the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 26 Jan 2024
Thermal-Driven Graupel Generation Process to Explain Dry-Season Convective Vigor over the Amazon
Abstract. Large-eddy simulations (LESs) are conducted for each day of the intensive observation periods (IOPs) of the Green Ocean Amazon (GoAmazon) field campaign to characterize the updrafts and microphysics within deep convective cores while contrasting those properties between Amazon wet and dry seasons. Mean Doppler velocity (Vdop) simulated using LESs are compared with 2-year measurements from a Radar Wind Profiler (RWP) as viewed by statistical composites separated according to wet and dry season conditions. In the observed RWP and simulated LES Vdop composites, we find more intense low-level updraft velocity, vigorous graupel generation, and intense surface rain during the dry periods than the wet periods. To investigate coupled updraft-microphysical processes further, single-day golden cases are selected from the wet and dry periods to conduct detailed cumulus thermal tracking analysis. Tracking analysis reveals that simulated dry-season environments generate more droplet-loaded low-level thermals than wet-season environments. This tendency correlates with seasonal contrasts in buoyancy and vertical moisture advection profiles in large-scale forcing. Employing a normalized time series of mean thermal microphysics, the simulated cumulus thermals appear to be the primary generator of cloud droplets. At the same time, ice crystals tend to be generated in inactive parts of clouds. Time series shows that thermals, however, entrain ice crystals and enhance riming due to large concentrations of droplets in the thermal core. This appears to be a production pathway of graupel/hail particles within simulated deep convective cores. In addition, less-diluted dry-case thermals tend to be elevated higher, and graupel grows further during sedimentation after spilling out from thermals. Therefore, greater concentrations of low-level moist thermals likely result in more graupel/hail production and associated dry-season convective vigor.
- Preprint
(2834 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-3', Anonymous Referee #1, 21 Feb 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-3/egusphere-2024-3-RC1-supplement.pdf
- AC1: 'Reply on RC1', Toshi Matsui, 10 Apr 2024
- AC2: 'Reply on RC1', Toshi Matsui, 10 Apr 2024
-
RC2: 'Comment on egusphere-2024-3', Anonymous Referee #2, 29 Feb 2024
Review of “Thermal-Driven Graupel Generation Process to Explain Dry-1 Season Convective Vigor over the Amazon” submitted to ACP by Matsui et al.
General comments:
This manuscript uses the large eddy simulations to study moist convection over Amazon in dry and wet seasons and compared to field observations during Go-Amazon. The unique results are from thermal analysis implying the microphysical processes throughout the development of thermals. This work is very valuable in understanding the moist convection process and should be shared with the community. However, some discussions are still in need of clarification. Therefore, I recommend accepting the manuscript with some minor-major revisions.
Major comments:
- Though there are a lot of great results in this manuscript, the statement of the major ice growth by riming on entrained ice particles into the thermal from top is not well supported by the figures. Figure 10 shows that ice seems coming from top of thermal. But it includes the effect of thermal growth and move upward into colder temperature and activate ice from top while it looks like ice falling into the thermal. This part should be carefully reexamined with environment temperature as a reference, similar like Figure 10, but using temperature range instead of lifetime.
- Normalizing through the life cycle of thermals may not be the best approach because it removes the dependence of time. Microphysics process is quite time sensitive, e.g. warm rain process, riming process. The time and path for hydrometeor growth is the key to interpret the differences in dry and wet moist convection. Current manuscript does discuss a little on this with Figure 9 (L592-602). But more can be added, such as velocity vs altitude in Figure 8, or the tracking thermal time at different temperatures.
- Maybe add some discussion on the limitations of how this result could apply globally. For example, dry season Amazon compared to the central plain of US, or Argentina.
Minor comments:
- L61, radiation warms temperature? Maybe surface and increase skin temperature
- L81, what do you mean by “vigorous raindrops”?
- L108, please be specific on stronger precipitation properties, e.g. larger raindrops, more raindrops, more cloud droplets, larger reflectivity?
