Upscaling a High-Resolution Reactive Transport Model for Predicting Inland Water CO2 Concentrations and Fluxes Across the Continental United States

Here, we implement a reach-scale reactive transport model across the continental US using a coupled river/lake/reservoir hydrography framework for millions of river/lake/reservoir reaches and compare it to statistical methods of predicting continental scale CO2 fluxes. Environmental fluxes of CO2 are an important part of the global carbon budget with inland waters playing an area-outsized role within the terrestrial carbon cycle. CO2 emissions from inland waters represent a significant portion of terrestrial fluxes and are of a similar magnitude to the net terrestrial CO2 sink; which is why many studies have sought to characterize the controls of CO2 fluxes in rivers and lakes. While the ability to predict CO2 concentrations and fluxes from inland waters has improved and can be modeled at small spatial scales, mechanistic controls have yet to be tested at regional or continental scales. This represents a critical next step in incorporating the fluxes from inland waters into global climate models. Our model is the first to incorporate hyporheic exchange, water column respiration, downstream advection, atmospheric exchange, and groundwater inputs of CO2 for both rivers and lakes in a coupled network. We use the model to evaluate regional patterns of CO2 including the proportion of fluxes from lakes and reservoirs compared to rivers and headwater streams. When compared to statistical methods, our model shows distinct differences in the regional distribution of fluxes with a strong east-west gradient. Overall, our results represent the first continental-scale mechanistic modeling of inland water CO2 fluxes and improve our understanding of the processes and regions that are most important.