WMOD00291

• Title: Bulk Parameterization of the Snow Field in a Cloud Model
• Date: 6/6/1983
• Weather Modification - Doc Type: Report

1084

JOU R N A L OF CLI MATE AND APPLI ED METEOROLOGY

VOLUME 22

3.0
E
u
~2.5
a::
w
f-
3!2.0
~
<[
a:: 1.5
Cl
w
t:W
.J
=>
::!:
30.5
u
<[

FIG. 9. The horizontal distribution of accumulated rainfall at
the surface at 48 min of simulated time for the three cases.

at 29 min and 44 min in Fig. 7b and 30 min (3 km
height) and 42 min (9.2 km height) in Fig. 6. The
decrease in net hail production around 33 to 36 min
for Cases I and 2 is caused by the melt of hail as it
falls below the OOC level. The fallout of precipitation
at this time leads to a stimulation of the lower right
portion of the cloud in Cases 1 and 2, which is re-
sponsible for the second surge in hail growth indicated
in Figs. 7b and 8.
A single hail growth peak shows up in Case 3 at
about 40 min (9.2 km height). Examination of Figs.
7 and 8 indicates Case 3 has rain forming nearly 12
min later than in Cases I and 2. However, hail for-
mation is delayed only about 6 min. Melting of hail
is the primary source of rain in all three cases, and
for Case 3 it is also the initiating mechanism for rain.
Rain in Cases I and 2 is initially produced by au-
!oconversion. It is the freezing of this earlier rain
which initiates the hail field in Cases I and 2 (slightly
earlier in Case I because of rain-snow interactions).
Hail in Case 3 is initiated by the aggregation of snow
which occurs 6 min later than the freezing of rain in
Cases 1 and 2.
The time period from 18 to 24 min in Cases I and
2 (Fig. 7a) indicates that the initial production of rain
is followed by a temporary loss before the major rain
production stage is reached. This behavior is due to
the initial formation of rain in the lower level of the
clouds, while the advection of rain to higher, colder
levels where accretion by hail (in Case 2) or by hail
and snow (in Case 1) depletes the rain. The decrease
in rain ceases when hail begins to melt and becomes
a dominant source of rain after 25 min.
Examination of Figs. 7 and 8 reveals that the pres-
ence of snow serves to deplete the rain field and to
keep its formation rate slower in Case I than in Case
2 (no snow case). The snow initiates the hail field
about 1 min sooner in Case 1, which results in earlier
depletion of the rain in Case I and a reduction in the
total amount of rain produced in the first surge.com-

pared to Case 2. The snow accelerates the formation
of hail in Case 1, which accounts for the slightly ear-
lier fallout of hail and associated melting to form
appreciable amounts of rain. Consequently, hail is a
sink for rain at first (as is snow), but a source later
as it melts below the OOC level.
Case 3 is much more inefficient in the production
of precipitation in the early stages (18-30 min) as is
clearly shown by the much larger amounts of cloud
water (Fig. 8). Although hail production is consid-
erably delayed in this case, the hail, once formed, is
produced in much larger quantities, being present in
abundant amounts throughout a greater portion of
the cloud. Rain production is delayed even more, but
follows the same trend as the hail in this case.
Fig. 9 shows the accumulated rain at the surface
at 48 min from all three cases. Table 2 provides ad-
ditional information on the results of the cases at 48
min. Some precipitation still remains to accumulate
from the main cells, but these results indicate the
trend of the differences. Returning to Fig. 8, we note
that Cases 2 and 3 have more rain and hail (efficient
precipitating forms) suspended aloft at 48 min than
does Case 1. If the simulations had been continued
beyond 48 min, it appears that Case 1 would remain
the least efficient of the three cases. It is unclear
whether Case 2 or 3 would end up producing the
greatest surface accumulations since, at 48 min, Case
2 has the most hail aloft, whereas Case 3 has more
rain. The no-snow case (Case 2) produces the greatest
surface accumulation, presumably due to the fact that
the snow process of Cases 1 and 3 depletes some of
the cloud water, the snow falling out as virga evap-
orating before reaching the ground. Compared to the
other cases, Case 2 has much larger quantities of
cloud ice, which in a sense indicates less efficiency.
The amount of cloud ice in Case 2, however, is less
than the combined amounts of snow and cloud ice
(both inefficient) in Cases I and 3. Case 3 produces
an intermediate value of rainout but the greatest sur-
face accumulation of hail [6.4 kT km-t vs. 2.0 for
Case 1, and 2.5 kT km-1 for Case 2]. This is due to
the delay in precipitation formation which causes.
much of the hail to form in the upper, colder parts
of the cloud; the hail thus formed must fall through

~

TABLE 2. Surface accumulations, total water vapor flux into
storm, and precipitation efficiency at 48 min. The units for the
precipitation amounts and water vapor are kT km-I and 109 g
k -I .
m .

Total Precip-
Rain Hail precip- Water vapor itation
at at itation at flux into efficiency
Case surface surface surface cloud (%)
Case I 67.66 2.06 69.71 658.46 10.6
Case 2 80.45 2.49 82.93 671.82 12.3
Case 3 72.29 6.42 78.71 650.44 12.1