Project with Rutledge group run by Brenda Dolan to look at rainfall size
distributions and link those to microphysical processes within multiple
cloud environements. My main contribution is to process former RAMS simulations
of various cloud environments highlighted below. Outputting Do, Dm, sigma,
logNw to compare to disdrometer data.

FIRST NOTE THAT ALL SIMULATIONS FOR THIS PROJECT SIT AT:
/precipitation/smsaleeb/DOE.MICRO-simulations/NOBAK

###############################################################################
Several Saleeby project directories sit at:
###############################################################################
proj.2017-doe.micro-multi-6.2.11-FLOYD = contains code to run:
 (1) DX=250m,DZ=100m ATEX stratocumulus, (2)namma idealized tropical 
 convection, (3)DX=1km Grant supercell runs to test RainDrop Breakup.

proj.2017-doe.micro-atex-hires-6.2.12-STRATUS = contains all code and scripts
 for running DX=200m,DZ=50m ATEX simulations which are higher resolution than
 the ones mentioned above. (Ran on DOE-STRATUS supercomputer)

proj.2017-doe.micro-supercell-hires-6.2.12-STRATUS = contains all code and
 scripts for running DX=500m and DX=300m Grant supercell simulations. (Ran on
 DOE-STRATUS supercomputer)

Auto.2017-doe.micro-archived-runs = contains all the ARCHIVED SIMULATIONS used
 in this study. Ran these here to mainly obtain Do rain drop frequency by
 bin for analyzing raindrop breakup.

Auto.2017-doe.micro-atex-hires-6.2.12 = analysis of the ATEX simulations run by
 proj.2017-doe.micro-atex-hires-6.2.12-STRATUS

Auto.2017-doe.micro-supercell-hires-6.2.12 = analysis of the GRANT SUPERCELL
 simulations run by proj.2017-doe.micro-supercell-hires-6.2.12-STRATUS

Auto.2017-doe.micro-multi-6.2.11 = analysis of simulations run within
 proj.2017-doe.micro-multi-6.2.11-FLOYD

###############################################################################
Here we are running REVU on multiple old RAMS simulations and some new ones.
This section summarizes the simulations in the database. New simulations
have more details at the bottom of the notes here.
###############################################################################

For simulations run with RAMS 6.1.11 and earlier we need to use the
executable "revu-6.2.09.reversearrays" since we have done an array
swap in newer RAMS/REVU to deal with row major / column major issues
in RAMS runtime and HDF5 output.

In older versions of RAMS, we also did not output GNU for each 
hydrometeor type in the header file, so we hardcoded this in REVU 
temporarily to get a compiled version of REVU with dm,d0,lowgn,sigma
for the value of GNU=2 for rain in the older simulations 
(revu-6.2.09.reversearrays.nu2). This applies to simulations run about
before 2015. Check header file for "gnu".

We run REVU for each simulations to output a number of variables as
group in different files.

All the rain-related variables are output to files with the "rain" label:
rain_gam_dm
rain_gam_d0
rain_gam_lognw
rain_gam_sigma
rain_m3
rain_concen_m3
pcprr
pcpvr

All the micro-budget variables are output to files with the "micro" label:
cld2raint
ice2raint
vapliqt
vapicet
melticet
rimecldt
rain2icet
aggregatet

Some extra variables are produced with "extra1" label:
vt_cld2raint
vt_ice2raint
vt_vapliqt
vt_vapicet
vt_melticet
vt_rimecldt
vt_rain2icet
vt_aggregatet
latheatvap
latheatfrz
latheatvapt
latheatfrzt
w
reflect_all

******************************************************************************
******************************************************************************
******************************************************************************
## Supercell using Leah Grant Setup HIRES - Saleeby Simulations ##
proj.2017-doe.micro-supercell-hires-6.2.12-STRATUS / runrams
RAMS VERSION 6.2.12 (run on DOE - STRATUS)

These are supercell simulations using Leah Grant's sounding from her paper
cited below. I am using the control RAMSIN from Leah Grant based on her JAS
papers in 2014,2015. We initially ran her tests at coarser spacing (DX=500m)
but decided to go to Leah's finer spacing (DX=300m). I'm using her physics 
specs, but ZROUGH=0.0001. Freeslip flag is set to on, so SFLUX surface terms 
are zero. In these simulations we varied the value of GNU for rain. This version 
of RAMS used DMB1 (minimum mean mass of 0.3mm). 

