the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 21 Oct 2021
Formulation and demonstration of an extended-3DVAR multi-scale data assimilation system for the SWOT altimetry era
Abstract. A state-of-the-art data assimilation system for a high-resolution model has been developed to address the opportunities and challenges posed by the upcoming Surface Water and Ocean Topography (SWOT) satellite mission. A new ‘extended’ three-dimensional variational data assimilation scheme (extended-3DVAR) is formulated to assimilate observations over a time interval, and integrated using a multi-scale approach (hereafter MSDA). The new MSDA scheme specifically enhances the efficacy of the assimilation of satellite along-track altimetry observations, which are limited by large repeat time intervals. This developed system is computationally highly efficient, and thus can be applied to a very high-resolution model. A crucial consideration of the system is first to assimilate routinely available observations, including satellite altimetry, sea surface temperature (SST) and temperature/salinity vertical profiles, to constrain large scales and large mesoscales. High-resolution (dense) observations and future SWOT measurements can then be effectively and seamlessly assimilated to constrain the smaller scales. Using this system, a reanalysis dataset was produced for the SWOT pre-launch field campaign that took place in the California Current System from September through December, 2019. An evaluation of this system with assimilated and withheld data demonstrates its ability to effectively utilize both routine and campaign observations. These results suggest a promising avenue for data assimilation development in the SWOT altimetry era, which will require the capability to efficiently assimilate large-volume datasets resolving small-scale ocean processes.
This preprint has been withdrawn.
-
Withdrawal notice
This preprint has been withdrawn.
-
Preprint
(4140 KB)
Interactive discussion
Status: closed
-
RC1: 'Comment on os-2021-89', Anonymous Referee #1, 25 Nov 2021
This manuscript introduces a high-resolution ocean data assimilation system for generating effective ocean state estimation with SWOT data. Authors adopt an extended 3DVAR and multi-scale data assimilation strategies, and mathematical background of those theories are described. They also evaluate the effectiveness of assimilating along-track satellite SSH data, and dense TS profiles obtained by an observational campaign.
In the manuscript, the current status of SWOT and its evaluation at the cross-over point are well described, which provides readers valuable information. They also show that the system is appropriately represent small mesoscale and submesoscale phenomena. However, I feel that some essential information on the configuration of the data assimilation system is missing (see major comment A), and therefore it is difficult to understand how the adopted data assimilation strategies work in their system. And they do not appropriately evaluate the impacts of introducing the extended 3DVAR (see major comment B). Therefore, I recommend to publish this manuscript after a major revision.
Major comments
A) Some essential information on the configuration of the data assimilation is missing in the manuscript. It is not clear what prediction variables in the ocean model are modified by the data assimilation. Does the system modify only TS and SSH, or does it also modify the velocity fields? And I did not find the description on how the DA system links the SSH increments to the TS increments. And there is no information how the analysis increments are inserted in the ocean model. Does the system adopt Incremental Analysis Updates or just direct insertion? And how often does the system perform the analysis?
Also, some information is missing on the observation and background errors. For example, I cannot find the information on the observation errors for the regular TS profiles. And, more importantly, authors should describe how they deal with the representativeness errors because a sophisticated setting is required for their multi-scale DA methods. For example, the standard deviations of the small mesoscale and submesoscale variations must be included in the representativeness errors of the regular TS profiles because the data are assimilated as the large mesoscale observations. The background errors for the three scales also have to be determined very carefully. These informations are required to evaluate the results shown in this manuscript correctly.
b) The impact of extended 3DVAR is not appropriately evaluated although authors insist it is the unique point of their system. In order to show the impacts of introducing extended 3DVAR, authors should compare the results with experiments in which a regular observation time window is used or inflation of the observation errors due to the sampling time errors are not applied. In addition, I would like to say that the extended 3DVAR cannot draw full values of observation data because larger observation errors are assigned to the data observed at the time deviated from the analysis time. It would be valuable if authors address why the extended 3DVAR is selected instead of FGAT, which is widely used in ocean data assimilation systems.
Minor comments
- Caption of Figure 1: It should be clearly explained that the swaths of SWOT are depicted by the thin light brown color zones and the cross over of two swaths are represented as the thicker light brown diamond areas.
