the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 06 Oct 2022
Preconditioning of block onset in the Southern Hemisphere: a perspective from static stability
Abstract. The horizontal and temporal variations of static stability prior to blocking onset are characterized through composite analysis of twenty blocking events in the Southern Hemisphere. It is found that, along with a low potential vorticity (PV) anomaly formation, a local minimum of static stability in the upper troposphere and on the tropopause is achieved over the block-onset region when blocking onset takes place. By partitioning the isentropic PV into the absolute vorticity and static stability contributions, it is found that they account for roughly 70 % and 30 % of low-PV anomaly formation over the block-onset region, respectively. A static stability budget analysis revealed that the decrease of static stability in the upper troposphere and on the tropopuase prior to blocking onset is attributable to horizontal advection of low static stability from subtropics to midlatitude as well as the stretching effect associated with upper-level convergence over the block-onset region, with the horizontal advection forcing being the primary contributor. On the other hand, the vertical advection of static stability tends to oppose the decreasing static stability through advecting more stable air downward such that it stabilizes the local air over the block-onset region. Furthermore, the direct effect of diabatic heating is negligible as its magnitude is generally an order of magnitude smaller than other effects in the static stability budget. Nevertheless, the indirect effect of diabatic heating, manifested as the advection of low static stability by diabatically forced upper-tropospheric outflow, greatly favors blocking onsets by destabilizing the air upstream block-onset region.
- Preprint
(7728 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2022-1038', Anonymous Referee #1, 26 Oct 2022
This work studies the static stability evolutions during blocking onset.
Reanalysis data is used to study 20 selected Southern Hemispheric blocking events during 1986-2008.
The work finds 30% of the anticyclonic PV anomaly attributed to weak static stability. This weak static stability is primarily attributed to horizontal advection and vertical stretching.The perspective from static stability is new and interesting. However, I have more than major reservations as follows. I suggest giving authors plenty of time (preferably no deadline) to undergo more-than-major revision.
Major comments
1. Figure 14: The static stability tendency budget is not closed. The difference is large and systematic. This cannot be explained by finite differencing (line 301). This cannot be compared with the difference seen in Teubler and Riemer (2016, their Fig. 6), which is much smaller and not systematic. This unclosed budget largely damages my confidence in the results.
Advection by rotational wind is suspicious because 32S 135W in Fig 12b shows positive tendency, but northerly should bring low static stability at that latitude (no matter JJA or DJF, Fig 7).
2. Different roles of static stability (low PV, high Eady growth rate) are confusingly presented (or not clearly distinguished). This makes the main finding unclear. Below is my understanding:
[Low PV] A detection criteria of blocking is low PV at the center of anticyclone. Low static stability at the center of anticyclone will help a system detected as blocking. This point is supported by Figure 11 and others.
[High Eady growth rate] Low static stability upstream can give high Eady growth rate and favors baroclinic eddies that maintain the blocking (line 358). This point is not supported by any figures. In fact, Figure 9 goes against this conjecture by showing high static stability upstream. I suggest largely cutting mentions of this conjecture (Lines 24-39, 321-322, 331, 351-365) and clearly saying that Figure 9 goes against this. Please also revise the title (avoid the word “preconditioning”), and rephrase line 14 and 348 (at least remove the word “upstream”), in order not to confuse with the unsupported/rejected conjecture.
3. My challenge to the low PV idea concerns the relevance of extreme weather conditions. It seems to me that blocking leads to extreme weather conditions (line 17) because of its wind anomaly, not static stability anomaly. In this sense, static stability will be relevant only if there is conversion to/from wind anomaly (or absolute vorticity). Is there such conversion (stretching term)? Or static stability and absolute vorticity are both doing their own thing without interaction?
4. Overall, description of results is not so balanced, not so scientific, and not so insightful. Some examples below:
Line 316-317: “Fig. 12(f)… positive values poleward side.” It might be unfair to highlight these positive values, which are much much weaker than the negative values equatorward.
Line 210: “lower left”->”southwest”? Also line 268 and 270.
Line 161: “became a cut-off low IPV anomaly on the following day.” The cut-off low anomaly might be referring to 50S 155W on 23 July (Fig 2d). Not sure if aforementioned understanding is right but this cut-off low measures less than 10 degrees in diameter and lasts only one day. Also, “cut-off” in anomaly field is not quite noteworthy (compared to cut-off in absolute field). Pointing to these fine details might not provide much insights.
5. Please focus on the role of static stability in giving low PV, by removing off-focus discussions. Examples below:
Equations 1-3: I don’t see the need to introduce sigma. You can directly introduce \partial \theta/\partial p, and use that in place of sigma.
Many figures: Rather than outlining the block-onset region as the wind reversal region, please try to highlight the low-PV region (e.g., where you detect PV anomaly).
6. A few previous papers are misinterpreted.
Line 123: Pelly and Hoskins (2003) were based on reversal of absolute field, not anomaly-based. I suggest removing the citation here.
Line 63-66: “The injection of diabatically processed anticyclonic PV is usually interpreted as the direct effect of latent heat release…” I suspect this is a misinterpretation of previous studies, at least of Teubler and Riemer (2016).
7. Figure 12: After closing the budget (comment 1), if panels b,d,e continue to be highly (anti-)correlated, please do a bit more discussion. I think horizontal convergence (panel e) correlates with sinking (panel d) because 300 mb is slightly below tropopause, so air is squeezed downward when it converges. Sinking correlates with equatorward motion (panel b) because air tend to move along isentropic surface, which is tilted in such way.
Having the dominate terms counteracting with each other does complicate the picture. Would it be better to use isentropic coordinate? - Also because it is low static stability on isentropic coordinate that helps low isentropic PV.
8. Fig 8f and 10c: Why zero lines differ in the two figures? Because of pressure coordinate vs. isentropic coordinate? Perhaps both should use isentropic coordinate (320K). Also Fig 9f vs 11c.
9. Figure 10abd: Why as low as -120%? Does the sign change? If the sign changes, it is perhaps infinitely more important than stability changes.
10. Figure 10abc: At 30S 160W, both panel b and c shows >40%, why panel a is <80%, not >96% (1.4*1.4=1.96)? Is mean(IPV) not equal to mean(vort)*mean(sta)? If numbers are confirmed to be correct, please add an explanation in caption.
11. Line 240: What does “long-term mean” mean? Since there is a great seasonal cycle (e.g., in static stability, Fig 7), would be good to use one season.
12. Table 1: Most case is in JJA or May or September, except case 16 is in March. I suggest removing case 16.
Minor comments
13. Line 18: For blocking and extreme weather, it might help to cite Kautz et al. [doi:10.5194/wcd-3-305-2022], a review article at WCD.
14. Line 58: Hauser et al. [doi:10.5194/wcd-2022-44] also confirmed the importance of divergent outflow aloft. It might help to cite that as well. Please also comment (e.g., on line 343, 349) whether their study agree or disagree with yours.
15. Line 66-72: The mentions of WCB, PRE and ET do not tie well to the paper. They only connect to latent heat release, but not blocking. Perhaps simply remove them all.
16. Line 130: “screened these blocking cases against PV-anomaly based blocking criteria.” Please give more details how that is done.
17. Table 1: If composite is done by overlapping the blocking centers (line 176), then table 1 should list the center, rather than the west boundary.
18. Line 166: The definition of block-onset region should be moved earlier, because it is already used in Figure 1. You may also include the definition in figure caption. (Also see comment 5 for suggested modification on definition.)
19. Figure 2 caption: Is contour interval 0.5 PVU instead of 1.0? Probably you don’t need to say it in caption. Though you do need to mention the unit of PVU.
20. Line 168: Instead of “blocking center”, please say 60S.
21. Figure 4: Please add tick labels for x-axis to show the scale.
22. Line 185: “originated from subtropics” - How do you see this?
23. Figure 6: Please modify the color scale so that white is 0.
24. Figure 12b: Vectors at high latitude are not parallel to contour. Maybe this is a map projection issue (one degree longitude measures different length at different latitude). Please either fix it, or add an explanation in caption.
25. Figure 12d: Upward motion not shown? Please mention in caption.
26. Line 353,354: “Rossy”->”Rossby”, “amply”->”amplify”? (Actually, I suggest removing the paragraph in comment 2.)
Citation: https://doi.org/10.5194/egusphere-2022-1038-RC1 -
CC1: 'Reply on RC1', Li Dong, 29 Oct 2022
We greatly appreciate your constructive comments. We are working on revising this manuscript now following your suggestions and will get back to you as soon as we have addressed all the major issues raised by the reviewer.
Citation: https://doi.org/10.5194/egusphere-2022-1038-CC1 -
CC2: 'Reply on RC1', Li Dong, 17 Nov 2022
This work studies the static stability evolutions during blocking onset.
Reanalysis data is used to study 20 selected Southern Hemispheric blocking events during 1986-2008.
The work finds 30% of the anticyclonic PV anomaly attributed to weak static stability. This weak static stability is primarily attributed to horizontal advection and vertical stretching.The perspective from static stability is new and interesting. However, I have more than major reservations as follows. I suggest giving authors plenty of time (preferably no deadline) to undergo more-than-major revision.
Major comments
- Figure 14: The static stability tendency budget is not closed. The difference is large and systematic. This cannot be explained by finite differencing (line 301). This cannot be compared with the difference seen in Teubler and Riemer (2016, their Fig. 6), which is much smaller and not systematic. This unclosed budget largely damages my confidence in the results.