Results from simulations were written up as:
 [Grant,L.D., and S.C. van den Heever, 2014: Microphysical and dynamical 
  characteristics of low-precipitation and classic supercells.
  J. Atmos. Sci., 71, 2604-2624.]

For simulations details see RAMSIN, Auto, and Proj directory files:
 Auto.2017-doe.micro-supercell-hires-6.2.12
 proj.2017-doe.micro-supercell-hires-6.2.12-STRATUS /
  runrams/A.RAMSINs /
   RAMSIN.supercell.grant.hires.gnu01.500dx
   RAMSIN.supercell.grant.hires.gnu02.500dx
   RAMSIN.supercell.grant.hires.gnu03.500dx
   RAMSIN.supercell.grant.hires.gnu04.500dx
   RAMSIN.supercell.grant.hires.gnu10.500dx
   RAMSIN.supercell.grant.hires.hucm-fast.500dx
   RAMSIN.supercell.grant.hires.hucm-full.500dx
   RAMSIN.supercell.grant.hires.gnu01.300dx
   RAMSIN.supercell.grant.hires.gnu02.300dx
   RAMSIN.supercell.grant.hires.gnu03.300dx
   RAMSIN.supercell.grant.hires.gnu04.300dx
   RAMSIN.supercell.grant.hires.gnu10.300dx
   RAMSIN.supercell.grant.hires.hucm-fast.300dx
   RAMSIN.supercell.grant.hires.hucm-full.300dx

Simulations can be found at:
/precipitation/smsaleeb/DOE.MICRO-simulations/NOBAK/
 doe.supercell.grant.hires.gnu01.500dx
 doe.supercell.grant.hires.gnu02.500dx
 doe.supercell.grant.hires.gnu03.500dx
 doe.supercell.grant.hires.gnu04.500dx
 doe.supercell.grant.hires.gnu10.500dx
 doe.supercell.grant.hires.hucm-fast.500dx
 doe.supercell.grant.hires.hucm-full.500dx
 doe.supercell.grant.hires.gnu01.300dx
 doe.supercell.grant.hires.gnu02.300dx
 doe.supercell.grant.hires.gnu03.300dx
 doe.supercell.grant.hires.gnu04.300dx
 doe.supercell.grant.hires.gnu10.300dx
 doe.supercell.grant.hires.hucm-fast.300dx
 doe.supercell.grant.hires.hucm-full.300dx

******************************************************************************
******************************************************************************
******************************************************************************
## ATEX-3D Stratocu HIRES - Saleeby Simulations ##
proj.2017-doe.micro-atex-hires-6.2.12-STRATUS / runrams
RAMS VERSION 6.2.12 (run on DOE - STRATUS)

These are the ATEX stratocumulus evolution simulations that are idealized
with periodic boundary conditions and run at 200m horizontal grid spacing and
50 meter vertical spacing. They are run 24-hrs and microphysics budgets
are updated every 5-min at the output interval of 5-min. The 5-min data micro
budget variables are values accumulated over the 5-min period. We are not
subsampling the data horizontally, but we are sub-sampling in time by only
providing analysis every 10-min. Micro budgets need to keep this in mind when
computing rates/time.

In these simulations we varied the value of GNU for rain. This version of RAMS
used DMB1 (minimum mean mass of 0.3mm). We ran a test at DMB1=0.1mm by
recompiling the code with this lower value to see how this impacts the rain drop
frequency distribution of Do. See simulation "doe.atex.0050ccn.hires.gnu03dmb1"

Note that the simulations using HUCM "doe.atex.0050ccn.hires.hucm-full" would
not produce a consistent stratocumulus deck at all. We resorted to even higher
resolution simulations that Adele Igel ran for us. This could also have to do
with HUCM removing and not regenerating aerosols. The regenerated aerosol
function was turned off in the code and commented out.

Results from SIMILAR ATEX simulations were written up as:
 [Saleeby et al. 2015: Impacts of cloud droplet-nucleating
  aerosols on shallow tropical convection.
  J.Atmos.Sci., 72, 1369-1385.]

