The Restoration of Ecosystems Degraded by Anthropogenic Development

December 13, 2021

A conversation with Dr. David Moreno Mateos, Assistant Professor in the Department of Landscape Architecture at the Graduate School of Design and the Department of Organismic and Evolutionary Biology at Harvard University.

Deforestation in the Brazilian Amazon Source: The Guardian

Elena Bouldin: Thank you very much for agreeing to speak with me. I wanted to ask you about your research, since ecosystem restoration is critical right now. When you study the health of an ecosystem, what variables do you study? What aspects of the ecosystem do you focus on? I say this because studying the whole ecosystem would obviously be very complicated, one cannot measure everything….

Dr. Moreno Mateos: Well, that is a very good question. One of the big issues we are trying to understand in ecology is how to measure change in an ecosystem. Any kind of change. In this case, specifically, first the change generated by human development for any economic activity involving the degradation of nature. And then, the change obtained in the recovery process. This is one of the most difficult aspects, because what we really measure are very simple things, which are not representative of what an ecosystem is. For example, biodiversity, or soil carbon, or things like that, are very, very, very basic abstractions of what the complexity of ecosystems really implies. So, what I am now devoting more attention to, is precisely how to measure change in an ecosystem, always within the recovery part, but somehow capturing the complexity of the ecosystem. We look at the interactions between plants, soil microbes, and how these expand to ecosystem functions. All of that, in a way, captures a little bit more of the complexity in ecosystems, but it's really still a simplification, because you think of the ecosystem as everything: you have: trees, plants, insects, birds, mammals, soil microbes, which are fungi and bacteria, and then micro invertebrates, nematodes.... Everything is interacting. All of those organisms are interacting with each other. When you come to a forest, cut it down, and turn it into a cornfield, all of that completely disappears. All of it. And we're talking about thousands and thousands, if not tens of thousands, of interactions, because it's really not very clear how many there are. And you destroy them in a day, without really knowing how long it's going to take to even recover that again. The place is never going to be the same again, but you want at least to get to something similar to what you had originally, of similar complexity, of similar functionality.

You try to find the measurements that give you an idea of the complexity of the system. I have focused on interactions between plants and soil microorganisms, which I hope to expand to other types of interactions. If you're measuring a whole plant community in an ecosystem that's not very complicated. How many microorganisms can you identify interactions with? We're talking about hundreds of possible interactions in a simple forest here in New England, which is not the most diverse in the world, but is much more diverse than those in Spain, in most cases. The point is that it's a small approximation, very simple, but a little bit less simple than the typical approximations of the species that are out there, which is called species composition or diversity.

Ecosystems have complex webs of interactions

Elena Bouldin: Of course, plants are at the base of food chains and biomass. It makes sense to study that first, because everything else comes from them, right?

Dr. Moreno Mateos: Yes. One of the reasons for choosing that type of interactions is that one of the most important structural and functional components of ecosystems are plants, so focusing on that and seeing how it expands upwards, with birds and insects, and downwards, with fungi and bacteria, for example, is the idea of all this. To keep expanding, as far as you can.

Elena Bouldin: How exactly do you study these relationships with plants? I suppose you take soil samples and study them chemically, but I don't know what else you do.

Dr. Moreno Mateos: Well, we are looking specifically for the interactions between plants and soil microorganisms. What we do is we take soil samples, as you say, and in those soil samples we look for the roots of the trees that are there. So what it's all about is identifying what organisms are interacting with the trees through the roots. We look for the different types of fungi that are interacting with the trees, which can be mycorrhizal fungi, and other types of endophytic fungi, which live inside the roots, or myco parasites, which are fungi that parasitize fungi, all of this to make that network more complex. And we also look for interactions with bacteria, for example, bacteria that eat fungi, or bacteria that affect the tree in a certain way as pathogens, for example, and those types of functions. Then we take that soil sample, remove the roots, and sequence them. And by sequencing that root, you can get the DNA of all the organisms present, either inside the root, or in what is called the rhizosphere, which is the root environment, just sticking to the root, so that you say, this bacterium is interacting with the tree. That would be a first step. The ideal would be to identify what the role of that interaction is over time. But that's much more difficult because individualizing the interactions is not easy. It's not easy to determine that this bacterium is doing this with this tree. That's more complicated. At the moment we are looking at the big picture of what the structure of the interactions looks like, so that we can gradually understand what effect those interactions have on the ecosystem.

Mycorrhizal fungi and bacteria are found n the soil and roots Source: Frontiers

Elena Bouldin: On your web page, where you talk about your research, you mention that you combine empirical research with meta-analysis. What is meta-analysis? or What does it require?