Thanks for bringing this up. Our Fig. 14 is for the composite result of 20 blocking cases at daily interval whereas the Fig. 6 of Teubler and Riemer (2016) is based on one single case at 6 hourly interval. In order to make the comparison more consistent, we plotted the analyzed and computed static stability tendency comparison figure for each blocking case at 3 hourly interval. We attached these figures here as the PDF file. In these figures, the analyzed static stability tendency is computed with central difference such that the tendency is the change of static stability over 6 hours. For each day, there are 8 time steps (00,03,06,09,12,15,18,21) as shown in the tick of the time axis. The attached figures for each blocking case demonstrate that the agreement between the analyzed and computed static stability tendency is fairly good, in general. In fact, we want to thank the reviewer for raising this important question as we realize that we need to update our Fig. 14 by using the selected single case figure instead of the composite figure since those averaging processes (case compositing as well as daily averaging) in the composite figure tend to degrade the agreement of analyzed and computed static stability tendency time series. We will make this revision accordingly in our revised manuscript.
Advection by rotational wind is suspicious because 32S 135W in Fig 12b shows positive tendency, but northerly should bring low static stability at that latitude (no matter JJA or DJF, Fig 7).
It is correct that in the climatological sense the northerly wind should bring low static stability at that latitude based on the climatological static stability pattern as shown in Fig.7. In addition, the climatological static stability does not change much along the longitude direction while it greatly changes along latitude. Nevertheless, in Fig. 12b, a zonal gradient of static stability exists at location of (135W, 32S) such that the rotational wind over that region, which is primarily westerly and northwesterly winds, tends to advect relatively high values of static stability easterward. Hence it leads to a positive advection by rotational wind there.
- Different roles of static stability (low PV, high Eady growth rate) are confusingly presented (or not clearly distinguished). This makes the main finding unclear. Below is my understanding:
[Low PV] A detection criteria of blocking is low PV at the center of anticyclone. Low static stability at the center of anticyclone will help a system detected as blocking. This point is supported by Figure 11 and others.
[High Eady growth rate] Low static stability upstream can give high Eady growth rate and favors baroclinic eddies that maintain the blocking (line 358). This point is not supported by any figures. In fact, Figure 9 goes against this conjecture by showing high static stability upstream. I suggest largely cutting mentions of this conjecture (Lines 24-39, 321-322, 331, 351-365) and clearly saying that Figure 9 goes against this. Please also revise the title (avoid the word “preconditioning”), and rephrase line 14 and 348 (at least remove the word “upstream”), in order not to confuse with the unsupported/rejected conjecture.
Thanks to the reviewer for raising this important point. We realize that we should have been more careful with emphasizing the specific stage of blocking when we summarized our findings and surmised our conjecture. In fact, for the finding that reads “static stability reached its local minimum over the block-onset region on the block-onset day”, this is primarily focused on the pre-blocking period, i.e. five days prior to block onset. This implies that during the pre-blocking stage the low-static-stability anomaly immediate upstream of block-onset region serves as a sort of wave generator which triggers block onset a few days later. Here “upstream” means the low-static-stability anomaly is upstream of the block-onset region before block onset takes place. Hence for our conjecture that reads “over upstream block-onset region, the static stability field should be relatively low such that the wave maker generates baroclinic eddies more efficiently to maintain the blocking structure”, it is supposed to refer to the pre-blocking stage instead of the blocking maintenance stage. Meanwhile we do realize that we have used “to maintain the blocking structure” in our conjecture, which is not correct indeed. So thanks to the reviewer for pointing this out. We should have made our conjecture consistent with the findings of our study. Therefore we plan to clarify our conjecture in the revised manuscript and rephrase it as “Prior to blocking onset, over upstream block-onset region the static stability field should be relatively low such that the wave maker generates more baroclinic eddies seeding to enter the block-onset region hereby initiating block onset.”
As for the reviewer’s suggestion upon avoid using “preconditioning” in the title, we agree on that as it appears somehow inaccurate. Now we tentatively modify the title as “Static Stability Variability and its Relation to Southern Hemisphere Blocking Onsets”.
- My challenge to the low PV idea concerns the relevance of extreme weather conditions. It seems to me that blocking leads to extreme weather conditions (line 17) because of its wind anomaly, not static stability anomaly. In this sense, static stability will be relevant only if there is conversion to/from wind anomaly (or absolute vorticity). Is there such conversion (stretching term)? Or static stability and absolute vorticity are both doing their own thing without interaction?
This is an interesting question and thanks for bringing it up. Blocking leads to extreme weather conditions such as heat waves (due to extremely high temperature), cold-air outbreak (due to extremely low temperature), flooding (due to extreme precipitation) and so on. Even though these extreme conditions are not explicitly represented by the static stability anomaly, this static stability anomaly is tightly integrated into the blocking onset procedure through both thermodynamic and dynamic processes. For instance, heat waves are commonly associated with extremely high temperature as well as low wind (partly due to adiabatic warming accompanied with strong sinking motion). The static stability anomaly could contribute to these temperature and wind anomalies through static stability advection or stretching processes, as the reviewer suggested. In addition, the static stability and absolute vorticity do interact with each other, even though the former is a thermodynamical indicator and the latter a dynamical indicator. For instance, the stretching term in the static stability tendency equation links the static stability and absolute vorticity perfectly in that both thermal stratification and convergence field (which is closely linked to absolute vorticity) work together to give rise to conditions favorable to block onset.
- Overall, description of results is not so balanced, not so scientific, and not so insightful. Some examples below:
Line 316-317: “Fig. 12(f)… positive values poleward side.” It might be unfair to highlight these positive values, which are much much weaker than the negative values equatorward.
Yes, we totally agree that these positive values are much weaker than the negative values on the equatorward side. But since we intended to focus on the block-onset region, we felt obliged to fully describe the 2-D distribution of static stability tendency attributable to the direct effect of diabatic heating over the block-onset region. In addition, the point that we attempted to deliver from Fig. 12(f) is that these positive values are much smaller than the negative values over the block-onset region in Fig. 12 (c), i.e. the indirect effect of the diabatic heating outweighs the direct effect significantly.
Line 210: “lower left”->”southwest”? Also line 268 and 270.
Thanks for pointing this out. We agree that we should have used more scientific terms in above circumstances. We have made the modifications following the reviewer’s suggestions.
Line 161: “became a cut-off low IPV anomaly on the following day.” The cut-off low anomaly might be referring to 50S 155W on 23 July (Fig 2d). Not sure if aforementioned understanding is right but this cut-off low measures less than 10 degrees in diameter and lasts only one day. Also, “cut-off” in anomaly field is not quite noteworthy (compared to cut-off in absolute field). Pointing to these fine details might not provide much insights.
Thanks for raising these concerns. We agree that the cut-off low PV anomaly is not significant in terms of both its size and magnitude. We have removed descriptions related to “cut-off” anomaly in the revised manuscript.
- Please focus on the role of static stability in giving low PV, by removing off-focus discussions. Examples below:
Equations 1-3: I don’t see the need to introduce sigma. You can directly introduce \partial \theta/\partial p, and use that in place of sigma.
The reason that we induced sigma, as shown in Eqs.1-3, is that we want to particularly refer to the static stability parameter in the QG height tendency equation as described in the textbook by Bluestein (1992). This static stability parameter has dual effects in the QG height tendency equation. Plus, Smith and Tsou (1988) used the generalized height tendency with the exact form of this static stability parameter to discuss the variations of static stability associated with cyclogenesis. As our work is closely related to the work of Smith and Tsou (1988), we intended to keep the form of the static stability parameter intact for easier comparison.
Many figures: Rather than outlining the block-onset region as the wind reversal region, please try to highlight the low-PV region (e.g., where you detect PV anomaly).
This is a great suggestion. In fact we had also noticed that there are several blocking cases in which we detected the block-onset region based on the geopotential height blocking index but found out that the defined block-onset region did not exactly overlap with the low-PV center. In the revised manuscript, we will update the figures by outlining block-onset region with the low-PV standard.
- A few previous papers are misinterpreted.
Line 123: Pelly and Hoskins (2003) were based on reversal of absolute field, not anomaly-based. I suggest removing the citation here.
Thanks for pointing this out. We will remove Pelly and Hoskins (2003) on Line 123 accordingly.
Line 63-66: “The injection of diabatically processed anticyclonic PV is usually interpreted as the direct effect of latent heat release…” I suspect this is a misinterpretation of previous studies, at least of Teubler and Riemer (2016).
Thanks for pointing this out. We will remove Teubler and Riemer (2016) on Line 63-66.
- Figure 12: After closing the budget (comment 1), if panels b,d,e continue to be highly (anti-)correlated, please do a bit more discussion. I think horizontal convergence (panel e) correlates with sinking (panel d) because 300 mb is slightly below tropopause, so air is squeezed downward when it converges. Sinking correlates with equatorward motion (panel b) because air tend to move along isentropic surface, which is tilted in such way.
This is great suggestion. We will provide more discussion regarding the connection among variables in Fig. 12. The reviewer is right about the sinking motion due to the convergence nearby tropopause, as shown in omega being greater than 4 Pa/s (in contour, positive means sinking motion) in Fig. 12(d). And this sinking motion is closely related to the southwesterly wind which moves toward equatorward since when air moves toward equatorward it would descent along the isentropic surface (the isentropic surface is tilted from Equator to Pole). By adding this discussion, it would make the whole picture more clear and complete. So we thank the reviewer for this great suggestion.
