the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 18 Sep 2024
Retrieval of Atmospheric Water Vapor and Temperature Profiles over Antarctica through Iterative Approach
Abstract. Retrieving atmospheric water vapor and temperature profiles presents considerable challenges over land surfaces using microwave radiometry due to uncertainties associated with estimating background surface emissions. In response, we have devised an approach that integrates the atmospheric retrieval algorithm with the background emission algorithm, establishing an iterative loop to refine the accuracy of atmospheric profiles. Leveraging optimal estimation techniques with sounding channels spanning from Ka- to G-band obtained from ATMS, we successfully retrieved atmospheric temperature and humidity profiles across space and time. These retrieved atmospheric profiles undergo continual updates throughout each iteration, exerting influence on subsequent surface retrievals. This iterative process persists until convergence is achieved in the atmospheric retrieval. The algorithm's novelty lies in its fusion of surface retrieval with atmospheric retrieval, thereby enhancing overall accuracy. We validated the retrievals against radiosonde data. Our iterative algorithm proved to be efficient and accurate in retrieving temperature profiles with surface emissivity and in detecting melting events. Though our algorithm was able to capture the water vapor variations, the results showed that to obtain accurate absolute values of the water content an independently retrieved surface emissivity is required.
- Preprint
(7349 KB) - Metadata XML
-
Supplement
(2392 KB) - BibTeX
- EndNote
Status: open (extended)
-
RC1: 'Comment on egusphere-2024-2578', Anonymous Referee #1, 18 Dec 2024
reply
Summary:
The paper presents retrievals of water vapor, temperature, and surface emissivity from microwave satellite observations over Antarctica using an optimal estimation approach. Surface emissivity and atmospheric profiles are iteratively retrieved jointly. The algorithm has a good performance in the lower frequency range but has some difficulties with the 183 GHz frequency which leads to some uncertainty in the water vapor estimation. The retrievals are shown in relation to their ability to detect temperature and humidity perturbations related to surface melt events in various regions in Antarctica.
General comments:
The paper is generally clearly written and focused. The technique is well described. Some changes are in my opinion necessary to improve the results.
One recommendation is to make the results more quantitative where possible. For example, at McMurdo, where radiosondes are available it would be good to see a profile of the RMS errors between satellite-derived and radiosondes temperature and humidity.
A second important aspect that should be discussed is how are clouds included in the retrieval algorithm? Do you retrieve LWP/IWP at the same time or you use a different LWP product in your optimal estimation? If so this should be well explained.
Finally, because melt events involve the surface it may be useful to focus at least part of the analysis on the mid-low troposphere. Especially a focus on assessing the quality of the retrievals below 5 km would be important.
Specific comments:
Abstract: In the title and/or abstract there should be a mention that these are satellite retrievals.
Abstract, line 12: Should the acronym ATMS on line 12 be spelled out?
Introduction, line 42: “With a total of 22 channels they are equipped…” The sentence needs revising.
Section 2.1, line 55: “…Sounder…are meant…” Should this be “is” meant?
Section 2.1, line 98: For the Jacobians shown here what is the vertical grid used? Is it the same as the MERRA mentioned later (73 layers, between surface and 70 km)? What vertical resolution? Is it uniform grid or it varies with height.
Section 2.2. I wonder if more discussion is needed here. What is used in the MERRA reanalysis product? It may be necessary to show at least the variability of the a priori profiles for a few locations. In my opinion in section 4 both MERRA profiles and these retrievals should be compared with radiosondes at McMurdo.
Section 2.3, line 143: “AWARE collected atmospheric profiles every minute”. I think there is some confusion here and also regarding the WAIS deployment. During both deployments 4-6 radiosondes were launched daily. Here you are probably referring to the interpolated sonde product where the profiles between radiosonde launches are interpolated into a 1-minute time grid according to some scheme and then scaled using the ground-based microwave radiometer integrated water vapor.
Section 3: In this section it is necessary to explain how clouds are included in the retrievals
Section 4: Here or perhaps in the retrieval session it would be good to explain the temporal resolution of the satellite dataset. How many satellite retrievals per day do you get? What time of day? Does the time of day of the retrievals varies depending on location?
Section 4.1, line 287: “…only 68% showing a difference within 30% of the actual profile”. This sentence is not clear. Why are the differences in specific humidity expressed as a percentage instead of g/kg (just like Kelvin for temperature?). It can be then specified that a given amount corresponds to 30%.
Section 4.1: As mentioned earlier in this section it would be good to show the vertical profile of the RMSE between retrievals and radiosondes and also between the a priori and radiosondes. Id it possible?
Fig. 8b and other similar figures: It would be good to keep the same color palette (viridis?) for temperature and humidity. Again, it is not clear what is the temporal resolution of the retrievals. In the caption it should be specified if those are interpolated sondes (1-minute time resolution).
Section 4.1, line 320: “From the average value during the specified period”. So, average between December 1 and March 1? Because this section and the next section refer to the same melt event it may be better to show both figures spanning the same time frame (Jan 1-30). This is also going to provide a comparable magnitude of the perturbation.
Section 4.2 title: This title is somewhat confusing. It says, “More locations near Ross ice shelf” but then there is a subsection titled “Ross ice shelf”. Perhaps “WAIS divide” can be 4.2 and Ross ice shelf could be section 4.3. The same for the Ross shelf. The 3 locations experienced a degree of melting event in January 2016, perhaps is better to show the same time frame (Jan 1-30) at the 3 locations.
Fig. 9a: It may be better to focus this on January 2016 as in figure 9b.
Fig. 11 caption: Is the average over the entire year? Because then you have the annual cycle embedded in the perturbations. See comment below
Figure 11 and 12 and 13 are somewhat difficult to see and their usefulness is not very clear. Perhaps a plot showing the retrieved emissivity vs skin temperature and retrieved emissivity vs. temperature/humidity perturbation may provide better insight. Also, I wonder if it is better to look at perturbations over summer/winter averages rather than all year average. Or you may want to low-pass filter (for example with a window of 2 months) the year time series to get the background field and subtract that from the single retrievals.
Thank you
Citation: https://doi.org/10.5194/egusphere-2024-2578-RC1
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 123 | 28 | 10 | 161 | 21 | 5 | 3 |
- HTML: 123
- PDF: 28
- XML: 10
- Total: 161
- Supplement: 21
- BibTeX: 5
- EndNote: 3
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1