say, and Cambodia, as opposed to say, you know, Vietnam, where there’s a little more intensive farming practices.

BRAD NEWBOLD 31:45
I just had a quick thought that was triggered by Chris’s question on that, is that yeah, if we’re seeing, yeah, arsenic being taken up in rice there in Cambodia, or other places, and he was asking about about, you know, the various farming practices, is there a potential concern then for for rice that is being sold? You know, in large scale, you know, bulk levels? Is there a concern for the, for that arsenic being in exported rice products you’ve talked about, about arsenic there in India? In, you know, south, you know, Southeast Asian, and those kinds of things? Is that, is that a potential concern that that other, you know, other areas away from, from the the ah these rice fields should be aware of?

MASON STAHL 32:28
Yes, so, so arsenic and exported rice is definitely a concern, actually. So there’s a couple a couple of things. It’s a great question. Definitely any export in rice can be high in arsenic, but also even domestically grown rice in the United States. And I don’t have the specifics like numbers on that there have been there’s been good studies on this looking at kind of surveys and buying up rice and measuring arsenic concentrations, but we grow a lot of rice in the US, in California, Texas, Arkansas. And there are issues. And I don’t want to say any specific numbers, because I don’t know off the top of my head. But I know that their concerns about arsenic and domestically grown rice in the United States as well. So it’s not just limited to the the kind of East Asia, South South Asia, where there’s a lot of rice growing. But I know there’s issues and concerns about rice that’s even grown here in the United States and in other parts of the world. And in large part that, you know, might stem from just the the conditions and rice fields are kind of very conducive to the geochemistry that’s necessary for arsenic release, and then ultimately uptake into into the crop. And so whether you’re growing rice, you often have soil conditions that are just favorable to the arsenic mobilization, and then ultimately uptake into the crop. So it is a concern. It is a concern more broadly.

BRAD NEWBOLD 33:57
Right. Right. You also mentioned using various sensors, I was just interested in, you know, what sensor, what was your instrumentation suite that you’re using? What variables were you looking at? In, in your research there?

MASON STAHL 34:10
Yeah, so what we’re doing right now, so we have several, of the ATMOS 41 weather weather stations that we’ve deployed. And so we have a few of those and to kind of get the, you know, you know, you know, every 15 minutes, you get air temperature, solar radiation, wind speed, precipitation which is a really important one for us, and let’s see and relative humidity. And so, so those key things and temperature and moisture are key are key controls on arsenic uptake. So there’s been very recent work from some folks at University of Washington really cool research coming out of there. Looking at how actually like heat, heat waves or heat spikes can affect arsenic uptake, and so measuring air temperature is really important. for thinking about that, we also have deployed in a number of fields. Across our study area, we have soil moisture sensors. So So in like about six fields, we have several sensors, and each of those fields, measuring soil moisture conditions. And so those measure every 15 minutes, we get soil moisture readings. Then we also have water level loggers and a number of fields. And that’s just to measure the actual height of when the field is ponded, or flooded to measure the height of the water in the field. And then we have cameras too. So just like field camps that take, you know, multiple, multiple pictures a day on some of our study fields, so we can track the growing season. And the growing cycle in the greenness of the crop in the field to see, you know, is the crop, you can tell from from the color of the crop in part if it’s water stressed are the nutrients stressed, and so that’s a useful thing. But also just the raw photo alone is really useful to seeing the when the crop was planted, because we can ask the farmers but it’s really nice to also have, you know, a nice digital record of when the crop was planted in its growth stage. And then when it’s harvested. So we have weather stations, we have water level loggers measuring continuous water levels in the field, we have soil moisture sensors, deployed in our fields and then we have cameras and as well. So the weather station, soil moisture, water level loggers and cameras are all measuring these parameters in the field. So those are the physical parameters. But then we’re also like, you know, has been talking about we really care about the chemistry. And so we have we install soil, soil water samplers in our fields, and then we can go and we can draw those and collect soil, the soil chemistry, so water chemistry, so we actually use these things called peepers. It’s basically like these, we basically drill a shallow, narrow hole into the field. And we can lower these dialysis tubing basically filled with water, and it exchanges with the soil water. And then we can retrieve them and collect the water and then measure depth profiles of chemistry of water chemistry in the field. We’re also collecting soil samples themselves and measuring them for the bulk chemistry. So getting elemental composition, as well as isotopic composition. So carbon is total carbon, total nitrogen and isotopes of carbon and nitrogen. And then we’ve collected a lot of soil, no sorry, grain and rice grain for, again, elemental analysis, which is crucial, because our key kind of outcome variable that we care about is arsenic in the grain. And then we collect rice leaf as well. And so we have, we have the physical hydrologic and climatological or meteorological variables that we’re measuring. And then we have a whole suite of chemical of the water of the soil and of the rice plant itself. And so bringing all that together, and we’re really hoping to see how the hydrology and climate impact the kind of that outcome of arsenic into the rice.

