the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
AI-derived 3D cloud tomography from geostationary 2D satellite data
Sarah Brüning
Stefan Niebler
Holger Tost
Abstract. Satellite instruments provide spatially extended data with a high temporal resolution on almost global scales. However, nowadays, it is still a challenge to extract fully three-dimensional data from the current generation of satellite instruments, which either provide horizontal patterns or vertical profiles along the orbit track. Following this, we train a neural network in this study to generate three-dimensional cloud structures from MSG SEVIRI satellite data in high spatio-temporal resolution. We evaluate the derived artificial intelligence-based predictions against the along-track radar reflectivity from the CloudSat satellite. By inferring the pixel-wise cloud column to the satellite’s full disk, our results emphasize that spatio-temporal dynamics can be delineated for the whole domain. Robust reflectivities are derived for different cloud types with a clear distinction regarding the cloud's intensity, height, and shape. Cloud-free pixels tend to be over-represented because of the high imbalance between cloudy and clear-sky samples. The average error (RMSE) spans about 7.5 % (3.41 dBZ) of the total value range enabling the advanced analysis of vertical cloud properties. Although we receive high accordance between radar data and our predictions, the quality of the results varies with the complexity of the cloud structure. The representation of multi-level and mesoscale clouds is often simplified. Despite current limitations, the obtained results can help close current data gaps and exhibit the potential to be applied to various climate science questions, like the further investigation of deep convection through time and space.
- Preprint
(8128 KB) - Metadata XML
- BibTeX
- EndNote
Sarah Brüning et al.
Status: open (until 12 Oct 2023)
-
RC1: 'Comment on egusphere-2023-1834', Anonymous Referee #1, 08 Sep 2023
reply
Review of "AI-derived 3D cloud tomography from geostationary 2D satellite data" by Brüning et al.
This paper presents a neural network that can reconstruct 3D cloud structures from 2D visible/IR satellite data. The authors describe a network architecture that maps the 2D observations to 3D fields. They then present a comparison to other machine-learning methods and compare the statistics of network-predicted and real observations.
The most interesting part of the paper to me is the network structure that generates the 3D fields. The U-Net structure is appropriate for the problem at hand, although it is not quite clear from the description how the transformation from 2D to 3D is performed.
While the network is interesting, I wonder about the usefulness of the model outputs since the model clearly suffers from regression to the mean and blurriness of the results. Also, the uncertainty of the outputs is not estimated. Could you comment a bit on which applications would find the current results useful? And how the abovementioned issues could be improved?
The language in the paper needs some improvement. There are awkward sentence structures throughout the paper, as well as misuse of vocabulary and terminology, which make the paper hard to understand at times. I point out some examples in the specific comments, but these are not an exclusive list; the whole paper would benefit from editing. In contrast, the figures in the paper are clear and well made.
Specific comments:
Lines 24-25: "Passive sensors such as geostationary satellites": the statement needs more precision, satellites are not sensors
Lines 36-37: "The large-scale generability of these methods is expandable since their 3D results are limited to the cloud’s spatial vicinity": I don't understand this sentence, please clarify.
Line 48: "time efficiency and feasibility": it's not clear to me what this means
Line 80: "orbiting the globe on a sinusoidal track": how does a satellite orbit on a "sinusoidal track"?
Lines 91-92: "resampled to a geographic grid": what kind of grid, a lat-lon one?
Section 2.1.2: CloudSat is on a sun-synchronous orbit, meaning it sees every location at the same local solar time. This might introduce some diurnal bias to the data; this should be acknowledged.
Line 109: The use of "XY" is confusing, I read this initially as "X times Y" but apparently you mean a diagonal transect through the image? Or did I misunderstand?
Line 125: "smoothing" should probably be "filtering"
Line 137: The use of the word "delineate" here and a couple of other places in the places seems incorrect
Line 159: I would like some more details on how the network structure maps the 2D input fields to the 3D output fields. Is the channels dimension used transformed to the Z dimension of the output?
Lines 163-164: How is the 3D scene predicted by the network compared to the CloudSat data during training? CloudSat only gets a 2D vertical cross section of the scene. Is only part of the scene selected for comparison? Also, what loss function do you use for training?
Line 176: "Both models": unclear which models this refers to
Line 181: "pictures" is used incorrectly here.
Lines 189-190: In the joined 2400 x 2400 pixel 3D prediction, is the field continuous at the borders of the 100 x 100 pixel tiles? Or do you see discontinuities or artifacts?
Line 220: "That said" seems out of place here - please revise.
Line 231: I don't understand that "denominational structure" means here.
Lines 262-263: High, thin ice clouds may also not be observed by CloudSat due to being under the minimum detectable reflectivity.
Figure 7: Maybe you could add a panel showing the difference of a and b to illustrate the biases better.
Line 288: "Leaving out the affected channels downgrades the overall performance": it would be good to see something to demonstrate this.
Lines 292-293: "In contrast to pixel-based DL methods like the CNN or CGAN, the Res-UNet utilizes a larger receptive field preserving the spatial dimensionality and global context information during the training routine." This is not a correct statement regarding the CNN or CGAN architectures. CNN architectures can also achieve large receptive fields and global context using downsampling. In fact the UNet itself is a type of CNN - its distinguishing feature is the addition of skip connections to preserve resolution. As for the CGAN, it refers to a certain training setup of generative models that could be implemented with either normal CNNs or with (Res-)UNets.
Lines 312-313: Approximately 1 km resolution is also already available from the GOES-R series and Himawari 8/9 satellites.
Lines 321-322: "Since it is independent of external or interconnected data sources, the bias within the data is reduced.": unclear sentence, I'm not sure how the latter follows from the former.
Citation: https://doi.org/10.5194/egusphere-2023-1834-RC1
Sarah Brüning et al.
Sarah Brüning et al.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 197 | 124 | 9 | 330 | 6 | 4 |
- HTML: 197
- PDF: 124
- XML: 9
- Total: 330
- BibTeX: 6
- EndNote: 4
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1