WRF-ARW model forecast track sensitivities of tropical cyclone Ernesto (2006) to various parameterizations, grid spacings, and initial conditions [e-mail] Advisor: Michael Morgan University of Wisconsin-Madison

1
WRF-ARW model forecast track sensitivities of
tropical cyclone Ernesto (2006) to various
parameterizations, grid spacings, and initial
conditions npbassill_at_wisc.eduAdvisor Michael
MorganUniversity of Wisconsin-Madison

• Overview of Ernestos Life Cycle1
• Ernesto originated from an African easterly wave
  on 24 August 2006 over the southern Leeward
  Islands.
• While in the eastern Caribbean, Ernesto was
  classified a tropical storm, and was forecast to
  become a powerful hurricane.
• However, Ernesto strongly deviated from its
  forecast track, surprising forecasters in the
  process.
• This deviation began while Ernesto was south of
  the Dominican Republic.
• NHC was aware of this possibility, stating in
  their 15 UTC 26 August forecast discussion
• THE FORECAST TRACK IS MORE PROBLEMATIC AFTER
  72 HR. THE LARGE-SCALE MODELS AGREE THAT THE
  MID-LEVEL RIDGE OVER THE GULF OF MEXICO WILL
  WEAKEN AS A SHORTWAVE TROUGH DIGS SOUTHEASTWARD
  THROUGH THE MISSISSIPPI VALLEY. HOWEVER ... THERE
  ARE DIFFERENCES IN HOW MUCH WEAKENING WILL OCCUR.
  THE ECMWF AND NOGAPS CALL FOR ERNESTO TO RECURVE
  OVER THE EASTERN GULF OF MEXICO ... WHILE THE
  UKMET SHOWS ENOUGH RIDGE TO KEEP ERNESTO MOVING
  WEST-NORTHWESTWARD
• The following is a study of the track and
  intensity sensitivity of Ernesto to different
  grid spacings, parameterizations, and initial
  conditions.
• Experiment 1
• Model Configuration

Experiment 1 Continued
Experiment 4 - Exp. 1 was redone using all
combinations of CPs and MPs with the same domain
with a grid spacing of 30 km and 31 vertical
levels (for a total of 9 simulations), holding
all else the equal. - However, now these
forecasts were initialized with GFS 1.0 data
from 12 UTC 26 August (instead of .5 data).

• Left Plotted is the difference in thickness of
  the 300-700 hPa layer for the two previously
  mentioned runs (west-moving minus north-moving).
  The fill interval is 10 m. The red (blue) numbers
  show relative sea level pressure minima for the
  north-moving (west-moving) run. From top to
  bottom, forecast hours 30, 60, and 90 are shown.
• Observations
• Shortly after hour 30, a significant positive
  thickness difference develops to the northwest of
  the cyclone(s), which would act to influence the
  track of Ernesto towards a more westward
  direction.
• This corresponds to the timing of the track
  split.
• Of note is that the simulation with the
  north-moving cyclone develops a tropical cyclone
  in the far eastern Pacific, while the west-moving
  cyclone simulation does not (this is typical of
  all simulations with these respective tracks).
• This Pacific cyclone was a real phenomenon, and
  is the analog of Hurricane John
• Although not a focus of this study, this
  suggests that the real-life northward movement of
  Ernesto allowed Hurricane John to form, likely
  because a westward track would create unfavorable
  shear over the potential Pacific cyclone.

Left NHC best track for Ernesto Right 15 UTC
26 August 5-day track forecast. NHC warns
Ernesto could become a potentially dangerous
hurricane as it moves across the northwestern
Caribbean Sea and the Gulf of Mexico

• Left Forecast tracks for 1.0 (dashed) and 0.5
  (solid, from Exp. 1). Every point represents 30
  hours, and each color represents a particular
  CP-MP combination. The actual track is shown in
  red.
• Exp. 4 Observations
• - There is a distinct northeastward shift of the
  forecast tracks with the 1.0 data.
• This would result in smaller forecast errors,
  even though 0.5 data theoretically has better
  initial conditions.

• Experiment 5
• As in Exp. 4, but now Exp. 3 is redone with GFS
  1.0 data.
• Right Track points and intensity of the 3 runs
  at 3 hour and 15 hour increments, respectively.
  The observed track and NHCs forecast track are
  also shown.
• Exp. 5 Observations
• Unlike Exp. 4, a smaller grid spacing in
  conjunction with 1.0 data prevents the storm
  from intensifying or taking a more realistic
  track this results in a larger grid spacing
  providing a more accurate forecast! (Exp. 4 vs.
  Exp. 5).

Right 90 hour accumulated precipitation
difference (in inches) of the west-moving run
minus the east-moving run.
- The positive precipitation anomalies over the
Mexican plateau and along the Ohio River likely
contribute to the above thickness anomaly through
latent heat release.

• Experiment 2
• An additional 9 simulations were conducted using
  a 60 km horizontal grid spacing, 31 vertical
  levels, and all possible combinations of CPs and
  MPs on a much larger domain (right), with all
  else the same as Exp. 1.
• This was done to determine if the boundary
  conditions used in Exp. 1 forced the above
  anomalies, and therefore forecast tracks.

