Unfortunately one of the agreed reviewers did not submit their review. In order to move the review process along, I decided in accordance with the journal's policies to provide my own review of the manuscript.

The manuscript presents a unique dataset of 1Hz solar irradiation data combined with additional context such as solar position, satelite cloud cover. The dateset may be useful to test forecast models related to solar energy production, where high frequency variation is important. 

Overall, the manuscript is well written and documents the dataset. I have a number of comments that should be included before final publication

1. True dataset resolution: While data is recorded at 1Hz, the sensor response time is lower. This is unfortunately a technical limitation and dataset users should be aware of this. I would recommend to better highlight the response times and to add response times to the abstract. The authors should also provide a clear definition of response time (i.e. first order response with time constant of ...).  Additionally, it would be good if the authors could expand somewhere in the manuscript on the the impacts of the response time on measurements and potential applications.

Expand description of data quality flagging. The data quality flagging is important and described very concisely. It might be good to exapand this a bit to make it easier to understand (or to add a flow chart).

Maintenance and sensor calibrations: L 115 mentions that sensor/site maintenance happens often. However are there any records on sensor calibrations, maintenance that could be added to the description?

L 112: Are these gap filled data marked in the quality flags?

L120: Specify all three components for clarity.

Figure 4 is very dense. Once could think about ways to make it less so, but it makse sense to overlay the different radiation components. The legend is confusion in te sence that it is not immediately clear to where the residuals map. Also the text on the bars is difficult to read due to white on light grey. Please revise the figures.

One naive question about the data in figure 4 is that is appears that cloud enhanced radiation conditions (SWTot>Clear Sky) dominate during cloud activity. If one were to integrat insolation over time, how would this look with respect to total irradiance vs clear sky irradiance

Figure 5: I am not sure, I understand this correctly, but according to the text shadow is defined as DNI < 120 W/m2. Around 8 UTC there is data classified as sunshine which does seem to fit the shadow definition from the description text. It would be good if the authors could explain this.