the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 17 Jan 2024
Combining a ground-based UV network with satellite maps: A case study for Germany
Abstract. A study for the comprehensive information of current UV exposure for the area of Germany, based on the method for near real time calculation of UV Index maps used in the framework of the Austrian UV Monitoring Network, is presented. For the area of Germany about 22.000 surface UV Index maps were calculated for the year 2022 via the radiative transfer model libRadtran by incorporating daily forecast data for ozone, albedo and aerosols from the Copernicus Atmospheric Monitoring Service and taking into account cloud information gathered from SEVIRI data of Meteosat Second Generation in the form of a cloud modification factor. Ground-based measurements of 17 stations of the German Solar Monitoring Network were then compared to the modelled maps. For most stations the correlation coefficient between measured and modelled UV Index (UVI) is above 0.92 and the mean difference of modelled UVI to measured UVI is smaller than 0.5 UVI. The modelling of the UVI at the high mountain station Schneefernerhaus is associated with higher uncertainties (correlation coefficient 0.85 and mean UVI difference 0.6 UVI) due to the small-scale topography with spatially highly variable albedo and clouds. Moreover, case studies for specific locations with respect to cloud conditions and topography are discussed, as well as a case study of the combination of ground-based measurements and modelled UVI maps in form of spatial correction factors.
This preprint has been withdrawn.
-
Withdrawal notice
This preprint has been withdrawn.
-
Preprint
(2916 KB)
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-3130', Anonymous Referee #1, 12 Feb 2024
A) General comments
The paper discusses modelled UV Index maps based on satellite imagery vs. ground-based UV measurements. This is a big topic in the community and several groups are working on those maps for near realtime forecasting or to provide high density spatial UV maps. The study is restricted to the area in Germany which provide a relatively homogeneous topography. The authors show that the derived modelled maps work well for most of the area but significant problems can be observed in mountainous regions. My first questions after studying this work is therefore why the map does include only one station in the Bavarian Alps and does not include the area of Austria with their high density of ground stations supervised by the authors themselves.
Another week point of the work is that it is based on a paper by Schenzinger et al. (2023) which has been rejected for publication. The presented work unfortunately did not give further light to the open questions raised by the previous work.
In general, the paper is well written, and the figures are of good quality. However following points must be considered to enhance the quality of the manuscript.
B) Specific scientific comments
- The paper discusses in various places problems occurring from the inaccuracy of the could mask derived from the satellite map. This seems to be one week point of the whole analysis of the derived UV-Index maps. The following additionally point should be discussed to enhance the quality of the manuscript.
- Line 109: Cloud filtering based on CLM (satellite): Comparison of the cloud map derived by the authors to different methods like standard deviation of the ground-based measurement (is low for clear sky and high for cloudy conditions) or auxiliary ground-based measurements (sky camera, direct solar irradiance measurements, etc.).
- Line 191: This point should be extended to an analysis of solar zenith angle on chosen pixel of cloud mask of the ground station.
- The cloud type is neglected at all in the analysis. Thin cirrus clouds reduce the UVI much less than cumulus clouds. A discussion about this effect is needed.
- Another point the authors study in the presented work (Section 3.3., line 199ff) is the problem of remote sensing in mountainous regions. As a minimum a figure showing the comparison of the cloud mask vs. the calculated albedo map (or satellite image) would help to illustrate the situation at ZS. A discussion about proposed solutions for the problem is missing. Second, as mentioned above data for the whole alpine region of Austria is available to the authors and at least a reference to publications addressed to this important topic must be added.
- 18 stations have been selected for the calculation of the UV Index maps. An analysis of the uncertainty caused by the number of stations is missing. Especially in section 3.4. the quality of the spatial interpolation is based on both, the number of data points supplied by each station and the density of stations.
- Line 83ff: Cutoff sza<70 - does affect the number of datapoint from North (higher latitude) to south. Add a sensitivity discussion.
- Line 89: Ground-based measurements uncertainties: Add “Sensitivity changes between calibrations”.
- In the paper a mix of the differences and ratios (modelled to measured data) is used. This must be explained why the one is favored for the other in the different cases.
- Figure 2: Shift towards negative is observed for clear sky for some stations. Therefore, the modelled data is underestimating the UVI in general. A discussion is needed.