BRAD NEWBOLD 38:03
What are some of your findings that that you’ve been seeing so far?

CHRIS CHAMBERS 38:07
Or what has surprised you most you know, it’s hard to tell in the middle of a project before you can run the stats and everything. But have you found anything surprising, or that you find is particularly interesting?

MASON STAHL 38:17
Yeah. So one of the things that we’ve been doing is we’ve, we’ve gone out, so I mentioned all these hydrologic and meteorological variables. And so those are a set of selected locations where we’ve installed sensors, right. But then we’ve also gotten gone and done a broad survey. And so this project started last January, so we’re just on a year in and we have gone and sampled at fields are a little bit over 80 fields now. Or so those are individual farms and collected the rice grain the soil, water were available, and then analyze the those, those those samples for elemental composition and isotopic composition. And so we have now this really, our goal is to get this really broad dataset where we sample across the many different environments, different soil texture, different flooding, timing, different duration of flooding, different farming practices, so that we can really kind of cover this whole variable space of different environments so we can understand how each of the environmental variables affects the end goal of arsenic in the rice. So the question and so one of the things that’s been a little surprising, and others have observed some, this too, is that when we look at so now we have, you know, over 80 fields where we have actual grain measurements of arsenic and a whole host of other things, and we have the soil from those same fields paired with it. What we see is we see clear spatial patterns and soil chemistry and that is very strongly related to the soil texture, proximity and a river. But what’s really makes this a particularly challenging question is that when you look at the chemistry of the rice, right Write it is much, there’s much less clear spatial patterns to it, there are some. And it’s much less it at least at first glance, when you just look at, say something like arsenic in the soil versus arsenic that ends up in the rice grain, there is a very weak relationship, and it actually in our data so far looks like a negative relationship. And so like, one’s intuition might be like high arsenic in the soil, high arsenic in the grain. And so that we actually look, it looks like there’s at least, you know, in our, in our, you know, 80 plus fields that we’ve sampled, the relationship is weak, and it actually looks negative. And so that suggests, right, that there’s, it’s not, it’s not going to just be an easy answer, like, okay, yep, the fields that generate really high arsenic in the rice grain, and that we need to be careful for are all hypothetically all this soil texture, and it’s always fields that have high arsenic in the soil. And that would be really convenient if that were true, because you could go, and you could measure soil pretty easily for arsenic across, you know, a wide area, and then you could say, I predict here, it’s going to be high arsenic in the grain and here is going to be low because of the soil arsenic. And so that’s a bit of a surprise, surprising finding. And that, you know, it’s the that is not that story is not so simple. And in fact, it’s, it’s really not the case that just because you have high, low arsenic in the in the soil material, that you’d have low arsenic in the grain, that’s often not true. And so that adds to this complexity of really looking at the hydrology of the fields and maybe other nutrients in the soil that might be driving this. But that’s been one of the things that’s really, you know, poses a challenge, because ultimately, we’d like to be able to make predictions, right, we can’t, we can’t sample 100,000 fields, we’d like to be able to sample 100 fields, say, and then link the outcome of arsenic in the grain to the environmental conditions, and then make predictions for 10,000 or 100,000 fields based on our mechanistic understanding, because it for instance, if we can say, Oh, this soil texture, and this flood timing, flooding, timing, and duration, is the key control on arsenic, then we can make predictions, like a whole map, right for a country and say, because those are things that we can measure remotely, or measure very easily without actually having to go to the fields. And so that’s our ultimate goal is to get this really ground truthing dataset by going and collecting more intensive data at a large number of fields, highly intensive physical measurements at a smaller subset of fields, like um sorry, physical hydrology measurements and a smaller subset of fields, and then use that to make predictions at a at a broader scale. So that’s kind of but, but like I was saying, it’s, it’s not so straightforward that we just see a clear arsenic in the soil arsenic in the grain. So that’s one surprising finding.

