the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 03 Apr 2023
Low-level Arctic clouds: A blind zone in our knowledge of the radiation budget
Abstract. Quantifying the role of clouds in the Earth radiation budget is essential for improving our understanding of the drivers and feedbacks of climate change. This holds in particular for the Arctic, the region currently undergoing the most rapid changes. This region, however, also poses significant challenges to remote-sensing retrievals of clouds and radiative fluxes, introducing large uncertainties in current climate data records. In particular, low-level stratiform clouds are common in the Arctic but are, due to their low altitude, challenging to observe and characterize with remote-sensing techniques. The availability of reliable ground-based observations as reference is thus of high importance. In the present study, radiative transfer simulations based on state-of-the-art ground-based remote sensing of clouds are contrasted to surface radiative flux measurements to assess their ability to constrain the cloud radiative effect. Cloud radar, lidar, and microwave radiometer observations from the PS106 cruise in the Arctic marginal sea ice zone in summer 2017 were used to derive cloud micro- and macrophysical properties by means of the instrument synergy approach of Cloudnet. Closure of surface radiative fluxes can only be achieved by a realistic representation of the low-level liquid-containing clouds in the radiative transfer simulations. The original, likely erroneous, representation of these low-level clouds in the radiative transfer simulations let to errors in the cloud radiative effect of 43 W m-2. The present study highlights the importance of jointly improving retrievals for low-level liquid-containing clouds which are frequently encountered in the high Arctic, together with observational capabilities both in terms of cloud remote sensing and radiative flux observations. Concrete suggestions for achieving these goals are provided.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(5371 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(5371 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-358', Anonymous Referee #1, 28 Apr 2023
Low-level stratiform clouds are a very common cloud type in the Arctic. However, capturing these clouds by remote sensing techniques still pose a challenge. The authors of this study address this very important topic by presenting an improved method to capture these near-surface clouds. In the presented case study, measurements from cloud radar, lidar and microwave radiometer are combined to retrieve cloud microphysical properties below the lowest cloud radar range gate, i.e. in the “radar blind zone”, which is not captured by the widely used Cloudnet cloud data products. In order to assess the improved representation of these low-level stratus clouds, a radiative closure study is performed, which reveals a much better closure with observed surface downward shortwave and terrestrial radiative fluxes. Including these very low-level stratus clouds has also a substantial impact on the cloud radiative effect (CRE): in this particular case, the net surface cooling effect by these clouds is substantially larger, if the clouds in the radar blind zone are taken into account.
The present study highlights the importance and also the challenge of observing Arctic low-level stratus clouds. The paper is in general well written and clear. My major concern is the short time period, i.e. just a case study, for which the improved method is presented. In order to clearly show the benefit of this new method, it needs to applied to an extended time period. I would have expected that after the presentation of the detailed case study (which is really nice), the application to a longer time series was shown. For example, it has been mentioned that for the PS106 cruise, these undetected low-level clouds occur in 25% of the time (Griesche et al., 2020). Why has the proposed method thus not been applied to the whole PS106 time series? How would the consideration of previously undetected low-level clouds change the CRE observed during the whole cruise? By means of a case study, it is difficult to draw some overall conclusions. I think that such an analysis will highlight the applicability of the new method and thus significantly add to the value of the manuscript. I thus recommend to include this piece of work before publication of this manuscript.
Minor comments:
lines 119-: Can you remind the reader how exactly LWC and reff is retrieved in Cloudnet?
lines 127-: “and the values for the HATPRO LWP and LWPLWC differ.”
In this case, LWPLWC is simply zero, since Cloudnet does not provide an LWC profile, right? Maybe this could be clarified in the text.line 144: There is no section 3.2. So subsection 3.1 is not needed. However, a section 3.2 could be added which includes an analysis of the whole PS106 time period, for example.
line 178: “In green,…” This sentence can be deleted.
lines 190-191: “Smaller deviations […] at around 5 UTC…”
I would not call the differences around 5 UTC small.Citation: https://doi.org/10.5194/egusphere-2023-358-RC1 - AC1: 'Reply on RC1', Hannes Griesche, 19 Jul 2023
-
RC2: 'Comment on egusphere-2023-358', Anonymous Referee #2, 19 May 2023
This study highlights the significance of enhancing retrievals for low-level liquid-containing clouds that are commonly found in the high Arctic, along with improving observational capabilities in cloud remote sensing and radiative flux observations. Difficulties in detecting low-level stratiform clouds introduce large uncertainties in climate observational records and simulations. The study uses ground-based remote sensing of clouds from PS106 to assess the ability to constrain the cloud radiative effect by comparing radiative transfer simulations with surface radiative flux measurements. It provides suggestions for improving the observational capabilities of cloud remote sensing and radiative flux observations in the Arctic region. This topic has a high impact as it provides important insights into a better representation of Arctic surface energy budget. The manuscript is well-written and easy to follow. However, I have major concerns about the study: 1) one case study is not sufficient to draw a robust conclusion; 2) the applicability of proposed adjusted classification scheme to other cases or ground-based observational field campaigns is not clear. Therefore, I would recommend this paper be published in this journal if my concerns can be addressed. Please see my detailed comments below.
