NWS HUN Training

1
Monitoring Satellite Data for Day- and Night time
Convective and Lightning Initiation John R.
Mecikalski1, Kristopher M. Bedka2 Simon J.
Paech1, Todd A. Berendes1, Wayne M.
Mackenzie1 1Atmospheric Science
Department University of Alabama in
Huntsville 2Cooperative Institute for
Meteorological Satellite Studies University of
Wisconsin-Madison Supported by NASA ASAP
Initiative NASA New Investigator Program (2002)
2
How this began


3
How this began

• Which cumulus will become a thunderstorm?
• GEO satellite seems to be well-suited to address
  this question.
• What methods are available?
• What changes to current, globally-developed
  codes are needed?
• Who can benefit from this research?
• What user groups are interested (e.g., 0-2 h
• nowcasting)

4
Where we are today

• Cumulus-filled land
• 0-1 h CI nowcasting
• CI climatology studies
• CI kinematic studies
• New research in 1-2 h Lightning nowcasting
• Testbed for short-term (0-6 h) prediction
• Satellite data assimilation research (6-24 h)

UAH
5
Background/Motivation

• There are many cumulus, and only a few are
  capable of growing to thunderstorms.
• Cumulus thus are identifiers of surface forcing
  for vertical motions (mass and moisture
  convergence), and the eventual formation of
  organized updrafts of significant scale O(1-5
  km)
• Especially in the Tropics, cumulus are the
  landscape, for which describing them allows one
  to say something about (e.g., quantify) the
  atmosphere in which they are in (see Riehl 1954).
• Satellites (especially geostationary)
• are able to monitor cumulus over time.
• Tracking cumulus, while
• monitoring their growth, should
• allow for some predictive
• skill of thunderstorm development.
• Limited to 1 km cumulus in clear-sky
• conditions.

6
Background/Motivation

• Numerical models have significant problems
  nowcasting location/intensity of convective
  weather phenomena in the 0-6 hour time frame
• This is especially true over oceanic regions
  where poor initialization results in incorrect
  location/intensity forecasts for convective
  storms
• Since little real-time satellite-derived data is
  available in airplane cockpits, coupled with NWP
  deficiencies, mid-flight convective storm
  initiation and growth represents a significant
  hazard for aviation interests
• A major portion of the accidents from aircraft
  turbulence encounters are within close proximity
  to atmospheric convection (Kaplan et al, 1999)
• The cost of diverted flight can be as high as
  150,000 and a cancellation close to 40,000,
  depending on the size of the plane (Irrgang and
  McKinney, 1992)

7
Evaluation of Pre-CI Satellite Signatures

• Build relationships between GOES-12 and WSR-88D
  imagery
• Studied numerous real-time and archived
  convective events with diverse mesoscale forcing
  regimes and thermodynamic environments
  continental (U.S. Great Plains) to sub-tropical
  (S. Florida)
• Identified GOES IR TB and multi-spectral
  technique thresholds and time trends present
  before convective storms begin to precipitate
• Leveraged upon documented satellite studies of
  convection/cirrus clouds Ackerman (1996),
  Schmetz et al. (1997), Roberts and Rutledge
  (2003)
• After pre-CI signatures are established, test on
  other independent cases to assess algorithm
  performance

8
Mesoscale Atmospheric Motion Vector Algorithm
Operational Settings
New Mesoscale AMVs (only 20 shown)

• We can combine mesoscale AMVs with sequences of
  10.7 ?m TB imagery to identify growing convective
  clouds, which represent a hazard to the aviation
  community

9
Mesoscale Atmospheric Motion Vector Algorithm
Operational Settings
New Mesoscale AMVs (only 20 shown)

• We can combine mesoscale AMVs with sequences of
  10.7 ?m TB imagery to identify growing convective
  clouds, which represent a hazard to the aviation
  community

10
Cloud-Top Cooling Rates for CI Assessment

• Study of co-located 10.7 ?m TB and radar
  reflectivity trends for stationary convection
  along the Colorado Front Range
• Found that sub-freezing 10.7 ?m TBs and
  4C/15mins (8 C/15mins) correspond to weak
  (vigorous) growth

15 min ?TB
By monitoring, via satellite, both the cloud
growth and the occurrence of sub-freezing
cloud-top temperatures, the potential for up to
30 mins advance notice of convective storm
initiation ( 35 dBz), over the use of radar
alone, is possible.
Roberts and Rutledge (2003), Wea. Forecasting
11
CI Nowcast Algorithm 4 May 2003
2000 UTC
CI Nowcast Pixels

• Satellite-based CI indicators provided 30-45 min
  advanced notice of CI in E. and N. Cent. KS

12
CI Nowcast Algorithm 12 June IHOP

• Since 5 min GOES-11 data was used, time trend
  thresholds are cut in half, resulting in noisy
  nowcasts for quasi-stationary convection in New
  Mexico
• TX Panhandle/OK convective development captured
  well

13
CI Nowcast Algorithm 3 August 2003

• Complex convective forcing from upper-level cold
  core cyclone, combined with lake breeze
  circulation
• Although noisy at first glance, CI over
  central/western IL identified up to 1 hour in
  advance
• Objective validation methodology very difficult
  to develop

kkoooooooookkkkkkkkkkk
1815 UTC
14
Interest Field Importance

• All eight interest fields are NOT necessarily
  important for predicting future
• locations of a 35 dBZ radar echo. They carry
  different weights.
• There is a strong convective regime dependence
  on what fields are most
• important.

15
Interest Field Importance

• Deep convection, dry upper troposphere.
• Best for high CAPE environments, and strong
  updrafts.
• Winter-time, Midlatitudes

16
Interest Field Importance

• Moist upper troposphere, warm-top convection.
  Shallow convection.
• Low CAPE environments (tropical, cold-upper
  atmosphere). f(?Physics)
• Optimal in Tropics during summer.

17
Limitations

• Algorithm has a difficult time with small
  cumulus on the order of less than 2km.
• If cirrus is present with in a pixel, it will be
  identified as cirrus and will not monitor that
  pixel.
• Not available operationally at night at this
  time.
• Has some difficulty monitoring previous
  convection for redevelopment (depends on
  convective cloud mask classification).

18
Advantages

• Excellent within a synoptically benign
  environment where cumulus are present.
• Excellent for monitoring areas of growing or
  towering cumulus.
• Cloud top cooling rates can provide some
  information into updraft strength.
• Easy monitoring of fast moving cumulus for
  development.

19
CI Validation
Conditional POD
Conditional FAR
20
Lightning Initiation

• Lightning (event) Initiation
• Linear regression of lightning trends against
  GOES interest fields (once lightning and
• GOES are in same reference frame parallax
  corrected) shows a 95 significant correlation
• between cloud-top cooling rates on 1 km scales
  and the amount of total lightning observed
• (in-cloud cloud-to-ground) 30-60 min nowcasted
  source density increases over 1 km pixels.

We are currently looking at using 10.7-3.9
(reflected component removed) difference to
determine cloud top phase which can aide in
Lightning Initiation.