CHRIS CHAMBERS 42:52
Wow.

MASON STAHL 42:53
Yeah,

CHRIS CHAMBERS 42:54
I was gonna get that key mobility.

BRAD NEWBOLD 42:55
I know, I was, I was trying to think from my, from my okay. I’m not here to give you suggestions, because I’m sure that you’ve thought of everything. But as you were talking about that, because it was, I was I was trying to think, you know, wrap my head around. Okay, what could you know, what could be at play? So, you mentioned, interplay, could it be, you know, also potentially interplay between surface and groundwater? I mean, is there something there? Could, I don’t know, if you’re dealing with any kind of, like, evaporative effects there in rice fields? I don’t know if, you know, if there’s anything there also potentially, like, you know, downstream upstream effects like, like, if you have, you know, arsenic in the soil, you know, upstream, is that then going to lend itself to to, you know, to send more arsenic in solution downstream that is then taken up, I don’t know, these are, these are, again, my uneducated suppositions here, but is it? Are any of those valid? Or does there be any potential for for any of that kind of stuff going on?

MASON STAHL 44:00
No, absolutely.

BRAD NEWBOLD 44:03
It’s okay to tell me, I’m out in the field here.

MASON STAHL 44:05
No no no, those are all really good suggestions. So I’ll say a couple of things. So your, your point about evaporation is a really good one. And in fact, actually, on a separate project on arsenic in surface water in the high plains, we actually have a clear evaporative signal. This is some ongoing and developing work, where we see as water moves downstream, a clear, strong evaporative signal and a significant increase in the concentration due to evaporative effects alone. It looks like so the question about evaporative concentration is a good one. And is definitely like, I was just actually giving a talk about this last week about some of our initial data from the US Northern Plains. And the talk was actually about teasing out the evaporative impact. So it’s a fantastic question, and it could be going on and rice paddies. But, and then your other question about surface water. I think you mentioned something about kind of upstream downstream effects and surface water is another great question. Point. And I’m actually ties in this was part of the motivation for this when we wrote the proposal thing to get funding for this project was we had, we last year we ah, Ben Bostick and a postdoc of ours, Craig Connolly, and then an undergrad who was at Union who was back to young, we published a paper on looking at just remote sensing images of flooding, occurrence and recurrence, basically, how frequently flooding happens, and how long flooding occurs for in Cambodia. So that’s something you can measure from a satellite. And then we linked that to measurements that we had actual measurements of groundwater arsenic for 1000s, of wells. And we found a really strong relationship. And so that motivated to kind of some of this work and help kind of lead into the proposal. But basically, kind of from that work we were looking at, can you look at just flooding, which is a surface occurrence, right, a surface phenomenon that you can see from a satellite. And then can you connect that to arsenic and groundwater, which is meters, 10s of meters down. And in fact, we couldn’t, there was a really strong relationship. And we were able to develop a predictive model for groundwater arsenic concentrations, based on just a flooding, occurrence and recurrence that you observe at the surface. And the model, we intentionally just use really, essentially just those variables, because we wanted to say, can we make predictions without having to go to a play, I mean, we to train this model, we had to, we had to have water samples that had been tested. But can we then use just satellite imagery alone without ever needing a geologic map or a soil map to make predictions of arsenic in the groundwater, but which is, you know, at death? And and it worked? It worked quite well. And so then our, our question, one of the, you know, many of the hydrologic variables we’re looking at, and one of the key ones that other people have observed, right, of course, flooding and rice fields matters. But now we’re really trying to look at to look and to your question is how does the flooding, duration and frequency affect the chemistry in the fields, and mobilization of arsenic and the uptake of arsenic and it could be, the timing of flooding might be really important, because we’re dealing, it’s not just that the arsenic needs to get into the soil water, but it also needs to get taken up into the plant. And so there’s plant physiology involved. So flooding, timing probably matters a lot. And we’re looking into this, but and so this is something we’re, we’re considering, because I’m going to kind of make something up here. But hypothetically, you might be like, Oh, if flooding, if the field is really heavily flooded during the first 30 days of growth, maybe that’s really important in terms of arsenic uptake, and maybe flooding from day 30 to 60, is less important, hypothetically speaking. And so these variables, right are really crucial, like this kind of flooding and surface water. And so so that is that is something we’re really looking into, and trying to tease out, using both our physical measurements from our soil moisture sensors, and our water level loggers. But again, we can’t put those across the country in every single field. So we’ll put them in a select set of fields. And then we can try to link what we get what we observe on the ground, right to satellite data that we can use to measure at a national or global scale. And we can basically use our physical measurements on the ground to ground truth and also to kind of connect these hydrologic processes to arsenic in the grain and then hopefully, use remotely sensed data to then make predictions at at scale. And so that’s, that’s an aspect of our project and our Co-PI on this project, Dan Susa at San Diego State. You know, that’s one of his specialties is, is remote sensing and or that’s a key specialty of his and so he’s really helping bring some of that like heavy like this really intimate knowledge of that and the cutting edge science on that aspect to this project as well, among other things.