For simulations details see RAMSIN, Auto, and Proj directory files:
 Auto.2017-doe.micro-atex-hires-6.2.12
 proj.2017-doe.micro-atex-hires-6.2.12-STRATUS /
  runrams/A.RAMSINs /
   RAMSIN.atex.0050ccn.hires.gnu01rain
   RAMSIN.atex.0050ccn.hires.gnu02rain
   RAMSIN.atex.0050ccn.hires.gnu03rain
   RAMSIN.atex.0050ccn.hires.gnu04rain
   RAMSIN.atex.0050ccn.hires.gnu10rain
   RAMSIN.atex.0050ccn.hires.hucm-full

Simulations can be found at:
/precipitation/smsaleeb/DOE.MICRO-simulations/NOBAK/
 doe.atex.0050ccn.hires.gnu01rain
 doe.atex.0050ccn.hires.gnu02rain
 doe.atex.0050ccn.hires.gnu03dmb1
 doe.atex.0050ccn.hires.gnu03rain
 doe.atex.0050ccn.hires.gnu04rain
 doe.atex.0050ccn.hires.gnu10rain
 doe.atex.0050ccn.hires.hucm-full

******************************************************************************
******************************************************************************
******************************************************************************
## ATEX-3D Stratocu - Saleeby Simulations ##
proj.2017-doe.micro-all-6.2.11-FLOYD / runrams.atex.stratocu.6.2.11
RAMS VERSION 6.2.11
simid = atex.0050ccn.gnu01rain
simid = atex.0050ccn.gnu02rain
simid = atex.0050ccn.gnu04rain

These are the ATEX stratocumulus evolution simulations that are idealized
with periodic boundary conditions and run at 250m horizontal grid spacing and
100 meter vertical spacing. They are run 24-hrs and microphysics budgets
are updated every 5-min at the output interval of 5-min. The 5-min data micro
budget variables are values accumulated over the 5-min period. We are not
subsampling the data horizontally, but we are sub-sampling in time by only
providing analysis every 10-min. Micro budgets need to keep this in mind when
computing rates/time.

Results from these simulations were written up as:
 [Saleeby et al. 2015: Impacts of cloud droplet-nucleating
  aerosols on shallow tropical convection.
  J.Atmos.Sci., 72, 1369-1385.]

Output from REVU are in files such as:
atex-clean-gnu01-g1-rain.h5
atex-clean-gnu02-g1-rain.h5
atex-clean-gnu04-g1-rain.h5

For simulations details see RAMSIN, Auto, and Proj directory files:
 Auto.2017-doe.micro-multi-6.2.11
 proj.2017-doe.micro-multi-6.2.11-FLOYD /
  runrams.atex.stratocu.6.2.11/A.RAMSINs /
   RAMSIN.atex3d.0050.gnu01
   RAMSIN.atex3d.0050.gnu02
   RAMSIN.atex3d.0050.gnu04

Simulations can be found at:
/precipitation/smsaleeb/DOE.MICRO-simulations/NOBAK/
 doe.atex.0050ccn.gnu01rain
 doe.atex.0050ccn.gnu02rain
 doe.atex.0050ccn.gnu04rain

******************************************************************************
******************************************************************************
******************************************************************************
## NAMMA - Tropical convection - Dust - Twohy/Saleeby Simulations ##
proj.2017-doe.micro-all-6.2.11-FLOYD / runrams.namma.tropconv.6.2.11
RAMS VERSION 6.2.11
simid = nammaconv.dustccnin
simid = nammaconv.dustinonly
simid = nammaconv.nodust

For the NAMMA simulations, we are using the simulations run by
Steve Saleeby that are idealized convection simulations using the
Dunion moist tropical sounding and using a convergence zone for
storm initiation. Simulations were run at 250m horizontal spacing
and with 108 vertical levels. Output is every 3-minutes and microphysics
budgets are accumulated during that time period. The name "DustCCNIN" 
indicates the simulations using the NAMMA dust and CCN profiles given by
Cindy Twohy. "DustINonly" means dust only acted as Ice Nuclei. "NoDust" means
we did not initialize with dust, but only the CCN profile from NAMMA.
Simulations were run 2hrs15min.