C.) Presentation
The manuscript is clearly structured. I found only the following two points the be considered to improve the quality of the paper:
- Figure 7, Caption, right cloud mask: Gray is "clear sky" or (white areas=clouds) only mentioned in the text.
- Figure 8: Define grey/black points in the caption (ratio of measurement to model)
Citation: https://doi.org/10.5194/egusphere-2023-3130-RC1 -
AC2: 'Reply on RC1', Barbara Klotz, 21 Mar 2024
We thank all the reviewers for the comments that reveal the issues of the manuscript. In the light of the required extensive "major" revisions to address all suggestions and also to deal with the rejection of our referenced paper (Schenzinger et al.) we believe that a re-submission without associated time constraints is the best way forward.
Citation: https://doi.org/10.5194/egusphere-2023-3130-AC2
- The paper discusses in various places problems occurring from the inaccuracy of the could mask derived from the satellite map. This seems to be one week point of the whole analysis of the derived UV-Index maps. The following additionally point should be discussed to enhance the quality of the manuscript.
-
RC2: 'Comment on egusphere-2023-3130', Anonymous Referee #2, 15 Feb 2024
The paper describes a study wherein ground-based measurements of the UV Index (UVI) are compared with about 22k calculated UVI maps based on a radiative transfer model with input data (like ozone) from CAMS, and SEVERI data is used to account for the presence of clouds.
The paper is well written and covers an interesting subject, but several clarifications and improvements are needed before publication.
General comments
While the authors describe their approach clearly, they do not discuss other sources of readily available satellite-based UVI data that might be used for comparison and why their approach using modelled UVI data is favourable. For example, the TEMIS UV data service ( https://www.temis.nl/uvradiation/ ) provides daily UVI maps worldwide derived from satellite-based ozone measurements.
Sect. 2 mentions how the cloud modification factor (CMF) is calculated, but only very briefly, and there is no mention of how accurate the CMF is. Looking at the results in Table 1, the mean difference in the UVI for clear sky is the result of uncertainties in model and measurement. For the cloudy cases the uncertainties in the CMF come on top of that. How much of the increased difference is caused by the CMF being uncertain? Seeing that the difference for cloudy cases is more negative than for clear, one could argue that the CMF is not accurate enough.
How about the other way around? What if the clear-sky average difference is systematic and expected to also be reached for cloudy cases, what is the CMF needed to achieve that and how does this related to the CMF used? Are there any seasonal / solar zenith angle (SZA) effects in imperfections in the CMF calculation?
Sect. 3.2 and Fig. 7 discuss the limitation to comparison due to the presence of clouds near the single pixel over the measurement used for the comparison. Does this not show that using cloud information for a single SEVIRI pixel is too limiting? After all, isn't it likely that instruments measure the total light coming from more than just the surface of the SEVERI pixel? How about always calculating the CMF from a set of 3x3, 4x4 or 5x5 SEVIRI pixels around the measurement combined? Would that not lead to a more useful CMF calculation and better agreement?
When the authors talk of a "pixel" this refers to a single pixel in the SEVIRI data (i.e. just the central dot in the right panel of Fig. 7), right? In Sect. 2.2, line 104, it is said that comparisons are done only for groundbased measurements that lie within 3 minutes of a SEVIRI measurement, so one can assume knowledge of the cloudiness to be more or less correct for the moment of measurement. Why then does the beginning of Sect. 3.2 say that the "relatively coarse temporal (15 minutes)" resolution is a source of uncertainty in the CMF?
Sect. 2.1 mentions that data from the year 2022 is used, but this not further specified. Is the data spread over the different months / seasons evenly? Are data throughout the day used, i.e. are all SZA, evenly covered? The seasonal variation, the variation with SZA, could be discussed using the clear-sky data. The effects in cloudy data would be obsured by imperfections in the CMF.
Differences UVI in e.g. Table 1 are given in absolute numbers as an average over all data of the year and multiple points during the day, i.e. an average over all SZA. Are the differences listed systematic for all SZA? After all: a, say, 0.5 UVI difference has relatively speaking a larger impact at low SZA than at high SZA, in winter than in summer.