- L133: It should be mentioned that the absolute dynamic topography is calculated from the sea surface height anomaly observed by satellite altimeters and the mean sea surface dynamic height estimated from other sources.
- L180: It would be better if authors kindly touches here that MW SST is assimilated in this study.
- L181: The subsession title “HYCOM” is confusing because “HYCOM” originally indicates an ocean model, not the reanalysis data. I suggest replacing the title by “HYCOM reanalysis data”.
- L204: It is not adequate that an equation is referred before it is written down in the text. Maybe “the cost function (5)” can be relaced by “A cost function that includes integration of the prediction model”.
- L235: ykot is not defined.
- L239, “In equations (2) and (3), however, the prediction model is not required to be the same model as in (3), but a different prediction model 240 can be chosen instead.”: I do not understand what this sentence means. In my understanding, the prediction model is just replaced by the persistence model as a simple approximation.
- L235: ykt is not defined and it is probably the same as ykot
- L252: I suggest inserting “,ekm ,” after “The first term, ",ekr ,” after “the second term, and “,ekmp ,” after “The last term”.
- L290, “15-day average”: Is it running average?
- L330: The small mesoscale observations include the large and small mesoscale components. And only the small mesoscale component should be input as the observation data into the cost function, that is, the large mesoscale component should be removed before using it in the analysis. It may be done by reducing the observed value by multiplying a coefficient less than one, or more practically by inflating the observation errors. Because of this reason, I think gk should be more than 1 even for day 0. In addition, the values of observation errors before the inflation are not shown for the small mesoscale.
- L340: How many grid points are required in the wavelength in order to represent the wave is a kind of controversy topics. Maybe some people think two points are enough and others insist more than three points are required. It should be nice if authors provide some reference that endorses their three-point criterion.
- L351: What “optimization of the MSDA implementation” indicates is not clear.
- L416: A period is missed at the end of the line.
- L430, “the MSDA-SWOT shows a significant improvement in SSH prediction.”: This cannot be affirmed. I suggest inserting “likely” between “MSDA-SWOT” and “shows”.
- L431, “The RMSD with the 2-day forecast”: Considering the typical time scale of the oceanic variability, I feel 2-day may be too short. Authors should compare the scores with the persistence forecasts. I also think it is very natural that this short forecast result is better than the free run result.
- L440: Considering the slow variations of oceanic fields (especially SSH fields), we cannot say that the along-track SSH data are completely independent from the 2-day forecast. The representativeness errors in SSH observations are generally correlated with the same data observed 2-days before, and I think it cannot be ignored if we consider the long time-scale of the oceanic variations.
- L515, “sporous”: this must be a misspelling of “spurious”.
- L544, “only in a pure linear system can observations be assimilated fully, reduce the error in the analysis”: Although I am not sure if I understand this sentence correctly, I think this sentence means that, only in a pure linear system, observations can be assimilated fully and can reduce the error in analysis. I agree that observation can be assimilated fully only in a pure system, but I think the analysis error can be reduced even if the system is nonlinear.
- L546: I agree that the reduction of the analysis errors does not ensure the reduction of forecast errors. But in the case of the predictions in this study, I think it is very natural that smaller analysis errors induce smaller forecast errors because the typical time scale of ocean variations are longer than the lead time of the forecasts.
- L549, “With a small analysis increment, the model performs more linearly in the time evolution related to the increment.”: I am skeptical on this discussion. I think the total of the large and small mesoscale and submesoscale increments is as large as the increments of conventional DA systems. And I think the small forecast errors are just caused by the small analysis errors.
Citation: https://doi.org/10.5194/os-2021-89-RC1 -
RC2: 'Comment on os-2021-89', Anonymous Referee #2, 03 Dec 2021
Review of "Formulation and demonstration of an extended-3DVAR multi-scale data assimilation system for the SWOT altimetry era" by Li et al
Summary
This paper claims to introduce a DA system suitable for assimilating wide swath altimetry data (SWOT). I found this a disappointing paper. First, the paper misleads the reader by giving the impression that assimilation of SWOT data will be explored or tested. In fact this is not the case and the discussion is insufficent to convince the reader that this system will be capable of assimilating SWOT. In particular, a system should be capable of assimilating data with the particular error characteristics of SWOT data. Particularly of concern are correlated errors in the swaths due to phase, roll and timing errors. Second, the experiments performed are unilluminating. The paper is really about demonstrating extended time window 3DVAR. This needs to be compared to a control experiment doing non-extended 3DVAR otherwise the results are of little interest. At the moment I cannot tell if the extended 3DVAR makes things better than your non-extended 3DVAR.