Having the dominate terms counteracting with each other does complicate the picture. Would it be better to use isentropic coordinate? - Also because it is low static stability on isentropic coordinate that helps low isentropic PV.
This is another great suggestion too. We checked one blocking case by converting the static stability budget figure (similar to Fig. 12) from 300mb to 320K and found that the general pattern is quite similar. We are processing composite Fig. 12 now on the 320K isentropic surface and will post the updated figure soon.
- Fig 8f and 10c: Why zero lines differ in the two figures? Because of pressure coordinate vs. isentropic coordinate? Perhaps both should use isentropic coordinate (320K). Also Fig 9f vs 11c.
Yes, the zero lines from Fig. 8f and Fig. 10c slightly differ in terms of their locations. The reviewer is right that this discrepancy is resulted from the different coordinates used in the figures as Fig. 8f is on 300mb isobaric surface while Fig. 10c is on 320K isentropic surface. At this moment we are updating Fig. 8 and Fig. 9 by converting these static stability anomaly from 300mb to 320K isentropic surface. We will post the updated figures soon.
- Figure 10abd: Why as low as -120%? Does the sign change? If the sign changes, it is perhaps infinitely more important than stability changes.
When the absolute vorticity is small, a moderate change of absolute vorticity can lead to a relative change over 100%. Hence the relative change of -120% occurs because the initial value of absolute vorticity is small and also this quantity changes sign after a moderate change. Regarding the change of sign, it means the relative vorticity changes from a small positive number (i.e. climatological value) to a negative number (anticyclonic vorticity). Overall, the focus of Fig. 10 is to demonstrate what Eq.(10) suggests, i.e. the sum of the relative change of static stability and absolute vorticity equals to the relative change of IPV. Thus Fig. 10 serves the main purpose.
- Figure 10abc: At 30S 160W, both panel b and c shows >40%, why panel a is <80%, not >96% (1.4*1.4=1.96)? Is mean(IPV) not equal to mean(vort)*mean(sta)? If numbers are confirmed to be correct, please add an explanation in caption.
The mismatch between 80% and 96% is primarily due to two reasons. One is that Eq. (10) is expressed in partial differential format which requires the time step to be as small as possible when using finite difference to approximate this partial difference. Here we used the 3-hourly reanalysis data to compute Eq. (10) and presented the daily averaged result in Fig. 10. We believe that the discrepancy between the two fields would decrease if we could further reduce the time interval when calculating the relative changes. Another reason is that the reanalysis data does contain observation errors which may contribute to this mismatch as well. Overall, based on Fig. 10(a) and Fig. 10(d), we feel that these two fields generally have a reasonable agreement. We will add some necessary explanation in the caption of this figure in the revised manuscript.
- Line 240: What does “long-term mean” mean? Since there is a great seasonal cycle (e.g., in static stability, Fig 7), would be good to use one season.
“Long-term mean” here refers to the climatological value of a particular month. For instance, for the July 1999 blocking case (block onset on July 25), the long-term mean static stability refers to the climatological static stability value for the month of July. Hence, the seasonal cycle has been removed indeed.
- Table 1: Most case is in JJA or May or September, except case 16 is in March. I suggest removing case 16.
Yes, we agree that case 16 (03/17/2003 blocking case) can be removed such that the remaining blocking cases took place around Southern Hemisphere winter season.
Minor comments
- Line 18: For blocking and extreme weather, it might help to cite Kautz et al. [doi:10.5194/wcd-3-305-2022], a review article at WCD.
Thanks for this suggestion. We checked this review article and found that it does have certain connection to our study. We will cite this work in our revised manuscript.
- Line 58: Hauser et al. [doi:10.5194/wcd-2022-44] also confirmed the importance of divergent outflow aloft. It might help to cite that as well. Please also comment (e.g., on line 343, 349) whether their study agree or disagree with yours.
Thanks for suggesting this article. We checked this paper and found it quite interesting. In fact it is tightly linked to our work here. This article confirmed that the moist process is important during blocking onset stage, which is consistent with our findings here. But it also revealed that the indirect effect of moist processes, i.e. the PV advection by the divergent flow, does not appear prominent from the Eulerian low-frequency perspective because the Eulerian approach only captures the local evolution of blocking onset whereas the upstream moist process is omitted from the Eulerian perspective. This is really an interesting finding to us. In our study, we did use the Eulerian perspective (but not the low-frequency perspective), and we found that the indirect effect of moist process is more significant than the direct effect, nevertheless the indirect effect is a secondary contributor to block onset compared to the dominant advection term. In this sense, our work agrees with Hauser et al (2022) ’s results. In the revised manuscript, we will add discussion on how our findings are linked to Hauser et al (2022).
- Line 66-72: The mentions of WCB, PRE and ET do not tie well to the paper. They only connect to latent heat release, but not blocking. Perhaps simply remove them all.
We agree that the material related to WCB, PRE and ET kind of distracts the readers from the main topic on blocking. We will remove them in the revised manuscript.
- Line 130: “screened these blocking cases against PV-anomaly based blocking criteria.” Please give more details how that is done.
Thanks for this suggestion. The PV-anomaly based blocking criteria refers to that the 320K low-IPV anomaly reaches at least 1 PVU and persists for at least 5 days. We used this criteria to double check blocking cases that are first detected with geopotential height reversal criteria. We will add these details in the revised manuscript.
- Table 1: If composite is done by overlapping the blocking centers (line 176), then table 1 should list the center, rather than the west boundary.
Yes, we agree that in Table 1 the block-onset regions can be represented by the center of the specific region instead of the west boundary. We will make the correction in the revised manuscript.
- Line 166: The definition of block-onset region should be moved earlier, because it is already used in Figure 1. You may also include the definition in figure caption. (Also see comment 5 for suggested modification on definition.)
Yes, this is a great suggestion. We will make the modifications in the revised manuscript as the reviewer suggested.
- Figure 2 caption: Is contour interval 0.5 PVU instead of 1.0? Probably you don’t need to say it in caption. Though you do need to mention the unit of PVU.
Thanks for pointing this out. The contour interval should be 0.5 PVU and we are sorry for making this typo in the caption. We will correct it in the revised manuscript.
- Line 168: Instead of “blocking center”, please say 60S.
Yes, we will make this correction in the revised manuscript.
- Figure 4: Please add tick labels for x-axis to show the scale.
Yes, we will make this correction in the revised manuscript.
- Line 185: “originated from subtropics” - How do you see this?
Thanks for pointing this out. As the reviewer suggested, this description is not accurate so we will remove this from the revised manuscript.
- Figure 6: Please modify the color scale so that white is 0.
Yes, we will make this correction in the revised manuscript.
- Figure 12b: Vectors at high latitude are not parallel to contour. Maybe this is a map projection issue (one degree longitude measures different length at different latitude). Please either fix it, or add an explanation in caption.
Thanks for bringing this up. In fact, it is primarily due to the compositing effect. We have checked the corresponding plot for single blocking case and confirmed that the rotational wind vectors are mostly parallel to the geopotential height contours even at high latitudes. Note that in Figure 12(b), the 300-mb rotational wind vector is overlaid with 500-mb geopotential height field instead of 300-mb geopotential height field.
- Figure 12d: Upward motion not shown? Please mention in caption.
Thanks for pointing this out. In Fig. 12(d), to highlight the sinking motion associated with convergence region, we only plotted the positive vertical velocity (omega) with magnitude greater than 2 Pa/s and omitted the rising motion. We will add these explanations in caption of the revised manuscript.
- Line 353,354: “Rossy”->”Rossby”, “amply”->”amplify”? (Actually, I suggest removing the paragraph in comment 2.)
We are sorry for these typos. We will follow the reviewer’s suggestion by removing that paragraph.
- AC2: 'Reply on RC1', Hairu Ding, 31 Jan 2023
-
CC1: 'Reply on RC1', Li Dong, 29 Oct 2022
-
RC2: 'Comment on egusphere-2022-1038', Anonymous Referee #2, 23 Nov 2022
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2022/egusphere-2022-1038/egusphere-2022-1038-RC2-supplement.pdf
-
CC3: 'Reply on RC2', Li Dong, 23 Nov 2022
We greatly appreciate this reviewer's constructive comments. We will get back to this reviewer with our detailed response in a couple of days. Please stay tuned. Thanks.
Citation: https://doi.org/10.5194/egusphere-2022-1038-CC3 -
AC1: 'Reply on RC2', Hairu Ding, 31 Jan 2023
Please find our reply below and in the supplement.
Major comments:
1) A major feature of this work is the treatment of vorticity and static stability as separate and mutually exclusive quantities. However, in reality this may not be the case. For example, the authors argue that the static stability can be created or destroyed by vertical stretching advection etc, however this is also true for vorticity, indicating that both static stability and vorticity will change simultaneously. For example, in a discussion around the evolution of static stability (Fig 8 and 9), the authors make the statement that the static stability evolves similarly to the IPV. Is this because the IPV and static stability cannot be decoupled? The authors should do more to convince the reader that of being able to treat static stability as a distinct quantity and that “the investigation of the local change of static stability would shed light on understanding the dynamic evolution of blocking events”.