What are you doing going forward right now? What is the future of this project? And I guess you could get delve into to, you know, your arsenic in particular research. Going? Yeah. Going forward here in the future.

Yeah, so…

BRAD NEWBOLD 49:24
What is next that’s, that’s right.

MASON STAHL 49:25
Yeah. Yeah. So for this project in Cambodia, one of the key next things is that we actually said we, we just completed some fieldwork back in January. And that was our wrapped up our first whole year of fieldwork. But what was nice is now we have two crops. So I mentioned that we’ve gone and sampled 80 fields or 80 plus fields. But for each of those fields, we’ve now returned at least twice. And so we now have two seasons of uh crops and so this is really useful and we’re running the samples right now. But our our hope is right, we had different kinds of weather conditions from the first crop to the crop we just collected. And so we’re really hoping now that we can dive in because we have time series data for every field, we’re really hoping now we can dive in, we can say, oh, for instance, the first year was a warmer and drier. And then the second year, which was cooler and wetter. And for every field, we have a crop for both of those years, and we can measure arsenic in the grain. And so we’d really like to now one of the key lifts will be can we connect, kind of from the climate and hydrology or the weather and hydrology from year one and year two, to say, what were the key variables? What are the key controls is the second year’s crop always higher in arsenic or always lower? Or is it more complex than that. And so that’s one of the key things kind of now using our time series data, which we’re actively running the samples now we have all of them with us. And we’re analyzing those at the moment. So that’s one of the key things. And then comparing, of course, across fields to, and that’s been ongoing, but the time series is a crucial thing. Another aspect now that we’re really working on is, so one thing that I is a kind of common thread through a lot of my work is sort of like using large datasets and machine learning has been a bigger part of much of my research is, is now trying to make predictive models. So we have a lot of this data. And sometimes, even before we might even know the mechanism, exactly, or modeling something with prescribing the model and knowing exactly how to model every process can be quite challenging or intractable. So we can at least apply certain, like machine learning algorithms to tease out Okay, can we see patterns in the data, are we able to identify the key variables that are controlling arsenic mobilization, and arsenic uptake. So using those kinds of data analytics approaches to identify key variables so that we can get better insight into the mechanism. So that’s right now kind of what my Co-PI’s and I and our students are working on, so that we can better develop these predictive models. So that’s a, that’s a key aspect, and then tied into that is really breaking in the remote sensing aspect to it, too. And so just this morning, we were we had our group meeting, and we were talking about using our cameras that we have on our fields, and connecting that to the greenness that we observed from satellite. So trying to find a connection there and then, or see how well those things agree. So how well can we basically can we trust the greenness that we measure from satellite imagery, which was an important measure of the crop health and growth cycle, and then can we link that to the arsenic concentrations that we’ve measured in all these fields, because now we have remote sensing data that we can link to every single field that we’ve sampled. And so those are some of the kind of ongoing kind of prongs of the research that we are the next steps that we’re doing. And then another key next step is we will have is returning to these fields. So we’ll make several trips a year, and getting the next season. So we can really, hopefully build out several years, not just two, but several years of harvests. And so we can have a nice time series. And that will give us a much more, I would say robust kind of data set for linking climate and hydrology to the arsenic in the grain. And so those are some of the kind of bigger picture things that we’re driving forward with