Dr. Moreno Mateos: A meta-analysis is basically a way of collecting information on a topic that you think is important, based on many, many studies. What you do is synthesize all that information and see what the general patterns are in the answer to your question. For example, what we have done for our meta-analysis is to look for an answer to the following question: Do ecosystems recover differently if you are looking at diversity or carbon cycling?

Then you want to see if they recover earlier or later, if there is a difference. So in some meta-analyses that we're doing now, for example, we're looking at that over time. We pick studies from a bunch of geographically different sites, we pull the data, and we put it into a common database and we look at the behavior of that data in a much more global way. Because it's one thing if you find, for example, that forests are accumulating a lot of carbon in Malaysia, and it's another thing if you find that forests in Malaysia, along with forests in Vietnam, South Africa, and Brazil, are accumulating a lot of carbon. These are very different things. A meta-analysis collects information from many sites, puts it together, and builds a single database that gives you a more general view of the questions you're asking. For example, you can say, okay this process that's happening in Malaysia, we're seeing it actually happening on a global scale, at least in the many studies that we've collected information from. With meta-analysis, what you do is you give more global answers to processes that have been studied more locally.

Elena Bouldin: So you spend a lot of time studying other studies and making connections between all those studies!

Dr. Moreno Mateos: Yes, that is a meta-analysis. But it's not that I spend a lot of time either, I spend a certain amount of time. Yes, I have spent a lot of time in the past. The truth is that when I was doing my post doc at Stanford and Berkeley I spent a lot of time with this and it was an infinite boredom. This looking for data in other people's studies is a deadly bore! But, the reality is, when you find these general patterns, it's very rewarding, because you say, wow, this is really interesting.

But now we do a lot of field work. We're still doing meta-analyses, in fact, we're working on two different ones right now. But we also do a lot of field work, because one of the nice things about doing meta-analysis, is that you say, okay, the general trend seems to be this, and once you understand that, you start to understand the overall process, and the first question that comes to you is: why is this happening? And that helps me a lot when it comes to defining empirical projects of going into the field to collect data and soils to understand why those general processes that you've identified in the meta-analysis are happening. It is a logical sequence of work, but what I like most is to do field work. I always enjoyed going to the countryside. I liked the countryside since I was very young and I still like it.

Source: AidGrade.org

Elena Bouldin: I understand perfectly well because I am the same way! Now, while I was listening to you, it occurred to me that maybe we could find a way to automate this meta-analysis that requires so many hours of work. Do you think it is possible to do some kind of automation, developing a program, or something with artificial intelligence, to identify these common processes or trends in the studies?

Dr. Moreno Mateos: Well, I don't know very well, you ask me some great questions! They are the same questions I have been asking myself. I would love it to exist, but it does not exist at the moment because you cannot clearly define very specific search parameters, even if you have them very clear in your mind. At the algorithm level it is far from obvious. I imagine that, in some time it could be done. And I have seen some approaches that have done similar things, but in much more concrete aspects. That is, you are looking for a very, very specific type of data, and a unique one. Yes, I have seen people who have used artificial intelligence algorithms, which is not really artificial intelligence but "machine learning", which is a lower level of complexity, because they are not algorithms that learn about themselves and improve, they are static algorithms that, yes, have some capacity for improvement, but it is not artificial intelligence. Anyway, yes, I have seen it sometimes, but for very specific things. It would be great. In fact, when I started with meta-analysis I thought, "if only I could put all the databases there and an algorithm could tell me which ones are the good ones!" But for the moment what we do is spend hours and hours, well, months, on this kind of things to get the results.

Elena Bouldin: Well, maybe in the not so distant future it can be done. I was recently talking to a professor here at Berkeley, I don't know if you know him, his name is Todd Dawson.

Dr. Moreno Mateos: Yes, yes, I know him. I was at Berkeley for three years.

Elena Bouldin: Yes, I know. I read it somewhere, that's why I thought maybe you knew him. Well, so I talked to him....

Dr. Moreno Mateos: He is the one who does echo physiology in trees, right?

Elena Bouldin: Yes, exactly. Well, he told me that, before, to measure all the parameters they had to measure in the trees, they went to the forest, they climbed the trees one by one, and of course, it took a lot of time. Now they can do that with drones. Sometimes they still climb, but drones are much faster and they can also analyze many more trees in much less time. So, well, you never know. Sometimes technology helps a lot and radically changes the way we operate.

Dr. Moreno Mateos: Yes, yes, it helps a lot. We have not found an artificial intelligence algorithm that allows us to do this, but for example, the issue of genomics as it has developed is an explosion of information that was unthinkable before. Now you can sequence basically all the microorganisms in the soil with a few samples of some brutal interaction networks, super easily constructed. There are still many things to improve, it is true, but this has been a brutal change that has allowed us to understand the complexity of things a little better. So yes, technology is essential, we must always try to be as up to date as possible with technology to get the best possible information in science.