Specific comments
Fig. 6: The horizontal axis is time of day (UTC), that should be mentioned in the caption.
Fig. 7: Same as Fig. 6 on time axis. Caption should also mention that the shaded (white) areas in the 9 subpanels on the right mark cloud-free (cloudy) pixels.
Citation: https://doi.org/10.5194/egusphere-2023-3130-RC2 -
AC1: 'Reply on RC2', Barbara Klotz, 21 Mar 2024
We thank all the reviewers for the comments that reveal the issues of the manuscript. In the light of the required extensive "major" revisions to address all suggestions and also to deal with the rejection of our referenced paper (Schenzinger et al.) we believe that a re-submission without associated time constraints is the best way forward.
Citation: https://doi.org/10.5194/egusphere-2023-3130-AC1
-
AC1: 'Reply on RC2', Barbara Klotz, 21 Mar 2024
-
RC3: 'Comment on egusphere-2023-3130', Anonymous Referee #3, 20 Feb 2024
GENERAL COMMENTS
The manuscript presents a method for retrieving UV radiation at ground level based on satellite maps and other modeled data over Germany. The first part of the manuscript compares the simulations with independent surface measurements from the German UV network, while the second part discusses the combination of measurements and models. The topic covered is important for the community, given its potential applications in public dissemination and biological studies.
This study essentially upscales a method previously applied in Austria, as described in Schenzinger et al., 2023 (not accepted). Compared to this latter, the manuscript lacks novel concepts, ideas, or tools. The authors should clarify the added value of this study compared to the Austrian case and how it contributes to improved knowledge. In addition, several shortcomings preventing the publication of the previous study persist in the current manuscript.
In my opinion, major revisions are necessary for acceptance.
SPECIFIC COMMENTS
- Structure: The manuscript structure is very weak. A separate 'Results' section is missing. Section 2.2 should be split into two parts: one describing methods (l. 87-114) and another presenting results (l. 115-135). The aim of Sect. 3.4 is unclear, with a mixture of discussion (l. 214-230) and method description (l. 231-247); the remaining lines describe the limitations of the interpolation method and raise 'the question of how and at what frequency corrections should be applied' (l. 265), without actually answering. Cloudy cases present even more issues (l. 268-274);
- Validation: The presence of diurnal and annual cycles can lead to high correlations between simulations and measurements. This issue is well known in the UV community and is acknowledge by the authors themselves (l. 121-122). Hence, while it is interesting to determine whether the difference between measurements and satellite-derived maps falls below a predefined threshold, there is a risk of misleading results wherein the comparison appears to yield better results than it actually does. I would suggest including a comparison of the retrieved and measured cloud modification factors (CMF) to better illustrate the algorithm performance. Another approach could involve demonstrating improvements compared to simpler methods, such as clear-sky simulations, climatology, or persistence;
- Algorithm: Consideration should be given to whether satellite radiance at a single wavelength (600 nm) provides sufficient contrast between clouds and snow. Why aren't MSG channels at other wavelengths considered? It is also important to discuss the reliability of the satellite product cloud mask (CLM) or provide references, and explain how the CAMS albedo product was validated over the area of interest. The relationship between the CAMS albedo product and the effective albedo required for radiative transfer calculations needs clarification;
- Parallax errors typical of satellites such as MSG are not mentioned. Failure to correct these errors can lead to a shift in both clouds and snow cover towards the north.
TECHNICAL CORRECTIONS
I have compiled a list of technical corrections that are required. Any other minor language-related observations should be addressed during the typesetting phase.