BRAD NEWBOLD 53:34
Putting your policymaker hat on. What are some of your, your thoughts or ideas or just from right now from interesting, I guess, both in Cambodia, but also in the in the Great Plains that you’ve seen? What are your thoughts or feelings towards mitigation and or remediation of of this specific arsenic? You also mentioned uranium, I think, but yeah, but these issues that were that we’re seeing here?

MASON STAHL 54:01
Yeah. Okay, so what so one thing is that we hope to come out of this project with is some, some understanding that is actually practical and actionable in terms of how what a farmer could do for minimizing arsenic uptake. So we’re not at that stage yet. But we are running some manipulation experiments. We’re manipulating conditions in the field, in a couple sets of experimental fields in Cambodia. And so we hope to see how how does changing certain conditions in the field, ultimately translating so I’m hoping in the next few years, we’ll have a really clear or a clear understanding of what are some simple things that a farmer could do that are consistent, you know, not like some really heavy lift that requires, you know, a substantial modification of their practices, but something that’s feasible for them to implement, that would improve hopefully yields and reduce arsenic concentration. So that’s what we’re working on. We’re hope, hoping to kind of make some headway there. But if you think about mitigating exposure. And so I’m definitely not an expert in the policy aspect, but I’ll say some some thoughts that I have. So, in some, some success stories that have occurred to. So arsenic, you know, when when you drink from a domestic well, both in the United States and elsewhere, you’re often not treating that water, you know, some people do have household filters, you know, commonly in the United States, you’ll have at least a water many places where they’ll have like a water softener to reduce hardness. And so there are things people do. But oftentimes, you’re drinking the raw groundwater. And that will, if you have arsenic in it, you know, then then exposure is likely. But what we’ve seen in the history, there’s some interesting history for Cambodia, is that, you know, it’s my understanding that exposures have gone down in Cambodia, because, and, and in Vietnam, actually, because of a switch over to centralized treated water. And so I’ve seen this more firsthand, and Vietnam, actually, well, actually, in both places, but in some of the places, so I started going to Vietnam, and my first trip was in 2010. And the place we were we did a lot of our research was outside Hanoi, so in the north of Vietnam, and in the kind of town, you know, just south of the capital, you know, as you know, kind of a more rural agricultural, a lot of people were on domestic wells. And then, you know, within five or six years after my first visit, many people have switched off of their domestic domestic well, so they might have been using it for like, you know, washing and external like just, you know, using outdoor water use, but they weren’t drinking from them anymore. And in fact that that town had a more centralized water supply, where they had some basic filtration in place, and arsenic mitigation practices in place. And so that’s definitely like a, you know, I would say, a tried and trued approach in there’s economies of scale, too, right. So if you, you know, a filtration system for 1000 people, is, you know, probably substantially cheaper than 1000 filter, small filtration systems for 1000 people. And so that’s definitely something that can work if there’s the, you know, ability to install that infrastructure, but sometimes there’s not, or there’s lack of funding or, you know, other impediments in place. And so, but, but I think that that’s kind of my perspective, is that a really good mitigation effort from from a water perspective is, you know, centralized treating a treated drinking water like we have in most parts of the United States. You know, most municipalities, unless you’re in a rural area, you might be on a domestic well, but most municipalities you’re on piped water, which is first withdrawn, either from surface water or groundwater sent to a water treatment plant, and then distributed through the, you know, pipe network to homes, businesses, and you know, residences. And so that’s, that’s sort of my take. And I’ll say to the, I don’t know, I know a little bit about the history here, but the capital, the capital of Cambodia, Phnom Penh actually has to have piped water that is potable and reasonably high quality, you know, compared to you know, many places on earth. And that’s a really a somewhat recent introduction in the last, you know, maybe 10, or 20, or 20 years. And it shifted from a system that was essentially failing and disrepair, provided water only a few hours a day, a very low quality, and was very unprofitable and losing money through a system. This is again, based on what I’ve what I’ve read, and in some basic experience, that they now have a system that works reasonably well meets water quality standards, reasonably well, and is a system that is profitable and, and sustainable and, and low cost for the average consumer to and so again, that that’s sort of this indicator that, you know, the centralized drinking water systems are feasible are tractable in not just here in the United States, but you know, in many different places and environments across kind of the, you know, income, global income scale. And so that’s, in some sense, I think, a tried and true approach that can work. So, you know, thinking about those kinds of centralized approaches, I think, is at least one way of managing for arsenic and rice. I mean, screening rice and knowing where if the concentrations are high is of course important for thinking about exposure. So just literally being, you know, having that information available so people can make, you know, adequate decisions on what rice to buy. And if they want to be consuming that rice, that would be that’s another aspect as well, and just raising awareness. Exactly, yeah. And the mitigation in the in the rice fields. So we can you can filter arsenic out of water. So the mitigation is a bit clearer there. For rice. That is, like I mentioned something we we want to understand better and other groups are doing this too and doing great work trying to look at well, how might you be able to, you know, manipulate conditions in a field, effectively so that you reduce arsenic uptake into the rice.