Elena Bouldin: Another question, a little bit different. I understand that you study both terrestrial and aquatic ecosystems. How do the recovery studies of these ecosystems differ? What do you measure in each type of environment?

An aquatic ecosystem

Dr. Moreno Mateos: They are different from each other because you cannot look at the same things. If I am interested in the interactions between organisms, for example, plants and soil microorganisms, I can do that in a terrestrial ecosystem, in quotation marks, more easily than in an aquatic ecosystem, because the interaction between plants in wetlands, for example, is much less important. There, plant growth is driven by the type of water that's there, the time of the water in the wetlands, the concentration of nutrients in the water, and that sort of thing. Mycorrhizal fungi, for example, don't exist in water. So there are really no organisms that interact as strongly with plants in a wetland as with plants in terrestrial ecosystems. At least it is not very clear yet. To begin to understand the recovery of the complexity of these aquatic systems, you would have to look at other types of organisms. For example, aquatic ecosystems have very complex interaction networks in the water, and since all the micro invertebrates that live in the water have such complicated trophic chains, with organisms with larger zooplankton, with small fish that eat them, larvae, insects, and all that, all that complexity existing in the water does not exist in terrestrial ecosystems, because in the air there is not all that complexity. So yes, you would have to look at other interaction networks to understand that complexity.

Elena Bouldin: I also wanted to ask you: an ecosystem that has been impacted by mining is very different from an ecosystem that has been impacted by deforestation, I guess. Obviously, there are other factors as well, but, just in general, have you found differences in the time required for recovery of ecosystems impacted by mining compared to ecosystems impacted by deforestation? Is there any way to quantify those impacts?

Dr. Moreno Mateos: That would be great. What happens is that we need a lot of data and we have not yet been able to get enough data to be able to quantify the impact. But we have seen small patterns, that is, it is clear that if an area is affected by mining, the recovery is different than if it is affected by logging; we have compared agriculture with timber extraction or logging, and yes, we have found certain differences. For example, in a meta-analysis we are finishing now, we found that the areas that were affected by agriculture recovered better than the areas that were affected by logging, surprisingly. It seems like it would have to be the opposite, but for the parameters that we looked at before, which were pretty simple, like diversity, carbon, and soil carbon, we saw that the effect of agriculture on recovery was more positive than the effect of logging, for some reason. Something that we're really not very clear about, because what one thinks is a little bit the opposite. But it seems to be that the effect of having more nitrogen speeds up the recovery processes and makes at least these very simple parameters recover faster. But, we did a study where we evaluated 8 impacts trying to see differences in those 8 impacts on recovery, and we really weren't able to find very clear differences. But there have to be. It's just that there's never been a database with enough data, in fact, to be able to test this kind of thing. But yes, the idea is to try to see what effects there are, and in fact in the meta-analysis we are doing we are comparing those two: logging and agriculture in forests, which are the two most important impacts in forests.

Land deforested for agriculture Source: The Guardian

Elena Bouldin: If regenerative agriculture is started, can it be considered a partial restoration of the ecosystem?

Dr. Moreno Mateos: Regenerative agriculture has things that are much better than conventional intensive agriculture, clearly. It would not be a restoration in the strict sense, but it is a recovery of certain ecosystem processes that are lost in conventional agriculture. I think that regenerative agriculture is a great idea, as long as it allows us to have enough production so that people do not starve.

One of the trends that favors nature conservation is the ultra-intensification of agriculture, although it may seem contradictory. Because if you intensify agriculture a lot and make small areas super productive, as long as you control the pollution that comes out of those areas, you are leaving many areas free for nature conservation and restoration. You put all the impact in smaller areas. So, it's complicated. There is no obvious answer to that, because regenerative agriculture, just like organic agriculture, is great, but both of them require large areas of land.

Elena Bouldin: But if you concentrate agriculture and make it very intensive only in certain areas, the soil will degrade so much that there will come a time when you will not be able to use it and you will have to go somewhere else….

Dr. Moreno Mateos: Right. You have to live with the knowledge that that same piece of land is going to be the same piece of land forever. You can't say okay, since this one has degraded, let's go somewhere else. No. The intensive agriculture that I am talking about is a type of agriculture that does not move. You develop a technology that allows constant growth on the site with technological tools. For example, there's the bell pepper type of farming in Holland. So they have these bell pepper farms with a productivity 15 or 20 times higher than the productivity of the normal pepper growth that is done in Spain. That is, it is ultra-intensive because everything is in greenhouses, everything is artificial, it has a very large energy impact, but the footprint in terms of area is minuscule compared to conventional agriculture. So, if you manage to maintain those greenhouses with renewable energies instead of oil to heat them in winter and all the pesticides you use and the fertilizers you use are organic, and there is a recycling of all that product, then that is the system, ideal in quotation marks. It's ultra-intensive, but you need a lot less land and if you do it in a sustainable way it's the perfect agriculture. Certainly that can't be applied to all crops. There are certain things, like wheat, that you can't have in ultra-intensive because what you have there are shelves. In fact, you don't really touch the soil. In most cases. You have different layers, which is often just water, and you have shelves where you grow peppers, tomatoes, beans, and other. But you can't have wheat on shelves.

