Comments on the manuscript:

Interesting study, focusing on the emission quantification of LNG facilities. Especially the use of the already emitted CO2 as a tracer is valuable. This study is therefore a good addition within this field of research. It also gives a good overview of the advantages and disadvantages of the mass balance and tracer approach. Furthermore, the statistical substantiation is well provided. The paper is written clear and a relevant topic, that contributes to a better understanding of methane emissions in the environment.  It is also helpful to the OGMP 2.0 implementation as a potential method.

Specific comments:

In the abstract, the authors present 2 approaches, downwind mass balance method (1) and the tracer correlation approach (2). In line 38 the authors mention the tracer correlation approach can be used, but here the mass balance approach is not mentioned. Maybe good to also mention the results of this approach in comparison to approach 2.

I miss information on the site description and size of the LNG facilities. I understand that they are anonymised, but later on some measurements are above water instead of land. This can have major influences on the dispersion and mixing of the plumes.

Line 488: For completeness, maybe add that also because of closer distances, plumes from differenced sources of the facilities will not be well mixed already. At closer distances, the individual sources can be detected, which may result in difference CH4:CO2 ratios. This is also shown in Figure 6, with the different sources having different CH4:CO2 ratios at closer distances (Figure 6 a, b and c) and are well mixed at further distances (8 km; Figure 6 d).

Line 538: Here the authors mentioned that measurements were often taken over water. This can also explain part of the poor mixing below 250 m (line 536). The mixing/dispersion of plumes behave different above water compared to above land (this is for example included in the EPA OCD (Offshore and Coastal Dispersion) model.