This research was published in the INSPIRED: A Publication of the New Hampshire Agricultural Experiment Station (Spring 2024)
Researchers: W. Wollheim, T. K. Harms, A. K. Robison, L. E. Koenig, A. M. Helton, C. Song, W. B. Bowden & J. C. Finlay
Akin to the circulatory systems within humans, river networks serve as vital conduits for our planet’s wellbeing. Just as veins and arteries work with organs to filter out impurities, these waterways remove pollutants from diverse sources before reaching potentially more fragile downstream ecosystems. However, not all river networks perform this task with equal efficiency. This study sheds light on a remarkable phenomenon: as the size of a watershed increases, its ability to filter pollution escalates exponentially, a process termed ‘superlinear scaling.’ The research has important implications for New England—a region characterized by its variety of watershed sizes from small coastal watersheds to the large regional Merrimack and Connecticut river watersheds. Findings from this study can be used to support land-use strategies, which are crucial in maintaining local ecosystem balance.
Methodology
The research used an innovative model to assess how different-sized watersheds process pollutants. This modeling integrated river network structure (shape), channel hydraulics (widths) and biogeochemical rates, and is analogous to aligning the metabolic process of an individual living organism to the entire system of those organisms based on a well-established set of factors. While differences in biogeochemical functions of river networks may differ from those in portions of that network, these differences do not significantly alter the capacity to predict system-wide outcomes from the modeling.
Results and Impacts
The models’ results indicate that larger watersheds with larger rivers are exponentially more efficient at filtering pollutants due to the intricate interplay among these characteristics.
The results show that pollution filtration does not increase linearly with watershed size but rather at a superlinear rate—disproportionally greater than the size of the watershed. This finding also suggests that carbon dioxide emissions from rivers in larger watersheds add relatively more to global emissions than smaller watersheds per unit of land (Fig. 1).
Additionally, the findings highlight the importance of managing smaller watersheds, as they are less equipped to naturally handle pollutants, especially when there is a high rate of pollutant entry (high flow) into the river. The research also sheds light on rivers’ roles in the global carbon cycle. Larger river watersheds have a disproportionately higher input into large-scale cycling. However, this also implies that larger watersheds potentially play similarly disproportionately larger roles in sequestering carbon.
Key Findings
- Watershed size significantly impacts a river network’s pollution filtration ability, with larger watersheds showing a superlinear increase—an increase unproportionally greater compared to the increase in watershed size—in filtration efficiency.
- Managing land use and mitigating non-point source pollution in smaller watersheds are priorities for protecting estuaries and oceans.
- Results indicate that larger watersheds may release more carbon back to the atmosphere because of aquatic processes.
About the Co-author

Wilfred Wollheim, Professor of Natural Resources and the Environment
Contact information: Wilfred.Wollheim@UNH.edu, 603-862-5022, FindScholars profile

Fig. 1. Logarithmic plot showing cumulative CO2 emissions from aquatic metabolism against watershed size, highlighting the superlinear scaling of pollution filtration and carbon release with increasing watershed area. Data derived from nine model scenarios emphasizing trends in aquatic ecosystems. Note: Cumulative CO2 derived from aquatic metabolism (cumulative ecosystem respiration (ER) – cumulative gross primary production (GPP), normalized to watershed area). Model results incorporate trends in the local rate of GPP and ER with watershed area for a rectangular river network at mean annual flow (500 mm yr−1). Median (solid line), 25th percentile and 75th percentiles (dashed lines) are derived from 9 model scenarios that reflect potential variation in hydraulic dimensions.
Implications for the Future
This study is an important contribution to environmental policy and land-use management. With a new understanding of watershed dynamics, policymakers and public agencies can create more effective strategies to enhance water quality and address climate change impacts.
The insights call for a recalibration of conservation efforts, particularly in New England where both nutrient pollution and carbon sequestration are factors, to leverage the natural filtration capabilities of river networks and better quantify their carbon release potential.
This work was supported by the National Science Foundation. Partial support was provided by the New Hampshire Agricultural Experiment Station through USDA National Institute of Food and Agriculture Hatch Project 0225006.