The validation and assessment of surface radiation flux products derived from satellite observations is difficult, especially in the Arctic region, where there are fewer observing stations. This study presents daily and monthly surface radiative fluxes (SRF) data collected at the Thule High Arctic Atmospheric Observatory (THAAO, 76.5° N, 68.8° W), in North-Western Greenland. These data are essential for the validation of satellite data. In the current state of the manuscript, some doubts need to be addressed.

1. I think it will be of interest to some readers to know why the THAAO was established at 76.5°N, 68.8°W. The addition of a diagram of the geographic location of the THAAO would also be helpful for reading.
2. Figure 3 shows the variation of the ratio of DSI with solar zenith angle for the spring and summer of 2012 for both instruments. What is the reason for selecting the data for 2012, and what is the variation of the ratio of DLI, USI, and ULI with solar zenith angle? What are the reasons for the large fluctuations in the ratio of the two observations at solar azimuths of 70 to 80?
3. What is the typical polar environment referred to in line 334? Is the environment in the area where the observatory is located different from other regions of the Arctic? What are the reasons for the largest differences in DLI in this environment? Also, the difference in Figure 4(b), where day of year is 78, appears to be uncorrected.
4. Does the phrase "extremely rare limits" in line 358 include limits due to meteorological conditions?
5. What is the Pearson correlation coefficient between temperature and radiative flux in figure 5?
6. In line 388, what is the sample size below which the sample size is too small? Could the authors please provide information on the sample size of valid observations during the observation period?
7. Lines 405-420 refer to temperature anomalies, and it is suggested that the time-series variation of temperature anomalies be presented in a graph. Have the temperature observations been validated against NCEP reanalysis information?
8. Does the maximum and minimum of the observed data refer to the average result of the observed maximum and minimum?
9. As global warming occurs, will the area and duration of snow cover in the Arctic decrease? Does this have an effect on the observation of shortwave radiative fluxes?
10. Is there a relationship between the monthly average expression of long-wave radiative fluxes and cloudiness? If so, what is the correlation between them?
11. What is the trend of the longwave and shortwave radiative fluxes over the observed time period? Are the trends due to meteorological conditions or changes in the surface type?
12. The paper has relatively little comparison and discussion with other similar studies known in the field. Can you provide more details about how your measurements compare to existing similar studies, highlighting where this dataset fills a gap?
13. Could you provide a more detailed description of the procedures associated with the processing of the original data, e.g., procedures for data quality control, uncertainty estimation, etc., to improve the accuracy and reliability of the study.
14. In the introduction, the authors mention that ground-based measurements can improve the discrepancy of climate models in modeling Arctic warming. Have you considered comparing measurements with simulations from existing models? Provide an in-depth analysis of the importance of existing measurements for model simulation assessment.
15. The introduction section needs to be strengthened.

This manuscript describes short- and long-wave surface radiation data obtained at the Thule High Arctic Observatory in Northern Greenland over several years. The data are a useful contribution to surface energy budget studies in the Arctic, where according observations are scarce. The data will gain importance with the progression of the observation period.

Regarding the measurements, the authors apply very thorough corrections with regard to different sources of offsets (e.g. thermal offset due to the lack of shielding; cosine correction). The applied procedures are described in detail and easy to understand. Also, the comparison of the applied pyranometers to instruments that are calibrated at PMOD is explained, pointing to the relation of the instruments to corresponding reference devices. The manuscript is an important basis for understanding the data quality of the data set and moreover for the listing of metadata like e.g. instrument model and serial numbers. As the metadata are not included in the downloadable data set, I recommend to include the link to the manuscript on the data download page, so data users are aware of it.

In the manuscript, time series of daily and monthly irradiances are shown, and single case studies presented in relation to meteorological conditions.

Overall, the manuscript is comprehensive and well written, and I recommend publication in ESSD upon addressing the comments listed below.

Comments:

L26:  DSI is absent (solar zenith angle≥90°) from 29 October to 13 February. Yet, the data files partly contain data before 13 Feb, showing negative radiation values. I suggest to flag these or discard them.

L118:  in the 1990s

L 493: Could the the lower A values in 2021 be related to liquid precipitation with consequent solid ice formation on the ground, being darker than snow?

Figure 10:  the relation between T anomalies and albedo isn’t obvious: preferably, the upper panel should show the absolute temperature (with indication of 0°C).  The same applies to Line 663, where this event is discussed. “11 K higher than ususal” doesn’t mean a lot here if no baseline is provided.

Figure 14:  unnecessary figure, the basic information is included in Figure 16.