Contribution of Shower Clouds to the Global Circuit (Alaka and Stolz)

Diurnal Variations of Global
Thunderstorms and Electrified Shower
Clouds and Their Contribution to the
Global Electrical Circuit
Liu et al. 2010
Review By:
Gus Alaka and Doug Stolz
20 February 2014
ATS 780
• C. T. R. Wilson’s Hypotheses: Thunderstorms AND electrified shower clouds are the “batteries” of
the global DC electrical circuit. Adopted (with permission) and later tested by F. Whipple.
Amplitude of global
thunderstorm area
over land is 2x that
of Carnegie Curve.
Phase lag between
diurnal maximum
chimney/TStormsAfrican chimney).
• Objective: To update and improve our understanding of the physical origins of the global circuit of
atmospheric electricity using the most extensive collection of simultaneous satellite observations
of precipitation and lightning in the Tropics (i.e., TRMM Precipitation Radar and Lightning Imaging
Sensor database).
Evidence For Electrified Shower Clouds
• Mach et al. (2009) and Blakeslee and Mach (2009) documented E-fields
between 500-1200 V m-1 in a tropical squall line during TOGA-COARE – the
absence of lightning was noted.
Takahashi (1973)
• Takahashi (1973) modeled the
hydrometeor charging in warm rain clouds.
• Rain, drizzle, and cloud droplets can charge
positively or negatively depending on the
stage of cloud development.
• Charged hydrometeors account for in-cloud
charge density of about 10-2 C km-3.
Evidence For Electrified Shower Clouds
Takahashi (1978)
“…At about 2100 LT, a cumulus cloud developed east
of the island and rain began to fall as the cloud moved
onshore and approached the observation site… The
electric potential gradient rapidly reversed in sign
from positive to negative as rainfall commenced at
the station. As the rain shower moved away from the
station, the potential gradient slowly recovered to
zero…Most of the raindrops measured were relatively
small and positively charged… There was no evidence
of lightning in any of the warm clouds studied in
Ponape…” –Takahashi (1978)
Feature Partitioning
• How do they segregate electrified shower clouds from shallow cumuli and
1. Invert the arbitrary 10% threshold
from Fig. 1: 90% of the PF’s with the 30
dBZ echo tops at -17C (-10C) over
ocean (land) DO NOT produce
2. Define: Electrified shower clouds are
those with PF’s without lightning and
30 dBZ echo top temperatures warmer
than the empirical temperature
thresholds (consistent with Fig. 8 of
Cecil et al. 2005).
The PF Breakdown
Thunderstorms and Electrified Shower Clouds account for an
overwhelmingly small portion of the total precipitation feature
population but they contribute roughly 40% of total rainfall.
Intuitive: thunderstorms are about 3x more common over land.
PF’s over ocean rain more (…the role of aerosols)?
Counter-intuitive: electrified shower clouds are also more
common over land.
So lets investigate…
…Thunderstorms and Electrified Shower Clouds…
…as drivers of the global electrical circuit on annual and seasonal
time scales.
Annual Average Diurnal Cycle
Total Rainfall
Rainfall over land matches the
max/min of the Carnegie electric
field (CEF)
Too much rain over the Maritime
Continent to explain CEF
Rainfall heavily weighted toward
ocean, which has weaker diurnal
Annual Average Diurnal Cycle
Rainfall in Thunderstorms
This is the best match
Diurnal variation is higher over
land and lower over the ocean
Rainfall heavily weighted toward
Annual Average Diurnal Cycle
Rainfall in Electric Shower Clouds
Agreement with CEF is
substantially worse
Diurnal cycle of ESC’s over ocean
is very weak
Most ESC’s are over land
-- (0.34% vs. 0.19%)
Annual Average Diurnal Cycle
Rainfall Fractions, Rain/Flash
Max rainfall in the Americas from
Min tstorm rainfall when the sun
is over the Pacific
Max rain/flash over Maritime
Blue arrow
This increase is
only 4%!
