ICAE International Commission on Atmospheric Electricity

ICAE 2003 Versailles

ICAE Home Page

General Information

Accommodation /
Accompanying Person

Benjamin Franklin

Conference Format

Notice to Chairpersons


Index to Authors




Monday 9th June



Session B2 Electrical Activity and Meteorology II

11:00 W. R. Burrows and C. Price
Statistical Models for 1-2 Day Lightning Prediction for Canada and the Northern United States

S. Soula, S. Coquillat, S. Chauzy, J. F. Georgis, and Y. Seity
Surface precipitation electric current produced by convective rains during MAP

11:30 J. E. Dye, J. C. Willett, W. D. Hall, E. Defer, S. Lewis, D. Mach, M. Bateman, H. Christian, C. A. Grainger, P. Willis and F. J. Merceret
The Decay of Electric Field in Anvils: Observations and Comparison with Model Calculations
11:45 M. G. Bateman, D. M. Mach, S. Lewis, J. E. Dye, E. Defer, C. A. Grainger, P. T. Willis, H. J. Christian, and F. J. Merceret
Comparison of in-situ Electric Field and Radar Derived Parameters for Stratiform Clouds in Central Florida
12:00 Y. Seity, S. Soula, and S.Tabary
Relationships between lightning flash production and microphysics observed in European thunderstorms
12 :15 Qie Xiushu, Zhang Guangshu, Kong Xiangzhen, Wang Huibin, Zhang Tong, Zhou Yunjun and Zhang Yijun
Observation on the Lightning Discharges in the Northeastern Verge of Tibetan Plateau


Statistical Models for 1-2 Day Lightning Prediction for Canada and the Northern United States

Meteorological Research Branch, Meteorological Service of Canada, Toronto, Ontario

Department of Geophysics & Planetary Sciences, Tel Aviv University


The Canadian Lightning Detection Network provides continuous detection to about 65° N in the west and 55° N in the east. Coverage is melded with the U.S. network to 35° N east of 100° W and to 40° N west of 100° W for use in Canadian weather offices. Studies of 1998-2000 lightning "climatology" were done by Burrows et. al. (2002) for this region, and by Orville et al. (2002) for the combined U. S. and Canadian detection networks. A complex pattern is revealed, showing strong latitudinal, seasonal, and diurnal dependencies, and significant influence by topography and land-water boundaries.

Development of statistical models to predict lightning to 48 hours for the north portion of the NALDN is underway in Canada. Models will be implemented at the Canadian Meteorological Center initially for use in the national automated public forecast production system there, and later for production of aviation forecast charts.

Models for 5 degree latitude-longitude boxes are built for each of the four 3-hour intervals from 0000 UTC and 1200 UTC and applied for each 24-hour period. Grid point resolution is currently 22 km. The predictand is 3-hour total "lightning report density". Potential predictors derive from output of the Canadian Meteorological Center's weather prediction model. These are: temperature, dew point, and geopotential height at several levels; CAPE; convective stability indices; a severe storm index; helicity; elevation; convective cloud depth; Price and Rind predicted flash rate; tropopause height and temperature; precipitable water (total and above 700 mb); thickness of 4 layers; total and convective rainfall; wet bulb potential temperature; 700 hPa vertical motion; land/water and vegetation designations. For every three-hour period, the average, maximum (or minimum) value, and change of meteorological potential predictors is found. Tree-based models are derived with Classification and Regression Trees (CART) (Brieman et al., 1984).

fVerification and analysis of available results will be presented.

Burrows, W. R., P. King, P. J. Lewis, B. Kochtubajda, B. Snyder, V. Turcotte, 2002: Lightning occurrence patterns over Canada and adjacent United States from lightning detection network observations. Atmosphere.-Ocean, 40(1), 59-80.
Orville, R. E., G. R. Huffines, W. R. Burrows, R. L. Holle, and K. L. Cummins, 2002: North American Lightning Detection Network (NALDN); First results: 1998-2000. Mon. Wea. Rev.., 130, 2098-2109.


TopFull program for B2 Session

Surface precipitation electric current produced by convective rains during map  

S. Soula, S. Coquillat, S. Chauzy, J.F. Georgis, and Y. Seity
Laboratoire dAérologie, UMR 5560 UPS/CNRS, OMP, 14 avenue Edouard Belin, 31400 Toulouse