CHRIS CHAMBERS 1:00:26
Mason, we really appreciate you being here. And I don’t want to take up any more of your valuable time. But I’m dying to just kind of shift gears a little bit and ask you one question, if you’ve got time. Sure. Okay. So recently, I came across the seasonal cycle of surface soil moisture in Journal of climate and like your central question there. Like kind of making the seasonal soil moisture regimes? You know, I found that question really interesting. And if you’d asked me, before I read the paper, I would have been like, oh, there’s hundreds, there’s got to be hundreds of seasonal soil moisture regimes, regimes across the planet Earth, right? Yep. And, you know, correct me if I’m wrong anywhere along the way, but use this machine learning model. And skipping the details. It arrived at five. And so how did how did you feel about that? Were you surprised by that? What’s the reception been to that? It has it? Has it gotten some good discussion? I

MASON STAHL 1:01:26
assume? Yeah, yes. Oh, so Yeah, great question. And happy to hear you read the paper. So yeah, so I was surprised. So I worked. The colleague, I was working with Cagin McCall, who’s a friend and colleague of mine. And so we’re looking at actual measurements of soil moisture, SMAP data. So this is remotely sensed from a satellite that’s measuring soil moisture. And we found that on the on the actual, you know, observational data, when we basically look at the seasonal cycle five, five, kind of distinct seasonal cycles, predominant across the globe. So almost every point on Earth, right, fell into one of those five cycles pretty well. That was surprising to me, honestly, prior to starting that I didn’t know what we were going to find, I wasn’t sure if you know, there’d be no clear distinct cycles or 50. Right. So there could be like, you could get more granular and look at it and say, oh, let’s take this cycle three and break it into two sub cycles. But broadly speaking, every place is falling pretty clearly into one of those cycles, right. So it was it was, it was a little surprising. When we looked at it further, we were able to then develop a physically based kind of very simple model that really only took two input parameters, solar radiation, and precipitation. And when we then looked at every plate point on Earth that we had classified, and we looked at the seasonal cycle of solar radiation and precipitation, we could then using an actual just basic water balance model, right. So physically based, not like a machine learning model. In that case, we could essentially replicate the soil moisture cycle, which was good, because we’re saying, Okay, there’s kind of two knobs in the environment that can get turned that control these distinct emergence of the cycle. So that gave us more confidence to. And that said, we can get these five emergent kind of seasonal cycles based on just differences and the timing of precipitation and the timing of solar radiation. So yeah, it was a little surprising. So one caveat, or thing that we we could look into further is that we’re looking at data. And this is remotely sensed data. And so the satellite returns. So a couple things about that, is that the spatial resolution, it’s taking an average, one of the satellites measuring soil moisture is averaging over like, basically like a, a block or grid cell, that’s about, say, 30 kilometers, but on each side. And so it’s definitely smoothing out, you know, that’s smaller spatial scale, scale variability. And so to your point of, are there many more cycles, if we’re looking at a more granular, like spatially granular resolution, there probably are many distinct cycles to within that, right. But kind of, in the aggregate over, you know, bigger spatial areas, we get these cycles. But like, imagine at a very small spatial scale, you’re on the hill, you go out to a hill near your house, you stand on it, and on the say, North Slope vs the south slope, right, they might have a little bit of a different cycle because of exposure to the sun, right. So things like that.

