The Insight Podcast
The Insight Podcast
Ep6. For peat's sake
In episode six Louise and Gráinne talk to Niall Ó Brolcháin about the case for rewetting and farming Ireland's peatlands; Prof Eileen Gibney explains how data science can revolutionise food and nutrition research; Prof Tomás Ward and Prof Brian Caulfield talk about Insight's Sensing and Actuation research challenge while explaining why success will resemble a Christmas cake.
Louise Holden 0:06
Hello and welcome back to The Insight Podcast. It's 2021 and Happy New Year to all our listeners and Happy New Year to you Gráinne
Grainne Faller 0:13
Thanks very much Louise. Did you have a good Christmas?
Louise Holden 0:15
I had a wonderful Christmas. Weird and wonderful like everybody else I guess
Grainne Faller 0:20
(laughs) weird.
Louise Holden 0:23
Yeah,
Grainne Faller 0:24
So it's good to be back. We've a great episode again lined up. I keep saying that but it's always true.
Louise Holden 0:29
We're here we're kicking off with Niall Ó Brolchain from G alway, who's going to be talking to us about the Care Peat program and about the transformative potential of rewetting our peatlands
Grainne Faller 0:38
Professor Eileen Gibney, one of our new FIs, is going to be talking about how data science can add to food and nutrition research.
Louise Holden 0:45
And in the first of our series on the Insight Research Challenges, we're going to be talking to Brian Caulfield and Tomas Ward, about the whole area of Sensing and Actuation, which is very interesting, trust me.
Grainne Faller 0:55
But first, did you know that Ireland's peatlands and Europe's peatlands could potentially be a powerful weapon in the fight against climate change? Well, the people in Insight at NUI Galway are doing a lot of work on this area. And we spoke to Niall O Brolchain, first about a European project called Care Peat, and then about the potential of peatlands in general in this fight.
Niall O Brolchain 1:17
It's a big European project, involving five different countries. It's an Interreg EU project. And there's other people in NUI Galway involved in that - the geography department we're collaborating with. And it's going very well, to be honest, I mean, we've had, we've had a lot of success so far. We just were involved in influencing certain votes in the European Parliament with the Common Agricultural Policy recently. And I suppose really, what we're talking about here, from the point of Insight is using data around climate science, to you know, this would include drone data, it would include readings from the peat land, in terms of water in terms of, obviously carbon emissions, and so on. And using that data to develop policy. So my role very much is I'm the policy lead on that, you know, so what we're doing is using evidence based policy, and we've got five different sites in Ireland, in the UK, in the Netherlands, Belgium, and France. And we're examining different types of people and and different types of scenario on different sites. So it's a practical project, it's actually physically saving carbon. But we're getting a lot of data from that. And the whole key thing to it is scalability across the European Union.
Louise Holden 2:37
When I first heard about this particular project on paludiculture, which I believe is the descriptor, Ihad never heard of it, and I have a son studying agricultural science and we had a conversation and he'd never heard of it either. And once we went to look into it, and we realized that the the the carbon sink capabilities of the peatlands was was was so significant and was comparable to, for example, tropical rainforest, we were both kind of flabbergasted. So it's a very interesting area. And I think a lot of people are not familiar with it.