- L140, replace “collected by” with “of”
- L170, I do not believe that you can calibrate the whole profile of radar measurements with surface disdrometer. Should this be ‘all radar measurements at near surface were calibrated…’
- L216, now everyone is using ERA5, but I do not expect there is a big difference in the forcing if interim is used. But it is still interesting to know how small the difference is just for one case.
- L269, can thermals merge? This 80% criteria could be problematic if somehow there is a merge if model resolution is not high enough.
- L278, what is the discard rate in your study, especially for those long-lasting ones? I am wondering how many thermal samples would not be follow the momentum budget, and why.
- L385-386, I am confused on this. Do you mean that terminal velocity is not included in the velocity shown in Figure 3? Then it should not be called Vdop. Please clarify.
- Figure 3, font of xytitle is too small to read.
- Figure 3 caption, What kind of samples are included? Any sample with certain Q? Convective region? Please clarify.
- L398-401, another important reason is that model overestimates the graupel and hail too fast at warmer temperatures, which leads to larger melted rain drops and high terminal velocity.
- Figure 6, caption, “domain mean” (include rain or no rain) or domain mean of convective cores?
- Figure 7. Have you tried to put streamline (or vector differences) of dry-wet in right panels?
- L552-553, the boundary layer eddies and the moist convective thermals are two different concepts and should not be mixed here. Larger boundary layer eddy could lead to a stronger initial velocity at cloud base. It should have minimum relationship to the initial thermal size.
- L558, differences not discussed here include the water load and where thermal starts. 8b shows that thermals start with the same size and growing up for dry and wet cases the same way, except dry starts at a higher altitude. Wet has more cloud water at 3 km due to longer path and weaker vertical velocity leading to a longer time for warm rain process. Dry has stronger vertical velocity to start and a shorter path so most of cloud droplets reach freezing level (8c) to help growing graupel and hail. Note that dry supposes to have higher vertical velocity at the beginning.
- Figure 8. I am very surprised that the vertical velocity is not included in Figure 8. This is almost the most important variable for whole study! Please add that in the revision.
- L606, can “have”
- Figure 9. R, W, and Entrainments are all a function of altitude/temperature. It would be great if CFADs can be made for them. For life time, it would be great if the proportion of time in > 0C can be added.
- Figure 10, Caption or y title. what is the unit of time lag here? What kind of thermal samples are included? Please clarify.
- L638-639, I disagree with this statement. If this statement is true, the ice would activate most outside of thermals. Then the follow up question would be why ice nuclei would not activate inside the thermal when it first encounters the cold temperature during rising and expanding. The composite includes thermals centers with upward motion. This could be just thermal transitioning from liquid to ice phase in growing/rising up. At least from this figure alone, this statement is not fully supported.
- L650, this could just be colder temperature leads to more riming process.
- L655, though I somewhat agree with the statement, but the evidence shown in the manuscript does not directly disapprove the tradition concept model. Note that Arakawa and Schubert were using models with much coarser resolutions and have no capability to resolve “thermals”, which has a very loose definition throughout last half century.
- L678-679, Figure 11, if the riming is mainly from entrainment of ice, you would see the enhanced Qi from side edge of thermals in Figure 10.
- L705, 0.8m/s as the definition of thermal could exclude weaker thermals in wet season. At least a discussion on the definition of thermals should be included here for discussion. What if you use 0.1 m/s, do we still see more thermals in dry season? Also, is this only from a golden case day? Or the whole time period? How representative is this to the whole dry and wet season?
- L716, this may not be fair to compare to L-O differences. The diurnal variations of land vs. ocean have one with thermals only in part of day and another spread the whole day.
- L728, it can be either numbers or stronger initial velocities satisfying 0.8m/s criteria. Note that larger eddies may only mean stronger initial velocity to start the thermal.
Citation: https://doi.org/10.5194/egusphere-2024-3-RC2 - AC3: 'Reply on RC2', Toshi Matsui, 10 Apr 2024
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 269 | 65 | 20 | 354 | 9 | 6 |
- HTML: 269
- PDF: 65
- XML: 20
- Total: 354
- BibTeX: 9
- EndNote: 6
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1