the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Mar 2024
Spatio-temporal filtering of jets obscures the reinforcement of baroclinicity by latent heating
Abstract. Latent heating modifies the jet stream by modifying the vertical geostrophic wind shear, thereby altering the potential for baroclinic development. Hence, correctly representing diabatic effects is important for modelling the mid-latitude atmospheric circulation and variability. Yet, the direct effects of diabatic heating remain poorly understood. For example, there is no consensus on the effect of latent heating on the cross-jet temperature contrast. We show that this disagreement is attributable to the choice of spatio-temporal filtering. Jet representations relying on filtered wind tend to have the strongest latent heating on the cold flank of the jet, thus weakening the cross-jet temperature contrast. In contrast, jet representations reflecting the two-dimensional instantaneous wind field have the strongest latent heating on the warm flank of the jet. Furthermore, we show that latent heating primarily occurs on the warm flank of poleward directed instantaneous jets, which is the case for all storm tracks and seasons.
- Preprint
(1998 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 01 May 2024)
-
RC1: 'Comment on egusphere-2024-597', Anonymous Referee #1, 01 Apr 2024
reply
This paper analyzes the spatial distribution of latent heating relative to the jet position when this position is estimated using different methodologies. The main conclusion of the study is that the heating is larger on the warm than on the cold side of instantaneous, non-zonal jets, which suggests that latent heating reinforces rather than depletes baroclinicity. This stormtrack self-maintenance suggests in turn that latent processes should increase rather than decrease zonal index persistence, contrary to what has been argued by previous studies.
I found the paper interesting and the results thought-provoking. I can see how the notion of stormtrack self-maintenance proposed by Hoskins and Valdes many years ago would point to a diabatic enhancement of zonal index persistence by latent heating, which I had not previously realized. At the same time, I am struggling to reconcile this result with the notion that meridional latent transport is important for decreasing the mean baroclinicity. Although I am not fully sold on the paper conclusions, the methodology seems sound (except for the concerns raised below) and the results reasonably convincing, so I support publication. However, I have issues with the terminology and some concerns with the methodology as described below.
Main comments:
1) My main issue is semantic. I think “averaging” is much more appropriate than “filtering” in the context in which the authors are using this word. Indeed, the 2D jet “filtering“ used in this paper is reminiscent of an isentropic or quasi-Lagrangian averaging, I think. While the conventional Eulerian mean circulation is thermally indirect, in isentropic coordinates the “upward” motion (diabatic heating) occurs at low latitudes, consistent with the authors’ findings for the 2D jet analysis. In my opinion, Eulerian mean is a more appropriate terminology than sector mean and noting the connection of the 2D jet averages with a quasi-Lagrangian/isentropic formalism would benefit the paper. But beyond this connection, even when Eulerian averages are considered the use of the word “filtering” for time averages seems awkward to me.
2) I am struggling to reconcile the synoptic notion that much of the condensation occurs along the warm conveyor belt (which the paper eloquently illustrates) with the climate notion that meridional latent heat transport helps decrease the meridional temperature gradient. I wonder if it might be necessary to consider the full 3D picture to reconcile these two views. The authors focus on where the heating occurs relative to the upper-level jet. But because the warm conveyor belt is 3D and fronts tilt with height, the condensation might in fact occur on the warm side of the jet but on the cold side of the lower-trosposphere temperature gradient. Moreover, as Fig. 1 shows that much of the cold front precipitation tends to occur along the actual front/upper level jet. Because the emphasis of the paper is the impact of the heating on the baroclinicity, I think it would be more appropriate to investigate where the heating occurs relative to the lower-troposphere baroclinicity rather than the upper level jet.
Other comments:
- The rationale for the procedure described starting in line 80 is not fully clear. Although this is described in detail in a previous paper, a clearer explanation here would help the reader.
- It is not clear to me if you allow for the existence of more than one jet when computing a sector mean, or you just look at the overall maximum.
- Relatedly, I am surprised by the fact that you get a Eulerian-mean thermally direct circulation when the jet is at very high latitudes. I wonder if there might be more than one jet in these circumstances so that you may be mixing their signals.
- Regarding the “dilution” of this pattern for 10d-Z jets compared to 1d-Z jets (line 130), I suspect the number of cases will be much smaller in the former case (the jet may not remain at such high latitudes for a long time), so these results may not be very robust.