Thank you for bringing up this point. The authors agree with the reviewer that vorticity and static stability are not fully independent quantities. They do interact with each other, even though the former is a thermodynamic quantity whereas the latter a dynamic quantity. For instance, the stretching term in the static stability tendency equation links the static stability and vorticity perfectly in that the stretching effect not only influences the atmospheric thermostratification but also influences vorticity via convergence/divergence changes. However, it should be clarified that the analysis done in the present study about the relative contributions of vorticity and static stability to the relative change of IPV does not require the two quantities to be “separate and mutually exclusive”. Vorticity and static stability do link to each other but they cannot solely determine one another. They are two important and different quantities in atmospheric dynamics with distinctly physical meanings. While the IPV change has traditionally been attributed to the change in vorticity, the present study clearly demonstrates the non-negligible contribution of static stability to IPV changes. The relevant statement has been modified in the revised manuscript to avoid possible confusion.
2) A concern of this work is the lack of closure of the static stability budget (e.g. Figure 10, 11 and 14). The authors need to do some more work to convince the reader that this does not affect their work or their results.
Thanks for bringing this up. Our Fig. 14 is for the composite result of 20 blocking cases at daily interval, which may not satisfactorily demonstrate the closure of the static stability budget, as the reviewer had concerned. Hence, in the revised manuscript we plotted the analyzed and computed static stability tendency comparison figure for each blocking case at 3 hourly interval, instead. We attached these figures here as the PDF file. In these figures, the analyzed static stability tendency is computed with central difference such that the tendency is the change of static stability over 6 hours. For each day, there are 8 time steps (00,03,06,09,12,15,18,21) as shown in the tick of the time axis. The attached figures demonstrate that the agreement between the analyzed and computed static stability tendency for each blocking case is fairly good, in general. In fact, we want to thank the reviewer for raising this important question as we realize that we need to update our Fig. 14 by using the selected single case figure instead of the composite figure since those averaging processes (case compositing as well as daily averaging) in the composite figure tend to degrade the agreement of analyzed and computed static stability tendency time series. We have made this revision accordingly in our revised manuscript.
3) There is generally a lack of dynamical reasoning or explanation in this work. The authors generally describe the figures and offer little dynamical explanation of the processes. For example: Analysis of the vertical structure of static stability (Fig. 13) is done without analysis of the vertical differences in static stability and what it means for blocking.
Thank you for pointing this out. Regarding the vertical structure of static stability of Figure 13, we do realize that some important dynamical explanations for this figure are missing. We have revised this part accordingly in the modified manuscript.
4) This work is structured by analysing a composite and a case study similarly. The work could be streamlined such that arguments are not made twice where there is generally consistency between the case study and climatological composites. This can get quite confusing for the reader. I would suggest that the structure of the main body of the work be comprised of the climatological composite and a short section at the end of the analysis with the case study (in less detail) be used to corroborate the climatology (or visa versa).
Thank you for bringing this up. We do realize that it may appear redundant to both present case study and climatological composite results with same details in the current form of the manuscript. We decided to take this reviewer’s suggestions by not making the arguments twice when a general consistency exists between case study and climatological composite results. In the revised manuscript, we present the case study first followed by a short section with the climatological composite result in less detail.
5) Use of language is very colloquial at times. Phrases such as “upper-left corner”, “a few days” and an anomaly that “shows up” are frequently used. The authors should read through the work carefully and correct their use of language and general grammar throughout the manuscript. Please check plurals and add “a” or “the” in places where required (eg. Line 18 “outbreak” -> outbreaks, Line 47 “… formation of low-PV anomaly …” -> … formation of a low-PV anomaly, Line 203: “… intersects stratosphere is characterized with strong meridional gradient of static stability.” -> …intersects the stratosphere is characterized by a strong meridional gradient of static stability
Thank you for your thorough proofreading and much appreciate your suggestions. In the modified manuscript, we have implemented all these suggestions.
Specific comments:
- Line 17-18: Some citations of blocking in cold air outbreaks, heatwaves and drought required.
Thanks for this suggestion. Relevant citations have been added in the modified manuscript.
- Line 131-136: “20 typical cases” – why are these 20 cases chosen? What makes them typical? Why not choose all cases detected by the algorithms employed?
Since we have used two criteria for blocking detection in this study, i.e. first by the geopotential height based criteria, and next by the PV anomaly criteria, it ended up with these 20 blocking cases that meet both criteria.
- Line 159-167: The descriptions of the various features being pointed out can be hard to follow. Consider improving this synoptic discussion and adding labels onto the relevant figures of the various important features.
Thanks for bringing this up. Another reviewer has also brought up similar comments on this paragraph. In the revised manuscript, we have rewritten this part by making it more clear and concise.
- Figure 3 and relevant discussion: It is unclear what features the authors are trying to show with the vertical cross-section that cannot be readily seen in the 2D plots (Figure 2). Would a longitudinal cross-section not make more sense so that the north-south dipole can be seen?
Figure 3 is a cross section of the IPV field for the July 1999 blocking case. The intention of this figure is to clearly demonstrate the elevated tropopause right above the blocking onset region. In addition, it is also capable of depicting the undulated tropopuase above high and low centers in the upper troposphere. Furthermore, it displays the vertical depth of the high and low PV anomaly structures within the entire troposphere and lower stratosphere.
- Figure 4: I recommend that the axes of these plots be changed such that they are “relative latitude and longitude” with the centre (0,0)
Thanks for pointing this out. In fact, originally we did use (0,0) to depict the center of the pseudo longitude/latitude. Later on, we changed the center of pseudo longitude to 180 as we desired to emphasize that most of these blocking cases in the SH take place over the Southeast Pacific. As for the pseudo latitude, since all blocking cases are composited between the 80S and 20S latitude band (i.e. their true latitude band), the pseudo latitude is indeed the true latitude in this case.
- Line 185: “… originating from the subtropics, gradually penetrating poleward …” – I do not see the process that is described unfolding in Figure 5. In Figure 5, I see the low-PV “trough” extending equatorward from the poles and growing in amplitude over Days -5 to 0. Please explain this contradiction.
Sorry for causing this confusion. We have modified the corresponding paragraph and made the descriptions consistent now.
- Figure 6: See comment on Figure 4.
Please refer to our response above regarding Figure 3.
- Line 199-200: “it is evident that the subtropics are dominated by tropospheric air associated with relatively low static stability while the high latitudes are dominated by stratospheric air characterized by large static stability” – this certainly looks true for DJF but looks more complicated for JJA. For example in JJA over the Australian region there seems to be a local maxima in static stability close to the subtropics and a minimum near the poles. Please clarify and explain this contradiction.
Thanks for bringing this up. In the revised manuscript, we added brief explanations of the local static stability maxima near Australia as well as the minimum near the poles.
- Line 209: “long-term mean” – annual, seasonal, daily mean? Please clarify.
“Long-term mean” here refers to the climatological value of a particular month. For instance, for the July 1999 blocking case (block onset on July 25), the longterm mean static stability refers to the climatological static stability value for the month of July. Hence, the seasonal cycle has been removed.
- Figure 8+9: I wonder if all the panels and the very detailed description of each panel (Line 210-225) is absolutely necessary? Consider a generalised discussion and skipping some panels.
Thanks for bringing this up. Since Figure 8 and 9 will be served to compare to Figure 10 and 11, respectively, we decided to keep Figure 8 and 9. Nevertheless, we do find the descriptions of Figure 8 and 9 are somehow too much detailed and redundant so we cut it short by following the reviewer’s suggestion in the revised manuscript.
- Figure 12 and Lines 249-257: The authors make the case for the changes in in local static stability in terms of Equation 6. Changes in vorticity (when analysing a vorticity budget equation) are governed by similar processes such as horizontal and vertical advection, stretching, tilting etc. Do the changes in vorticity by these processes correspond to the changes in static stability? Are they coupled. Such an analysis may help motivate the authors arguments for analysing changes in static stability.
These are great suggestions! Thank for bringing them up. Since we are mainly focused on the static stability budget in the present study, we did not compute the vorticity budget. Nevertheless, we plan to do it as a part of the future work as it would help determine to what extent vorticity and static stability are coupled.
- Line 259-261: day-5 “features the most representative patterns” – why are they representative and what do they represent? Do you mean they are closest to the results in Tables 2 and 3? How much do the other days results differ as they there seems to be a large difference in the results as we move from day-5 to day 0 in Table 2 and 3).
“day-5” is regarded as the most representative day because the spatial patterns displayed in Figure 12 are most prominent for day-5, even though other days are accompanied with similar patterns but with smaller magnitude. The Table 2 and 3 only provide a rough and quick estimation by taking the area-average such that the detailed spatial patterns are not fully reflected by the area-averaged values.
- Figure 13: Difficult to see the relative contributions easily since the x-axes difference between panels. I suggest merging all panels onto a single graph with same x-axes limits.
Thanks for bringing this up. In fact, we did plot them all into one single figure in the first place, but found out that the magnitudes of each contribution span over a great range such that some of the smaller contributing terms did not really show up (near zero). In order to distinguish both primary and secondary contributing terms, we decided to display each term as a single panel.
- Line 296: “The findings are once again confirmed”. What findings? There seems to be very little analysis of Figure 6 and the differences in the various terms in the vertical. Please discuss.
Thanks for bringing this up. We agree that more dynamical descriptions for Figure 13 are needed. We have rewritten this paragraph in the revised manuscript.
- Figure 14: Labeling (1,2,…6) on plot should match the time lags (-5,-4,….0).
Thanks for pointing this out. We have made the correction in the attached figure. Please refer to the attached figure details.
