Bernard and co-authors’ manuscript is a concise, clear description of an updated version of the well-known GIEMS inundation product called GIEMS-MethaneCentric (GIEMS-MC). Modifications include separating the areas of open water using the Global Lakes and Wetlands Database version 2 (GLWDv2) and rice paddies from wetland areas, and using filters for coastal zones to avoid ocean artifacts and for regions with snow cover. GIEMS-MC spans the period 1992-2020 on a monthly timescale at 0.25° x 0.25° spatial resolution, and includes one time-series of flooded and saturated wetlands and one for wetlands and peatlands, plus static areas of lakes, rivers and reservoirs, seasonal rice paddies and dominant vegetation.  The updated product is compared with the Wetland Area and Dynamics for Methane Modeling (WAD2M) product globally.  Regional comparisons are also done for the Siberian Lowlands, the Sudd, the Amazon and South-East Asia.  GIEMS-MC is likely to be used by global and regional modeling efforts and should improve estimates of temporal variations in wetland emissions of methane.

Although the authors state that quantification of the uncertainties of the GIEMS-MC variables is beyond the scope of this study, several aspects of the approach and its validation are in need of further information. 

• Though the remote sensing basis and limitations of GIEMS and GIEMS-2 are described in a series of publications, a brief summary of these two items would benefit readers of this report.
• The comparisons of WAD2M and GIEMS-MC are primary illustrated in coarse global figures and in a table. Statistical analyses of global and regional differences between these products is needed.  The figures showing temporal variability also need quantitative analysis and comparison with other products, when possible.
• The use of GLWDv2 to separate lake, river and reservoir areas is reasonable for many regions, but not for the widespread and extensive floodplains of tropical and northern rivers because open water areas in these systems vary considerably on a seasonal timescale. A discussion of the consequences of this issue should be added.
• The regional comparison for the Amazon basin should mention the evaluation of several inundation products published in Fleischmann et al. (2022). How much inundation occurs in the Amazon River basin? Remote Sensing of Environment. 278, 113099. doi.org/10.1016/j.rse.2022.113099.

Given the limitations of the sensors used by GIEMS and the well-validated results from synthetic aperture radar in seasonally inundated forests, this publication provides valuable information about some of the uncertainties inherent in the GIEMS products.

• The Boreal–Arctic Wetland and Lake Dataset (BAWLD) would seem appropriate for comparison in northern regions.
• Given that peatlands add a large area to the GIEMS-MC results, an evaluation of the veracity of the peatland products is needed. For example, the approach used by Gumbricht et al. (2017) has serious problems, at least for the Amazon basin. In general, remote sensing of peatlands is difficult.
• Though the ERA product provides a global estimate of the occurrence of snowcover, a comparison of these estimates with the SNODAS products for parts of North America or with other snowcover products would be useful.
• The European Space Agency Land Cover dataset has limitations when applied to seasonally varying wetland vegetation, and these uncertainties need to be mentioned.

In this manuscript, the authors presented a new dataset of inundation dynamics, GIEMS-MethaneCentric. The authors improved the previous dataset by using updated input data and a revised systematic data production process. They compared the new data with other data (WAD2M based on SWAMPS) and obtained consistent results.

Major comments

Accurate inundation (methane-emitting aquatic surface) maps are undoubtedly important for the emission evaluation of methane, a potent greenhouse gas and short-lived climate forcer. However, uncertainties in the inundation dataset have been a serious problem in the global methane budget. The dataset presented in this study is remarkable, because it captured spatial heterogeneity by using updated satellite data and land surface maps. The GIEMS-2 data also covers a long period from 1992 to 2020, allowing us to assess interannual to decadal dynamics of inundation and resultant methane emissions. Moreover, in producing the dataset, the authors used new freshwater and paddy field data to avoid double counting, which is a serious problem in the global methane budget. The dataset is clearly useful for wetland and methane researchers; the spatial resolution (quarter degree) and time step (monthly) may be coarse for field studies but useful for regional to broader assessments.

              I have a minor concern about the quality of the dataset. Namely, the results (e.g., Figure 3) show that peatlands made a substantial contribution to the global extent especially in tropical and northern latitudes. Nevertheless, the peatland extent was derived from static maps like PeatMap, and therefore interannual variability in peatland inundation could be underrepresented in the dataset. For example, in Figure 8, the anomalies were not largely different between GIEMS-MCISW and GIESM-MCISW+P except for variability due to snow cover in northern areas (Ob). In section 5.2.3., the authors discussed a problem with peatland integration but focused on the separation of inundated and saturated areas. As discussed in section 5.2.2. about rice paddy fields, the authors should discuss the temporal variability of peatlands; this can be serious in Southeast Asia (see Figure 7), where peatlands are prevailing and meteorological variability like ENSO is influential.

              Overall, the manuscript is well prepared. The methodological description is adequate, and the dataset is presented nicely. I conclude that the manuscript is acceptable for publication after minor revision.

Minor comments

- I agree that the ERA5 is widely used meteorological dataset, but I am not sure the quality of snow density and depth in the dataset. Did you check it by comparing with observational data?

- Line 223: I agree to apply a clearing process to reduce the artifacts in coastal areas. However, I suspect that it resulted in the removal of riverine estuaries where are potentially important methane sources. Is my understanding correct?

- Figure 3: Please show the period for the data used in the figure.

- Line 484: Can you give a rough estimation of how much the new dataset improves the evaluation of global wetland methane emissions? For example, a +10% larger inundation area may lead to correspondingly larger emissions.

This study presents an updated global inundation area dataset spanning 1992 to 2020, providing monthly data at a spatial resolution of 0.25°×0.25°. The dataset differentiates coastal areas and rice paddies, thereby mitigating the overestimation of wetland areas and improving data accuracy. This product has significant value for quantifying methane emissions from global wetlands. Overall, the article is concise and well-structured, accordingly I have only several comments requiring further clarification:

Both peatland area and methane emission intensity are highly influenced by soil moisture saturation and water table. Thus, water table playing a crucial role in peatland area variation. How does the study address this issue?

As shown in Figure 3, peatlands contribute significantly to wetland areas in high-latitude and tropical regions. In high-latitude areas, snow cover changes can partially explain the interannual variability of peatlands, but this approach is clearly unsuitable for low-latitude regions. How can the variability in low-latitude peatlands be better understood? Could this issue be discussed further?

Would subtracting coastal area directly from wetland areas result in the underestimate of wetland area by excluding riverine estuaries?

Rainfed rice exhibits seasonal and interannual variations in their inundation status, though its contribution to total wetland area is relatively minor. Could this be discussed further?

Finally, could the article delve into the specific advantages of using an update dataset that classifies inundation areas into distinct types for accurately estimating global wetland area and its methane emissions?