- l. 1: Should the authors use the term 'irradiance' instead of 'exposure' here? Additionally, consider replacing 'the method' with 'a method';
- l. 4: I would recommend always using the expression 'effective surface albedo';
- l. 11-12: the last sentence is not clear;
- l. 14: the effects of solar radiation on human health depend on the 'level of exposure', but this might not be the only important factor. I think that the initial words 'Depending on the level of exposure' can be omitted or the sentence can be rephrased;
- l. 22: add altitude as a variable;
- l. 23: it is not clear what the authors mean by 'continuous' here;
- l. 25: replace 'is depending' with 'depends';
- l. 26: 'but there are methods of...' --> 'but only indirectly by...';
- l. 28-30: the utility of the UV Index as a valuable variable for public communication could be briefly introduced. Additionally, it can be anticipated that the method will be described in Sect. 2;
- l. 34-38: this paragraph seems more appropriate in the conclusions or in the abstract rather than in the introduction. Please, specify the altitude of the Schneefernerhaus station;
- l. 47: 'the Angstrom coefficient \beta';
- l. 52: the SEVIRI instrument should perhaps be introduced (e.g., native spatial resolution, etc.);
- l. 58: reference to the 'mix-max scaling' is not useful if the technique is not explained in detail;
- l. 73: the criteria for the selection should be anticipated at l. 69 where it is stated that only a part of the stations were chosen;
- l. 74: 'site acronym with 4 letters in Fig. 1'. However, the figure and the table should be splitted. A column describing the type of instrument could be added to improve readability;
- l. 76: specify the coverage factor (k) and include bibliographic references about the traceability/calibration of the ground-based network. Have the measurement devices ever been compared with each other?
- l. 93: 'is introducing' --> 'introduces';
- l. 134-135: a comparison between the two methods would be indeed instructive, at least to provide an rough estimate of the performances. Differing cloudiness in the respective years can be mentioned as a limiting factor;
- Table 2 (caption): define U0.5 and U1.0 here as well;
- l. 144-145: it is unclear why the deviations should be inequivocally attributed to ground measurements;
- l. 156: correct 'SZ' to 'ZS';
- l. 160: is cloud cover taken from the satellite cloud mask?
- l. 174: 'partly' --> 'partial'
- l. 216: what does 'continuous' mean here? Aren't geostationary satellites able to provide 'continuous' estimates in time?
- l. 233: is cloud cover taken from the satellite cloud mask?
- l. 237: what are the most important 'spatially coarse input parameters' causing outliers?
- l. 242: include bibliographic references for the interpolation method;
- l. 246: 'the fundamental issue does not lie...' unclear;
- l. 271: 'is relying' --> 'relies';
- l. 299: 'and vice-versa', the intended meaning of the authors here is unclear;
- l. 308-309: does this mean that operating at a national scale is insufficient and that the algorithm should be applied at a European level?
Citation: https://doi.org/10.5194/egusphere-2023-3130-RC3 -
AC3: 'Reply on RC3', Barbara Klotz, 21 Mar 2024
We thank all the reviewers for the comments that reveal the issues of the manuscript. In the light of the required extensive "major" revisions to address all suggestions and also to deal with the rejection of our referenced paper (Schenzinger et al.) we believe that a re-submission without associated time constraints is the best way forward.
Citation: https://doi.org/10.5194/egusphere-2023-3130-AC3
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-3130', Anonymous Referee #1, 12 Feb 2024
A) General comments
The paper discusses modelled UV Index maps based on satellite imagery vs. ground-based UV measurements. This is a big topic in the community and several groups are working on those maps for near realtime forecasting or to provide high density spatial UV maps. The study is restricted to the area in Germany which provide a relatively homogeneous topography. The authors show that the derived modelled maps work well for most of the area but significant problems can be observed in mountainous regions. My first questions after studying this work is therefore why the map does include only one station in the Bavarian Alps and does not include the area of Austria with their high density of ground stations supervised by the authors themselves.
Another week point of the work is that it is based on a paper by Schenzinger et al. (2023) which has been rejected for publication. The presented work unfortunately did not give further light to the open questions raised by the previous work.
In general, the paper is well written, and the figures are of good quality. However following points must be considered to enhance the quality of the manuscript.
B) Specific scientific comments
- The paper discusses in various places problems occurring from the inaccuracy of the could mask derived from the satellite map. This seems to be one week point of the whole analysis of the derived UV-Index maps. The following additionally point should be discussed to enhance the quality of the manuscript.
- Line 109: Cloud filtering based on CLM (satellite): Comparison of the cloud map derived by the authors to different methods like standard deviation of the ground-based measurement (is low for clear sky and high for cloudy conditions) or auxiliary ground-based measurements (sky camera, direct solar irradiance measurements, etc.).
- Line 191: This point should be extended to an analysis of solar zenith angle on chosen pixel of cloud mask of the ground station.