- Figure 15 and Lines 305-318: In previous sections and figures, diabatic heating is shown to have a small role in the local changes in static stability. It remains unclear as to what this analysis is trying to elucidate. Please clarify why this analysis is important and the points you are trying to make here.
Thanks for pointing this out. We have made the modification in this paragraph by emphasizing the role of upstream cyclogenesis-induced latent heat release in favoring blocking onset. By following the reviewer’s suggestions, we have made our explanations more clear and concise.
-
CC3: 'Reply on RC2', Li Dong, 23 Nov 2022
Status: closed
-
RC1: 'Comment on egusphere-2022-1038', Anonymous Referee #1, 26 Oct 2022
This work studies the static stability evolutions during blocking onset.
Reanalysis data is used to study 20 selected Southern Hemispheric blocking events during 1986-2008.
The work finds 30% of the anticyclonic PV anomaly attributed to weak static stability. This weak static stability is primarily attributed to horizontal advection and vertical stretching.The perspective from static stability is new and interesting. However, I have more than major reservations as follows. I suggest giving authors plenty of time (preferably no deadline) to undergo more-than-major revision.
Major comments
1. Figure 14: The static stability tendency budget is not closed. The difference is large and systematic. This cannot be explained by finite differencing (line 301). This cannot be compared with the difference seen in Teubler and Riemer (2016, their Fig. 6), which is much smaller and not systematic. This unclosed budget largely damages my confidence in the results.
Advection by rotational wind is suspicious because 32S 135W in Fig 12b shows positive tendency, but northerly should bring low static stability at that latitude (no matter JJA or DJF, Fig 7).
2. Different roles of static stability (low PV, high Eady growth rate) are confusingly presented (or not clearly distinguished). This makes the main finding unclear. Below is my understanding:
[Low PV] A detection criteria of blocking is low PV at the center of anticyclone. Low static stability at the center of anticyclone will help a system detected as blocking. This point is supported by Figure 11 and others.
[High Eady growth rate] Low static stability upstream can give high Eady growth rate and favors baroclinic eddies that maintain the blocking (line 358). This point is not supported by any figures. In fact, Figure 9 goes against this conjecture by showing high static stability upstream. I suggest largely cutting mentions of this conjecture (Lines 24-39, 321-322, 331, 351-365) and clearly saying that Figure 9 goes against this. Please also revise the title (avoid the word “preconditioning”), and rephrase line 14 and 348 (at least remove the word “upstream”), in order not to confuse with the unsupported/rejected conjecture.
3. My challenge to the low PV idea concerns the relevance of extreme weather conditions. It seems to me that blocking leads to extreme weather conditions (line 17) because of its wind anomaly, not static stability anomaly. In this sense, static stability will be relevant only if there is conversion to/from wind anomaly (or absolute vorticity). Is there such conversion (stretching term)? Or static stability and absolute vorticity are both doing their own thing without interaction?
4. Overall, description of results is not so balanced, not so scientific, and not so insightful. Some examples below:
Line 316-317: “Fig. 12(f)… positive values poleward side.” It might be unfair to highlight these positive values, which are much much weaker than the negative values equatorward.
Line 210: “lower left”->”southwest”? Also line 268 and 270.
Line 161: “became a cut-off low IPV anomaly on the following day.” The cut-off low anomaly might be referring to 50S 155W on 23 July (Fig 2d). Not sure if aforementioned understanding is right but this cut-off low measures less than 10 degrees in diameter and lasts only one day. Also, “cut-off” in anomaly field is not quite noteworthy (compared to cut-off in absolute field). Pointing to these fine details might not provide much insights.
5. Please focus on the role of static stability in giving low PV, by removing off-focus discussions. Examples below:
Equations 1-3: I don’t see the need to introduce sigma. You can directly introduce \partial \theta/\partial p, and use that in place of sigma.
Many figures: Rather than outlining the block-onset region as the wind reversal region, please try to highlight the low-PV region (e.g., where you detect PV anomaly).
6. A few previous papers are misinterpreted.
Line 123: Pelly and Hoskins (2003) were based on reversal of absolute field, not anomaly-based. I suggest removing the citation here.
Line 63-66: “The injection of diabatically processed anticyclonic PV is usually interpreted as the direct effect of latent heat release…” I suspect this is a misinterpretation of previous studies, at least of Teubler and Riemer (2016).
7. Figure 12: After closing the budget (comment 1), if panels b,d,e continue to be highly (anti-)correlated, please do a bit more discussion. I think horizontal convergence (panel e) correlates with sinking (panel d) because 300 mb is slightly below tropopause, so air is squeezed downward when it converges. Sinking correlates with equatorward motion (panel b) because air tend to move along isentropic surface, which is tilted in such way.
Having the dominate terms counteracting with each other does complicate the picture. Would it be better to use isentropic coordinate? - Also because it is low static stability on isentropic coordinate that helps low isentropic PV.
8. Fig 8f and 10c: Why zero lines differ in the two figures? Because of pressure coordinate vs. isentropic coordinate? Perhaps both should use isentropic coordinate (320K). Also Fig 9f vs 11c.
9. Figure 10abd: Why as low as -120%? Does the sign change? If the sign changes, it is perhaps infinitely more important than stability changes.
10. Figure 10abc: At 30S 160W, both panel b and c shows >40%, why panel a is <80%, not >96% (1.4*1.4=1.96)? Is mean(IPV) not equal to mean(vort)*mean(sta)? If numbers are confirmed to be correct, please add an explanation in caption.
11. Line 240: What does “long-term mean” mean? Since there is a great seasonal cycle (e.g., in static stability, Fig 7), would be good to use one season.
12. Table 1: Most case is in JJA or May or September, except case 16 is in March. I suggest removing case 16.
Minor comments
13. Line 18: For blocking and extreme weather, it might help to cite Kautz et al. [doi:10.5194/wcd-3-305-2022], a review article at WCD.
14. Line 58: Hauser et al. [doi:10.5194/wcd-2022-44] also confirmed the importance of divergent outflow aloft. It might help to cite that as well. Please also comment (e.g., on line 343, 349) whether their study agree or disagree with yours.
15. Line 66-72: The mentions of WCB, PRE and ET do not tie well to the paper. They only connect to latent heat release, but not blocking. Perhaps simply remove them all.
16. Line 130: “screened these blocking cases against PV-anomaly based blocking criteria.” Please give more details how that is done.
17. Table 1: If composite is done by overlapping the blocking centers (line 176), then table 1 should list the center, rather than the west boundary.
18. Line 166: The definition of block-onset region should be moved earlier, because it is already used in Figure 1. You may also include the definition in figure caption. (Also see comment 5 for suggested modification on definition.)
19. Figure 2 caption: Is contour interval 0.5 PVU instead of 1.0? Probably you don’t need to say it in caption. Though you do need to mention the unit of PVU.
20. Line 168: Instead of “blocking center”, please say 60S.
21. Figure 4: Please add tick labels for x-axis to show the scale.
22. Line 185: “originated from subtropics” - How do you see this?
23. Figure 6: Please modify the color scale so that white is 0.
24. Figure 12b: Vectors at high latitude are not parallel to contour. Maybe this is a map projection issue (one degree longitude measures different length at different latitude). Please either fix it, or add an explanation in caption.
25. Figure 12d: Upward motion not shown? Please mention in caption.
26. Line 353,354: “Rossy”->”Rossby”, “amply”->”amplify”? (Actually, I suggest removing the paragraph in comment 2.)
Citation: https://doi.org/10.5194/egusphere-2022-1038-RC1 -
CC1: 'Reply on RC1', Li Dong, 29 Oct 2022
We greatly appreciate your constructive comments. We are working on revising this manuscript now following your suggestions and will get back to you as soon as we have addressed all the major issues raised by the reviewer.
Citation: https://doi.org/10.5194/egusphere-2022-1038-CC1 -
CC2: 'Reply on RC1', Li Dong, 17 Nov 2022
This work studies the static stability evolutions during blocking onset.
Reanalysis data is used to study 20 selected Southern Hemispheric blocking events during 1986-2008.
The work finds 30% of the anticyclonic PV anomaly attributed to weak static stability. This weak static stability is primarily attributed to horizontal advection and vertical stretching.The perspective from static stability is new and interesting. However, I have more than major reservations as follows. I suggest giving authors plenty of time (preferably no deadline) to undergo more-than-major revision.
Major comments
- Figure 14: The static stability tendency budget is not closed. The difference is large and systematic. This cannot be explained by finite differencing (line 301). This cannot be compared with the difference seen in Teubler and Riemer (2016, their Fig. 6), which is much smaller and not systematic. This unclosed budget largely damages my confidence in the results.
Thanks for bringing this up. Our Fig. 14 is for the composite result of 20 blocking cases at daily interval whereas the Fig. 6 of Teubler and Riemer (2016) is based on one single case at 6 hourly interval. In order to make the comparison more consistent, we plotted the analyzed and computed static stability tendency comparison figure for each blocking case at 3 hourly interval. We attached these figures here as the PDF file. In these figures, the analyzed static stability tendency is computed with central difference such that the tendency is the change of static stability over 6 hours. For each day, there are 8 time steps (00,03,06,09,12,15,18,21) as shown in the tick of the time axis. The attached figures for each blocking case demonstrate that the agreement between the analyzed and computed static stability tendency is fairly good, in general. In fact, we want to thank the reviewer for raising this important question as we realize that we need to update our Fig. 14 by using the selected single case figure instead of the composite figure since those averaging processes (case compositing as well as daily averaging) in the composite figure tend to degrade the agreement of analyzed and computed static stability tendency time series. We will make this revision accordingly in our revised manuscript.