Niall O Brolchain 3:02
Yes, indeed, I was talking Department of Agriculture official, a couple of days ago who wasn't familiar with the term paludiculture culture, which I find quite startling, because what paludiculture is, it was voted in the European Parliament, I think two days ago, it's now included, it's basically wetland farming. So you know, the sorts of products that you're talking about, there's many different products that have been developed, the one that people might be most familiar with known as cat tails, I suppose, you know, in European parlance, but bullrushes is what we tend to call them. You know, they're slightly different in terms of breed. But basically, they can be used for building materials. There's things like horsetails can be grown wetlands, which are very, very high value product. And there's many different wetland plants that can be used for cosmetics can be used for food can be used for various different things. And we we haven't really tapped into that whole market. I mean, Bord na Mona, for example, I think the sort of the average, you know, yield from dairy farming for a hectare of peatland would be around 800 euros per annum. Whereas, Bord na Mona actually discovered that they could sell horsetails for about 3000 euros per hectare per annum, which is significantly greater. So you know it's a very specialized market and the market hasn't really developed in Ireland. But we'd strongly suggest that this sort of market should be because palludiculture does two things. One is it allows for very high value yield, but it also opens up for another yield in terms of carbon credits because if you keep peatlands wet, you're actually saving a hell of a lot of carbon. Because people don't realize and I use this finger now that I found that every fortnight the amount of carbon which is emitted from drained peatlands in Ireland, is equivalent to the total weight of the Irish population. That might sound like a crazy thing, but you know, every two weeks, could you imagine the whole every single person in the world or sorry, in Ireland going up in terms of weight into the sky. I mean, That's literally what we're doing right now. It's not something we should be doing. And it's something we can actually reduce to net zero. European Commission says that we can achieve this by by 2035, you know, which is about, what 15 years time. So, I mean, the potential for using our peatlands, which are 20% of our country to reduce our overall emissions is enormous to reduce agricultural emissions in particular, which I think the Department of Agriculture needs to wake up to. It's also enormous because you know, the pressure is on in terms of reducing the national dairy herd, surely the pressure should be on to do something positive, instead of something negative, you know, and farming organizations are beginning to wake up to this, but I don't think Department of Agriculture has really cottoned on to this. I don't think Teagasc is really looking at thi, to the extent they could, but you know, there's huge potential from this.
Grainne Faller 5:50
And is it, in terms of the Department of Agriculture and Teagasc and that, is it just that they're not aware? I mean, the kinds of things that you're saying they're they're kind of extraordinary transformative, potentially, it seems surprising this decision makers aren't on top of this.
Niall O Brolchain 6:04
Well, I suppose decision makers, who are they, I mean, public servants aren't necessarily decision makers. They're more sort of administrators. And you know, administrators tend to administration, they tend to keep doing things the way they're comfortable with the way they know. But the point is the way they know is changing. It's changing. Literally, it changed this week, do you know in the European Parliament, and it is going to change whether they'd like it or not. And the reality is, we're heading towards the the 1 trillion euro Green Deal. And you know, that's happening next year, you know, there's going to be hugely transformative changes, we can either decide, look, we're going to put our head in the sand and keep looking for the same sort of dairy farming, beef farming subsidies that we've always looked for. Or we can we can say, well, hang on, why don't we go where the money is, you know, because I mean, farmers and farmers are very clear on this. Farmers want a good income more than they want anything else. They're not sort of religiously wedded to sort of old fashioned farming practices, farmers aren't the issue here. Farmers are basically businessman, do you know, and common sense people and most of them really have a great grá for the land, they've grá for ecosystems. And if they could have a decent landscape and a decent ecosystem, which didn't sort of produce with huge amounts of carbon into the air, they do it. And I mean, I haven't really met a farmer who disagrees with us, if they can get a better income, and better income for their families, they can they can sort of protect the environment at the same time. And, you know, that's, that's what they're looking for, you know, so why why people just are religiously sort of session on sticking with what they know. And the old fashioned thing, I suppose it's just the nature of governance and the nature of administration. It likes to plod along and do things in a slow way, in a conservative way. And that's that's it. But, you know, we're researchers, we're looking to the future. And government also has to look to the future and government is taking this on board.
Louise Holden 7:58
After reading some of your inputs, I've got lovely visuals in my mind of water buffalos, roaming around Ireland and so forth. I'd like to understand the practicalities of making those kind of transitions, we would obviously need a lot of support from Europe if we were going to rewet agricultural land and introduce entirely new flora and fauna, wouldn't we?