This study deals with the role of the precipitation in the charge transfer between the thundercloud and the ground. Data were collected during the Mesoscale Alpine Program (MAP) Special Observing Period (SOP) in Autumn 1999 in the North of Italy. Ground measurements of electric field, precipitation current density, individual drop charges, and rainfall parameters were locally performed at the site of the Lago Maggiore. This site was included in the Lago Maggiore Target Area (LMTA) and thus, the events documented with the ground measurements were also covered with radar facilities. Three ground-based Doppler radar allowed to retrieve 3-D precipitation and wind fields. Several days of the period provided substantially charged rainfall of both polarities. A case of deeply convective cell occurring on September 17th 1999 and several shallow convective cells passing over the experimental site on October 3rd are especially analyzed. For both considered events, the precipitation carried charge at the ground, during close duration for each polarity. In the case of the deeply convective cell, the charge carried by the rain was mainly negative and the current density corresponding was firstly positive, reached more than 100 nA m-2, and changed its polarity when the rainfall was maximum with a value close to 200 mm h-1. The thundercell producing the rainfall was strongly developed with an echo top at about 12 km, and a large number of lightning flashes occurred as indicated by the electric field evolution locally measured. In the second case, eight convective cells passed over the site and produced short rain showers, and six of them produced substantial precipitation currents and electric field increases. The dynamical analysis shows that the vertical velocity (averaged over 1 km 1 km mesh) was weaker within the two cells that did not produced any evolution of the electric parameters with a value of only 0.5 m s-1, while it reached 1.5 m s-1 within the other cells. The six charged cells did not produced any lightning flashes. Both charge polarities were also observed on the rain produced by these electrified cells, first the negative one and then the positive one. The analysis of the individual drop charges at the ground shows that most of the time both polarities are not mixed, even at the reverse of the precipitation current polarity. Large proportions of charged drops (> 2 pC in absolute value) are found, exceeding 50 % in many cases.

For all these observations, a very tight correlation between surface electric field and precipitation current was observed, displaying the mirror image effect. The ground electric field was due to the cloud charge of opposite polarity to that carried down to the ground by the rainfall. In order to reproduce the field evolution created by the cell passage, we tested different models of charge distribution. According to these models, the net charge of the cloud above the site is chronologically positive and negative, which can be the result of the evacuation of an opposite charge by the rain.


TopFull program for B2 Session

The Decay of Electric Field in Anvils:
Observations and Comparison with Model Calculations

J.E. Dye, W.D. Hall, S. Lewis
NCAR PO Box 3000, Boulder, CO 80307

J.C Willett
PO Box 41,Garrett Park MD

E. Defer
15 ave Paul Herbe, 92390 Villeneuve la Garenne, France

D. Mach, M. Bateman, H. Christian
NASA MSFC, Huntsville AL

C.A. Grainger
Univ. of No. Dakota, Grand Forks ND

P. Willis
NOAA/Hurricane Research Division, Miami FL

F.J. Merceret
NASA/KSC, Kennedy Space Center, Florida


Airborne observations of electric fields and associated microphysics have been made in anvils of active and decaying thunderstorms over or near Kennedy Space Center, Florida from the Univ. of No. Dakota Citation II jet aircraft to examine the decay of electric fields with time and space. The aircraft observations were coordinated with simultaneous radar coverage from the Patrick Air Force Base WSR74C 5 cm radar and the Melborne NEXRAD 10 cm radar.

The microphysical observations were made with several different instruments which spanned particle sizes from a few microns to several millimeters, thus from frozen cloud droplets to large aggregates. When electric fields are strong (> 20 kV/m) the entire size distribution of particles is significantly greater than when electric fields are weak (<1 kV/m). As the aircraft flew from near the convective core of a storm toward the downwind edge of the anvil, particle concentrations in all size ranges tend to gradually decrease. However, the electric field measurements show more variability and decrease much more abruptly than the decrease in particle concentrations. The complex nature of the electric field structure and changes of polarity even when flying at constant altitude show that the charge distribution in these anvils is not a simple uniform layering of charge.

In a companion paper at this conference [Willett and Dye, "A Simple Model to Estimate Electrical Decay Times in Anvil Clouds"] the observed particle size distributions are used to calculate theoretical decay times from 50 KV/m for a uniform layer of positive charge with negative screening layers above and below. The calculations suggest that particles in the size range from 200 microns to ~2.5 mm are dominant in determining the length of time for decay to weak fields. In portions of the anvils with electric fields ~40 to 50 KV/m the microphysical observations showed a surprising consistency of particle concentrations in this size range from day to day. When the electric fields are large (regions with high concentrations (>25 per liter) of 0.2 to 2.5 mm particles), the predicted electrical decay times tend to be long, and when the fields are low (regions with reduced particle concentrations), the predicted decay times tend to be short. Calculations using size distributions from regions with strong fields with high concentrations of particles yield decay times of 1 1/2 to 2 hours. Even though the observed electric fields showed more complexity than the uniform charge layers assumed in the model, the observed electrical decay times are of this same order. Examples of the observations will be presented and will be compared with the theoretical decay times.


