This study utilizes ICESat-2 data from November 2018 to November 2019 to generate a new DEM over Greenland, and validates the newly generated DEM with IceBridge ATM data obtained in May 2019. This study also presents a comparison of different DEM products. I think that this paper is interesting and the topic is suitable to ESSD. However, several major comments should be seriously considered.

Major comments:

1. The present language quality is not good enough and needs to be improved throughout. For example, Page 1 Line 10-11, “but long temporal coverage introduced additional time uncertainty to scientific research”, what does time uncertainty mean? Line 11-12 “with a definite time”, what does it mean? Page 2 Line 49, “Hence”? Page 3 Line 97, “by different beams”? Page 9 Line 253, “Of these” should be “Among them”.

2. The resample resolution of IceBridge is about 25 m, how to validate the ICESat-2 DEM with different spatial resolution (namely, 500 m, 1 km, 2 km). Please clarify.

3. Please rewrite the Method part in Page 6, Line 164-173, which is different to understand. When the ICESat-2 data were gridded to fine resolution (i.e. 500 m), there would be many gaps. These gaps will be filled with values from coarse-grid data (i.e. 1 km or 2 km)? “We set the minimum number of grid points to 10 and the minimum timestamp to 2 months ...”? What does this sentence mean and how the values of 10 and 2 were determined? “In addition, we introduced thresholds to remove outliers, which are RMSE≥10 m, the uncertainty of elevation change ≥10 m ...” Please clarify how these thresholds were calculated?

4. Please rewrite the conclusions in Section 5.2 and Section 6. Due to time discrepancy between different DEM products, I don’t think it is possible to validate ICESat/GLAS DEM (2003~2005), ArcticDEM (2015~2018), TanDEM (2011~2014), CryoSat-2 DEM (2011~2014) using IceBridge data acquired in May 2019 and current results can support the conclusion that the ICESat-2 DEM showed significant improvements in accuracy compared with other altimeter-derived DEMs (in Page 1 Line 20~25). If possible, I suggest selecting areas with little elevation changes and doing the comparison. 

5. Since the ICESat-2 data is available from 2018 to present, two years’ DEM products could be generated and compared. Otherwise, the specific time should be added to the title, for example, “A new Greenland digital elevation model derived from ICESat-2 during 2018-2019”. 

General comments:

Page 2 Line 37: “The previously published Greenland DEM dates back to the 1980s ...”. “previously” would be “first”, since “previously published” would have several DEM products?

Page 2 Line 38-40: “However, the data acquisition was limited by the low-visibility contrast between snow and ice surfaces (Noh and Howat, 2015), which introduced large time uncertainty into the DEM.” How the low-visibility contrast introduced large time uncertainty. Please clarify.

Page 3 Line 78-79: “IceBridge data were used to evaluate the accuracy for all of Greenland and for different basins.”

According to Figure 1 in Page 16, the IceBridge data didn’t coverage all the Greenland, please rewrite this sentence.

Page 3 Line 91-92: “However, for strong and weak beams in the ATL06 product, both beams in one pair show similar performance, with a median difference of -0.08 cm and -0.13 cm for strong beam2 and weak beam1”

The statement is confusing. “a median difference”, compared with what? The results is for strong beam2 and weak beam1, what about the other beams? Please clarify.

Page 4 Line 107-112, what do α, β, αs,n, αw,e, β0 stand for? Please clarify.

Page 5 Line 145-150, what does h, a0 to a4 stand for? Please clarify how to get a0 to a4 and dh/dt.

DEMs are the basic datasets for digital hydrology, watershed analysis, quantification of remote sensing, glacier change, etc. Compared with the optical stereo images (photogrammetric method) and radar interferometry (microwave band), the DEM data obtained by laser altimetry satellite data has higher reliability, especially for the surface elevation of glaciers and snow cover. In this paper, a new Greenland DEM is derived from ICESAT-2 withe a definite time (13 months), which is very meaningful and practical. However, the paper still has many problems, and substantial revisions are required before the acceptance of this manuscript can be recommended. 



Specific comments:

1. There are some problems with cryosat-2 satellite orbit descriptionin the article. Some parameter values have inconsistencies or errors, please check carefully. For example, “ which is a great improvement over CryoSat-2's along-track distance of 1.5 km and cross-track distance of 3 km” .
2. Each parameter in the formula needs to be defined, but there are no explanations for many parameters in the paper, such as formula (1)-(4).In formula (5), what does “h” mean and how to input its parameter values?
3. There is no need to write specific formulas for commonly used parameters in the paper, such as slope and aspect.
4. How to use ICESAT-2 laser point cloud data to simulate DEM data at 500m grid scale is the focus of this paper, but the paper does not describe it clearly. How to achieve “To improve ICESat-2 data utilization, DEMs with 1 km and 2 km resolution across all of Greenland and an additional 5 km resolution in southernmost Greenland were used to fill the DEM gaps. Kriging interpolation was used to 15 fill the remaining 2% of void grids that were insufficiently observed by ICESat-2 measurements.” And how to get calculated grids and interpolated grids in Figure 6. Therefore, it is recommended to expand the content of section 3.1.
5. “We set the minimum number of grid points to 10 and the minimum timestamp to 2 months”, are you sure it is 2 months timestamp here? Generally speaking, the revisit period of icesat-2 is 91 days.
6. Some descriptions in the paper are not clear. For example, in Figure 4(b) and (d), how is the elevation uncertainty calculated and which index is used?
7. In general, the labels of the figures are arranged in the paper from small to large, please check the order of Figure 4 and Figure 5.
8. The structure of the conclusion and discussion section is a bit confusing, please reorganize it. Generally speaking, accuracy verification is part of the results.