- The cloud type is neglected at all in the analysis. Thin cirrus clouds reduce the UVI much less than cumulus clouds. A discussion about this effect is needed.
- Another point the authors study in the presented work (Section 3.3., line 199ff) is the problem of remote sensing in mountainous regions. As a minimum a figure showing the comparison of the cloud mask vs. the calculated albedo map (or satellite image) would help to illustrate the situation at ZS. A discussion about proposed solutions for the problem is missing. Second, as mentioned above data for the whole alpine region of Austria is available to the authors and at least a reference to publications addressed to this important topic must be added.
- 18 stations have been selected for the calculation of the UV Index maps. An analysis of the uncertainty caused by the number of stations is missing. Especially in section 3.4. the quality of the spatial interpolation is based on both, the number of data points supplied by each station and the density of stations.
- Line 83ff: Cutoff sza<70 - does affect the number of datapoint from North (higher latitude) to south. Add a sensitivity discussion.
- Line 89: Ground-based measurements uncertainties: Add “Sensitivity changes between calibrations”.
- In the paper a mix of the differences and ratios (modelled to measured data) is used. This must be explained why the one is favored for the other in the different cases.
- Figure 2: Shift towards negative is observed for clear sky for some stations. Therefore, the modelled data is underestimating the UVI in general. A discussion is needed.
C.) Presentation
The manuscript is clearly structured. I found only the following two points the be considered to improve the quality of the paper:
- Figure 7, Caption, right cloud mask: Gray is "clear sky" or (white areas=clouds) only mentioned in the text.
- Figure 8: Define grey/black points in the caption (ratio of measurement to model)
Citation: https://doi.org/10.5194/egusphere-2023-3130-RC1 -
AC2: 'Reply on RC1', Barbara Klotz, 21 Mar 2024
We thank all the reviewers for the comments that reveal the issues of the manuscript. In the light of the required extensive "major" revisions to address all suggestions and also to deal with the rejection of our referenced paper (Schenzinger et al.) we believe that a re-submission without associated time constraints is the best way forward.
Citation: https://doi.org/10.5194/egusphere-2023-3130-AC2
- The paper discusses in various places problems occurring from the inaccuracy of the could mask derived from the satellite map. This seems to be one week point of the whole analysis of the derived UV-Index maps. The following additionally point should be discussed to enhance the quality of the manuscript.
-
RC2: 'Comment on egusphere-2023-3130', Anonymous Referee #2, 15 Feb 2024
The paper describes a study wherein ground-based measurements of the UV Index (UVI) are compared with about 22k calculated UVI maps based on a radiative transfer model with input data (like ozone) from CAMS, and SEVERI data is used to account for the presence of clouds.
The paper is well written and covers an interesting subject, but several clarifications and improvements are needed before publication.
General comments
While the authors describe their approach clearly, they do not discuss other sources of readily available satellite-based UVI data that might be used for comparison and why their approach using modelled UVI data is favourable. For example, the TEMIS UV data service ( https://www.temis.nl/uvradiation/ ) provides daily UVI maps worldwide derived from satellite-based ozone measurements.
Sect. 2 mentions how the cloud modification factor (CMF) is calculated, but only very briefly, and there is no mention of how accurate the CMF is. Looking at the results in Table 1, the mean difference in the UVI for clear sky is the result of uncertainties in model and measurement. For the cloudy cases the uncertainties in the CMF come on top of that. How much of the increased difference is caused by the CMF being uncertain? Seeing that the difference for cloudy cases is more negative than for clear, one could argue that the CMF is not accurate enough.
How about the other way around? What if the clear-sky average difference is systematic and expected to also be reached for cloudy cases, what is the CMF needed to achieve that and how does this related to the CMF used? Are there any seasonal / solar zenith angle (SZA) effects in imperfections in the CMF calculation?
Sect. 3.2 and Fig. 7 discuss the limitation to comparison due to the presence of clouds near the single pixel over the measurement used for the comparison. Does this not show that using cloud information for a single SEVIRI pixel is too limiting? After all, isn't it likely that instruments measure the total light coming from more than just the surface of the SEVERI pixel? How about always calculating the CMF from a set of 3x3, 4x4 or 5x5 SEVIRI pixels around the measurement combined? Would that not lead to a more useful CMF calculation and better agreement?