The paper, in my view, goes into too much detail in aspects that were not novel and glosses over aspects that were potentially novel. An example of this is spending a lot of time on the characteristics of the existing observing network for example SST and at the same failing to go into sufficient detail on how the "time sampling" error is determined. Similarly the discussion of the results is superficial.
I recommend rejecting this paper for the above reasons. I give some more detailed comments below.
More detailed comments
The title is misleading it is not convincing that this system addresses the assimilation of SWOT data. Just removing "for the SWOT altimetry era" from the title would avoid misleading the reader. By "extended" you mean "extended time window" it would be clearer to write that.
In the abstract you write your system "is first to assimilate routinely available observations" is misleading. Is the first for your system? It is certainly not the first in world to do this.Line 93: Error in the citation Li et al (2019). There is a Li et al (2019a) and Li et al (2019b) which one is it?
It is highly misleading to call the system MSDA-SWOT when you don't appear to have tested it with SWOT data. I feel strongly that it is not acceptable to do this.Figure 1b with the Google Earth background is very unclear.
Section 2.3 First sentence. Use of gridded and along track altimeter would mean you are assimilating the same data twice because the along track data is used to generate the gridded data. This needs clarification here.
Figure 2. explain in the caption what the ring of 6 black circles at about 36 N 125 W
Line 148. Gridded products will also distort the positions of eddies. They are produced using along track altimeter data. Indeed they may be more likely to do so since they do not use a model to evolve the eddy positions from a previous analysis.
Figure 4. A plot of the observations is really unnecessary here. The reader will be thinking “when am I going to see something about SWOT data”? The answer of course is never.
Section 2.4 line 178. You should write what you do in this paper that is not done in Li 2019b. It is unclear at this point whether the data assimilated or just used for validation in the work presented here and is it IR SST and/or Microwave SST (line 180).
Section 3. lines 191-95. It is not clear why a 1 km or finer resolution is necessary to assimilate SWOT. I suppose you might say a km scale model is needed to take full advantage of SWOT but it is not ncessary to assimilate SWOT data. Anyway your model does not meet the claimed requirement having a resolution of ~3 km.
Section 3. lines 212-215. and Section 3.1.2. It is a good idea to account for the sampling time error but note this is not a new approach (see below). It should be noted first that errors associated with this are not random or Gaussian since they are associated with not accounting for model evolution. Second the impact of this will be to reduce the weight given to observations far from the analysis time which may consequently have only a small impact on the analysis. Assimilating such data if not carefully done may degrade the analysis in some cases.
Section 3.1.2. The sampling time error is far from a new type of observation error - Oke et al (2005) and Martin et al (2007) both increase the observation error with based on the difference of observation and analysis time. Additionally, since this is one of key novel aspects in your system this section requires much more detail. Questions not answered are what is the form of the time sampling error how does it vary with the analysis observation time difference? How are any parameters determined?
Section 4.1. Second sentence. Delete "Here we give a description". Not necessary since you are giving a description in the next sentence.
Section 4.2. There seems to be a fundamental misunderstanding in what the background error is. It does not relate to the observation scales instead it is the scales in the model forecast (the background) error. I would concede this is affected by previous observations assimilated in previous assimilation cycles, but not in the straightforward way you claim here. It is not clear what you mean by in lines 311-313. This should be clarified and perhaps explained with a specific example.
Section 4.2 line 325. It would help the reader to describe in words what the terms in equation 7 mean. What is the justification for (7) particularly including the measurement/instrumental error in the calculation of the sampling time error? This error is due to not including the model time evolution in the calculation of H(x) and this should not be affected by the instrumental error. I suppose there is a possible argument for time sampling errors being associated with the representativeness error since this relates to how well the model equivalent represents the observations. This may be higher where the model is more variable and therefore where bigger errors would be due to assuming the model is the constant in the cost function. This whole discussion could do with more detail since this is what makes this work different to previous work you have done. Describing exactly how lambda_k is determined would be interesting.