Advection by rotational wind is suspicious because 32S 135W in Fig 12b shows positive tendency, but northerly should bring low static stability at that latitude (no matter JJA or DJF, Fig 7).
It is correct that in the climatological sense the northerly wind should bring low static stability at that latitude based on the climatological static stability pattern as shown in Fig.7. In addition, the climatological static stability does not change much along the longitude direction while it greatly changes along latitude. Nevertheless, in Fig. 12b, a zonal gradient of static stability exists at location of (135W, 32S) such that the rotational wind over that region, which is primarily westerly and northwesterly winds, tends to advect relatively high values of static stability easterward. Hence it leads to a positive advection by rotational wind there.
- Different roles of static stability (low PV, high Eady growth rate) are confusingly presented (or not clearly distinguished). This makes the main finding unclear. Below is my understanding:
[Low PV] A detection criteria of blocking is low PV at the center of anticyclone. Low static stability at the center of anticyclone will help a system detected as blocking. This point is supported by Figure 11 and others.
[High Eady growth rate] Low static stability upstream can give high Eady growth rate and favors baroclinic eddies that maintain the blocking (line 358). This point is not supported by any figures. In fact, Figure 9 goes against this conjecture by showing high static stability upstream. I suggest largely cutting mentions of this conjecture (Lines 24-39, 321-322, 331, 351-365) and clearly saying that Figure 9 goes against this. Please also revise the title (avoid the word “preconditioning”), and rephrase line 14 and 348 (at least remove the word “upstream”), in order not to confuse with the unsupported/rejected conjecture.
Thanks to the reviewer for raising this important point. We realize that we should have been more careful with emphasizing the specific stage of blocking when we summarized our findings and surmised our conjecture. In fact, for the finding that reads “static stability reached its local minimum over the block-onset region on the block-onset day”, this is primarily focused on the pre-blocking period, i.e. five days prior to block onset. This implies that during the pre-blocking stage the low-static-stability anomaly immediate upstream of block-onset region serves as a sort of wave generator which triggers block onset a few days later. Here “upstream” means the low-static-stability anomaly is upstream of the block-onset region before block onset takes place. Hence for our conjecture that reads “over upstream block-onset region, the static stability field should be relatively low such that the wave maker generates baroclinic eddies more efficiently to maintain the blocking structure”, it is supposed to refer to the pre-blocking stage instead of the blocking maintenance stage. Meanwhile we do realize that we have used “to maintain the blocking structure” in our conjecture, which is not correct indeed. So thanks to the reviewer for pointing this out. We should have made our conjecture consistent with the findings of our study. Therefore we plan to clarify our conjecture in the revised manuscript and rephrase it as “Prior to blocking onset, over upstream block-onset region the static stability field should be relatively low such that the wave maker generates more baroclinic eddies seeding to enter the block-onset region hereby initiating block onset.”
As for the reviewer’s suggestion upon avoid using “preconditioning” in the title, we agree on that as it appears somehow inaccurate. Now we tentatively modify the title as “Static Stability Variability and its Relation to Southern Hemisphere Blocking Onsets”.
- My challenge to the low PV idea concerns the relevance of extreme weather conditions. It seems to me that blocking leads to extreme weather conditions (line 17) because of its wind anomaly, not static stability anomaly. In this sense, static stability will be relevant only if there is conversion to/from wind anomaly (or absolute vorticity). Is there such conversion (stretching term)? Or static stability and absolute vorticity are both doing their own thing without interaction?
This is an interesting question and thanks for bringing it up. Blocking leads to extreme weather conditions such as heat waves (due to extremely high temperature), cold-air outbreak (due to extremely low temperature), flooding (due to extreme precipitation) and so on. Even though these extreme conditions are not explicitly represented by the static stability anomaly, this static stability anomaly is tightly integrated into the blocking onset procedure through both thermodynamic and dynamic processes. For instance, heat waves are commonly associated with extremely high temperature as well as low wind (partly due to adiabatic warming accompanied with strong sinking motion). The static stability anomaly could contribute to these temperature and wind anomalies through static stability advection or stretching processes, as the reviewer suggested. In addition, the static stability and absolute vorticity do interact with each other, even though the former is a thermodynamical indicator and the latter a dynamical indicator. For instance, the stretching term in the static stability tendency equation links the static stability and absolute vorticity perfectly in that both thermal stratification and convergence field (which is closely linked to absolute vorticity) work together to give rise to conditions favorable to block onset.
- Overall, description of results is not so balanced, not so scientific, and not so insightful. Some examples below:
Line 316-317: “Fig. 12(f)… positive values poleward side.” It might be unfair to highlight these positive values, which are much much weaker than the negative values equatorward.
Yes, we totally agree that these positive values are much weaker than the negative values on the equatorward side. But since we intended to focus on the block-onset region, we felt obliged to fully describe the 2-D distribution of static stability tendency attributable to the direct effect of diabatic heating over the block-onset region. In addition, the point that we attempted to deliver from Fig. 12(f) is that these positive values are much smaller than the negative values over the block-onset region in Fig. 12 (c), i.e. the indirect effect of the diabatic heating outweighs the direct effect significantly.
Line 210: “lower left”->”southwest”? Also line 268 and 270.
Thanks for pointing this out. We agree that we should have used more scientific terms in above circumstances. We have made the modifications following the reviewer’s suggestions.
Line 161: “became a cut-off low IPV anomaly on the following day.” The cut-off low anomaly might be referring to 50S 155W on 23 July (Fig 2d). Not sure if aforementioned understanding is right but this cut-off low measures less than 10 degrees in diameter and lasts only one day. Also, “cut-off” in anomaly field is not quite noteworthy (compared to cut-off in absolute field). Pointing to these fine details might not provide much insights.
Thanks for raising these concerns. We agree that the cut-off low PV anomaly is not significant in terms of both its size and magnitude. We have removed descriptions related to “cut-off” anomaly in the revised manuscript.
- Please focus on the role of static stability in giving low PV, by removing off-focus discussions. Examples below:
Equations 1-3: I don’t see the need to introduce sigma. You can directly introduce \partial \theta/\partial p, and use that in place of sigma.
The reason that we induced sigma, as shown in Eqs.1-3, is that we want to particularly refer to the static stability parameter in the QG height tendency equation as described in the textbook by Bluestein (1992). This static stability parameter has dual effects in the QG height tendency equation. Plus, Smith and Tsou (1988) used the generalized height tendency with the exact form of this static stability parameter to discuss the variations of static stability associated with cyclogenesis. As our work is closely related to the work of Smith and Tsou (1988), we intended to keep the form of the static stability parameter intact for easier comparison.
Many figures: Rather than outlining the block-onset region as the wind reversal region, please try to highlight the low-PV region (e.g., where you detect PV anomaly).
This is a great suggestion. In fact we had also noticed that there are several blocking cases in which we detected the block-onset region based on the geopotential height blocking index but found out that the defined block-onset region did not exactly overlap with the low-PV center. In the revised manuscript, we will update the figures by outlining block-onset region with the low-PV standard.
- A few previous papers are misinterpreted.
Line 123: Pelly and Hoskins (2003) were based on reversal of absolute field, not anomaly-based. I suggest removing the citation here.
Thanks for pointing this out. We will remove Pelly and Hoskins (2003) on Line 123 accordingly.
Line 63-66: “The injection of diabatically processed anticyclonic PV is usually interpreted as the direct effect of latent heat release…” I suspect this is a misinterpretation of previous studies, at least of Teubler and Riemer (2016).
Thanks for pointing this out. We will remove Teubler and Riemer (2016) on Line 63-66.
- Figure 12: After closing the budget (comment 1), if panels b,d,e continue to be highly (anti-)correlated, please do a bit more discussion. I think horizontal convergence (panel e) correlates with sinking (panel d) because 300 mb is slightly below tropopause, so air is squeezed downward when it converges. Sinking correlates with equatorward motion (panel b) because air tend to move along isentropic surface, which is tilted in such way.
This is great suggestion. We will provide more discussion regarding the connection among variables in Fig. 12. The reviewer is right about the sinking motion due to the convergence nearby tropopause, as shown in omega being greater than 4 Pa/s (in contour, positive means sinking motion) in Fig. 12(d). And this sinking motion is closely related to the southwesterly wind which moves toward equatorward since when air moves toward equatorward it would descent along the isentropic surface (the isentropic surface is tilted from Equator to Pole). By adding this discussion, it would make the whole picture more clear and complete. So we thank the reviewer for this great suggestion.
Having the dominate terms counteracting with each other does complicate the picture. Would it be better to use isentropic coordinate? - Also because it is low static stability on isentropic coordinate that helps low isentropic PV.
This is another great suggestion too. We checked one blocking case by converting the static stability budget figure (similar to Fig. 12) from 300mb to 320K and found that the general pattern is quite similar. We are processing composite Fig. 12 now on the 320K isentropic surface and will post the updated figure soon.
- Fig 8f and 10c: Why zero lines differ in the two figures? Because of pressure coordinate vs. isentropic coordinate? Perhaps both should use isentropic coordinate (320K). Also Fig 9f vs 11c.