Output from REVU are in files such as:
nammaconv-dustccnin-g1-rain.h5
nammaconv-dustinonly-g1-rain.h5
nammaconv-nodust-g1-rain.h5

For simulations details see RAMSIN, Auto, and Proj directory files:
 Auto.2017-doe.micro-multi-6.2.11
 proj.2017-doe.micro-multi-6.2.11-FLOYD /
  runrams.namma.tropconv.6.2.11/A.RAMSINs /
   RAMSIN.nammaconv.dustccnin
   RAMSIN.nammaconv.dustinonly
   RAMSIN.nammaconv.nodust

NAMMA simulations can be found at:
/precipitation/smsaleeb/DOE.MICRO-simulations/NOBAK/
 doe.nammaconv.dustccnin
 doe.nammaconv.dustinonly
 doe.nammaconv.nodust

******************************************************************************
******************************************************************************
******************************************************************************
## BSISO - Ben Toms MJO Basin Scale Luzon Thesis Simulations ##
FROM SIMULATION ARCHIVE
RAMS VERSION 6.2.07
simid = bsiso.phase1
simid = bsiso.phase2
simid = bsiso.phase3
simid = bsiso.phase4
simid = bsiso.phase5
simid = bsiso.phase6
simid = bsiso.phase7
simid = bsiso.phase8

These are MJO / BSISO long time scale simulations from Ben Toms.
These simulations were run over the region of Luzon. We are subsetting the
regions to be over Luzon in a 5-deg x 5-deg box with a domain that is about
50% land and 50% ocean area. We are using latitudinal points 495-635 and
longitudinal points 447-578. There are 8 days of data at 10-minutes interval
LITE files. The analysis files were output hourly, so accumulated microphysics
budgets were not reset until on the hour. To get at micro budgets in LITE files,
we would need to subtract values between LITE file times. This just adds another
step to the analysis. We have 1 day of data for each MJO phase. Aerosol profiles
were kept static across the domain over time. Horizontal grid spacing is 4km.

Results from these simulations were written up as:
 [Toms et al. 2019: The boreal summer Madden-Julian oscillation and 
  moist convective morphology over the maritime continent.
  J.Atmos.Sci.)

Output from REVU are in files such as:
bsiso-phase1-g1-rain.h5 for 7/12/2016
bsiso-phase2-g1-rain.h5 for 7/17/2016
bsiso-phase3-g1-rain.h5 for 7/21/2016
bsiso-phase4-g1-rain.h5 for 7/25/2016
bsiso-phase5-g1-rain.h5 for 7/28/2016
bsiso-phase6-g1-rain.h5 for 8/03/2016
bsiso-phase7-g1-rain.h5 for 8/08/2016
bsiso-phase8-g1-rain.h5 for 8/16/2016

For simulations details see RAMSIN files:
 Auto.2017-doe.micro-archived-runs /
  RAMSIN.bsiso

Simulations can be found at:
/precipitation/smsaleeb/DOE.MICRO-simulations/NOBAK/
 doe.bsiso.phase1
 doe.bsiso.phase2
 doe.bsiso.phase3
 doe.bsiso.phase4
 doe.bsiso.phase5
 doe.bsiso.phase6
 doe.bsiso.phase7
 doe.bsiso.phase8

******************************************************************************
******************************************************************************
******************************************************************************
## CAPPM - NASA - Squall line - Saleeby Simulations ##
FROM SIMULATION ARCHIVE
RAMS VERSION 6.1.06
simid = cappm.squall.control

This is the CAPPM Control simulation. These squall line simulations have not
been published but have been included in multiple conference presentations
and presentations for Wei-Kuo Tao at NASA Goddard. These simulations are of
the MC3E May 20, 2011 squall line at 1.0km grid spacing on Grid-3 using grid
nesting. We are not subsampling the data. We did not output micro budgets.
Data were natively output every 30min on Grid-3. Simulations start at
00Z May 20, 2011 and were run 39 hours.
(output in HDF5 reversed arrays)

Output from REVU are in files such as:
cappm-2011may20-ctrl-g3-rain.h5

For simulations details see RAMSIN files:
 Auto.2017-doe.micro-archived-runs /
  RAMSIN.cappm.squall

Simulations can be found at:
/precipitation/smsaleeb/DOE.MICRO-simulations/NOBAK/
 doe.cappm.squall.control