The paragraph lines 336-342 is very unclear. I think my problem relates to my fundamental issue with this whole section. I do not see that you can just specify the background error correlation scales based on baseline requirements from SWOT it should related to what the actual background error is in your system.
Table 2 is confusing. You list observations, but then it appears that many of the types are not assimilated in the experiments here. In that case they should not be a table with the caption "observations assimilated".
Table 2. It does not seem good practice to assimilate maps created from along track SSH and then assimilate the along track SSH data. You are in affect assimilating the same thing twice since the along track SSH data was used to generate the map. You should at least highlight potential issues with this approach not least overfitting to the observations.
Section 5. To convincingly demonstrate the utility of extended-3DVar but you need to run a control experiment with non-extended-3DVar. It would also be useful to have another experiment to assess the impact of the optimisations you mention versus a control run.
Section 5.1. Assessing the assimilation analysis against observations it has assimilated to produce the analysis is not very useful. The results may look good but you may in fact just be overfitting to the observations and their associated errors.
Figure 6. Describe the significance of the SWOT baseline line compared to the nadir altimeters observations. It has more power at larger wavenumbers/ smaller wavelengths. Is this signal or noise? Is there a significance where the lines cross over? Why is the power so much less at smaller wavenumbers/longer wavelengths?
Figure 7. Anomaly correlations are much more interesting that correlations this largely just tells me that the mean fields match. This needs another experiment to compare to. Explain why the RMSE grows with time. At the moment I might think that you started your DA experiment with an analysis from your old system and made the results worse with the extended-3DVar. Perhaps seasonal effects are causing the increase in error with time but I cannot tell from the results you show.
Line 408 A change from 2.6 cm to 2.8 cm does not seem substantial to me.
Line 436. Saying the domain average RMSD is as large as 10.0 cm is imprecise language just say what it is averaged over the comparison time period.
Line 440. "As much as 14%" (is it 14% or not?)
Figure 8. Again what is the little black ring of circles in each plot. It should be stated in the caption in each figure it appears in.
Figure 9. I really think you need to give anomaly correlations for SST since it is quite easy to match the climatology and achieve a very high correlation. The (non-anomaly) correlations will be very high even for a non-assimilating run. This plot also needs another experiment to compare to. Again the error grows with time so with no other experiment the reader may conclude that your changes may be making the results worse.
Figure 10. Not really discussed in detail. What is the significance of the shape of the histogram, for example? It again could really do with another experiment to compare the results with. Also the anomaly correlation should be used as the correlation is not very useful to SST as I explained previously.
Section 5.4 I think this idea of the "campaign area circulation" is a potentially interesting one. But the exploration of it here is superficial. I think this would be a good place to add more figures. It would be interesting to show an example of this and how your work has reduced such errors. It seems a bit trivial to say if you assimilate other observation types and keep your analysis close to the truth then the increments from campaign observations will be smaller. An illustration of this with results (perhaps showing surface currents) from a run where you exclude another data type so that analysis is not as close along with your campaign data would be useful to see.
Lines 475 I’m not really sure the comparison to no data assimilation is particularly interesting it is unlikely that someone would assimilate campaign data and fail to assimilate other observational data.
Line 492 Briefly explain here again what the DA Cal experiment is. What does “Cal” stand for?
Figure 11. You are comparing against observations you assimilate I’m not sure this plot is particularly useful since it is quite easy for a DA system to fit data it is assimilating. It certainly doesn’t illuminate on the “campaign area circulation” idea.
Figure 12. This figure is a mess of overlapping lines there doesn’t appear to be any consistent differences between the experiment results. The discussion of this figure does not help in at all to explain what the reader is supposed to conclude from this figure.
References
Oke, P.R., Schiller, A., Griffin, D.A. and Brassington, G.B. (2005), Ensemble data assimilation for an eddy-resolving ocean model of the Australian region. Q.J.R. Meteorol. Soc., 131: 3301-3311. https://doi.org/10.1256/qj.05.95
Martin, M.J., Hines, A. and Bell, M.J. (2007), Data assimilation in the FOAM operational short-range ocean forecasting system: a description of the scheme and its impact. Q.J.R. Meteorol. Soc., 133: 981-995. https://doi.org/10.1002/qj.74
Citation: https://doi.org/10.5194/os-2021-89-RC2 -
EC1: 'Comment on os-2021-89: Manuscript Rejected', Bernadette Sloyan, 24 Feb 2022
This manuscript has been rejected for publication as the authors have advised that they are unable to provide a reply to the reviewer's concerns and a revised manuscript.
