This paper began with a simple but challenging question: why do some river networks show predictable patterns in carbon dynamics, while others show sudden and unexpected changes? Rivers may appear simple - a stream flowing through a landscape toward the ocean-but they are among the most dynamic parts of the global carbon cycle. Along their journey, rivers receive carbon from forests, soils, wetlands, groundwater, and the atmosphere. This carbon is transformed, stored, transported, and sometimes released back into the atmosphere as a greenhouse gas. Understanding how these different sources and processes combine across landscapes has been a long-standing challenge in freshwater science.
For many years, two different perspectives have emerged. Some studies have shown that carbon changes gradually as streams become larger rivers, following predictable patterns across the river network. Other studies have demonstrated that local landscape features such as wetlands, lakes, peatlands, and groundwater connections can create strong changes over very short distances.
So which view is correct?
Our new study "Variable scale domains reconcile continuous and patchy carbon dynamics in river networks", published in Nature Water by Laudon et al, 2026 suggests that both are.
A 40-year journey through the Krycklan Catchment Study
This work was made possible through the unique opportunity provided by the Krycklan Catchment Study (KCS) in northern Sweden. For more than four decades, researchers have collected observations across a network of streams, rivers, and surrounding landscapes representing boreal ecosystems. These long-term measurements have created one of the most detailed records of how freshwater systems function and change over time. Long-term research infrastructures are valuable because environmental processes often unfold slowly. A short-term study may capture a snapshot, but decades of observations allow us to identify persistent patterns, understand variability, and discover processes that only become visible over longer timescales. The Krycklan dataset allowed us to move beyond individual streams and ask a broader question:
What controls carbon dynamics across an entire river network?
A new way of viewing river carbon dynamics
To address this question, we developed the Variable Scale Domains (VSD) framework.
The central idea is that river carbon dynamics are controlled by multiple processes operating across different spatial scales. Some controls are connected to the structure of the river network itself. As water moves downstream, changes in catchment size, hydrology, and connectivity create patterns that can often be predicted. However, rivers are also shaped by local landscape features. A wetland connected to a stream, a groundwater discharge area, or a lake upstream can act as a hotspot where carbon enters, changes, or is stored differently from the surrounding landscape. A useful way to think about this is that a river network is influenced both by the major pathways that connect the landscape and by the unique characteristics of individual locations along the way. The VSD framework brings these perspectives together. Rather than asking whether river carbon dynamics are controlled by continuous downstream processes or isolated landscape patches, it recognizes that both occur simultaneously. The dominant control depends on the type of carbon and the environmental setting.
Different forms of carbon tell different stories
Carbon in rivers is not a single component. Different forms of carbon respond to different environmental processes. Dissolved organic carbon (DOC) is strongly connected to soils, vegetation, and organic matter entering streams from the surrounding landscape. Dissolved inorganic carbon (DIC) is often influenced by groundwater interactions, weathering, and biological respiration. Carbon dioxide (CO₂) and methane (CH₄) reflect transformations occurring within aquatic ecosystems and their connections to surrounding environments. The VSD framework helps explain why these different carbon forms can show different patterns across the same river network.
Why does this matter?
Freshwater ecosystems are increasingly recognized as important parts of the global carbon cycle. Rivers do not simply transport carbon from land to the ocean they actively process it along the way. Climate change, changing precipitation patterns, land-use change, and ecosystem disturbances can alter where carbon originates, how it moves through landscapes, and how much is released back into the atmosphere. To predict future changes, we need approaches that capture the complexity of natural systems. The VSD framework provides a way to understand how processes operating across multiple scales combine to shape freshwater carbon dynamics.
A discovery built on collaboration and long-term observations
This paper represents more than a new conceptual framework. It represents decades of collaboration, field measurements, data collection, and scientific curiosity. The Krycklan Catchment Study highlights the importance of maintaining long-term environmental observations. The discoveries we make today are possible because previous generations of researchers invested in collecting consistent measurements over many years. As environmental change accelerates, long-term research infrastructures will become increasingly important. They provide the foundation needed to understand complex Earth system processes and improve predictions of how ecosystems will respond in the future. We hope the Variable Scale Domains framework encourages researchers to view river networks differently—not as systems controlled by a single scale, but as interconnected landscapes where multiple processes interact across space and time. Understanding these interactions is essential for predicting the future of freshwater ecosystems and their role in the global carbon cycle.