Specific comments:
The study concluded that representation of these low-level clouds in the radiative transfer simulations let to errors in the cloud radiative effect of 43Wm−2, leading to an improved representation of surface radiation budget. All conclusions are drawn from one case study. To make conclusions more robust, it is necessary to perform calculations in different time and location.
Is proposed adjusted classification scheme applicable to other conditions or field campaigns?
In addition to 13 July 2017 case, do you have other cases to quantify the impact of adjusted Cloudnet classification scheme on surface radiative fluxes? How representative is this case to the summertime cloud conditions in the Arctic?
Line 34-48: A more recent study also performed radiative transfer simulations using measurements from MOSAiC field campaign to quantify the uncertainties in Arctic surface radiation budget.
Reference:
Huang, Y., P.C. Taylor, F.G. Rose, D.A. Rutan, M.D. Shupe, and M.A. Webster. (2022): Towards a more realistic representation of surface albedo in NASA CERES satellite products: a comparison with the MOSAiC field campaign. Elementa: Science of the Anthropocene, 10 (1): 00013.
Section 2.3: Can you mark the location of the observations taken on 13 July 2017 in a map?
Line 120: Please correct this sentence: “The cloud radar measurements, however, are limited to altitudes above 165m above the ground.”
Line 206-216: Is CRE calculation improved with adjusted Cloudnet classification scheme? Can you compare simulated CRE with the observations?
Line 229: “no qualitative assessment of the microphysical properties of these low-level clouds has been conducted.” Do you want to say “quantitative” here?
Section 4 Discussion and Conclusions: It would be better to re-organize this section. The challenges previously identified in field campaigns and studies would be better suited for the motivation section. The future direction of the study should follow the conclusions drawn from the current research.
Citation: https://doi.org/10.5194/egusphere-2023-358-RC2 - AC2: 'Reply on RC2', Hannes Griesche, 19 Jul 2023
-
CC1: 'Comment on egusphere-2023-358', Luca Lelli, 22 May 2023
This commentary is not meant to be a complete review of the work, but rather is meant to comment on a statement by the authors in the text.
In both the introduction (lines 54-56: "Yet ... missing.") and the conclusion (lines 298-301: "In conclusions ... clouds") it is stated that clouds, whose altitude is very low (~150 m, if I am not mistaken), cannot be characterized by means of satellite retrievals.This is stated without at least mentioning the different sensitivities of remote sensing techniques from satellite. While it is well known that space-based lidars can lose sensitivity to low clouds due to pulse attenuation in the presence of substantial tropospheric opacity and that equally missions such as those of CloudSAT can suffer from lack of contrast between the surface and lower layers of the atmosphere, I would like to point the authors to some recent studies which have have proven otherwise. That is, the detectability of clouds below 200 m.
The recent paper by Vinjamuri et al (AMT, 2023, https://doi.org/10.5194/amt-2022-312, accepted for final publication) has shown that over the North Slope Alaska site low cloud can be retrieved and compared to those collected by ground-based instrumentation. I paste here below a project-internal plot proving this, upon colocation with the ARM station.
The spaceborne record in the plot is labeled CCI and referes to the ESA Climate Change Initative record based on AVHRR measurements, spanning 1982-2019 (Stengel et al, ESSD, 2020, https://doi.org/10.5194/essd-12-41-2020).
Consequently, this AVHRR record has been used to assess pan-Arctic, regional and seasonal cloud radiative effect (CRE), or forcing (CRF) in Lelli et al. (ACP, 2023, https://doi.org/10.5194/acp-23-2579-2023) and feedbacks in Philipp et al. (JCLIM, 2020, https://doi.org/10.1175/JCLI-D-19-0895.1).
In these two papers all-cloud CRE and low-level clouds feedbacks are explicitly analyzed. But then very-low clouds, as investigated in the this work, are present together with their radiative influence, albeit conflated.
Certainly much more analysis and algorithmic development will be needed in the future to solve the inherently ill-posed problem of spaceborne remote sensing of very low clouds. From this perspective, the atmospheric profiling superiority of ground-based instrumentation is surely unmatched by satellite-based instruments. And likely for a long time to go yet.