CHRIS CHAMBERS 1:04:43
Definitely, really, that really kind of gets like a microclimate thing instead of instead of a climate thing. But I really thought the approach was interesting how we’re going to let how we’re going to let the algorithm you know, we’re not going to bring in all these preconceptions, and just kind of let the algorithm I pick the number of bins. Right? And so I just, it was it. I just really appreciate your thoughts on that, that it. Using this approach, I thought it was a nice, simple approach to use to have kind of an objective starting place rather than having people throw all of these all of their preconceptions in and you guys do a great job in the paper of addressing limitations too. But yeah, I just thought it was really neat.

MASON STAHL 1:05:30
And oh, thank you. Yeah, it’s a real. So the approach that we use is unsupervised clustering. It’s an unsupervised machine learning approach. And it was it’s called clustering. And basically, right, yeah, it looks and it says like, yeah, what are the different groupings. Like, so I, I don’t have to tell it how many groups initially, or I don’t have to tell it like, these are the groups now sort everybody into them. But rather here are the signals, basically environmental signals, so you know, your measurements. And can you identify the key the groups that are in or the patterns in the data. And so it’s really handy, because if we had 10 locations, I could just look at the time series. And I could say, Okay, there’s three groups. But we with with this is the thing that the unsupervised learning is really handy for is it lets you identify patterns, so that then you can explain the mechanism, because we had, when we’re looking at this global soil moisture data, there were 40,000 pixels or locations on Earth, where we had good enough data. And so then it’s impossible for anyone to look at 40,000 pixels right, and the time series, so the algorithm can do that in a matter of minutes, and say, Hey, these look like groupings. And then it doesn’t tell you what those groupings mean. But now we can look at them. And we can say, Okay, now we can explain them and understand, you know, the mechanisms, you know, much better, because we just wouldn’t have known, even if I knew there were five clusters, I couldn’t go through 40,000, you know, time series on by hand and do that. So it’s really nice. Yeah, it’s a really handy approach. And we use that a lot in our research just for figuring out what what are the key groups in a data set that’s really complex.

CHRIS CHAMBERS 1:07:06
Yeah. Well, I opened a can of worms there now.

BRAD NEWBOLD 1:07:08
That’s fine.

CHRIS CHAMBERS 1:07:09
That we can probably do a whole hour but uh..

BRAD NEWBOLD 1:07:11
Yeah, I was gonna say we could have you back to talk about machine learning and predictive models.

CHRIS CHAMBERS 1:07:16
Thank you so much for your time. This. This is great. I learned a lot.

BRAD NEWBOLD 1:07:20
Yeah any final thoughts? Mason, anything, we didn’t cover anything you’d like to, anything you’d like to pitch or anything along those lines?

MASON STAHL 1:07:29
No, I think you guys, I mean, you asked great questions. I really enjoyed the conversation. Thank you so much for having me.

BRAD NEWBOLD 1:07:35
Our time’s up for today. Thanks again, Mason. We really appreciate you taking the time to talk with us. It has been a super fascinating conversation. And if you in the audience have any questions about this topic or want to hear more, feel free to contact us at metergroup.com or reach out to us on Twitter @meter_env. And you can also view the full transcript from today in the podcast description. That’s all for now. Stay safe, and we’ll catch you next time on We Measure the World.