Global environmental crises

They are now considered to be the most far-reaching global environmental crises – and a twin crisis at that: climate change and biodiversity loss. On the one hand, humanity is currently struggling to limit the rise in global temperatures in order to prevent the worst effects of climate change on the planet. At the same time, the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES) estimates that one million species worldwide could become extinct in the next few decades. Scientists around the world are working to quantify the impact of these dramatic changes. This means building a body of data-based evidence to understand the challenges and identify ways in which humanity can address them. One of these scientists is Miguel Mahecha, a professor at Leipzig University and Head of the Environmental Data Science group at the Remote Sensing Centre for Earth System Research (RSC4Earth). “Global warming and changes in biodiversity go hand in hand. And even though the mechanisms behind the two trends are partially different, their efects are interrelated. If we are to understand this connection better, we need a collaborative approach,” he says.

Mahecha, a remote sensing expert, works at the Institute for Earth System Science and Remote Sensing at Talstraße 35, which he has headed since October 2022, two and a half years after he started his professorship. His office at the end of the corridor is freshly refurbished, the floorboards gleaming, the walls still rather bare, and a desk and two chairs will have to be enough for now. He has obviously not had much time to set things up. But that fits into the picture. Global warming and species extinction are accelerating at an alarming rate, humanity must react quickly and science wants to deliver. Mahecha’s contribution is to work analytically with large amounts of data in order to be able to make quantitative statements. “Thanks to remote sensing, citizen science and monitoring programmes, we now have a lot of global data at our disposal. This also makes it possible to understand global crises,” he explains.

The 44-year-old’s personal eureka moment came during his geoecology studies at the University of Bayreuth in the early 2000s, he says, a cup of coffee in his hand: “I was involved in a biodiversity project in Ecuador back then, and my job was to visualise the data in Excel. But it wasn’t working.” One afternoon in the library, Mahecha read an article about dimensionality reduction – an approach that can be used to compress large amounts of data in order to make it easier to identify trends or statements in very large data sets. “I found this fascinating, and my professor at the time let me focus more on this, and then we just kept experimenting and compressing data.” In this way, he identified a gradient in vegetation succession in the Ecuador data set that had previously gone undetected. “That’s when I understood that you can learn new things from data without conducting (as before) traditional hypothesis-driven research,” he says. “This moment of surprise was exciting. And since then, I have realised that data-driven research will change the world.”

Focus on interdisciplinarity

While the use of data science was avant-garde more than 20 years ago, it is now considered part of the standard research repertoire. “Data science has become mainstream, and we are well positioned here at Leipzig University, especially thanks to the establishment of ScaDS.AI,” concludes Mahecha. After completing his doctorate at ETH Zurich in 2009, he spent more than eleven years as a researcher at the Max Planck Institute for Biogeochemistry in Jena, where he headed the Empirical Inference in the Earth System research group. He has been in Leipzig for three years now. “What we are doing here is a new kind of research and teaching that is data-based, interdisciplinary and methodologically new,” says Mahecha, describing the research being carried out at his institute. This is evident in many ways, including the backgrounds of the research staff: “My team is as interdisciplinary as possible. For example, my colleagues come from geography, applied mathematics, theoretical physics and bioinformatics, and there is also a data journalist on the team. What matters to me are the skills that my staff bring with them. The team can only be successful if they complement each other,” he says.

This interdisciplinarity is also reflected in the Remote Sensing Centre for Earth System Research, which was founded by the Faculty of Physics and Earth System Sciences and the Helmholtz Centre for Environmental Research (UFZ) in 2020. Mahecha is one of the four principal investigators (PIs); the other three are experts in remote sensing of the water cycle (Professor Jian Peng), remote sensing of biodiversity (Professor Hannes Feilhauer) and remote sensing of soils (Professor Michael Vohland). The centre brings together Leipzig’s researchers from a variety of disciplines to work together on issues such as the impact of ecosystems and water resource dynamics in the context of global change across spatial and temporal scales. “Everyone at the centre has a different research specialisation, but methodologically we speak a similar language that focuses on modelling and the use of artificial intelligence,” says Mahecha. It is like a jigsaw puzzle. Each PI contributes their expertise, which is then pieced together to form the bigger picture.

Interdisciplinary teaching

It is clear that this interdisciplinary and data-driven research needs to be brought into teaching. And this is happening. In the winter semester 2022/23, for example, the faculty launched the English-language master’s programme Earth System Data Science and Remote Sensing under the direction of Hannes Feilhauer, a programme that aims to teach students technologies and methods of environmental data science and remote sensing. “In the first year, students from all over the world with backgrounds in engineering, oceanography, biology, computer science and geography already applied to the programme – without us even advertising it. The degree programme was filled almost immediately,” says Mahecha, for whom teaching is important and one of the reasons why he moved from a Max Planck Institute to the world of higher education. “I care about the next generation of researchers. They are excellent students, and we provide them with comprehensive training here. This is also why I feel obliged to pass on something valuable to them,” he explains. The beauty of this is that it benefits not only the students, but also the lecturers – including Mahecha himself.