TopFull program for B2 Session

Comparison of in-situ Electric Field and Radar Derived Parameters for Stratiform Clouds in Central Florida

M.G. Bateman
Universities Space Research Association, Huntsville, AL

D.M. Mach
The University of Alabama in Huntsville

S. Lewis, J.E. Dye, E. Defer
NCAR, Boulder, CO

C.A. Grainger
Atmospheric Sciences Dept, The University of North Dakota, Grand Forks

P.T. Willis
NOAA/Hurricane Research Division, Miami, FL

H.J. Christian
NASA/MSFC, Huntsville, AL

F.J. Merceret
NASA/KSC, Kennedy Space Center, FL


Airborne measurements of electric fields and particle microphysics were made during a field program at NASA's Kennedy Space Center. The aircraft, a Cessna Citation II jet operated by the University of North Dakota, carried six rotating-vane style electric field mills, several microphysics instruments, and thermodynamic instruments. In addition to the aircraft measurements, we also have data from both the Eastern Test Range WSR-74C (Patrick AFB) and the Weather Service WSR-88D radars (primarily Melbourne, FL). One specific goal of this program was to try to develop a radar-based rule for estimating the hazard that an in-cloud electric field would present to a vehicle launched into the cloud.

Based on past experience, and our desire to quantify the mixed-phase region of the cloud in question, we have created several algorithms for integrating radar reflectivity data in and above the mixed-phase region. A successful radar algorithm is one that can accurately predict the presence or absence of significant electric fields. We have compared the calculations from various algorithms with the measured in-cloud electric field strength in an attempt to develop a radar rule for assessing launch hazard. We will present the results of these comparisons.


TopFull program for B2 Session

Relationships between lightning flash production and microphysics observed in European thunderstorms 

Laboratoire dAérologie, France

Centre détude des Environnements Terrestre et Planétaires, France

Centre National de Recherche en Météorologie, France


This study is based on two distinct dataset. The first one was collected during the Mesoscale Alpine Program (MAP) Special Observing Period (SOP) in Autumn 1999 over the Alps in the North of Italy. It consists of data from the local Cloud to Ground (CG) flash detection network and data from three Doppler and/or Polarimetric cm-radars. The CG flash data provide their location, rate, polarity, and multiplicity. The polarimetric radar is the American S-Pol and its data allow us to discriminate 11 kinds of particle types in the cloud volume. Data from two other Doppler radar, the French Ronsard and the Swiss Monte-Lema, retrieve reflectivity and wind 3D-fields within the thundercells. The Intense Observing Period (IOP) 2a was especially active since it represents 75 % of the total CG activity measured over the Lago Maggiore Target Area (LMTA) during the whole SOP. Several thundercells were strongly vertically developed (sometimes more than 12 km of height) and produced large amounts of rainfall and some hail. Some of them exhibited exceptional CG rates at 12 min-1. A correlation study was made at global scale including the whole thundercell lifetimes. We observe very strong correlation between CGs of both polarities and the presence of Graupel-Hail mixture, what is in good agreement with the non inductive charging mechanism. We especially note at this scale the correlation of the presence of large radar relectivities within the thundercloud, at low altitude (2-5 km) with the negative CG production on one hand, and at high altitude (10-12 km) with the positive CG production on the other hand. The correlation study was also considered at the scale of the thundercell during its lifetime. We thus observe low CG rates associated with the presence of hail in the thundercell. Large positive CG proportions are observed to be associated with severe weather, especially with the presence of hail and strong vertical velocities.

The second dataset, for a duration of one year (2000), consists of total flashes data over the Ile-de-France area (France) and data from the cm-radar located in Trappes (close to Paris) included in the french Aramis network. CG flashes were detected thanks to the Meteorage/Meteo-France network and total lightning flashes (Intra-Cloud (IC) + CG) thanks to a SAFIR system. By considering several hail-producing thundercells in order to study their electrical behavior, we observe very low IC and CG rates associated with hail. We also observe high positive CG proportion, low negative CG multiplicity and peak current associated with hail. Theoretical aspects will be discussed in order to try to explain this singular electrical behavior.


TopFull program for B2 Session

 Observation on the Lightning Discharges in the Northeastern Verge of Tibetan Plateau

X. Qie, X. Kong, G. Zhang, Y. Zhang, H. Wang, Y. Zhou, and T. Zhang
Cold and Arid Regions Environmental and Engineering Research Institute,
Chinese Academy of Sciences, Lanzhou, Gansu 730000, P. R. China


In the summer of 2002, a comprehensive observation on natural lightning discharges was conducted in the northeastern verge of Qinghai-Tibetan Plateau. The thunderstorm is usually shows an inverted bipolar charge structure. The stepped leader develops downward to ground with multi-branches with a 2-D velocity of 0.8-1.2×10 5 m/s and more than one striking point on the ground was often observed as a result. The pictures from high-speed digital camera confirmed the bi-level structure of IC discharge. The IC discharge initiates between upper negative charge region and lower positive charge region, and developed downward and upward to the lower positive and upper negative region, respectively.

TopFull program for B2 Session

Last Update : June 3, 2003
[ ICAE 2003 Home Page] - [ ICAE Contacts ]