GATropicsDevelopment_SCMexperiments - ACCESS-NRI/accessdev-Trac-archive GitHub Wiki
#!div style="max-width: 800px; margin: 0 auto;color: red;"
[[PageOutline]]
# **Single Column Model (SCM) Experiments**
#!div style="max-width: 800px; margin: 0 auto;"
[[PageOutline]]
[{
#!html
<h1 style="color: blue; font-size: 90%">(Abhik Santra, Christian Jakob, Hongyan Zhu, Harry Hendon, Matthew Wheeler and Jing-Jia Luo)</h1>
In this section, Unified Model (UM) Single Column Model (SCM) is employed to understand the various physical processes, especially the model's convection.
UM SCM is an idealized setting, allows complete control over the large-scale and lower boundary (e.g. sea surface temperature, land-surface parameters) conditions. Other than these, various advective tendencies of heat and moisture fluxes are also specified during the simulation.
All model runs, default (CTRL) and experiments (EXP), are 15 days long (360 hours), starting from 9th of January,1993 at 16:56 local time. The model is run over the Western tropical pacific at 2^o^S, 156^o^E, based on TOGA-COARE observations. The model time-step used in all simulations is 12 minutes, while the boundary layer and convection schemes have sub-timestep of one third of this.
EXP 1: Reduction of Deep Convective Entrainment Coefficients
The deep convective fractional entrainment coefficients, aekp14 and aekp34, are reduced to 10% of their respective default values. The idea behind this experiment is to enhance the precipitation amount by reducing the dry air entrainment in the column. The corresponding entrainment profile is shown here.\ * Results\ The evolution of the deep convective precipitation (mm hr^-1^) at various periods are shown here.
Precipitation intensity is increased and it's too noisy.
EXP 2: Increase of Deep Convective Entrainment Coefficients
In this experiment, aekp14 and aekp34 are increased by 1.5 times of their respective default values. It is noteworthy to mention that similar change was implemented in GA3.0 by Bush et al. (Q. J. R. Meteorol. Soc., 141:311-326. 2015) over the tropical regions of large precipitation bias.\ * Results\ The evolution of the deep convective precipitation (mm hr^-1^) at various periods are shown here\ Precipitation intensity is reduced, however, it persists for longer duration.
EXP 3: Alternative deep entrainment profile
The deep convective entrainment profile is expressed as e = f,,dp,,3A,,E,,(p/p,,,,)^ndp^1/p,,,,\ where ndp>=1.0. Values of ndp > 1.0 give profiles in which the entrainment rate decreases more rapidly with height.\ In this experiment, sensitivity of ndp on deep convective precipitation is examined. The corresponding updraft entrainment profiles are available here.\ * Results\ The evolution of the deep convective precipitation (mm hr^-1^) for ndp # 2 and ndp 3 are compared to CTRL with ndp = 1.0. \ Precipitation intensity appears to be proportional to the values of ndp.
EXP 4: Reduction of Cloud Base Mass-flux
The deep convective cloud base mass-flux is reduced by increasing the CAPE time-scale. In this experiment, the CAPE time-scale is doubled to its default value to get half of the default cloud base mass-flux.\
- Results\ The evolution of the deep convective precipitation (mm hr^-1^) relative to CTRL is shown here\
Precipitation intensity decreases and it smoothly evolves in time.
#!html
<center>
<a href="http://cawcr.gov.au"><img src="http://cawcr.csiro.au/wp-content/uploads/2015/08/CAWCR-logos-e1440147415279.png" alt="CAWCR Logo" height=50></a>
<a href="https://go8.edu.au"><img src="https://go8.edu.au/sites/default/files/images/institution/monash.png" alt="Monash Logo" height=50></a><br/>
<a href="http://www.climatescience.org.au"><img src="http://www.climatescience.org.au/sites/all/themes/coeclean/images/coe/logo_coecss.jpg" alt="ARCCSS Logo" height=50></a>
</center>
</div>