Left Resultant 9 forecast tracks. Every point
represents 30 hours of forecast. This plot also
shows the full domain that was used.

• Left Difference of 800 hPa vorticity (1.0- 0.5
  s-1), and location of Ernesto at initial time
  (Red for 1.0, Blue for 0.5).
• Lower Left Zoomed in 9-hour accumulated rainfall
  difference (1.0-0.5 in inches), sea level
  pressure difference at 9 hours (1.0-0.5 in mb),
  and location of Ernesto at 9 hours (Red for 1.0,
  Blue for 0.5) using Kessler runs.
• Further Observations
• 0.5 data initializes Ernesto as a more compact
  system, whereas 1.0 data smears out initial
  vorticity (see squares, above left).
• This allows for more organized initial
  convection (with 0.5 data), which produces more
  latent heat release and allows the storm to
  intensify more realistically from the beginning.
• Unlike runs using larger grid spacings, in this
  case inner core dynamics do influence forecast
  track.
• More intense storms as in Exp. 3 are steered to
  the northwest whereas weaker storms in Exp. 5 are
  steered by the low level flow to the west.
• Using a smaller grid spacing makes the
  differences between 1.0 and 0.5 data more
  important.

Left Resultant 48 track forecasts. Every point
represents 30 hours of forecast. Actual track is
red, while NHC forecast is grey. Times for both
are shown. This plot also shows the full domain
that was used.

• Observations From Experiments 1 and 2
• - Choice of MPs and CPs are most important in
  determining forecast track.
• - Choice of horizontal and vertical grid spacing
  modifies intensity, but not track philosophy.
• - Larger domain does not alter forecast track
  philosophy.
• - At these grid spacings, the Kessler MP scheme
  performs best.
• Westward moving Ernesto runs prevent realistic
  formation of Hurricane John.
• Track difference appears to be a result of
  large-scale steering differences, and not a
  result of inner-core dynamics.
• - Eta/Ferrier MP scheme is generally unable to
  intensify Ernesto (below).

Table Average minimum central pressure of storms
entering the Bay of Campeche for various grid
spacings. - Even though the tracks and synoptic
environments for these storms are nearly
identical, the use of the Eta/Ferrier MP seems to
hinder intensification.
?x, ?y Eta/Ferrier WSM6 ?MSLP
30 km 989 mb 952.5 mb 36.5 mb
45 km 1001 mb 981.3 mb 19.7 mb
60 km 1002.7 mb 991.3 mb 11.4 mb

• Experiment 6
• As in Exp. 4, but now using 00 UTC 26 August GFS
  1.0 data.
• Right Tracks for 00 UTC 1.0 data (solid) and 12
  UTC 1.0 data (dashed). Track points are
  normalized to be at the same forecast time (ex.
  42 hours from 00 UTC, 30 hours from 12 UTC).
• Exp. 5 Observations
• Generally similar forecasts, although some 00
  UTC forecasts are better able to forecast the
  initial jump to the north.
• Similar intensity issues when using the
  Eta/Ferrier MP scheme.

• Experiment 3
• 3 additional runs using a smaller domain and
  smaller grid spacing (4 km, 31 levels) .
• No CP scheme, 3 previously mentioned MP schemes
  used, all else as in Exp. 1 and Exp. 2.

• - It was presumed that some type of large-scale
  steering difference is the cause of the split.
• To examine why this split occurs, comparisons
  between a representative simulation of each was
  conducted.
• Left These are the two runs chosen. Both use a
  30 km horizontal grid spacing, 31 vertical
  levels, and use the Kain-Fritsch CP for
  consistency. The red track uses the Kessler MP
  and the blue uses the WSM6 MP.
• 1Information in this section was obtained from
  the National Hurricane Centers website
  (http//www.nhc.noaa.gov/)
• 2Curiously, any combinations using 54 vertical
  levels and the Betts-Miller-Janjic CP fail due to
  unknown boundary condition problems, reducing the
  number of runs to 48.

• Left Track points and intensity of the 3 runs at
  3 hour and 15 hour increments, respectively. The
  observed track and NHCs forecast track are also
  shown. This is the actual domain used.
• Exp. 3 Observations
• - Smaller grid spacing allows WSM6 MP scheme to
  perform best
• This could be a result of the ability to resolve
  graupel producing updrafts with this grid
  spacing, which should favor WSM6 over other MP
  schemes
• Smaller grid spacing does not allow Eta/Ferrier
  scheme to comparably deepen the cyclone, even
  without the involvement of CP schemes
• Note Ernesto is initialized much too weak

• Concluding Thoughts
• - Perhaps modeling focus should be on improving
  parameterizations, for two primary reasons
• These experiments demonstrate that choice of MP
  and CP influence track and intensity more than
  choice of grid spacing or vertical resolution (at
  least for this case at the times chosen).
• Using a smaller grid spacing can be detrimental
  to model forecast with certain initial
  conditions.
• Future Work
• - Test other MPs, initial conditions,
  initialization times, and tropical cyclones to
  verify these results.