Bell Peppers growing in a Netherlands greenhouse Source: FreshPlaza

Dr. Moreno Mateos: Do we want to depend on technological solutions for everything? Well, I don't know. In terms of nature conservation it's much better to do that, honestly, because, of course, you leave a lot of area free for elephants to roam the world again and all that sort of thing. But you're relying on very intensive systems that could potentially be more vulnerable to, for example, resistant pests. Imagine if a resistant bell pepper pest starts hitting all the greenhouses in the world. Then you are screwed, but it is true that you can also develop technologies for that, because if you detect a resistant pest, the good thing about greenhouses is that they are closed systems. You can close the greenhouse and the pest will not enter, unless you put it in. Then, if you control the origin of the seeds, etc., those pests can be controlled to a great extent, but I can always find one that is not, such as fungi, because fungi are extremely mobile.

These are somewhat philosophical questions, sometimes, but they are really happening right now. In other words, ultra-intensification is really happening and has been happening for many years. In Spain, for example, it exists throughout the southeast. There are large mega-productive systems that use a minuscule amount of land and have very high productivity. It is complicated. That is why I say that regenerative agriculture is great, yes, in principle. Conceptually it sounds like paradise, but regenerative agriculture needs a lot of land and that is not very positive for nature conservation.

Elena Bouldin: Yeah, that's true. I hadn't thought of it that way.

Dr. Moreno Mateos: It's complicated. Everything has a price. We must think about the consequences of the actions we take. It is the same as nuclear energy. Is nuclear energy very dangerous? Well, honestly, at present it is not very dangerous because it has multiple safety systems. People have a somewhat irrational fear based on certain accidents that have happened in history, but that can no longer happen.

Elena Bouldin: Compared to others, it has actually a lesser impact, right?

Dr. Moreno Mateos: Of course, so, today I think that, until the energy issue is solved at a global level with mega-productive energies such as fusion, this is the best solution for a while. I don't like mortgaging certain parts of the world with radioactive waste, but I think it's much better than burning oil.

Elena Bouldin: The last thing I wanted to ask you was about landscape architecture. Do you determine what is necessary to recover a specific ecosystem ? Do you say, here we are going to do this and here we are going to do that? I am not sure I understand what a landscape architect does!

Dr. Moreno Mateos: The reality is that I am not a landscape architect! I am an ecologist, I am a scientist. What happens is that I was hired in the Department of Landscape Architecture, because there is an enormous interest in integrating ecology into landscape architecture and developing more ecosystem restoration. Here in the United States at least, the role of landscape architects in restoration is becoming more and more important. Always, or in most cases, when there are restoration projects, there are landscape architects behind them. The department has long seen the importance of considering ecology at a deeper level in landscape architecture and reinforcing the role of restoration. That’s why they brought me here. But I really had almost no knowledge of landscape architecture when I arrived. I've learned a tremendous amount in these two years and I understand a little bit more of the potential that it has in all of this, which is huge.

Harvard Graduate School of Design Source: Harvard GSD

Dr. Moreno Mateos: Landscape architecture is a discipline that works at a scale that no other discipline works at. You are dealing with large areas where you have to consider a lot of different perspectives, because you work with ecology, you work with soil, you work with plants, you work with people, you work with space, you work with cities. It's a lot of factors. Your role is tremendously important, and the department has understood that. So, my role here is to continue doing my research but, at the level of teaching the students, I have to teach them to understand the importance of ecology in urban planning and landscape design. My role is to show them how to integrate ecology in all stages of the decision making process of these projects and, on the other hand, to stress to them the importance of restoration in the context of landscape architecture. It is true that most of my students are landscape architects, but I also have biology students who are interested in conservation issues. So I teach a class called Ecosystem Restoration, which starts in the Spring semester, and I have a little bit of everything: I have students from landscape architecture, biology, and also from other fields. I've had students from history, politics, all kinds, because it's something that people are becoming more and more interested in. That course is always at full capacity.

Elena Bouldin: It sounds very interesting. I read the description and I really liked it a lot.

Dr. Moreno Mateos: Ah, you read it, good.

Elena Bouldin: Well, I know we have been here a long time and you probably have a lot of work to do. I don't want to abuse your kindness.

Dr. Moreno Mateos: No, I've loved talking to you!

Elena Bouldin: Thank you. It has been very interesting for me.

Dr. Moreno Mateos: I'm glad. If you have any more questions, let me know.