I sincerely thank the reviewers for their time to provide a thorough review the manuscript.
Citation: https://doi.org/10.5194/os-2021-89-EC1
Interactive discussion
Status: closed
-
RC1: 'Comment on os-2021-89', Anonymous Referee #1, 25 Nov 2021
This manuscript introduces a high-resolution ocean data assimilation system for generating effective ocean state estimation with SWOT data. Authors adopt an extended 3DVAR and multi-scale data assimilation strategies, and mathematical background of those theories are described. They also evaluate the effectiveness of assimilating along-track satellite SSH data, and dense TS profiles obtained by an observational campaign.
In the manuscript, the current status of SWOT and its evaluation at the cross-over point are well described, which provides readers valuable information. They also show that the system is appropriately represent small mesoscale and submesoscale phenomena. However, I feel that some essential information on the configuration of the data assimilation system is missing (see major comment A), and therefore it is difficult to understand how the adopted data assimilation strategies work in their system. And they do not appropriately evaluate the impacts of introducing the extended 3DVAR (see major comment B). Therefore, I recommend to publish this manuscript after a major revision.
Major comments
A) Some essential information on the configuration of the data assimilation is missing in the manuscript. It is not clear what prediction variables in the ocean model are modified by the data assimilation. Does the system modify only TS and SSH, or does it also modify the velocity fields? And I did not find the description on how the DA system links the SSH increments to the TS increments. And there is no information how the analysis increments are inserted in the ocean model. Does the system adopt Incremental Analysis Updates or just direct insertion? And how often does the system perform the analysis?
Also, some information is missing on the observation and background errors. For example, I cannot find the information on the observation errors for the regular TS profiles. And, more importantly, authors should describe how they deal with the representativeness errors because a sophisticated setting is required for their multi-scale DA methods. For example, the standard deviations of the small mesoscale and submesoscale variations must be included in the representativeness errors of the regular TS profiles because the data are assimilated as the large mesoscale observations. The background errors for the three scales also have to be determined very carefully. These informations are required to evaluate the results shown in this manuscript correctly.
b) The impact of extended 3DVAR is not appropriately evaluated although authors insist it is the unique point of their system. In order to show the impacts of introducing extended 3DVAR, authors should compare the results with experiments in which a regular observation time window is used or inflation of the observation errors due to the sampling time errors are not applied. In addition, I would like to say that the extended 3DVAR cannot draw full values of observation data because larger observation errors are assigned to the data observed at the time deviated from the analysis time. It would be valuable if authors address why the extended 3DVAR is selected instead of FGAT, which is widely used in ocean data assimilation systems.
Minor comments
- Caption of Figure 1: It should be clearly explained that the swaths of SWOT are depicted by the thin light brown color zones and the cross over of two swaths are represented as the thicker light brown diamond areas.
- L133: It should be mentioned that the absolute dynamic topography is calculated from the sea surface height anomaly observed by satellite altimeters and the mean sea surface dynamic height estimated from other sources.
- L180: It would be better if authors kindly touches here that MW SST is assimilated in this study.
- L181: The subsession title “HYCOM” is confusing because “HYCOM” originally indicates an ocean model, not the reanalysis data. I suggest replacing the title by “HYCOM reanalysis data”.
- L204: It is not adequate that an equation is referred before it is written down in the text. Maybe “the cost function (5)” can be relaced by “A cost function that includes integration of the prediction model”.
- L235: ykot is not defined.
- L239, “In equations (2) and (3), however, the prediction model is not required to be the same model as in (3), but a different prediction model 240 can be chosen instead.”: I do not understand what this sentence means. In my understanding, the prediction model is just replaced by the persistence model as a simple approximation.
- L235: ykt is not defined and it is probably the same as ykot
- L252: I suggest inserting “,ekm ,” after “The first term, ",ekr ,” after “the second term, and “,ekmp ,” after “The last term”.
- L290, “15-day average”: Is it running average?
