Conservation and Causal Diagrams

Felix Weitkämper – German University of Digital Science

Maintaining healthy ecosystems is widely recognised as one of the greatest challenges faced by mankind in the 21st Century. As of 2025, the IUCN Red List records more than 48,600 endangered species of plants, animals and fungi. Endangered organisms range from tiny algae to mighty elephants, with widely differing ecological niches and often profound impacts on the plants and animals around them.

Preserving such staggering biodiversity requires an in-depth understanding of the conditions under which endangered species can survive, considering both their natural conditions and the changes brought about by human activity. This challenge lies at the heart of modern conservation science and ecosystem research.

Understanding Complex Ecosystems in Conservation Science

However, the huge range of different habitat conditions and specific niches occupied by plants and animals means that every organism and local region needs to be studied individually, and decisions used to be made based on only limited available data. After all, if the organism were abundant, it would not be a conservation concern.

This means that any model needs to incorporate all the information that is available, and that the expertise of various disciplines should be brought to bear on the issues at hand. In modern environmental research, scientists increasingly rely on causal diagrams and data-driven ecological modelling to better understand these complex relationships.

A Case Study: The False Tamarisk in Alpine River Ecosystems

I recently had the opportunity to assist a group of biologists and foresters with understanding the conditions affecting survival of an endangered alpine shrub growing along rivers in northern Italy.

The False Tamarisk, also known as the German Tamarisk (scientific name Myricaria germanica), is an important species in mountain regions ranging from the European Alps and the Caucasus to the Central Asian mountain ranges. Its strong and sturdy roots allow it to grow on the fast-changing gravel river banks in these alpine areas, allowing it to establish itself in areas hostile to other large river plants such as willows.

As their root systems make the gravel river banks more suitable for other species to establish alongside it later, the False Tamarisk is considered a “pioneer species” central to its local river ecosystem.

In the Alpine region, the False Tamarisk has decreased dramatically over the last decades. This is widely considered to be due to human activity such as straightening and dam building and its impact on gravel turnover. Recently, there has been an upsurge of interest in conserving surviving populations by restoring rivers to a state more closely resembling their natural condition.

Using Causal Diagrams to Understand Ecological Systems

The aspects that make this a particularly challenging problem include the limited availability of data, since there are only very few rivers that still have or had populations of the False Tamarisk over the last 15 years, and the complex interplay of the different contributing factors.

To gain a holistic understanding of this system, we used the technique of causal diagrams, an analytical framework promoted by Judea Pearl, who won the 2011 Turing Award (often considered the equivalent of the Nobel Prize for computer scientists).

The (slightly simplified) causal diagram that resulted from combining common sense, expert knowledge and statistical forecasting is depicted below.

Slightly simplified causal diagram showing the factors that ultimately influence the survival of False Tamarisk. Straight lines represent physical or biological effects, while wavy lines show the influence of human behavioural patterns.

Causal diagram explaining factors influencing False Tamarisk survival in alpine river ecosystems.

The diagram shows that even though so many factors are all correlated with the survival of False Tamarisk juveniles, ultimately only the width of the river channel and the presence of river bank protection has a direct impact. All the other factors are only important because they themselves impact channel width or the state of the river bank.

For example, rivers tend to be wider as they are further from the source. Hence, False Tamarisks are more likely to survive at lower altitudes. However, if a site has the same channel width and level of river bank protection, then the expected survival rate will be the same.

Interdisciplinary Collaboration in Conservation Research

Causal diagrams also help to tackle the key issues of studying such complicated and specific systems.

While the effects of river channel width and river bank protections on False Tamarisk can only be studied on rivers that have or recently had False Tamarisk populations, the relationships deeper into the diagram are more general.

For instance, the effects of altitude, river bank protections and protected conservation status can be assessed on any sufficiently similar river system, even if there have not been any False Tamarisk populations there at all.

The diagram also helps researchers from different disciplines to contribute their different perspectives to a unified understanding of the system.

Here, for example:

• Biologists provide expertise on False Tamarisk ecology
• Hydrologists understand river channel dynamics
• Civil engineers analyse river bank protections
• Social scientists examine human behavioural patterns

Thus, a single diagram can focus the efforts of disparate teams of researchers with very distinct methods.

The Future of Causal Inference in Environmental Science 

Ultimately, such combined insight will help policy-makers focus their efforts when trying to conserve individual species with very specific needs, such as False Tamarisk.

However, our use of causal diagrams barely scratches the surface of what is made possible by explicitly modelling cause-and-effect relationships in complex systems.

If you are curious to know more about this approach, I wholeheartedly recommend Judea Pearl and Dana Mackenzie’s The Book of Why, published by Penguin and Basic Books in 2018. In my view, it is one of the most insightful pieces of popular science explaining modern causal inference.

If you would like to learn about all the details of the study presented here, stay tuned for the special issue on non-model plants in the Applications in Plant Sciences journal, due for publication in mid-2026, where our article is due to appear.

Reference

Kailin Sun, Kilian Rückschloß, Tommaso Sitzia, Bruno Michielon and Felix Weitkämper: Pearl’s Causality for integrating ecological datasets: a case study on Myricaria germanica in northern Italy. Applications in Plant Sciences, in press.