- It is clear from the 10d-2D results that any relation to the actual jet has been lost in this analysis, and all you can see is noise added to a background latitudinal distribution. I would remove this analysis.
- Line 181. Figs. 6c and 6d look quite different to me. They only agree in the weak latitude-dependent background.
- I believe there is a mistake and the label “Poleward” should be placed on the right of the panels (as in Fig. 2c). This is also confusing, it may be helpful to add an “Equatorward” label on the left axis.
Minor
61. The second clause is not a full sentence. I suggest using a colon sign instead of a period to separate the first two clauses.
Eq. 1. I don’t think n should be hatted in the denominator
80. I believe a U is missing in front of the second n-derivative in the equation
The transition from line 89 to 90 is too abrupt. As written, an example is construed as general. You should say something like “To illustrate the differences between the jet definitions, Fig. 1 shows…”
209. Sect. A should be Appendix A
Citation: https://doi.org/10.5194/egusphere-2024-597-RC1 -
RC2: 'Comment on egusphere-2024-597', Anonymous Referee #2, 23 Apr 2024
reply
Summary
This manuscript presents an analysis of the observed relation between the upper-tropospheric jet location and midlatitude mid-tropospheric latent heating. The motivation is to examine whether latent heating occurs mostly on the warm/cold side of the jet, thus increasing/decreasing the baroclinicity around the jet peak. Specifically, the authors aim to reconcile contrasting results from previous studies, some of which found that latent heating tends to reduce the baroclinic shear of the zonal-mean or time-mean jet, while others found that overall latent heating tends to reinforce the baroclinicity at midlatitudes. Using spatio-temporal filtering, the authors show that while latent heating occurs mostly poleward of the zonal-mean time-mean jet (Fig.3l), it occurs mostly on the right (i.e., warm) side of the instantaneous local jet (Fig.5g). They further stratify the data according to the instantaneous local jet direction and show that latent heating occurs mostly to the east of poleward-directed jets (Fig.7c). These results are shown for the winter North Atlantic jet, with some result shown also for other longitudinal sectors in both hemispheres, showing qualitative similarity to the North Atlantic jet results (Figs.8,9).
I think these results would be interesting for the Weather and Climate Dynamics readers. The manuscript adds another piece to the puzzle of the interaction between the circulation and moisture (specifically latent heat release) in midlatitudes. I have some comments that I think should be addressed before the manuscript is accepted for publication.
Major comments
- The description of the analysis method in section 4 could be improved. Currently it is somewhat confusing. These are the points that weren’t clear to me:
- Lines 98-101: This description is not clear. I understood it only after I continued to read. Perhaps it would help to explain here what the horizontal and vertical axes in figure 2a represent. It took me a while to understand that “the lowest jet latitude states are on the left and highest jet latitude states on the right” refers to the horizontal axis in figure 2a.
- In the description of figure 2b the authors write that (lines 105-108) “we rotate the two-dimensional jets so that they appear as pure westerlies. With this rotation, the cold flanks of the jets are on the positive- and the warm flanks on the negative side of the vertical axis”. In the description of figure 2c they write that (lines 110-112) “…we make use of the jet cross sections but this time organize them by the direction of the jet rather than their latitude. Their original orientation in the latitude-longitude space is preserved and the cross sections are thus organized in semicircles according to their orientation”. I couldn’t understand why in figure 2c (and later in figure 7) the cross sections are organized in semicircles and not in circles. How are westward winds represented? This seems like a critical point for the interpretation of figure 7. If there is latent heating to the right of a westward jet, then this means latent heating on the poleward side of the cyclone.
- I suggest to replace the wording of the “warm/cold” side of the jet throughout the manuscript to the “right/left” side. There are no figures showing the temperature anomalies in the manuscript. Though we could expect the right side to be warmer than the left side, this is not strictly speaking what the right/left sides are. Specifically, when interpreting the results for the zonal-mean jet (e.g., lines 129-130), it could be that the equatorward (right) side of the jet is not necessarily the warmer side, when the 3h local variables are considered.
- The interpretation of the results:
- The analysis of the zonal-mean jet (figures 3,4) shows a diagonal pattern, which I assume represents a constant latitude for the analysis point (for example a point at latitude 40 would be poleward of the jet when the jet is at latitude 20 and equatorward of the jet when the jet is at latitude 50). It would aid the interpretation to add to these figures a diagonal line showing the slope of a constant-latitude line. In this context, I am not sure why the authors claim that “… a banded latitudinal structure … indicates that is has a strong geographical preference” (lines 148-151, and similarly in line 183). If there is a geographical preference for a specific variable, shouldn’t it appear as a diagonal structure in figure 3?