Yes, the zero lines from Fig. 8f and Fig. 10c slightly differ in terms of their locations. The reviewer is right that this discrepancy is resulted from the different coordinates used in the figures as Fig. 8f is on 300mb isobaric surface while Fig. 10c is on 320K isentropic surface. At this moment we are updating Fig. 8 and Fig. 9 by converting these static stability anomaly from 300mb to 320K isentropic surface. We will post the updated figures soon.
- Figure 10abd: Why as low as -120%? Does the sign change? If the sign changes, it is perhaps infinitely more important than stability changes.
When the absolute vorticity is small, a moderate change of absolute vorticity can lead to a relative change over 100%. Hence the relative change of -120% occurs because the initial value of absolute vorticity is small and also this quantity changes sign after a moderate change. Regarding the change of sign, it means the relative vorticity changes from a small positive number (i.e. climatological value) to a negative number (anticyclonic vorticity). Overall, the focus of Fig. 10 is to demonstrate what Eq.(10) suggests, i.e. the sum of the relative change of static stability and absolute vorticity equals to the relative change of IPV. Thus Fig. 10 serves the main purpose.
- Figure 10abc: At 30S 160W, both panel b and c shows >40%, why panel a is <80%, not >96% (1.4*1.4=1.96)? Is mean(IPV) not equal to mean(vort)*mean(sta)? If numbers are confirmed to be correct, please add an explanation in caption.
The mismatch between 80% and 96% is primarily due to two reasons. One is that Eq. (10) is expressed in partial differential format which requires the time step to be as small as possible when using finite difference to approximate this partial difference. Here we used the 3-hourly reanalysis data to compute Eq. (10) and presented the daily averaged result in Fig. 10. We believe that the discrepancy between the two fields would decrease if we could further reduce the time interval when calculating the relative changes. Another reason is that the reanalysis data does contain observation errors which may contribute to this mismatch as well. Overall, based on Fig. 10(a) and Fig. 10(d), we feel that these two fields generally have a reasonable agreement. We will add some necessary explanation in the caption of this figure in the revised manuscript.
- Line 240: What does “long-term mean” mean? Since there is a great seasonal cycle (e.g., in static stability, Fig 7), would be good to use one season.
“Long-term mean” here refers to the climatological value of a particular month. For instance, for the July 1999 blocking case (block onset on July 25), the long-term mean static stability refers to the climatological static stability value for the month of July. Hence, the seasonal cycle has been removed indeed.
- Table 1: Most case is in JJA or May or September, except case 16 is in March. I suggest removing case 16.
Yes, we agree that case 16 (03/17/2003 blocking case) can be removed such that the remaining blocking cases took place around Southern Hemisphere winter season.
Minor comments
- Line 18: For blocking and extreme weather, it might help to cite Kautz et al. [doi:10.5194/wcd-3-305-2022], a review article at WCD.
Thanks for this suggestion. We checked this review article and found that it does have certain connection to our study. We will cite this work in our revised manuscript.
- Line 58: Hauser et al. [doi:10.5194/wcd-2022-44] also confirmed the importance of divergent outflow aloft. It might help to cite that as well. Please also comment (e.g., on line 343, 349) whether their study agree or disagree with yours.
Thanks for suggesting this article. We checked this paper and found it quite interesting. In fact it is tightly linked to our work here. This article confirmed that the moist process is important during blocking onset stage, which is consistent with our findings here. But it also revealed that the indirect effect of moist processes, i.e. the PV advection by the divergent flow, does not appear prominent from the Eulerian low-frequency perspective because the Eulerian approach only captures the local evolution of blocking onset whereas the upstream moist process is omitted from the Eulerian perspective. This is really an interesting finding to us. In our study, we did use the Eulerian perspective (but not the low-frequency perspective), and we found that the indirect effect of moist process is more significant than the direct effect, nevertheless the indirect effect is a secondary contributor to block onset compared to the dominant advection term. In this sense, our work agrees with Hauser et al (2022) ’s results. In the revised manuscript, we will add discussion on how our findings are linked to Hauser et al (2022).
- Line 66-72: The mentions of WCB, PRE and ET do not tie well to the paper. They only connect to latent heat release, but not blocking. Perhaps simply remove them all.
We agree that the material related to WCB, PRE and ET kind of distracts the readers from the main topic on blocking. We will remove them in the revised manuscript.
- Line 130: “screened these blocking cases against PV-anomaly based blocking criteria.” Please give more details how that is done.
Thanks for this suggestion. The PV-anomaly based blocking criteria refers to that the 320K low-IPV anomaly reaches at least 1 PVU and persists for at least 5 days. We used this criteria to double check blocking cases that are first detected with geopotential height reversal criteria. We will add these details in the revised manuscript.
- Table 1: If composite is done by overlapping the blocking centers (line 176), then table 1 should list the center, rather than the west boundary.
Yes, we agree that in Table 1 the block-onset regions can be represented by the center of the specific region instead of the west boundary. We will make the correction in the revised manuscript.
- Line 166: The definition of block-onset region should be moved earlier, because it is already used in Figure 1. You may also include the definition in figure caption. (Also see comment 5 for suggested modification on definition.)
Yes, this is a great suggestion. We will make the modifications in the revised manuscript as the reviewer suggested.
- Figure 2 caption: Is contour interval 0.5 PVU instead of 1.0? Probably you don’t need to say it in caption. Though you do need to mention the unit of PVU.
Thanks for pointing this out. The contour interval should be 0.5 PVU and we are sorry for making this typo in the caption. We will correct it in the revised manuscript.
- Line 168: Instead of “blocking center”, please say 60S.
Yes, we will make this correction in the revised manuscript.
- Figure 4: Please add tick labels for x-axis to show the scale.
Yes, we will make this correction in the revised manuscript.
- Line 185: “originated from subtropics” - How do you see this?
Thanks for pointing this out. As the reviewer suggested, this description is not accurate so we will remove this from the revised manuscript.
- Figure 6: Please modify the color scale so that white is 0.
Yes, we will make this correction in the revised manuscript.
- Figure 12b: Vectors at high latitude are not parallel to contour. Maybe this is a map projection issue (one degree longitude measures different length at different latitude). Please either fix it, or add an explanation in caption.
Thanks for bringing this up. In fact, it is primarily due to the compositing effect. We have checked the corresponding plot for single blocking case and confirmed that the rotational wind vectors are mostly parallel to the geopotential height contours even at high latitudes. Note that in Figure 12(b), the 300-mb rotational wind vector is overlaid with 500-mb geopotential height field instead of 300-mb geopotential height field.
- Figure 12d: Upward motion not shown? Please mention in caption.
Thanks for pointing this out. In Fig. 12(d), to highlight the sinking motion associated with convergence region, we only plotted the positive vertical velocity (omega) with magnitude greater than 2 Pa/s and omitted the rising motion. We will add these explanations in caption of the revised manuscript.
- Line 353,354: “Rossy”->”Rossby”, “amply”->”amplify”? (Actually, I suggest removing the paragraph in comment 2.)
We are sorry for these typos. We will follow the reviewer’s suggestion by removing that paragraph.
- AC2: 'Reply on RC1', Hairu Ding, 31 Jan 2023
-
CC1: 'Reply on RC1', Li Dong, 29 Oct 2022
-
RC2: 'Comment on egusphere-2022-1038', Anonymous Referee #2, 23 Nov 2022
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2022/egusphere-2022-1038/egusphere-2022-1038-RC2-supplement.pdf
-
CC3: 'Reply on RC2', Li Dong, 23 Nov 2022
We greatly appreciate this reviewer's constructive comments. We will get back to this reviewer with our detailed response in a couple of days. Please stay tuned. Thanks.
Citation: https://doi.org/10.5194/egusphere-2022-1038-CC3 -
AC1: 'Reply on RC2', Hairu Ding, 31 Jan 2023
Please find our reply below and in the supplement.
Major comments:
1) A major feature of this work is the treatment of vorticity and static stability as separate and mutually exclusive quantities. However, in reality this may not be the case. For example, the authors argue that the static stability can be created or destroyed by vertical stretching advection etc, however this is also true for vorticity, indicating that both static stability and vorticity will change simultaneously. For example, in a discussion around the evolution of static stability (Fig 8 and 9), the authors make the statement that the static stability evolves similarly to the IPV. Is this because the IPV and static stability cannot be decoupled? The authors should do more to convince the reader that of being able to treat static stability as a distinct quantity and that “the investigation of the local change of static stability would shed light on understanding the dynamic evolution of blocking events”.
Thank you for bringing up this point. The authors agree with the reviewer that vorticity and static stability are not fully independent quantities. They do interact with each other, even though the former is a thermodynamic quantity whereas the latter a dynamic quantity. For instance, the stretching term in the static stability tendency equation links the static stability and vorticity perfectly in that the stretching effect not only influences the atmospheric thermostratification but also influences vorticity via convergence/divergence changes. However, it should be clarified that the analysis done in the present study about the relative contributions of vorticity and static stability to the relative change of IPV does not require the two quantities to be “separate and mutually exclusive”. Vorticity and static stability do link to each other but they cannot solely determine one another. They are two important and different quantities in atmospheric dynamics with distinctly physical meanings. While the IPV change has traditionally been attributed to the change in vorticity, the present study clearly demonstrates the non-negligible contribution of static stability to IPV changes. The relevant statement has been modified in the revised manuscript to avoid possible confusion.