One day, he says, a computer science student came to his office to talk about environmental data. Together they came up with the idea of an interactive application that could visualise climate data in the form of a cube with two spatial dimensions and a time dimension. Working with colleagues from the Institute of Computer Science, doctoral researcher Maximilian Söchting developed the 3D visualisation lexcube.org, which processes terabyte-sized satellite observations and displays them in a three-dimensional cube. You can use your mobile phone, for example, to see a very clear visual representation of the state of the forest in the Hainich National Park traced back over the past years. “Since we can see much faster whether the data is okay or where there are errors and irregularities, this visualisation is very useful for quality assurance of data sets,” says Professor Mahecha, who shows us how, with just a few quick clicks of the mouse, he can see global temperature trends for the past 40 years on his screen. “Such unexpected collaborations are a regular occurrence at our university. I think this is a great asset,” he says.

Intelligent big data management

Data management is one of Mahecha’s core competencies, but he sometimes sees the abundance of this type of data as a double-edged sword. On the one hand, the satellite data used at the Remote Sensing Centre, for example, is an essential basis for the research there. Thanks to artificial intelligence methods, there seem to be no limits to data analysis, allowing researchers to bring new findings to light. On the other hand, he says, more data does not always mean more information because the data can be redundant and sometimes contains erroneous information. And along with the global, it is also necessary to analyse local data. “If I want to understand global trends, I also need to know about local processes,” he says. For example, his colleague Dr Sebastian Wieneke is researching at the arboretum in Großpösna to determine how the photosynthetic rates and fluorescence of different tree species vary depending on stress and radiation. “Only by understanding how fluorescence works in individual trees can we understand the corresponding global satellite data and subsequently conclude how entire ecosystems respond under stress.”

Managing big data intelligently is therefore crucial, especially when your team is as diverse and interdisciplinary as Mahecha’s – as the group’s research portfolio shows. Xaida (eXtreme events: Artificial Intelligence for Detection and Attribution), for example, is the name of an EU project in which Mahecha’s working group, which includes researchers from 15 international research institutions, analyses extreme events. The aim of the work at Leipzig University is to understand which climate conditions, and which losses of ecosystem services, lead to humanitarian disasters such as crop failure and famine. Khalil Teber, a doctoral researcher, uses machine learning methods to try to determine how linked extreme events, such as extreme droughts followed by floods, can have particularly catastrophic effects. “Such international projects are important not only because we can maintain contacts, but also because we can draw on the expertise of other working groups,” he says. He was particularly pleased that two outstanding early career researchers also knew about the joint research in the Xaida project at the Leipzig site and that they then decided to move here from abroad. Since the summer semester 2023, junior professors Marlene Kretschmer and Sebastian Sippel have been working at Leipzig University’s Institute for Meteorology on climate extremes and attribution, i.e. how extreme weather events can be attributed to climate change. They have significantly expanded the work being done on climate extremes at Leipzig.

Huge research project launched

Their expertise will also be used in the planned Breathing Nature cluster – a huge project headed by meteorologist Professor Johannes Quaas, with which Leipzig University has applied for a Cluster of Excellence in the current round of the Excellence Strategy of the federal and state governments. A total of 25 PIs are involved as well as a further 31 professors from the fields of biodiversity, climate and social research, along with many other research institutions. The aim is to gain a better understanding of the effects of climate change on the environment and society. “As a remote sensing centre, we bring expertise in monitoring and data analysis,” says Mahecha, one of the 25 PIs. This is particularly interesting because the researchers are not only studying typical cases such as global changes in forests and their impact on the atmosphere but also socio-economic data. “If Leipzig University is awarded the cluster, it will mean new professorships, new working groups and new degree programmes. That would be a huge step forward.”

But this is all still a long way off – as is the full proposal for a Collaborative Research Centre to be submitted to the German Research Foundation, which Mahecha is preparing with his colleagues. The focus here is on the processes that can exacerbate the occurrence of climate extremes such as droughts and heatwaves, that is, how biodiversity buffers the influence of climate extremes – or how a change in biodiversity can affect the atmosphere in a way that makes climate extremes more likely. It may sound like a bold statement, but for Mahecha it’s logical: “Biodiversity doesn’t only refer to the number of species. A change in biodiversity often means the loss of a particular function,” he explains. And that function could be important for the whole ecosystem. For example, if a certain plant species disappears, then the mobilisation of nutrients, the absorption of carbon dioxide or the regulation of temperature can change. A high level of biodiversity therefore makes sense because only in this way can an ecosystem with a large functional diversity respond more effectively to external influences.

A basis for social and political decision-making

It is clear that there is a lot going on at Leipzig University in the fields of climate and biodiversity research, remote sensing and data science. If all goes well, this area of research in Leipzig could receive an enormous boost. But science is one thing; the question is to what extent the newly generated knowledge will reach society and policy-making. “First of all, it is very satisfying to be able to quantify certain processes like the effects of climate change on ecosystems because this creates a basis for social and political decision-making and action,” says Professor Mahecha. Politicians have a good understanding of these problems, but the existing knowledge is still not being used enough. “Social discourse suggests that the issue of climate change is negotiable, but, of course, it’s not,” he says. What we have are uncertainties in prognosis and diagnosis. “But climate change itself has been here for a long time, and there is no disagreement about it in the scientific community.”