When the authors talk of a "pixel" this refers to a single pixel in the SEVIRI data (i.e. just the central dot in the right panel of Fig. 7), right? In Sect. 2.2, line 104, it is said that comparisons are done only for groundbased measurements that lie within 3 minutes of a SEVIRI measurement, so one can assume knowledge of the cloudiness to be more or less correct for the moment of measurement. Why then does the beginning of Sect. 3.2 say that the "relatively coarse temporal (15 minutes)" resolution is a source of uncertainty in the CMF?
Sect. 2.1 mentions that data from the year 2022 is used, but this not further specified. Is the data spread over the different months / seasons evenly? Are data throughout the day used, i.e. are all SZA, evenly covered? The seasonal variation, the variation with SZA, could be discussed using the clear-sky data. The effects in cloudy data would be obsured by imperfections in the CMF.
Differences UVI in e.g. Table 1 are given in absolute numbers as an average over all data of the year and multiple points during the day, i.e. an average over all SZA. Are the differences listed systematic for all SZA? After all: a, say, 0.5 UVI difference has relatively speaking a larger impact at low SZA than at high SZA, in winter than in summer.
Specific comments
Fig. 6: The horizontal axis is time of day (UTC), that should be mentioned in the caption.
Fig. 7: Same as Fig. 6 on time axis. Caption should also mention that the shaded (white) areas in the 9 subpanels on the right mark cloud-free (cloudy) pixels.
Citation: https://doi.org/10.5194/egusphere-2023-3130-RC2 -
AC1: 'Reply on RC2', Barbara Klotz, 21 Mar 2024
We thank all the reviewers for the comments that reveal the issues of the manuscript. In the light of the required extensive "major" revisions to address all suggestions and also to deal with the rejection of our referenced paper (Schenzinger et al.) we believe that a re-submission without associated time constraints is the best way forward.
Citation: https://doi.org/10.5194/egusphere-2023-3130-AC1
-
AC1: 'Reply on RC2', Barbara Klotz, 21 Mar 2024
-
RC3: 'Comment on egusphere-2023-3130', Anonymous Referee #3, 20 Feb 2024
GENERAL COMMENTS
The manuscript presents a method for retrieving UV radiation at ground level based on satellite maps and other modeled data over Germany. The first part of the manuscript compares the simulations with independent surface measurements from the German UV network, while the second part discusses the combination of measurements and models. The topic covered is important for the community, given its potential applications in public dissemination and biological studies.
This study essentially upscales a method previously applied in Austria, as described in Schenzinger et al., 2023 (not accepted). Compared to this latter, the manuscript lacks novel concepts, ideas, or tools. The authors should clarify the added value of this study compared to the Austrian case and how it contributes to improved knowledge. In addition, several shortcomings preventing the publication of the previous study persist in the current manuscript.
In my opinion, major revisions are necessary for acceptance.
SPECIFIC COMMENTS
- Structure: The manuscript structure is very weak. A separate 'Results' section is missing. Section 2.2 should be split into two parts: one describing methods (l. 87-114) and another presenting results (l. 115-135). The aim of Sect. 3.4 is unclear, with a mixture of discussion (l. 214-230) and method description (l. 231-247); the remaining lines describe the limitations of the interpolation method and raise 'the question of how and at what frequency corrections should be applied' (l. 265), without actually answering. Cloudy cases present even more issues (l. 268-274);
- Validation: The presence of diurnal and annual cycles can lead to high correlations between simulations and measurements. This issue is well known in the UV community and is acknowledge by the authors themselves (l. 121-122). Hence, while it is interesting to determine whether the difference between measurements and satellite-derived maps falls below a predefined threshold, there is a risk of misleading results wherein the comparison appears to yield better results than it actually does. I would suggest including a comparison of the retrieved and measured cloud modification factors (CMF) to better illustrate the algorithm performance. Another approach could involve demonstrating improvements compared to simpler methods, such as clear-sky simulations, climatology, or persistence;
- Algorithm: Consideration should be given to whether satellite radiance at a single wavelength (600 nm) provides sufficient contrast between clouds and snow. Why aren't MSG channels at other wavelengths considered? It is also important to discuss the reliability of the satellite product cloud mask (CLM) or provide references, and explain how the CAMS albedo product was validated over the area of interest. The relationship between the CAMS albedo product and the effective albedo required for radiative transfer calculations needs clarification;
- Parallax errors typical of satellites such as MSG are not mentioned. Failure to correct these errors can lead to a shift in both clouds and snow cover towards the north.