2) A concern of this work is the lack of closure of the static stability budget (e.g. Figure 10, 11 and 14). The authors need to do some more work to convince the reader that this does not affect their work or their results.
Thanks for bringing this up. Our Fig. 14 is for the composite result of 20 blocking cases at daily interval, which may not satisfactorily demonstrate the closure of the static stability budget, as the reviewer had concerned. Hence, in the revised manuscript we plotted the analyzed and computed static stability tendency comparison figure for each blocking case at 3 hourly interval, instead. We attached these figures here as the PDF file. In these figures, the analyzed static stability tendency is computed with central difference such that the tendency is the change of static stability over 6 hours. For each day, there are 8 time steps (00,03,06,09,12,15,18,21) as shown in the tick of the time axis. The attached figures demonstrate that the agreement between the analyzed and computed static stability tendency for each blocking case is fairly good, in general. In fact, we want to thank the reviewer for raising this important question as we realize that we need to update our Fig. 14 by using the selected single case figure instead of the composite figure since those averaging processes (case compositing as well as daily averaging) in the composite figure tend to degrade the agreement of analyzed and computed static stability tendency time series. We have made this revision accordingly in our revised manuscript.
3) There is generally a lack of dynamical reasoning or explanation in this work. The authors generally describe the figures and offer little dynamical explanation of the processes. For example: Analysis of the vertical structure of static stability (Fig. 13) is done without analysis of the vertical differences in static stability and what it means for blocking.
Thank you for pointing this out. Regarding the vertical structure of static stability of Figure 13, we do realize that some important dynamical explanations for this figure are missing. We have revised this part accordingly in the modified manuscript.
4) This work is structured by analysing a composite and a case study similarly. The work could be streamlined such that arguments are not made twice where there is generally consistency between the case study and climatological composites. This can get quite confusing for the reader. I would suggest that the structure of the main body of the work be comprised of the climatological composite and a short section at the end of the analysis with the case study (in less detail) be used to corroborate the climatology (or visa versa).
Thank you for bringing this up. We do realize that it may appear redundant to both present case study and climatological composite results with same details in the current form of the manuscript. We decided to take this reviewer’s suggestions by not making the arguments twice when a general consistency exists between case study and climatological composite results. In the revised manuscript, we present the case study first followed by a short section with the climatological composite result in less detail.
5) Use of language is very colloquial at times. Phrases such as “upper-left corner”, “a few days” and an anomaly that “shows up” are frequently used. The authors should read through the work carefully and correct their use of language and general grammar throughout the manuscript. Please check plurals and add “a” or “the” in places where required (eg. Line 18 “outbreak” -> outbreaks, Line 47 “… formation of low-PV anomaly …” -> … formation of a low-PV anomaly, Line 203: “… intersects stratosphere is characterized with strong meridional gradient of static stability.” -> …intersects the stratosphere is characterized by a strong meridional gradient of static stability
Thank you for your thorough proofreading and much appreciate your suggestions. In the modified manuscript, we have implemented all these suggestions.
Specific comments:
- Line 17-18: Some citations of blocking in cold air outbreaks, heatwaves and drought required.
Thanks for this suggestion. Relevant citations have been added in the modified manuscript.
- Line 131-136: “20 typical cases” – why are these 20 cases chosen? What makes them typical? Why not choose all cases detected by the algorithms employed?
Since we have used two criteria for blocking detection in this study, i.e. first by the geopotential height based criteria, and next by the PV anomaly criteria, it ended up with these 20 blocking cases that meet both criteria.
- Line 159-167: The descriptions of the various features being pointed out can be hard to follow. Consider improving this synoptic discussion and adding labels onto the relevant figures of the various important features.
Thanks for bringing this up. Another reviewer has also brought up similar comments on this paragraph. In the revised manuscript, we have rewritten this part by making it more clear and concise.
- Figure 3 and relevant discussion: It is unclear what features the authors are trying to show with the vertical cross-section that cannot be readily seen in the 2D plots (Figure 2). Would a longitudinal cross-section not make more sense so that the north-south dipole can be seen?
Figure 3 is a cross section of the IPV field for the July 1999 blocking case. The intention of this figure is to clearly demonstrate the elevated tropopause right above the blocking onset region. In addition, it is also capable of depicting the undulated tropopuase above high and low centers in the upper troposphere. Furthermore, it displays the vertical depth of the high and low PV anomaly structures within the entire troposphere and lower stratosphere.
- Figure 4: I recommend that the axes of these plots be changed such that they are “relative latitude and longitude” with the centre (0,0)
Thanks for pointing this out. In fact, originally we did use (0,0) to depict the center of the pseudo longitude/latitude. Later on, we changed the center of pseudo longitude to 180 as we desired to emphasize that most of these blocking cases in the SH take place over the Southeast Pacific. As for the pseudo latitude, since all blocking cases are composited between the 80S and 20S latitude band (i.e. their true latitude band), the pseudo latitude is indeed the true latitude in this case.
- Line 185: “… originating from the subtropics, gradually penetrating poleward …” – I do not see the process that is described unfolding in Figure 5. In Figure 5, I see the low-PV “trough” extending equatorward from the poles and growing in amplitude over Days -5 to 0. Please explain this contradiction.
Sorry for causing this confusion. We have modified the corresponding paragraph and made the descriptions consistent now.
- Figure 6: See comment on Figure 4.
Please refer to our response above regarding Figure 3.
- Line 199-200: “it is evident that the subtropics are dominated by tropospheric air associated with relatively low static stability while the high latitudes are dominated by stratospheric air characterized by large static stability” – this certainly looks true for DJF but looks more complicated for JJA. For example in JJA over the Australian region there seems to be a local maxima in static stability close to the subtropics and a minimum near the poles. Please clarify and explain this contradiction.
Thanks for bringing this up. In the revised manuscript, we added brief explanations of the local static stability maxima near Australia as well as the minimum near the poles.
- Line 209: “long-term mean” – annual, seasonal, daily mean? Please clarify.
“Long-term mean” here refers to the climatological value of a particular month. For instance, for the July 1999 blocking case (block onset on July 25), the longterm mean static stability refers to the climatological static stability value for the month of July. Hence, the seasonal cycle has been removed.
- Figure 8+9: I wonder if all the panels and the very detailed description of each panel (Line 210-225) is absolutely necessary? Consider a generalised discussion and skipping some panels.
Thanks for bringing this up. Since Figure 8 and 9 will be served to compare to Figure 10 and 11, respectively, we decided to keep Figure 8 and 9. Nevertheless, we do find the descriptions of Figure 8 and 9 are somehow too much detailed and redundant so we cut it short by following the reviewer’s suggestion in the revised manuscript.
- Figure 12 and Lines 249-257: The authors make the case for the changes in in local static stability in terms of Equation 6. Changes in vorticity (when analysing a vorticity budget equation) are governed by similar processes such as horizontal and vertical advection, stretching, tilting etc. Do the changes in vorticity by these processes correspond to the changes in static stability? Are they coupled. Such an analysis may help motivate the authors arguments for analysing changes in static stability.
These are great suggestions! Thank for bringing them up. Since we are mainly focused on the static stability budget in the present study, we did not compute the vorticity budget. Nevertheless, we plan to do it as a part of the future work as it would help determine to what extent vorticity and static stability are coupled.
- Line 259-261: day-5 “features the most representative patterns” – why are they representative and what do they represent? Do you mean they are closest to the results in Tables 2 and 3? How much do the other days results differ as they there seems to be a large difference in the results as we move from day-5 to day 0 in Table 2 and 3).
“day-5” is regarded as the most representative day because the spatial patterns displayed in Figure 12 are most prominent for day-5, even though other days are accompanied with similar patterns but with smaller magnitude. The Table 2 and 3 only provide a rough and quick estimation by taking the area-average such that the detailed spatial patterns are not fully reflected by the area-averaged values.
- Figure 13: Difficult to see the relative contributions easily since the x-axes difference between panels. I suggest merging all panels onto a single graph with same x-axes limits.
Thanks for bringing this up. In fact, we did plot them all into one single figure in the first place, but found out that the magnitudes of each contribution span over a great range such that some of the smaller contributing terms did not really show up (near zero). In order to distinguish both primary and secondary contributing terms, we decided to display each term as a single panel.
- Line 296: “The findings are once again confirmed”. What findings? There seems to be very little analysis of Figure 6 and the differences in the various terms in the vertical. Please discuss.
Thanks for bringing this up. We agree that more dynamical descriptions for Figure 13 are needed. We have rewritten this paragraph in the revised manuscript.
- Figure 14: Labeling (1,2,…6) on plot should match the time lags (-5,-4,….0).
Thanks for pointing this out. We have made the correction in the attached figure. Please refer to the attached figure details.
- Figure 15 and Lines 305-318: In previous sections and figures, diabatic heating is shown to have a small role in the local changes in static stability. It remains unclear as to what this analysis is trying to elucidate. Please clarify why this analysis is important and the points you are trying to make here.
Thanks for pointing this out. We have made the modification in this paragraph by emphasizing the role of upstream cyclogenesis-induced latent heat release in favoring blocking onset. By following the reviewer’s suggestions, we have made our explanations more clear and concise.
-
CC3: 'Reply on RC2', Li Dong, 23 Nov 2022
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 632 | 192 | 33 | 857 | 30 | 30 |
- HTML: 632
- PDF: 192
- XML: 33
- Total: 857
- BibTeX: 30
- EndNote: 30
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1