Parameters affecting condensation/evaporation rates (CR/ER) in trade wind cumulus
clouds were analyzed using LES model simulations. The model was initialized
with data observed during the RICO field project, and simulated in a rather
large

The results revealed rather remarkable relationship between integral condensation/evaporation rate and integral upward mass flux. Identified relathionship may be useful for parameterization of subgrid latent heat in meso and large-scale models.

The latent heat release plays an important role in predicting the
thermodynamic structure of cumulus convection. Its accurate formulation is
challenging in meso and large-scale models, not least because of sub-grid
scale microphysical variability. The goal of our LES study is to investigate
the phase transition process which is the source of latent heat release in
convective clouds, and, specifically, its dependance on cloud
thermo-dynamical variables. The LES model we use (SAMBM) employs the
dynamical core of the System for Atmospheric Modeling (SAM, Khairoutdinov
and Randall, 2003) and the Bulk Microphysics tuned for shallow Cu convection
(BM, Kogan, 2013). The observations from the RICO field campaign (vanZanten
et al., 2011) were used for initializing the LES simulations conducted in a
rather large

Over the course of the 32 h run, the simulation data was saved every 30 min; from this archive a total of 2031 clouds were selected for analysis. Our initial attempt of using the “brute force” statistical approach to relate phase transition rates to the dynamical parameters did not succeed because of the complexity of the cloud system, consisting of clouds at various stages of their development. A better approach proved to be separation of the entire dataset into subsets stratified by the cloud size, maturity, parameters of precipitation.

Specifically, we employed the method used previously for PDF parameterization development (Kogan and Mechem, 2014, 2016); namely, the dataset was sorted by cloud top height and divided into four groups G1–G4, each of which condenses approximately equal amount of water vapor per second. The groups G1–G2 represented small clouds (1346 in G1 and 483 in G2). mostly at the growing stage, while groups G3–G4 (137 clouds in G3 and 65 in G4) contained mature or decaying clouds.

Mean and standard deviation of cloud physical parameters in each of
the four groups.

Figure 1 shows mean and standard deviation of selected physical and
precipitation cloud parameters in each group. G1 clouds are most numerous;
they are also the smallest with cloud tops varying in the range from 1.34 to
2.3 km. Their mean projected surface area is on average less than 2 km

Mean condensation/evaporation rates (CR/ER) shown in Fig. 1b are more in line
with the increase in volume (note the probability scale on the

Even larger, exponential increase is seen when analyzing PR. For clouds in G1 PRs are very small, only 0.6 mm/h. Clouds in G2 are three time larger in volume, but their PRs are six times larger. The clouds in G3 have about four times larger volumes than G2 clouds, but their PRs increase more than seven times. Even more dramatic difference exists for G4 clouds, where nine times increase in volume leads to 18 times larger precipitation rates.

While Fig. 1a–b show mean cloud parameters in each group, the Fig. 1c shows

Correlation of condensation rate (CR) with parameters denoted in plot
legends. Group G2

Together G4 and G3 account for three quarter of total precipitation; G1 and G2 contribute, respectively 8 % and 17 %. As already mentioned, these groups precipitate less than condense, i.e., they are still growing, while G3, and especially G4 clouds precipitate about 70 % and 100 % more than condense, that is, they are losing water and, therefore, at the stage of decay.

Analysis of correlation between CR/ER and the thermodynamical parameters which
may affect CR/ER was conducted for clouds in each group separately. The analyzed
parameters were integrated over the whole cloud volume. These include: up
and down mass flux MF (defined as air density

Correlation plots of:

Figure 2 shows, as an example, results of the correlation analysis for clouds in G2; the results for clouds in other groups are similar. One can note exceptionally high correlation between condensation rate and upward (Plus) mass flux (MFP). Correlation between CR and upward buoyancy flux (BFP) is also high, but weaker than with MFP. Similar strong correlation exists with cloud water QC. As one might expect, the correlation between condensation and downward (Minus) fluxes (MFM, BFM), as well as rain parameters (QR) is weaker. For other cloud groups results are similar with the same conclusion: the strongest correlation is between CR and MFP.

The evaporation rate (ER) (Fig. 3) also has stronger correlation with the upward (MFP) than with downward mass flux (MFM) (Fig. 3a–b). Buoyancy flux determines ER worse than upward mass flux (Fig. 3c–d), however, its downward component (BFM) correlates with ER slightly stronger than its upward component BFP. All in all, it is rather remarkable that the upward mass flux is the parameter which defines both condensation and evaporation in a cloud.

What is also remarkable, is that the slope of the linear fit approximating
the correlation between CR/ER and MFP only slightly depends on cloud group, i.e., on
cloud top height. The scatter plots in Fig. 3e–f illustrate this fact which
can be expressed as a linear relationship between phase transition rate
(PTR) and upward mass flux:

Based on the LES model data, we analyzed condensation/evaporation
parameters, and their correlation with thermodynamical parameters of
trade-wind cumulus convective clouds. A very strong correlation was found
not only between the condensation, but also evaporation rate and upward mass
flux (all parameters were integrated over the whole cloud volume). While
good correlation between the upward mass flux and condensation is not
surprising (obviously due to larger supersaturation in stronger updrafts),
the very high correlation coefficient (

The software code for data analysis was developed by the author and is available upon request.

Analysis data is available upon request from the author.

The author declares that there is no conflict of interest.

This article is part of the special issue “Applied Meteorology and Climatology Proceedings 2020: contributions in the pandemic year”.

This investigation was supported by ONR Grant N00014-20-1-2050. The author is grateful to Vaughan Phillips and anonymous reviewer for constructive comments. The computing for this project was performed at the OU Supercomputing Center for Education and Research (OSCER) at the University of Oklahoma.

This research has been supported by the Office of Naval Research (grant no. N00014-20-1-2050).

This paper was edited by Emily Gleeson and reviewed by Vaughan Phillips and one anonymous referee.