Niall O Brolchain 8:17
Well, absolutely. But the beauty of what you see is that there's when you examine exactly what's going on at the moment, if you look at dairy well, sheep farming in particular, because we're talking about peatlands, we're not talking about prime agricultural land here. We're talking about peatlands, which I think I think only about 7% of the agricultural land is actually peatland. So, you know, of the existing agricultural land, I mean, but a lot of a lot of the existing land is owned by farmers, which is a different thing. Because it previously it wasn't designated, necessarily designated as agricultural land. So to try and answer your question, you know, a lot of the income that farmers get, for example, from sheep farming on peatlands is actually subsidies, if you actually look at the income they get, it's almost a sort of, I suppose, a zero sum game, if you remove the subsidies, they're almost making a loss from the activity. So you know, simply replacing one set of income with another is the way to go. And there's no reason why we can't sort of have slightly different animals or from them in a slightly different way. And indeed, still get a certain proportion of the income. You mentioned, water buffaloes there, for example. We're not talking, as I said, about prime agricultural land, we're talking about peat land, which is naturally wetland, so it should be wet, it shouldn't be drained. So that, you know, there's huge costs and efforts and put into draining the land in order to make it suitable for cows. But if you simply let it be wet, and maybe had a slightly different type of cow, the income might be slightly different, might be reduced. But on the other hand, there'd be other forms of income to supplement that and indeed, maybe surpass that. So you know, I mean, I think farmers farmers are involved. To the Netherlands. There's been kangaroo farms. There's been ostrich farms. There's been, you know, llamas and alpacas and all sorts of things brought into Ireland. Most of them haven't worked, you know, but I mean, there's, there have been various different success stories as well over the years.
Grainne Faller 10:16
Can you tell me about the policy end of things? You said, you've been having some success at European level?
Niall O Brolchain 10:22
Sure. Yes, we, you know, along with a number of other I suppose, knowledge institutes and research centers around Europe, we we collaborated in terms of, I suppose, getting information for the Common Agricultural Policy. So we were asked to do that. And we provided information to the European Commission, we provided information and talk directly with the Environment Committee and the Agriculture Committee in the European Parliament. And in particular, it's now a massive transformation in the Common Agricultural Policy, they've just, as I said, voted on it this week. And basically what it is is, peatlands can now be recognized as wetlands in the context of designated farmland. Now, that might not seem like much, but what it actually means is it opens the door for many Irish farmers to actually designate em, you know, many Irish firms would have both peat land and other land, other types of land within their sort of parcel of land. And they can now actually include an peatlands potentially in there, or the will be once this once this is fully implemented. And they be able to include proper wetlands in their inventory of land for the single farm payments, and so on. So you know, it's it's, it's, it is a transformative thing, because it means that the, I suppose the emphasis on draining land, always draining land is no longer necessarily, you know, the way the way it is anymore. So it means that farmers will be able to make money on wetlands and retain subsidies.
Grainne Faller 12:00
Professor Eileen Gibney is one of Insight's newer FIs. She's based in Insight at UCD and her work is based around food and nutrition. We started out by asking her how data science can contribute to that area.
Eileen Gibney 12:12
When we talk about diet, one of the key things we have to do is to be able to measure someone's diet. And to measure it accurately and correctly. We need to know what people are eating accurately so that we can then you know, compare that to risk of particular diseases, we need to know where the nutrients are coming from within the diet. And we need all of that information for public health purposes. Now, when we ask someone about their diet, we can do it in a number of different ways we can ask them to think about what they ate yesterday, or we can ask them to collect information on foods that they consume over the next few days. But there's inherent issues with dietary data intake assessment in the US, people don't always tell us the truth. And it's a very complex data set to try and collect. It's very detailed, and we need to get down to incredibly granular detail. And do you know what it's just difficult and a little bit boring to do? So what we need to do is to really look at that data and determine what are the key aspects that we need to collect, and are there ways that we can make this easier for an individual so that when we do collect it, we can collect large amounts of that really easily. You need large datasets for this because there's so much variation in the dyers, there's so much variation, and it's linked to disease. So we do need to be able to take datasets from Ireland, from Germany, from Australia, America and combine them. There's lots of major research projects around the world doing this, but they're still using quite traditional methods of collecting dietary intake data. So I'm kind of taking on a little bit of that challenge in Insight. And what I'm doing is trying to look at this in a slightly different way. So at the moment, we ask individuals, when they talk about, you know, breakfasts, they list every individual food, and what I'm trying to do is to see can we get them to describe their diets that are very generic meal level, but have background data that we have derived statistically, from existing data sets to give precise information about that, but without asking the individual for that precise information. And if we can do that, what we can then do is be able to ask individuals to collect very easily their information at a meal base level. And we can then utilize that information to to get a bigger picture of their overall diet and to see what kind of foods they're eating when and where, why I'm interested in this, again, is that we're very much moving towards the digital side where a lot of people like to take pictures of their meals and to take images. But what I'm sort of saying is if we can do this at a meal level, I could take a picture of that meal and try and identify it as a fried breakfast rather than determining all the individual components of it. And that may itself increase the accuracy of these methods because at the moment when you take an image of a meal and try and translate it into nutrients, the accuracy is really, really low, you know, there's an inherent need for detail. But there is a difficulty in collecting that detail either by an individual or through an image. And I'm just trying to see if there are clever ways we can use the data that exists to make that easier to collect that information. And if it's easier, we can collect a lot more of it. So for me, that's where data science and nutrition kind of meet.