TECHNICAL CORRECTIONS
I have compiled a list of technical corrections that are required. Any other minor language-related observations should be addressed during the typesetting phase.
- l. 1: Should the authors use the term 'irradiance' instead of 'exposure' here? Additionally, consider replacing 'the method' with 'a method';
- l. 4: I would recommend always using the expression 'effective surface albedo';
- l. 11-12: the last sentence is not clear;
- l. 14: the effects of solar radiation on human health depend on the 'level of exposure', but this might not be the only important factor. I think that the initial words 'Depending on the level of exposure' can be omitted or the sentence can be rephrased;
- l. 22: add altitude as a variable;
- l. 23: it is not clear what the authors mean by 'continuous' here;
- l. 25: replace 'is depending' with 'depends';
- l. 26: 'but there are methods of...' --> 'but only indirectly by...';
- l. 28-30: the utility of the UV Index as a valuable variable for public communication could be briefly introduced. Additionally, it can be anticipated that the method will be described in Sect. 2;
- l. 34-38: this paragraph seems more appropriate in the conclusions or in the abstract rather than in the introduction. Please, specify the altitude of the Schneefernerhaus station;
- l. 47: 'the Angstrom coefficient \beta';
- l. 52: the SEVIRI instrument should perhaps be introduced (e.g., native spatial resolution, etc.);
- l. 58: reference to the 'mix-max scaling' is not useful if the technique is not explained in detail;
- l. 73: the criteria for the selection should be anticipated at l. 69 where it is stated that only a part of the stations were chosen;
- l. 74: 'site acronym with 4 letters in Fig. 1'. However, the figure and the table should be splitted. A column describing the type of instrument could be added to improve readability;
- l. 76: specify the coverage factor (k) and include bibliographic references about the traceability/calibration of the ground-based network. Have the measurement devices ever been compared with each other?
- l. 93: 'is introducing' --> 'introduces';
- l. 134-135: a comparison between the two methods would be indeed instructive, at least to provide an rough estimate of the performances. Differing cloudiness in the respective years can be mentioned as a limiting factor;
- Table 2 (caption): define U0.5 and U1.0 here as well;
- l. 144-145: it is unclear why the deviations should be inequivocally attributed to ground measurements;
- l. 156: correct 'SZ' to 'ZS';
- l. 160: is cloud cover taken from the satellite cloud mask?
- l. 174: 'partly' --> 'partial'
- l. 216: what does 'continuous' mean here? Aren't geostationary satellites able to provide 'continuous' estimates in time?
- l. 233: is cloud cover taken from the satellite cloud mask?
- l. 237: what are the most important 'spatially coarse input parameters' causing outliers?
- l. 242: include bibliographic references for the interpolation method;
- l. 246: 'the fundamental issue does not lie...' unclear;
- l. 271: 'is relying' --> 'relies';
- l. 299: 'and vice-versa', the intended meaning of the authors here is unclear;
- l. 308-309: does this mean that operating at a national scale is insufficient and that the algorithm should be applied at a European level?
Citation: https://doi.org/10.5194/egusphere-2023-3130-RC3 -
AC3: 'Reply on RC3', Barbara Klotz, 21 Mar 2024
We thank all the reviewers for the comments that reveal the issues of the manuscript. In the light of the required extensive "major" revisions to address all suggestions and also to deal with the rejection of our referenced paper (Schenzinger et al.) we believe that a re-submission without associated time constraints is the best way forward.
Citation: https://doi.org/10.5194/egusphere-2023-3130-AC3
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 264 | 90 | 30 | 384 | 18 | 18 |
- HTML: 264
- PDF: 90
- XML: 30
- Total: 384
- BibTeX: 18
- EndNote: 18
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