But on the other hand, in situ (ground, ship, airborne) campaings will never cover the the Arctic in its full width.
I think that the authors of this work could relax their statements and conclusions, especially if they have not conducted themselves first-hand a sensitivity and frequency assessment of very-low clouds within the available Arctic cloud records (inferred with different techniques and sensors spanning the full electro-magnetic spectrum).
Citation: https://doi.org/10.5194/egusphere-2023-358-CC1 - AC3: 'Reply on CC1', Hannes Griesche, 19 Jul 2023
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-358', Anonymous Referee #1, 28 Apr 2023
Low-level stratiform clouds are a very common cloud type in the Arctic. However, capturing these clouds by remote sensing techniques still pose a challenge. The authors of this study address this very important topic by presenting an improved method to capture these near-surface clouds. In the presented case study, measurements from cloud radar, lidar and microwave radiometer are combined to retrieve cloud microphysical properties below the lowest cloud radar range gate, i.e. in the “radar blind zone”, which is not captured by the widely used Cloudnet cloud data products. In order to assess the improved representation of these low-level stratus clouds, a radiative closure study is performed, which reveals a much better closure with observed surface downward shortwave and terrestrial radiative fluxes. Including these very low-level stratus clouds has also a substantial impact on the cloud radiative effect (CRE): in this particular case, the net surface cooling effect by these clouds is substantially larger, if the clouds in the radar blind zone are taken into account.
The present study highlights the importance and also the challenge of observing Arctic low-level stratus clouds. The paper is in general well written and clear. My major concern is the short time period, i.e. just a case study, for which the improved method is presented. In order to clearly show the benefit of this new method, it needs to applied to an extended time period. I would have expected that after the presentation of the detailed case study (which is really nice), the application to a longer time series was shown. For example, it has been mentioned that for the PS106 cruise, these undetected low-level clouds occur in 25% of the time (Griesche et al., 2020). Why has the proposed method thus not been applied to the whole PS106 time series? How would the consideration of previously undetected low-level clouds change the CRE observed during the whole cruise? By means of a case study, it is difficult to draw some overall conclusions. I think that such an analysis will highlight the applicability of the new method and thus significantly add to the value of the manuscript. I thus recommend to include this piece of work before publication of this manuscript.
Minor comments:
lines 119-: Can you remind the reader how exactly LWC and reff is retrieved in Cloudnet?
lines 127-: “and the values for the HATPRO LWP and LWPLWC differ.”
In this case, LWPLWC is simply zero, since Cloudnet does not provide an LWC profile, right? Maybe this could be clarified in the text.line 144: There is no section 3.2. So subsection 3.1 is not needed. However, a section 3.2 could be added which includes an analysis of the whole PS106 time period, for example.
line 178: “In green,…” This sentence can be deleted.
lines 190-191: “Smaller deviations […] at around 5 UTC…”
I would not call the differences around 5 UTC small.Citation: https://doi.org/10.5194/egusphere-2023-358-RC1 - AC1: 'Reply on RC1', Hannes Griesche, 19 Jul 2023
-
RC2: 'Comment on egusphere-2023-358', Anonymous Referee #2, 19 May 2023
This study highlights the significance of enhancing retrievals for low-level liquid-containing clouds that are commonly found in the high Arctic, along with improving observational capabilities in cloud remote sensing and radiative flux observations. Difficulties in detecting low-level stratiform clouds introduce large uncertainties in climate observational records and simulations. The study uses ground-based remote sensing of clouds from PS106 to assess the ability to constrain the cloud radiative effect by comparing radiative transfer simulations with surface radiative flux measurements. It provides suggestions for improving the observational capabilities of cloud remote sensing and radiative flux observations in the Arctic region. This topic has a high impact as it provides important insights into a better representation of Arctic surface energy budget. The manuscript is well-written and easy to follow. However, I have major concerns about the study: 1) one case study is not sufficient to draw a robust conclusion; 2) the applicability of proposed adjusted classification scheme to other cases or ground-based observational field campaigns is not clear. Therefore, I would recommend this paper be published in this journal if my concerns can be addressed. Please see my detailed comments below.
Specific comments:
The study concluded that representation of these low-level clouds in the radiative transfer simulations let to errors in the cloud radiative effect of 43Wm−2, leading to an improved representation of surface radiation budget. All conclusions are drawn from one case study. To make conclusions more robust, it is necessary to perform calculations in different time and location.
Is proposed adjusted classification scheme applicable to other conditions or field campaigns?