Louise Holden 15:27
okay, and what's the end use for these large data sets and for the information you derive from them?
Eileen Gibney 15:33
So a lot of these datasets are used in what we call epidemiological research, where if you can collect detailed dietary information or any dietary information for individuals, you can determine if those individuals either prospectively included have particular diseases or disorders, you know, like coronary heart disease, those linked to, to diet or particular cancers, or, you know, more specific, say, hypertension, or linked to obesity, so watch the eating patterns, and dietary patterns that are linked to these particular diseases or disorders. And if we know that at a public health level, then we're able to offer solutions and changes to that or tailor public health programs to support something that could make it an effective change within that, they can also be utilized very much, you know, to help us understand what people are eating when and why and the food industry like to know that information, what nutrients are particular foods, you know, coming from, so where is the saturated fat coming from? Is there a way that we could reformulate to reduce that, you know, and that's where the food industry and public health come work together with those. And then ultimately, the consumer wants to know this information a little bit more on how as well, the consumer is very interested in monitoring themselves. And I can see very much the the tech technique that I'm working towards, and the technologies that we can apply this to, could make this a simple tool for individuals to collect their dietary intake data and give, you know, potentially give feedback at an individual level to say, you know, if you change from this particular, you know, meal to this meal, you could improve your overall nutrient intake of whatever nutrients that they need to improve. So it's allowing the consumer to be slightly empowered to or to track their own intake and to make more informed decisions, I suppose with their own diet and health
Louise Holden 17:31
Could ask you Eileen, one of the benefits or values I can see coming down the line from this is an ability to, to to stratify recommendations a bit more, because you know, the food pyramid and so on is very helpful, but it's a bit of a blunt instrument, you've worked a lot with particular groups particularly worked with older people and their dietary needs, which are specific in some cases to that group. And are you continuing your focus on on these specific groups, or have you moved away now with these data, data sets and deliberations.
Eileen Gibney 17:57
So when we're working with the data, I suppose the ultimate aim for me is to be able to capture that data easily, but then also to be able to use it appropriately. So it is important that when we gather that data, that we can get the right feedback. And I've done quite a bit of work on this in previous projects, where we have taken dietary intake data, and we've been able to, to look at an individual level to see where an individual's nutrients are coming from and know their recommendations and say what they need to change to and why. And I would feel that this approach could also become part of that, and why I suppose I'm passionate about collecting at a meal level is that we eat meals, you know, and if we eat, we don't we eat foods we when we get but we get nutrient recommendations. So I might say to you or your high end saturated fast, but then I have to tell you where your saturated fat is coming from in your diet. And I have to offer you a solution to remove that and to change it. So what I'm looking at is how can we collect the data in a way that would allow us to make those recommendations a little bit more tangible for an individual. So I believe if you collect the data in a particular way, you can give the feedback in a very similar way. And I think giving feedback at a meal based level would allow an individual to say, Well, look, I commonly eat these three breakfasts. But if I switched to these three breakfasts, this would make my diet a little bit more healthier. And it offers that solution. And I mean, if you want to join it right the way through, if it's not a meal based level, I can very easily create a you know, a baskets for an individual for online shopping and join up that digital pathway for consumers to completely fix them or, you know, reach out to what restaurants are offering and draw that data that exists in our everyday life in an online form, to you know, to create that end to end solution for individuals. So that's my take over the world vision. But I do believe we're there, I think I think we can get there. But you know, we've got to do the segments properly. But I suppose the main bit is, we could get there very quickly. But as a nutritionist, I want to get there accurately. And that's the big thing.