- L330: The small mesoscale observations include the large and small mesoscale components. And only the small mesoscale component should be input as the observation data into the cost function, that is, the large mesoscale component should be removed before using it in the analysis. It may be done by reducing the observed value by multiplying a coefficient less than one, or more practically by inflating the observation errors. Because of this reason, I think gk should be more than 1 even for day 0. In addition, the values of observation errors before the inflation are not shown for the small mesoscale.
- L340: How many grid points are required in the wavelength in order to represent the wave is a kind of controversy topics. Maybe some people think two points are enough and others insist more than three points are required. It should be nice if authors provide some reference that endorses their three-point criterion.
- L351: What “optimization of the MSDA implementation” indicates is not clear.
- L416: A period is missed at the end of the line.
- L430, “the MSDA-SWOT shows a significant improvement in SSH prediction.”: This cannot be affirmed. I suggest inserting “likely” between “MSDA-SWOT” and “shows”.
- L431, “The RMSD with the 2-day forecast”: Considering the typical time scale of the oceanic variability, I feel 2-day may be too short. Authors should compare the scores with the persistence forecasts. I also think it is very natural that this short forecast result is better than the free run result.
- L440: Considering the slow variations of oceanic fields (especially SSH fields), we cannot say that the along-track SSH data are completely independent from the 2-day forecast. The representativeness errors in SSH observations are generally correlated with the same data observed 2-days before, and I think it cannot be ignored if we consider the long time-scale of the oceanic variations.
- L515, “sporous”: this must be a misspelling of “spurious”.
- L544, “only in a pure linear system can observations be assimilated fully, reduce the error in the analysis”: Although I am not sure if I understand this sentence correctly, I think this sentence means that, only in a pure linear system, observations can be assimilated fully and can reduce the error in analysis. I agree that observation can be assimilated fully only in a pure system, but I think the analysis error can be reduced even if the system is nonlinear.
- L546: I agree that the reduction of the analysis errors does not ensure the reduction of forecast errors. But in the case of the predictions in this study, I think it is very natural that smaller analysis errors induce smaller forecast errors because the typical time scale of ocean variations are longer than the lead time of the forecasts.
- L549, “With a small analysis increment, the model performs more linearly in the time evolution related to the increment.”: I am skeptical on this discussion. I think the total of the large and small mesoscale and submesoscale increments is as large as the increments of conventional DA systems. And I think the small forecast errors are just caused by the small analysis errors.
Citation: https://doi.org/10.5194/os-2021-89-RC1 -
RC2: 'Comment on os-2021-89', Anonymous Referee #2, 03 Dec 2021
Review of "Formulation and demonstration of an extended-3DVAR multi-scale data assimilation system for the SWOT altimetry era" by Li et al
Summary
This paper claims to introduce a DA system suitable for assimilating wide swath altimetry data (SWOT). I found this a disappointing paper. First, the paper misleads the reader by giving the impression that assimilation of SWOT data will be explored or tested. In fact this is not the case and the discussion is insufficent to convince the reader that this system will be capable of assimilating SWOT. In particular, a system should be capable of assimilating data with the particular error characteristics of SWOT data. Particularly of concern are correlated errors in the swaths due to phase, roll and timing errors. Second, the experiments performed are unilluminating. The paper is really about demonstrating extended time window 3DVAR. This needs to be compared to a control experiment doing non-extended 3DVAR otherwise the results are of little interest. At the moment I cannot tell if the extended 3DVAR makes things better than your non-extended 3DVAR.
The paper, in my view, goes into too much detail in aspects that were not novel and glosses over aspects that were potentially novel. An example of this is spending a lot of time on the characteristics of the existing observing network for example SST and at the same failing to go into sufficient detail on how the "time sampling" error is determined. Similarly the discussion of the results is superficial.
I recommend rejecting this paper for the above reasons. I give some more detailed comments below.
More detailed comments
The title is misleading it is not convincing that this system addresses the assimilation of SWOT data. Just removing "for the SWOT altimetry era" from the title would avoid misleading the reader. By "extended" you mean "extended time window" it would be clearer to write that.
In the abstract you write your system "is first to assimilate routinely available observations" is misleading. Is the first for your system? It is certainly not the first in world to do this.Line 93: Error in the citation Li et al (2019). There is a Li et al (2019a) and Li et al (2019b) which one is it?
It is highly misleading to call the system MSDA-SWOT when you don't appear to have tested it with SWOT data. I feel strongly that it is not acceptable to do this.Figure 1b with the Google Earth background is very unclear.