******************************************************************************
******************************************************************************
******************************************************************************
## MC3E - DOE - Squall line - Saleeby/Marinescu Simulations ##
FROM SIMULATION ARCHIVE
RAMS VERSION 6.1.01
simid = mc3e.squall.clean.5min

This is a MC3E squall line control simulations. We are using the rerun 
5-min output data for the May 20, 2011 Oklahome squall line case at 1.2km grid 
spacing on Grid-3 from the CLEan pollution simulation. The 5-min data micro 
budget variables are values accumulated over the 5-min period. We are not 
subsampling the data horizontally, but we are sub-sampling in time by only 
providing analysis every 10-min. Micro budgets need to keep this in mind when 
computing rates/time.
(output in HDF5 reversed arrays)

Results from these simulations were written up as:
 [Saleeby et al. 2016: Aerosol indirect effects on the anvil 
  characteristics of mesoscale convective systems. 
  J.Geo.Res., 121, 10880-10901.]
 [Marinescu et al. 2016: The microphysical contributions to and 
  evolution of latent heating profiles in two MC3E MCSs. 
  J.Geo.Res., 121, 7913-7935.]

Output from REVU are in files such as:
mc3e-2011may20-clean-g3-rain.h5

For simulations details see RAMSIN files:
 Auto.2017-doe.micro-archived-runs /
  RAMSIN.mc3e.squall.clean

Simulations can be found at:
/precipitation/smsaleeb/DOE.MICRO-simulations/NOBAK/
 doe.mc3e.squall.clean.5min

******************************************************************************
******************************************************************************
******************************************************************************
## Seabreeze - Leah Grant Simulations ##
FROM SIMULATION ARCHIVE
RAMS VERSION rams6-20121207 (not sure of version number)
simid = seabreeze.cpoll
simid = seabreeze.ctl

For the Seabreeze simulations, we are using simulations from Leah
Grant. Simulations were idealized, run at 1km grid spacing. The domain
included here is a subsample mostly over land area from grid points
180 to 500 in x-direction. Land further west was ocean area. Use all of
y-direction which had periodic boundaries. Analysis included times from
12-21 UTC at 5-min increments. We are examining the control and most polluted
simulations with 75% initial soil moisture. Microphysics budgets are
accumulted every 5-min between analysis files.
(output in HDF5 reversed arrays and without GNU in header files)

Results from these simulations were written up as:
 [Grant and vandenHeever 2014: Aerosol-cloud-land surface
  interactions within tropical sea breeze convection.
  J.Geo.Res., 119, 8340-8361.]

Output from REVU are in files such as:
seabreeze-cpoll-g1-rain.h5
seabreeze-ctl-g1-rain.h5

For simulations details see RAMSIN files:
 Auto.2017-doe.micro-archived-runs /
  RAMSIN.seabreeze.cpoll (most polluted)
  RAMSIN.seabreeze.ctl   (control)

Seabreeze simulations can be found at:
/precipitation/smsaleeb/DOE.MICRO-simulations/NOBAK/
 doe.seabreeze.cpoll
 doe.seabreeze.ctl

******************************************************************************
******************************************************************************
******************************************************************************
## Supercell - HiRes Tornadic Scale - Sean Freeman Simulations ##
FROM SIMULATION ARCHIVE
RAMS VERSION 6.1.18
simid = supercell-freeman.gnu1
simid = supercell-freeman.gnu2
simid = supercell-freeman.gnu4

For the supercell simulations, we are using Sean Freemean's 
simulations run at 250m grid spacing and output every 1-min. These
were his Gamma GNU test simulations that he has published. We are 
subsampling every 2 points in the horizontal and every 5-min. The 
micro budget variables are accumlulated for 1-min periods. The use of gamma
distribution shape parameter gnu=1,2,4 varies for RAIN only. (From Sean's
originally named simulations such as "SUPER_R4GS250RBG".