Grainne Faller 20:19
In the course of the next few weeks and months, we're going to be exploring Insight's new research challenges. Today, we're going to be looking at Research Challenge One sensing and actuation. Louise, can you tell us a bit more about that?
Louise Holden 20:30
Sensors are all around us these days, and they're collecting vast amounts of data. Here at Insight, there's a lot of work being done making sense of that data right across the centre and we're going to talk today to Tomas Ward and Brian Caulfield, we started out by asking them what is sensing and actuation.
Brian Caulfield 20:45
So sensing and actuation is, is actually an extension of in Insight 1, what was personal sensing. And I think we have kind of three main pillars in sensing and actuation. The first thing is, we're interested in developing the next generation of sensing technologies, using a lot of new types of sensing platforms, going after sensing targets that are not currently available to us. The other thing we are trying to do is using existing sensor technologies and the new ones, as they come through the pipeline, try and understand how we can either better capture human performance, mostly human performance, as people move throughout their daily lives and understand human behavior and human performance as people move through their daily lives. Regardless of whether the context is in health or wellness or sports or, you know, any other aspect of behavior. And then finally, we're interested in seeing how the things we learn about human behavior can be used to enhance or optimize human behavior. That's kind of a in a nutshell, it's one of those parts of insight that brings a lot of people from a lot of different disciplines together.
Tomás Ward 21:58
Yeah, that's,that's pretty good. Brian, I like that description. Because it also, it allows us to move from maybe previously under Insight 1, there might have been a particular emphasis on wearable sensing, and while wearable sensing is still probably, you know, the the huge focus and insight to we brought that we brought in this idea of sensing it kind of sensing in our, in our environment, right. And that's everything from maybe, you know, you know, smart living technologies, but even kind of more larger scale sensing in your local region, where you live in your town, in your city, in your campus, in your country. And because these all affect how we live, and they all affect how we how the qualities of our lies in everything from just personal health to just, you know, I suppose, you know, our mental health, if you like, you know, how we enjoy living. So insight, I think insight is, has nicely pivoted or broadened from a world where we were very much focused on sensing of the individual to maybe know, efforts where we're looking at the collective a bit more and particularly knowing the context of COVID. You can imagine how important that is. Because it's only as a kind of a community and working together that we can, you know, I suppose, improve our current circumstances, you know,
Brian Caulfield 23:19
Yeah. And another, another extension, this is something we frequently do, we start to talk about this as we we actually forget about all the things we are actually doing, until we start talking about another extension from Insight 1 is that in Insight one, when we spoke about optimizing behavior change, we were primarily focusing on how information or knowledge could help people change behavior. And in Insight 2 we're actually getting into trying to directly actuate or influence behavior by means of, you know, leveraging, bridging that gap between sensing people's behavior, and then using that information to actuate, for example, soft robotics platforms that can help people directly help people change the way they move.
Unknown Speaker 24:06
That's, yeah, that's another very interesting thing to reflect on. Because that was, in Insight 1 that sort of research or, you know, previously, we wouldn't have done so much work in that area before. So that's a real innovation within Insight 2 this time, Brian, isn't it? And again, the full scale of things, the full scale of I suppose behavioral change takes place, as Brian says, from the very direct and that at the wearable, that'd be a wearable system, Brian, right. To again, at the collective scale, where we can present feedback at a community level, you know. So yeah, I think so, that reflects as far as the, the growth in ambition, I suppose, and scale of activities. That insight can now address given the size of the center. I mean, it's one of Europe's largest data analytics research centers. And so it's, I'm delighted to see that we have we have that ambition to operate that sort of scale and, and have a bit of a breath and a vision of problem sets that are that are commensurate with as well as the huge investment. That's that, that the Irish taxpayers made us, you know, and Irish industry are continuing to back, you know, that's great.
