Very good product with easy-to-use data. Potentally of wide interest and utility but needs small changes before publication in ESSD.

The dataset and associated documentation generate a value added. However, some of the assumptions made are not very plausible and lack motivation. In addition, some methods used are non-standard and it is not clear why standard state-of-the-art techniques are not used. In combination, these issues make it difficult for me to endorse publication of the current manuscript and data. Below I describe the concerns in more detail, followed by some minor comments, mostly of editorial nature.

The analysis makes a very strong assumption that in the industry sector, gas is substituted with electrically powered heat pumps (p. 3 l. 75) which seems not to be grounded in actual data. In reality, many companies, in particular in the first year after the Russian invasion into Ukraine, switched to other fossil fuels such as oil and coal to avoid disruptions of supply and suffering from high cost of gas. Also, a lot of applications that natural gas is used for in industry require high temperature levels that cannot be provided by heat pumps.

The ”empirical temperature-gas-consumption (TGC) curves” introduced in the Methods section (p. 3 l. 99) appears to miss the well-established concept of heating degree days (HDDs) that are typically used to correct heating energy consumption for climatic variations across different years (or other time periods). HDDs in addition to temperature variations also make use of a reference building, taking into account at which temperatures heating is actually needed.

I don’t understand why results of the gas demand reduction (p.7, l. 206) are presented as an average across both winters, not separately per winter as the situation clearly changed. Gas prices were clearly much higher in the 2022/23 winter than in the 2023/24 winter and also adaptation measures such as increasing LNG import capacity had been scaled up to significantly higher levels by the 2^nd winter. More generally, prices as an explanatory variable seem to be not adequately considered in the analysis

In the power sector analysis (p. 5 l. 169), increased electricity imports don’t seem to be included as a substitute for domestic gas-powered electricity generation. Moreover, policy interventions across different EU27 countries and the UK were quite different, leading to different incentives for power producers to substitute gas with other fuels. For example, in Spain gas-fired power generation was taken out of the general merit order approach and producers were compensated via different instruments to reduce electricity price spikes.

Minor comments:

• The precision with which some of the results are reported appear to exceed reasonable expectations regarding their uncertainty (e.g., 87.8% or 976.8 TWh per winter – p. 1 l. 22).
• While not wrong, some of the units used are not those commonly used in energy research (e.g., 10^9 m3 instead of bcm = billion cubic meters – p.2 l. 38).
• The GHG balance of pipeline and LNG transportation of natural gas (p. 10 l. 344) depends on many factors, but importantly on the transport distance. Leakage aside, on shorter distances pipeline transport tends to come with lower emissions but with increasing transport distance the energy use of compressor stations increases proportionally to the distance while the biggest chunk of emissions by LNG tanker originates from the liquefaction process.
• The writing is somewhat repetitive and could benefit from being thoroughly edited.

Zhou et al updates and develops datasets characterizing EU sectoral and country-based gas supply, storage, transmission, and consumption statistics to assess the dynamics in gas supply and demand following the energy crisis triggered by the 2022 Russian invasion of Ukraine. The paper is well written and provides interesting insights into the ramifications of the dramatically reduced Russian gas imports following the invasion in Ukraine. I have a few comments worth addressing before the paper can be published.

Line 22 – Since these are based on estimates, 3 significant figures seem too precise given the modeling uncertainties. Suggest reporting all metrics to no more than 2 significant figures here and throughout the paper. In addition, any result based on models or estimates need to have uncertainties reported (e.g., 1 standard deviation or 95% confidence intervals).

Line 23 – regarding LNG imports, would be helpful to name the countries dominating the increase in LNG exports to the EU.

In line 295, when reporting the p-value, is significance established at p < 0.05? If so, this should be stated in the methods. Note that significance can also be established at p < 0.01.

There are several places where the authors make statements without providing supporting statistics in the main text, which are needed for clarity. For example, in section 4.3.1, what is the mean winter temperature (line 265) and what is meant by “warmer temperature anomalies” (line 267-268). In addition, the study states that there has been an increase in the adoption of heat pumps in the EU (line 374), without providing supporting statistics on the magnitude of such an increase. Similarly, in lines 376-377 – by how much did electricity prices increase? What is meant by “lower start heating temperatures” in line 378?

There are several typos in the manuscript that should be corrected (e.g., line 38, 318)

Figure 5 (a). Font sizes in the figure legend and too small.