Results from these simulations were written up as:
 [Freeman et al. 2019: Relative sensitivities of rainfall prediction
  to fixed shape parameters and collection efficiences.
  Q.J.Roy.Met.Soc., 145, 2181-2201.
  DOI:10.1002/qj.3550]

Output from REVU are in files such as: 
supercell-freeman-gnu1-g1-rain.h5
supercell-freeman-gnu2-g1-rain.h5
supercell-freeman-gnu4-g1-rain.h5

For simulations details see RAMSIN files:
 Auto.2017-doe.micro-archived-runs / 
  RAMSIN.supercell-freeman.gnu2

SUPERCELL runs can be found at:
/precipitation/smsaleeb/DOE.MICRO-simulations/NOBAK/
 doe.supercell-freeman.gnu1
 doe.supercell-freeman.gnu2
 doe.supercell-freeman.gnu4

******************************************************************************
******************************************************************************
******************************************************************************
## Tropical Convection - Leah Grant Simulations ##
FROM SIMULATION ARCHIVE
RAMS VERSION 6.1.18
simid = tropconvsys.ctl-clst
simid = tropconvsys.ctl-lin

For the Tropical Convective simulations we are using simulations from Leah
Grant. Simulations were idealized RCE type simulations. The labels "CTL-CLST"
indicates the control cluster convection run and "CTL-LIN" indicates the 
control linear convection run. The cluster simulation data are from hours 
10-20 and for the linear convection from hours 8-16. Simulations data are in
"LITE" files with "a-L" prefix. Grid spacing is 250m, so we subsample
every 2 points from x-direction points 100 to 1900 within the nudging zone.
Analysis files writes were every hour, so accumulated microphysics budgets 
were not reset until on the hour. To get at micro budgets in LITE files, we
would need to subtract values between LITE file times. This just adds another
step to the analysis. Output of LITE files is every 10-min so we sample at this
frequency.

Results from these simulations were written up as:
 [Grant et al. 2018: The role of cold pools in tropical 
  oceanic convective systems.
  J.AtmosSci., 75, 2615-2634.]

Output from REVU are in files such as:
tropconv-ctlclst-g1-rain.h5
tropconv-ctllin-g1-rain.h5

For simulations details see RAMSIN files:
 Auto.2017-doe.micro-archived-runs  /
  RAMSIN.tropconv.ctl-clst
  RAMSIN.tropconv.ctl-lin

Tropical convective simulations can be found at:
/precipitation/smsaleeb/DOE.MICRO-simulations/NOBAK/
 doe.tropconvsys.ctl-clst
 doe.tropconvsys.ctl-lin

###############################################################################
###############################################################################
NEW SIMULATIONS EXTRA DETAILS
###############################################################################
###############################################################################

*******************************************************************************
proj.2017-doe.micro-multi-6.2.11-FLOYD / runrams.namma.tropconv.6.2.11
*******************************************************************************

This project involves using the Loftus et al. (2008) convergence method
for generating idealized convection. I am using the Dunion (2011) composite
tropical soundings to generated tropical convection in a moist tropical (MT)
environment.

Initially used a warm bubble to initiate convection, then Sue decided that
using convergence would be better. Apparently some in the field no longer
like using thermals to generate idealized convection. Perhaps using 
convergence will eliminate the model shock and generation of gravity waves
that often results from using a warm bubble. That being said, I wanted to
save the model configurations for the warm bubble tests. There were 9 tests
that are outlined in my notes. They are as follows:

-------------------------------------------------------------------------------
CONVERGENCE TESTS INFORMATION
-------------------------------------------------------------------------------
1. Forced Convection using surface convergence (Loftus et al. 2008)
2. Initial moist tropical (MT) sounding (Dunion 2011)

To generate convection via convergence, there are a number of parameters
that have to be tuned that impact the nature of the convection.
1. radius of convergence
2. depth of convergence
3. amplitude of convergence
4. threshold for stopping convergence either by meeting time or vertical
   velocity criteria

From initial tests and from Loftus et al. (2008) it seems that representative
convection is generated for convergence width of 5km (represented by number
of grid points in RAMSIN) and depth of 1km. These need to be tested and
varied depending upon the sounding.

Chose to try Radius of convergence of 7km to see if I can get a larger 
area of convergence. This did make a larger area with larger anvil and
slightly stronger updrafts. Might go with this size for further experiments.

This set of tests look rather good in terms of updraft and anvil production.
Max W varies from 20s up to 42m/s.

Interpolated Cindy Twohy's aerosol profiles to the RAMS levels and made
appropriate conversions between her data (#/cm3) and RAMS (#/kg).