Section 2.3 First sentence. Use of gridded and along track altimeter would mean you are assimilating the same data twice because the along track data is used to generate the gridded data. This needs clarification here.
Figure 2. explain in the caption what the ring of 6 black circles at about 36 N 125 W
Line 148. Gridded products will also distort the positions of eddies. They are produced using along track altimeter data. Indeed they may be more likely to do so since they do not use a model to evolve the eddy positions from a previous analysis.
Figure 4. A plot of the observations is really unnecessary here. The reader will be thinking “when am I going to see something about SWOT data”? The answer of course is never.
Section 2.4 line 178. You should write what you do in this paper that is not done in Li 2019b. It is unclear at this point whether the data assimilated or just used for validation in the work presented here and is it IR SST and/or Microwave SST (line 180).
Section 3. lines 191-95. It is not clear why a 1 km or finer resolution is necessary to assimilate SWOT. I suppose you might say a km scale model is needed to take full advantage of SWOT but it is not ncessary to assimilate SWOT data. Anyway your model does not meet the claimed requirement having a resolution of ~3 km.
Section 3. lines 212-215. and Section 3.1.2. It is a good idea to account for the sampling time error but note this is not a new approach (see below). It should be noted first that errors associated with this are not random or Gaussian since they are associated with not accounting for model evolution. Second the impact of this will be to reduce the weight given to observations far from the analysis time which may consequently have only a small impact on the analysis. Assimilating such data if not carefully done may degrade the analysis in some cases.
Section 3.1.2. The sampling time error is far from a new type of observation error - Oke et al (2005) and Martin et al (2007) both increase the observation error with based on the difference of observation and analysis time. Additionally, since this is one of key novel aspects in your system this section requires much more detail. Questions not answered are what is the form of the time sampling error how does it vary with the analysis observation time difference? How are any parameters determined?
Section 4.1. Second sentence. Delete "Here we give a description". Not necessary since you are giving a description in the next sentence.
Section 4.2. There seems to be a fundamental misunderstanding in what the background error is. It does not relate to the observation scales instead it is the scales in the model forecast (the background) error. I would concede this is affected by previous observations assimilated in previous assimilation cycles, but not in the straightforward way you claim here. It is not clear what you mean by in lines 311-313. This should be clarified and perhaps explained with a specific example.
Section 4.2 line 325. It would help the reader to describe in words what the terms in equation 7 mean. What is the justification for (7) particularly including the measurement/instrumental error in the calculation of the sampling time error? This error is due to not including the model time evolution in the calculation of H(x) and this should not be affected by the instrumental error. I suppose there is a possible argument for time sampling errors being associated with the representativeness error since this relates to how well the model equivalent represents the observations. This may be higher where the model is more variable and therefore where bigger errors would be due to assuming the model is the constant in the cost function. This whole discussion could do with more detail since this is what makes this work different to previous work you have done. Describing exactly how lambda_k is determined would be interesting.
The paragraph lines 336-342 is very unclear. I think my problem relates to my fundamental issue with this whole section. I do not see that you can just specify the background error correlation scales based on baseline requirements from SWOT it should related to what the actual background error is in your system.
Table 2 is confusing. You list observations, but then it appears that many of the types are not assimilated in the experiments here. In that case they should not be a table with the caption "observations assimilated".
Table 2. It does not seem good practice to assimilate maps created from along track SSH and then assimilate the along track SSH data. You are in affect assimilating the same thing twice since the along track SSH data was used to generate the map. You should at least highlight potential issues with this approach not least overfitting to the observations.
Section 5. To convincingly demonstrate the utility of extended-3DVar but you need to run a control experiment with non-extended-3DVar. It would also be useful to have another experiment to assess the impact of the optimisations you mention versus a control run.
Section 5.1. Assessing the assimilation analysis against observations it has assimilated to produce the analysis is not very useful. The results may look good but you may in fact just be overfitting to the observations and their associated errors.
Figure 6. Describe the significance of the SWOT baseline line compared to the nadir altimeters observations. It has more power at larger wavenumbers/ smaller wavelengths. Is this signal or noise? Is there a significance where the lines cross over? Why is the power so much less at smaller wavenumbers/longer wavelengths?
