Tuesday, Feb. 19, 2013
click here to download today's notes in a more printer friendly format

The 3rd homework assignment was collected today and a new assignment was handed out.  This new assignment is due next Tuesday, Feb. 26 and will be the last assignment before the Midterm Exam scheduled for Tue., Mar. 5.

This lecture will be devoted to qualitative descriptions of cloud electrification processes. 

The picture below is from the supplementary lecture on cold cloud/ thunderstorm cloud structure.  The tripolar charge structure is what a viable electrification process needs to explain.

The main charge centers are first an upper positive charge center at 1(a) and smaller lower positive charge centers (1c).  There is more of a layer of negative charge at (1a) that seems always to be found at temperatures between -10 C and -30 C.

Screening layers are found at the top and sides of the cloud (2a and 2b in the figure).  These form because of the abrupt drop in conductivity as you move from outside the cloud into the cloud aren't part of the electrification process.

The figure below presents some of the evidence supporting the idea that the negative charge is found between -10 and -30 C in thunderstorm clouds.

The figure shows the locations of negative charge neutralized during cloud-to-ground and intracloud discharges in clouds in Florida, New Mexico, and Japan.  We'll look at how this is done in our next lecture.  In each case the negative charge is found in the same temperature range despite differences in cloud heights, cloud base altitudes, and cloud thicknesses.  (source: Krehbiel, Paul R., "The Electrical Structure of Thunderstorms," Ch. 8 in The Earth's Electrical Environment, National Academy Press, Washington, 1986)





The intent of this figure is to show that the distribution of charge in thunderstorms is sometimes much more complex than the simple tripolar model discussed above. (source: Stolzenburg, M., W.D. Rust, and T.C. Marshall, "Electrical Structure in Thunderstorm Convective Regions 3. Synthesis," J. Geophys. Res., 103, 14097-14108, 1998)

We'll start with the convective process of cloud electrification. 

The convective theory starts with the positive charge found in the air above the ground during fair weather (Fig. A).  In Fig. B an updraft and a cloud have started to form and positive charge is carried upward into the cloud where the charge attaches to cloud particles.  Negative charge carriers in the surrounding air are drawn to the positive charge in the top of the cloud and form a screening layer.  Cloud edge motions shown in Fig. C then begin to carry this negative charge down and into the middle center of the cloud to form the main negative charge center.  The electric field at the ground intensifies because of the close proximity of negative charge in the cloud.  Objects on the ground go into corona discharge and "spray" positive charge into the air (Fig. D).  This is one way of accelerating the charging process.

While cloud motions can clearly have an effect on the distribution of charge inside a thunderstorm, the convective theory is generally not considered to be a viable mechanism for the initial electrification of thunderstorms.  For example, why would the cloud motions always concentrate the negative charge in the -10 to -30 region in the cloud?

In the inductive process an existing (initially fair weather) field induces charges in precipitation particles (in the same way charges were induced on a conducting sphere in a uniform field).



The downward pointing fair weather E field induces equal amounts of negative and positive charge on the tops and bottoms of different sized precipitation particles.  Because the particles are moving in different directions or at different speeds, they collide.  During the collision some of the charge on each particle is neutralized.  Each particle is left with net charge following the collision.  Positive charge moves upward, negative charge downward.

The direction of charge motions is consistent with the direction and strengthening of in-cloud electric fields, something I think I confused in class.



The field at the ground under a thunderstorm points upward, the opposite of the fair weather field.  The direction of the in cloud field is the same as the fair weather field.  Thus the directions of charge movement in the inductive process is consistent with thunderstorm charge distribution. 


There are a variety of types of particles that can collide.
water - water
 ice - ice
water - ice
 ice - riming ice

When two water droplets collide, they will often stick together.  If make just a glancing collision, there won't be any charge neutralization because there isn't any charge induced at the particle's equators.  Thus a collision between two water droplets would not seem like it could separate significant amounts of charge.