-------------------------------------------------------------------------------
FINAL DECISION ON CONTROL TESTS
-------------------------------------------------------------------------------
RESOLUTION: DX=250m DZ=50-250m, NNXP=350 NNYP=350 NNZP=108, DTLONG=4

At this point I like the 7km radius, 4.5e-6 conergence and convergence stop at
3600s. This give an Max W just under 40m/s, 50min of simulation with W>10m/s,
a good sized elliptical anvil, good dust removal, a single convective pulse,
good looking cross-sections of condensate and W.

Reducing the radius of convergence, the amplitude, and/or the duration of 
applied convergences gives smaller storms, weaker W, much less anvil, less 
dust processing. 

Used only the "MT" sounding for production runs to look at dust
as microphysically active CCN and/or INP.

1. 07km.4.50e-6conv.nodust.mt        - No dust
2. 07km.4.50e-6conv.dustbg.mt        - Background (5%) dust
3. 07km.4.50e-6conv.dustccnin.mt     - Dust as CCN and INP
4. 07km.4.50e-6conv.dustin.mt        - Dust as INP only
5. 07km.4.50e-6conv.dustccn.mt       - Dust as CCN only
6. 07km.4.50e-6conv.dustin.mt.tlimit - Dust as INP only, 
                                       dont apply DeMott below 236K

###############################################################################

*******************************************************************************
proj.2017-doe.micro-multi-6.2.11-FLOYD / runrams.supercell.grant.6.2.12
*******************************************************************************

These are tests of the rain self-collection routine to examine how rain drop
breakup occurs if we change the collection efficiency curve and negative E
cutoffs relative to the default of E >= -5.

Using the control RAMSIN from Leah Grant based on her JAS
papers in 2014,2015. But my tests here are much coarser spacing
for simplicity. I'm using her physics specs, but ZROUGH=0.0001.
Freeslip flag is set to on, so SFLUX surface terms are zero.

Leah Grant sounding (JAS 2014,2015, Weisman-Klemp type sounding)

1.supercell.grant.orig-revised = CONTROL experiment with:
  a. standard collection efficiency coefficient (0.1326e7)
  b. DMB1 (rain) = 0.3mm
  c. E (min) = -5.0
  d. GNU (rain) = 2

The tests below are same as the CONTROL except for the described change:

2.supercell.grant.min-5.0.gnu2.dmb0.1 = same as (1) but with DMB1 (rain) = 0.1mm

3.supercell.grant.effxy-0.80e6 = changing shape of "eff" curve by changing
  exponential coefficient from 0.1326e7 to 0.80e6. This pushes the curve outward
  so that is crosses the zero line at larger size drops.

4.supercell.grant.effxy-0.50e6 = changing shape of "eff" curve by changing
  exponential coefficient from 0.1326e7 to 0.50e6. This pushes the curve outward
  so that is crosses the zero line at larger size drops.

5.supercell.grant.min-1.0 = changing E(min) from -5.0 to -1.0 to limit the strength
  of breakup. The choice of -5.0 seems arbitrary and there is no documentation 
  for why to use that value.

6.supercell.grant.min-0.5 = changing E(min) from -5.0 to -0.5 to limit the strength
  of breakup.

7.supercell.grant.min-0.3 = changing E(min) from -5.0 to -0.3 to limit the strength
  of breakup.

8.supercell.grant.min-0.1 = changing E(min) from -5.0 to -0.1 to limit the strength
  of breakup.

9.supercell.grant.min-0.0 = changing E(min) from -5.0 to -0.0 to limit the strength
  of breakup.

10.supercell.grant.min-1.0.gnu1 = changing E(min) from -5.0 to -1.0 and GNU (rain)
  from 2 to 1.

11.supercell.grant.min-0.1.gnu1 = changing E(min) from -5.0 to -0.1 and GNU (rain)
  from 2 to 1.

12.supercell.grant.min-0.3.gnu1 = changing E(min) from -5.0 to -0.3 and GNU (rain)
  from 2 to 1.

13.supercell.grant.min-0.3.gnu4 = changing E(min) from -5.0 to -0.3 and GNU (rain)
  from 2 to 4.

14.supercell.grant.min-0.3.gnu3 = changing E(min) from -5.0 to -0.3 and GNU (rain)
  from 2 to 3.