Aho et al. presents a large dataset of gas exchange velocities (k600) and gas exchange rates (K600) along with hydraulic geometries from 22 stream and river sites across the United States. This dataset is based on rigorous field experiments and measurements and is the largest gas exchange dataset I have ever seen. The value of the dataset is to facilitate key research questions such as riverine greenhouse gas (GHG) emissions and river metabolism and help understand the river ecological processes and biogeochemical fluxes. The manuscript is well-organized and straightforward to follow. The field experiments, data collection, data processing, and data description are all described in great details, although more information is needed for some aspects. Generally, I endorse the publication of this work in ESSD. I have the following specific comments need to be addressed.

L64: The term “𝑘600” and “𝐾600” should be defined and/or explained as they appear first time in the main text. Those two parameters look very similar (differ in non-capital and capital K) and they may confuse the readers. An explanation here would be very helpful.

L76: Why use SF₆ as tracer gas? SF₆ is a very potent GHG that has 23500 times greater global warming potential than CO₂. Such large-scale experiments may cause environmental burdens.

L83: How was the tracer gas sample collected? Using headspace equilibrium method? More details are needed.

L87: How was the tracer SF₆ samples analyzed?

L90: Can you plot the discharge hydrograph and marked when the tracer experiments were conducted? Histograms of discharge with tracer experiment marks on them are also acceptable. These graphs, which can be put in the supplement, will clearly show the representativeness of the experiments with flow regimes.

L368: Scaling relationships between k600 and hydraulics parameters such as velocity and channel slope would also be very useful for future users to estimate k600. I suggest the authors provided the equations between k600 and velocity and channel slope if that’s feasible.

L418: Can you go a step forward and provide these predictive models of gas exchange in this paper? With such comprehensive dataset in-hand, those models should be easy to fit (see the comment above).

This dataset presents a large compilations of gas exchange related estimates based on gas loss experiments in streams across a range of discharges. It will be an important resource for the aquatic biogeochemistry community in general. Such compilation is necessary in order to better quantify the role and contribution of fluvial systems to greenhouse gas emissions, as well as to better identify current geographic gaps. The paper is structured in a good way that allows readers to follow easily. I have only few suggestions and minor comments in the hope that these can be helpful to the authors. Thanks to the authors for this important contribution to the field, and I’d like to see this work could be published in ESSD.

Line by Line Comments:

L31: I noticed the authors cited Rocher-Ros et al., 2023, so I wonder if the authors are able to provide some discussion about missing information of ebullitive k for CH₄.

L77: It seems like these experimented streams are in small size. This raises the question of whether the methodologies and conclusions drawn from this study can be reliably extrapolated to the context of big rivers.

Please indicate the dates or seasons of experiments somewhere in the methods, because these are key information regarding discharge, velocity, water depth, etc.