Ice particle - ice particle collisions don't neutralize much charge because the particles don't remain in contact for very long and the lower electrical conductivity of ice means charge doesn't flow as readily as it does through water.

Water - ice particle and ice - riming particle collisions might work, but the general feeling still is that the inductive process is not able to start with a fair weather E field and turn it into a thunderstorm strength field.

Before we leave the inductive process behind a quick demonstration of an apparatus that depends on induction (and positive feedback) to produce a surprising amount of charge from tap water.  The apparatus is a Kelvin Water Dropper or Kelvin Electrostatic Generator.  A picture of the apparatus used in class is shown below at left











Water from a reservoir (at the top of the photo) travels down two sections of tubing and out of two plastic nozzles positioned near the middle of the photo.  The falling streams of water fall through two metal rings (colored red and green) and into two plastic cups at the bottom of the picture.  The cup on the left is electrically connected to the ring at right (the red ring), the cup on the right is connected to the left (green) ring.  The two cups are also electrically connected to 5 neon lamps wired together in series.  The right photo above is a close up of the neon lamps.

The operation (as I understand it) is explained below.

Let's assume that the water in the left cup (Pt. 1) contains some + charge (it is equally likely that it contains - charge).  The left cup is electrically connected to the right (red) ring, Pt. 2.  The positively charged red ring will cause (by induction) negative charge to build up at the bottom end of the falling stream of water.  The stream then breaks up into drops which carry the negative charge to the cup on the right (Pt. 3).  This is connected to the left ring at Pt. 4.  As shown in the close up at Pt. 5 positive charge is induced at the tip of the stream of water.  As the stream breaks into drops the drops carry more positive charge to the cup on the left.  Charge builds up until a spark flashes across the spark gap (or until the bank of neon lamps flashes in the apparatus used in class).

There are a lot of videos on YouTube.  This first video (from Reinhard Schumacher at Carnegie Mellon University) shows an apparatus that produces a visible spark and also contains a sensitive electroscope.  Here's a second video (from Thomas Kim) that produces a visible and audible spark (note how the sound of the dripping water also stops a second or two before the spark).

The non-inductive process, also called the Reynolds, Brook, Gourley process is generally thought to be the most viable explanation for the rapid initial electrification of thunderstorms.  It doesn't require a existing field.  The process is shown in general terms below.  We will look at some of the details later.


Basically graupel collides with a snow crystal, in the presence of supercooled water droplets, and then depending on the environmental temperature, the graupel ends up with negative or positive charge and the ice crystal ends up with the opposite polarity.  At temperatures colder than about -15 C the snow crystal ends up with positive charge and the graupel is negative.  The different sizes and fall velocities of the two types of particles means they will tend to separate after the collision.  The negatively charged graupel tends to accumulate in the middle of the cloud and the snow crystal is carried up to higher parts of the cloud.  At temperatures warmer than about -15 C the polarities reverse.  We will see that this charge reversal temperature can vary depending on the cloud liquid water content.  The letters A, B, and C show how this process can account for the three main charge centers in a thunderstorm (A & C are the upper and lower positive charge centers, respectively; B is the main negative charge center)

The following figure shows the experimental apparatus used to make the initial measurements of charging.  [source: S.E. Reynolds, M. Brook, and Mary Foulks Gourley, "Thunderstorm Charge Separation", J. Meteorology, 14, 426-436, 1957)

Two metal balls at the ends of arms mounted on a rotating shaft simulate graupel particles in a cloud (pt. 1).  These balls become electrically charged the electrical path from the balls to a sensitive electrometer has been highlighted in yellow.  A little bit wider view of the apparatus is shown below.


Temperature and liquid water content (LWC) are measured near the spinning arms (pt. 2).  Liquid water content is really just a measure of the concentration of supercooled water droplets. 