In addition to 13 July 2017 case, do you have other cases to quantify the impact of adjusted Cloudnet classification scheme on surface radiative fluxes? How representative is this case to the summertime cloud conditions in the Arctic?
Line 34-48: A more recent study also performed radiative transfer simulations using measurements from MOSAiC field campaign to quantify the uncertainties in Arctic surface radiation budget.
Reference:
Huang, Y., P.C. Taylor, F.G. Rose, D.A. Rutan, M.D. Shupe, and M.A. Webster. (2022): Towards a more realistic representation of surface albedo in NASA CERES satellite products: a comparison with the MOSAiC field campaign. Elementa: Science of the Anthropocene, 10 (1): 00013.
Section 2.3: Can you mark the location of the observations taken on 13 July 2017 in a map?
Line 120: Please correct this sentence: “The cloud radar measurements, however, are limited to altitudes above 165m above the ground.”
Line 206-216: Is CRE calculation improved with adjusted Cloudnet classification scheme? Can you compare simulated CRE with the observations?
Line 229: “no qualitative assessment of the microphysical properties of these low-level clouds has been conducted.” Do you want to say “quantitative” here?
Section 4 Discussion and Conclusions: It would be better to re-organize this section. The challenges previously identified in field campaigns and studies would be better suited for the motivation section. The future direction of the study should follow the conclusions drawn from the current research.
Citation: https://doi.org/10.5194/egusphere-2023-358-RC2 - AC2: 'Reply on RC2', Hannes Griesche, 19 Jul 2023
-
CC1: 'Comment on egusphere-2023-358', Luca Lelli, 22 May 2023
This commentary is not meant to be a complete review of the work, but rather is meant to comment on a statement by the authors in the text.
In both the introduction (lines 54-56: "Yet ... missing.") and the conclusion (lines 298-301: "In conclusions ... clouds") it is stated that clouds, whose altitude is very low (~150 m, if I am not mistaken), cannot be characterized by means of satellite retrievals.This is stated without at least mentioning the different sensitivities of remote sensing techniques from satellite. While it is well known that space-based lidars can lose sensitivity to low clouds due to pulse attenuation in the presence of substantial tropospheric opacity and that equally missions such as those of CloudSAT can suffer from lack of contrast between the surface and lower layers of the atmosphere, I would like to point the authors to some recent studies which have have proven otherwise. That is, the detectability of clouds below 200 m.
The recent paper by Vinjamuri et al (AMT, 2023, https://doi.org/10.5194/amt-2022-312, accepted for final publication) has shown that over the North Slope Alaska site low cloud can be retrieved and compared to those collected by ground-based instrumentation. I paste here below a project-internal plot proving this, upon colocation with the ARM station.
The spaceborne record in the plot is labeled CCI and referes to the ESA Climate Change Initative record based on AVHRR measurements, spanning 1982-2019 (Stengel et al, ESSD, 2020, https://doi.org/10.5194/essd-12-41-2020).
Consequently, this AVHRR record has been used to assess pan-Arctic, regional and seasonal cloud radiative effect (CRE), or forcing (CRF) in Lelli et al. (ACP, 2023, https://doi.org/10.5194/acp-23-2579-2023) and feedbacks in Philipp et al. (JCLIM, 2020, https://doi.org/10.1175/JCLI-D-19-0895.1).
In these two papers all-cloud CRE and low-level clouds feedbacks are explicitly analyzed. But then very-low clouds, as investigated in the this work, are present together with their radiative influence, albeit conflated.
Certainly much more analysis and algorithmic development will be needed in the future to solve the inherently ill-posed problem of spaceborne remote sensing of very low clouds. From this perspective, the atmospheric profiling superiority of ground-based instrumentation is surely unmatched by satellite-based instruments. And likely for a long time to go yet.
But on the other hand, in situ (ground, ship, airborne) campaings will never cover the the Arctic in its full width.
I think that the authors of this work could relax their statements and conclusions, especially if they have not conducted themselves first-hand a sensitivity and frequency assessment of very-low clouds within the available Arctic cloud records (inferred with different techniques and sensors spanning the full electro-magnetic spectrum).Citation: https://doi.org/10.5194/egusphere-2023-358-CC1 - AC3: 'Reply on CC1', Hannes Griesche, 19 Jul 2023
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 462 | 180 | 32 | 674 | 16 | 16 |
- HTML: 462
- PDF: 180
- XML: 32
- Total: 674
- BibTeX: 16
- EndNote: 16
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Hannes Jascha Griesche
Carola Barrientos-Velasco
Hartwig Deneke
Anja Hünerbein
Patric Seifert
Andreas Macke
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(5371 KB) - Metadata XML