-- Clean marine air supports bigger droplets
Min rain/flash over Africa
-- Aerosol-rich continental air supports longlasting, more mature clouds
-- slightly higher for Americas
In the Americas, rain/flash is similar to Africa, but
Americas also have a higher rain fraction, so…
more flashes in the Americas??
We know this isn’t true…
Seasonal Average Diurnal Cycle
Flash Counts
Very pronounced diurnal cycle for flash
Same problem as in Whipple (1929) since
most tstorms occur over land.
Africa dominates flash counts year
Seasonal Average Diurnal Cycle
Rainfall Over Land
A better match…
Max variation (21Z) matches fairly
Abundance of non-electrified
rainfall in Maritime Continent
mutes diurnal cycle
-- especially DJF
Americas dominate rainfall most
of the year
Seasonal Average Diurnal Cycle
Rainfall in Thunderstorms
Once again, best match
Weakest variation in DJF
Might be due to lower quality of electric
field over Vostok due to weather
Tstorm rainfall variations are more
consistent region to region
Seasonal Average Diurnal Cycle
Rainfall in Electrified Shower Clouds
Weaker diurnal variation for ESCs
in all seasons
More ESCs in Americas than in
ESCs are not a good measure of
the fair weather electric field
Liu et al.’s Elusive Conclusion About
Electrified Shower Clouds
“Nevertheless, with all things considered, it seems unlikely that the
contribution of the electrified shower clouds to the global circuit
will be either completely dominant or entirely negligible in
comparison with the thunderstorm contribution. This result is in
keeping with the predictions of Wilson (1920).”
Continued Uncertainty…
“In a hierarchical global population of convective clouds, one
expects…more electrified shower clouds than thunderclouds…”
• The method employed by Liu et al. represents an attempt at characterizing the
relative contributions of thunderstorms and electrified shower clouds to the
global DC circuit, but their arbitrary partitioning scheme leaves room for
• Are electrified shower clouds lumped into thunderstorm PF’s so that the populations
are inaccurately represented?
• Strength in numbers – separately electrified shower clouds may make small
contributions, but together they represent an appreciable component of the
DC global circuit. More observations/modeling studies of hydrometeor
charging are called for.
Food For Discussion
• Liu et al.’s major conclusion (or suggestion) is that the global DC circuit
responds mostly to precipitation, used as a proxy for charge separation herein.
Knowing what you do about the currents associated with common lightning
discharges and patterns in global rainfall,
• How might the sampling strategy or sampling capability of TRMM LIS
contribute to the first of the two main discrepancies between lightning trends
and the Carnegie Curve (i.e., that the amplitude of the diurnal lightning signal
is 2x that of the Carnegie Curve)?
More Take Home Points
• Electric shower clouds reduce the amplitude variation of the global circuit
• Contribute negative charge to surface via precipitation
• Difference from Whipple (1929):
• ESCs are more prevalent over land than ocean
• Africa dominates flash rates
• Americas dominate electrified rainfall (more flashes?)
• Majority of oceanic rainfall associated with non-electrified convection
What did they miss by their methods for selecting ESCs?
How reliable is 6-hr NCEP reanalysis fields interpolated to the scale of a PF?
Lightning flashes could be from dissipating MSCs
Sub-pixel classification
• 1 thunderstorm cell is likely surrounded by several ESC cells
• Overall, the findings of Wilson (1920) and Whipple (1929) are supported.
Extra stuff
Sub-PF Scale Considerations
• TRMM PR has a horizontal footprint of 4.5 km x 4.5 km which is larger than
many shower clouds that may be electrified (one caveat in its own right).
• Grouping raining pixels into PF’s potentially includes electrified shower
clouds in thunderstorms.
= Raining
= Raining/T30dbz
= Raining/T30dbz/Lightning
• Both of the above PF’s would are classified as thunderstorms when the
example on the left is likely to contain electrified shower clouds.
Carnegie Curve
Annual Average Diurnal Cycle
Carnegie Curve
Seasonal Average Diurnal Cycle
Carnegie Curve
Seasonal Average Diurnal Cycle

similar documents