- The comparison with the results of Xia and Chang (2014) in lines 152-156: It should be noted that the data analyzed here is for winter, while Xia and Chang (2014) analyzed summer data. Note that the relation between the latitude of maximum diabatic heating and the jet latitude depends on the season (Lachmy and Kaspi, 2020).
- The structure of the 10d-2D results: I couldn’t really understand what this analysis represents and why the patterns in the right column of figure 5 seem to depend only on the jet latitude. Looking at the orange dashed line in figure 1b, I wouldn’t a-priori guess that the analysis variables would depend on the latitude of this line rather than the latitude of the analysis point. If the 10d-2D results are to be included in the paper, they should be explained. Otherwise, consider removing them.
- The interpretation of the 3h-2D results: The authors claim that these results “indicate an enhancement of baroclinicity at all latitudes in the extratropics” (lines 186-188), and contrast these results with previous studies which “highlighted detrimental effects of latent heating on the jet”. I think this interpretation adds unnecessary confusion. The instantaneous local (3h-2D) “jet” has a different meaning compared to the zonal-mean time-mean (10d-Z) jet. In the context of wave-mean flow theory, the 3h-2D jet represents mostly the eddies. The results of the current manuscript show that latent heating contributes to the baroclinicity within individual eddies (in other words, to the eddy available potential energy), but not the zonal-mean time-mean baroclinicity (the mean available potential energy). Specifically, it seems that the maximum latent heating appears in the poleward-moving warm conveyor belt inside cyclones. Perhaps the warming occurs on the poleward side of the cyclone (if figure 7 would be a full circle rather than a semi-circle then this would be visible). Therefore, I think that referring to both the 3h-2d and the 10-Z wind maxima by the same word – “jet” – is confusing. These are essentially different components of the flow. The same comment applies to lines 196-199, 230-235 and other places in the manuscript.
- Citations: There are a few relevant citations, which are not discussed in this manuscript:
- Madonna et al. (2014) – It would add some insight to relate the results in the current manuscript to the view of the warm conveyor belt. This subject is mentioned in lines 194-195, but without this reference, which I think is the most relevant. I would guess that most of the latent heating found in the 3h-2D analysis comes from warm conveyor belts.
- Lachmy and Kaspi (2020) – They analyze the observed climatological zonal-mean structure of the midlatitude diabatic heating and how it is related to the zonal-mean eddy-driven jet. I think this is relevant for the interpretation of the 10d-Z results.
- Pauluis et al. (2010) – This observational analysis looks at diabatic heating (upward motion in potential temperature coordinates) when it is averaged over moist isentropic surfaces, which capture the diabatic heating within eddies (as the 3h-2D analysis here). I think it would add another insight to compare with their results.
- Yamada and Pauluis (2017) – While this is a numerical study and not observational analysis, I think it would be useful to compare with their results, which show the role of latent heating in the development of baroclinicity during an eddy life-cycle.
Minor comments
- Line 45: “explore to the extent which” -> “explore the extent to which”.
- I would suggest to add a plot for the wind speed in figure 7, to make it comparable to the previous figures.
References
- Lachmy, O., & Kaspi, Y. (2020). The role of diabatic heating in Ferrel cell dynamics. Geophysical Research Letters, 47(23), e2020GL090619.
- Madonna, E., Wernli, H., Joos, H., & Martius, O. (2014). Warm conveyor belts in the ERA-Interim dataset (1979–2010). Part I: Climatology and potential vorticity evolution. Journal of climate, 27(1), 3-26.
- Yamada, R., & Pauluis, O. (2017). Wave–mean-flow interactions in moist baroclinic life cycles. Journal of the Atmospheric Sciences, 74(7), 2143-2162.
- Pauluis, O., Czaja, A., & Korty, R. (2010). The global atmospheric circulation in moist isentropic coordinates. Journal of climate, 23(11), 3077-3093.
Citation: https://doi.org/10.5194/egusphere-2024-597-RC2 - The description of the analysis method in section 4 could be improved. Currently it is somewhat confusing. These are the points that weren’t clear to me:
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 145 | 35 | 10 | 190 | 4 | 4 |
- HTML: 145
- PDF: 35
- XML: 10
- Total: 190
- BibTeX: 4
- EndNote: 4
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1