Sensitivity of Supercell Tornado Simulations to Variations in Microphysical Parameters

1
Sensitivity of Supercell Tornado Simulations to
Variations in Microphysical Parameters Nathan
Snook and Ming Xue School of Meteorology,
University of Oklahoma, Norman OK, U.S.A.

• Introduction
• Tornadoes spawned by supercell thunderstorms are
  a major severe weather hazard in the central
  United States, causing multiple fatalities and
  millions of dollars in damage each year.
• Accurate numerical simulation of tornadic
  supercells remains a challenge, as the solution
  is affected by grid resolution and model
  parameters.
• Commonly used microphysical schemes in NWP models
  assume a dropsize distribution based on that
  observed by Marshall and Palmer (1948) for some
  or all hydrometeor species explicitly predicted.
• Observational studies of Marshall-Palmer
  intercept parameters for rain, snow, and hail
  have yielded values that vary by several orders
  of magnitude (Gilmore et al., 2004).

Results
Table of 100 m Experiments

• Coarse resolution simulations revealed that cold
  pool intensity was most sensitive to rain and
  hail intercept parameters, and less sensitive to
  snow intercept parameter and hail density, as
  seen in Fig. 1.
• Simulations with large raindrops and hailstones
  produced weak cold pools, while small raindrops
  and hailstones produced strong cold pools due to
  enhanced evaporational cooling.

Fig. 1) Plot of cold pool intensity for 1 km
resolution simulations. Explanation of
experiments not conducted at 100 m resolution s7
and s8 have an increased snow intercept
parameter, d400 has decreased hail density.
Kessler uses warm rain microphysics.
Large Raindrops (r5) Maximum intensity
f2 Duration 9 min.
Control (CON) Maximum intensity f2 Duration 4
min.

• Simulations favoring large hydrometeors (weak
  cold pools) were observed to be most favorable
  for formation of long-lived tornadoes.
• Tornadic spinups in simulations favoring small
  hydrometeors (strong cold pools) were weaker and
  more short-lived than those in simulations
  favoring large hydrometeors (weak cold pools).
  Long-lived surface tornadic vortices are noted in
  Fig. 2.

• Conclusions
• Of the variables studied, the rain intercept
  parameter appears to have the most influence on
  supercell dynamics, followed by the hail
  intercept parameter.
• Organizational mode, storm propagation, gust
  front location, and tornado potential were all
  strongly influenced by variation in microphysical
  parameters.
• Simulations with weaker cold pools produced more
  vertically oriented updrafts, while simulations
  with strong cold pools tended to produce updrafts
  that tilted westward with height.
• Dropsize distributions favoring large raindrops
  and large hailstones (small rain and hail
  intercept parameters respectively) result in
  weaker cold pools and greater potential for
  long-lived tornadoes due to more favorable
  updraft orientation and vertical alignment of
  low- and mid-level vorticity maxima.
• Varying intercept parameters alone is enough to
  affect whether or not tornadoes form.

Fig. 2) Timeseries of maximum vertical vorticity
in the lowest 2 km of the atmosphere for
high-resolution simulations. Long-lived surface
tornadic circulations are noted, along with their
f-scale intensity and duration.

• Objectives
• Investigate the sensitivity of supercell storm
  dynamics to variation in Marshall-Palmer
  intercept parameters for rain, hail, and snow
  dropsize distributions, and hail density.
• Cold Pool Intensity
• Organizational Mode
• Precipitation Distribution and Intensity
• Explore the impacts of these effects on tornado
  potential and tornado formation.

• In simulations with stronger cold pools, the gust
  front was stronger and propagated eastward more
  quickly, often advancing several kilometers ahead
  of the storm.
• A more linear storm mode was favored in the
  simulation with the strongest cold pool (h6r7,
  pictured on the right of Fig. 3a).

Fig. 3b) Zoomed-in view of the tornadic vortex
circulation in r5. Plotted are radar
reflectivity (color-fill), vertical vorticity
(contour), and surface wind vectors.
Fig. 3a) Plots of cold pool strength (shaded),
vertical vorticity (color-fill), radar
reflectivity (contour) and wind vectors for a
simulation favoring large raindrops (r5, left)
and one favoring small raindrops and hailstones
(h6r7, right). The black box in the left plot
indicates the area plotted in Fig. 3b.

• Data and Methods
• The Advanced Regional Prediction System (ARPS)
  was used to numerically simulate supercell storms
  initialized using a thermal perturbation
  superimposed on a horizontally homogeneous base
  state derived from a sounding associated with the
  May 20, 1977 tornadic supercell near Del City,
  Oklahoma.
• 18 runs were conducted at coarse (1 km)
  horizontal grid resolution to determine which
  parameters were most influential in supercell
  dynamics.
• 7 runs varying the most influential parameters
  within the range of published observations were
  conducted at uniform 100 m horizontal grid
  resolution.

• The positioning of the gust front in simulations
  favoring large hydrometeors (weak cold pools)
  allowed for stronger updrafts with a more
  vertical orientation than in simulations favoring
  small hydrometeors (strong cold pools).
• Simulations with strong cold pools exhibited more
  pulse-like updraft behavior and fewer supercell
  characteristics, as seen on the right of Fig. 4.

Acknowledgement This research was conducted
as part of the Center for Collaborative Adaptive
Sensing of the Atmosphere (CASA), and wasfunded
in part by NSF grant EEC-0313747 of
the Engineering Research Center Program. For
further information, contact Nathan Snook at
nsnook_at_ou.edu
Fig. 4) Vertical cross-sections of radar
reflectivity (color-fill), cold pool intensity
(shaded), and wind vectors for a simulation
favoring large raindrops (r5, left) and one
favoring small raindrops and hailstones (h6r7,
right).
Gilmore, M. S., J. M. Straka, and E. N.
Rasmussen, 2004 Precipitation uncertainty due to
variations in precipitation particle parameters
within a simple microphysics scheme. Mon. Wea.
Rev., 132, 2610-2626. Marshall, J. S., and W. M.
Palmer, 1948 The distribution of raindrops with
size. J. Meteor., 5, 165-166.
References