Figure 7. Anomaly correlations are much more interesting that correlations this largely just tells me that the mean fields match. This needs another experiment to compare to. Explain why the RMSE grows with time. At the moment I might think that you started your DA experiment with an analysis from your old system and made the results worse with the extended-3DVar. Perhaps seasonal effects are causing the increase in error with time but I cannot tell from the results you show.
Line 408 A change from 2.6 cm to 2.8 cm does not seem substantial to me.
Line 436. Saying the domain average RMSD is as large as 10.0 cm is imprecise language just say what it is averaged over the comparison time period.
Line 440. "As much as 14%" (is it 14% or not?)
Figure 8. Again what is the little black ring of circles in each plot. It should be stated in the caption in each figure it appears in.
Figure 9. I really think you need to give anomaly correlations for SST since it is quite easy to match the climatology and achieve a very high correlation. The (non-anomaly) correlations will be very high even for a non-assimilating run. This plot also needs another experiment to compare to. Again the error grows with time so with no other experiment the reader may conclude that your changes may be making the results worse.
Figure 10. Not really discussed in detail. What is the significance of the shape of the histogram, for example? It again could really do with another experiment to compare the results with. Also the anomaly correlation should be used as the correlation is not very useful to SST as I explained previously.
Section 5.4 I think this idea of the "campaign area circulation" is a potentially interesting one. But the exploration of it here is superficial. I think this would be a good place to add more figures. It would be interesting to show an example of this and how your work has reduced such errors. It seems a bit trivial to say if you assimilate other observation types and keep your analysis close to the truth then the increments from campaign observations will be smaller. An illustration of this with results (perhaps showing surface currents) from a run where you exclude another data type so that analysis is not as close along with your campaign data would be useful to see.
Lines 475 I’m not really sure the comparison to no data assimilation is particularly interesting it is unlikely that someone would assimilate campaign data and fail to assimilate other observational data.
Line 492 Briefly explain here again what the DA Cal experiment is. What does “Cal” stand for?
Figure 11. You are comparing against observations you assimilate I’m not sure this plot is particularly useful since it is quite easy for a DA system to fit data it is assimilating. It certainly doesn’t illuminate on the “campaign area circulation” idea.
Figure 12. This figure is a mess of overlapping lines there doesn’t appear to be any consistent differences between the experiment results. The discussion of this figure does not help in at all to explain what the reader is supposed to conclude from this figure.
References
Oke, P.R., Schiller, A., Griffin, D.A. and Brassington, G.B. (2005), Ensemble data assimilation for an eddy-resolving ocean model of the Australian region. Q.J.R. Meteorol. Soc., 131: 3301-3311. https://doi.org/10.1256/qj.05.95
Martin, M.J., Hines, A. and Bell, M.J. (2007), Data assimilation in the FOAM operational short-range ocean forecasting system: a description of the scheme and its impact. Q.J.R. Meteorol. Soc., 133: 981-995. https://doi.org/10.1002/qj.74
Citation: https://doi.org/10.5194/os-2021-89-RC2 -
EC1: 'Comment on os-2021-89: Manuscript Rejected', Bernadette Sloyan, 24 Feb 2022
This manuscript has been rejected for publication as the authors have advised that they are unable to provide a reply to the reviewer's concerns and a revised manuscript.
I sincerely thank the reviewers for their time to provide a thorough review the manuscript.
Citation: https://doi.org/10.5194/os-2021-89-EC1
Data sets
Reanalysis and sensitivity experiments using a multi-scale data assimilation system during a field campaign in the California current system Zhijin Li, Matthew Archer, Jinbo Wang, Lee-Lueng Fu https://doi.org/10.5281/zenodo.4602095
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 762 | 293 | 40 | 1,095 | 37 | 34 |
- HTML: 762
- PDF: 293
- XML: 40
- Total: 1,095
- BibTeX: 37
- EndNote: 34
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Cited
2 citations as recorded by crossref.
- 4DVarNet-SSH: end-to-end learning of variational interpolation schemes for nadir and wide-swath satellite altimetry M. Beauchamp et al. 10.5194/gmd-16-2119-2023
- Reconstructing Fine‐Scale Ocean Variability via Data Assimilation of the SWOT Pre‐Launch In Situ Observing System M. Archer et al. 10.1029/2021JC017362