Louise Holden 25:25
And can I ask, obviously, one of the objectives is to consolidate a lot of that work that's happening under this this theme across the different sites? How is that collaboration, looking at the moment, obviously, taking into consideration but just generally, because you're geographically spread?
Brian Caulfield 25:41
I suppose one of the contradictions in terms of our own COVID is that it's sent us all back into our homes. But it equally has stopped us from traveling all around the world. So we've actually had a lot more opportunities to, to speak to each other than we may have done in the past. Because in the past, I think we tended to rely a lot on face to face, you know, there were opportunities where we got face to face. And we didn't use platforms like this so much as we as we've learned to use them now. So I know that we've had a lot more interaction across sites and a lot more good discussions about collaborative activities in the last eight months than we would have had in the eight months leading up to it, I think. And there's a lot of really interesting activities that I think are gathering up ahead of steam. For example, on some of these platform research initiatives that have been sparked off and incise, in recent times have been a great vehicle for for bringing people together to go after common challenges. And in sensing and actuation. There's a couple of them that people are coalescing around. Tomas has just recently kicked off the platform Research Initiative around development of novel sensors for, for health, for personal performance and health. That's been kicked off by Margaret McColl, which I think is really interesting was just trying to link the neck, the development of next generation of sensors, through to the applied research that's been done in other parts of insights. So yeah, there's quite a lot of things that are kicking off. And Tomas and myself have been continuing our collaborations that we've, that we started in Insight 1.
Grainne Faller 27:35
What will success in this research challenge look like?
Tomás Ward 27:38
What I like, and Brian was just discussing it there. And I like the fact that we are there's a focus now on cross challenge activities that predict through these these platform research initiatives that are bringing people together. And I know maybe from the outside people say why there's a bunch of, you know, you know, data scientists, computer scientists, engineers, and, you know, movement scientists, aren't they all want this, right. But even within these, you know, it's amazing how siloed people get even within, you know, and there's sub disciplines and disciplines, within sub disciplines, and so on, it's very easy for people not to talk to each other. And then the theorize are definitely interesting. So what I'd like to see in a few years, that we're able to tell stories, there's a bit of a narrative where we can tell stories of the success of data analytics, it was specific initiatives and projects that the average public can understand. And that are tangible and relatable to. And when we examine them, and try and pick apart the contributions, I want us to struggle to pick apart those contributions. So I don't want you I want us to see the stories, we won't be able to say, this is all research, challenge one or research challenge two, it'll be it'll be like a Christmas game, you know, it'll be at a massive, every challenge just mashed together. I want that, you know, so that's what I'd like to say, where everybody will be putting their hand up and owning their success, you know, or potential failure. There has to be risk, there has to be risks, right? When I bring that in, we're doing we're engaged in research. It has to have risk. There's no risk. There's no there's no game. It's not cutting edge, right. So yes, there will be some failures, I imagine. But we react very, very, we pivot very responsibly to failures and generate success, everything. But anyway.
Brian Caulfield 29:32
Yeah. And I think another thing that I would say is that, like, I don't want to sound like I'm coming across being, you know, saying something that's almost like, you're a no some type of beauty pageant here, but we're looking for world peace. But ultimately, I think if we have a set of graduates, you know, PhDs graduates and people who have come through as postdocs in inside at the end of it. You want them to look back and You know, if they came in having a computer science background or they come in with a movement, science background, wherever that they, they go out having that a really interdisciplinary understanding. They've gone really, really deep in their own area, but they also understand some of the other areas that they've worked with, and that they come out of incise haven't been really proud of the fact that they actually came through insight, you know, that they, they, they, they feel like that has been a real benefit to them and their career development. I think that that's the true success of insight.
Grainne Faller 30:37
That's it for episode six of The Insight podcast, do rate review and subscribe. Wherever you get your podcasts, we'd really do appreciate it. Thanks so much for listening. And thank you to our contributors Niall O Brolchain, Eileen Gibney, Tomas Ward and Brian Caulfield. We'll see you next week for another round of data science straight to your ears.
Louise Holden 30:57
Bye.
This has been a snoring dog production on behalf of the insights SSI Research Center for data analytics.
Transcribed by https://otter.ai