Warm water at pt. 3 is the source of water vapor.  Cooling coils (pt. 4) cool the air inside the chamber.  Cold air will sink and probably establish a vertical temperature gradient.  The spinning metal ball could be raised or lower to change the temperature of the surrounding air.  Replicas of ice crystals that form can be collected at pt. 5 for later examination.

This would not be an easy experiment to conduct.  It would be difficult to create, measure, and monitor the cloud environment in the chamber and the charging that does occur is probably very weak and a sensitive electrometer would be needed.

The following information was copied from the www.SnowCrystals.com website.   This is an information packed website and you should definitely have a look at it if only to view the photomicrographs of snow crystals.  The information below is longer available.  It briefly discusses the quasi-liquid layer that is found at the boundary between ice and air. 



The quasi liquid layer seems to play an important role in charging in the non inductive process.


The two colliding particles in the non-inductive process are each thought to be surrounded by a quasi liquid layer.  The particle that is growing faster will have a thicker quasi liquid layer and will transfer mass to the other particle at the time of the collision.  The particle that gains mass ends up with negative charge.

Now we'll look at some of the details from more recent studies of the non-inductive electrification process.  Results from some of the work done by a group at the University of Manchester Institute of Science and Technology (UMIST) are summarized on the graph below (source: Saunders, C.P.R. and S.L. Peck, "Laboratory Studies of the Influence of the Rime Accretion Rate on Charge Transfer during Crystal/Graupel Collisions," J. Geophys. Res., 103, 13949-13956, 1998).



The + and - polarities in the figure refer to the charge on the metal target that simulates a graupel particle in these laboratory experiments.  The charge depends on the environmental temperature and on the cloud liquid water content as shown in the figure above.  EW on the vertical axis is effective liquid water content and is the product of liquid water content and collision efficiency (what fraction of the supercooled water droplets colliding with the target stick and eventually freeze to the target). 

Here is my interpretation of what happens in regions A and B.  The temperature is low in both cases.  The difference is the liquid water content, the concentration of supercooled water droplets.


Region A
Supercooled droplets are colliding with and sticking to the graupel particle at a relatively low rate.  That is because of the low effective liquid water content.  The droplets warm as they release latent heat and try to freeze.  The droplets warm but there isn't enough latent heat energy release to warm the graupel particle itself.  So both the graupel and the ice crystal are cold and the rates of deposition of water vapor to the graupel and the ice crystal are about equal (deposition is the water vapor to ice phase change).  The graupel gets some additional deposition from the warm water droplets on its surface (the thin purple arrows from the droplets to the graupel particle in the figure).  As a result the graupel is growing more quickly and has the thicker quasi liquid layer.  Mass transfer during a collision with an ice crystal is to the ice crystal.  The ice crystal ends up with negative charge, the graupel particle with positive charge.


Region B
The effective liquid water content is higher and there are more supercooled water droplets colliding with and sticking to the graupel particle.  Now as they release latent heat and try to freeze they are able to warm the graupel particle.  The net rate of deposition from the surroundings to the graupel particle is reduced (2 arrows are shown in B versus 3 arrows in A).  Even with the deposition from the droplets on the surface of the graupel particle, the ice crystal grows more quickly and has the thicker quasi liquid layer.  Mass transfer during a collision goes to the graupel particle and the graupel particle ends up with negative charge.

Region C
Both the ice crystal and the graupel particle are in warmer surroundings and the rate of deposition to each is reduced.  The graupel is warmer than it was in Region B.  Now however the deposition from the accreted water droplets is increased because it takes them longer to freeze.  The graupel particle grows faster and has the thicker quasi liquid layer.  Mass transfer is to the ice crystal again and it ends up with negative charge.

Most thunderstorms have the dipolar (or tripolar) charge distribution as shown below at left.  The majority of cloud to ground (CG) discharges from these types of storms carry negative charge to ground.  Just a few percent of the CG discharges are positively charged (we'll see though that these positive CG discharges sometimes have very large peak currents).

A much larger percentage (50% or more) of positive CG discharges have been observed coming from some Central Plains storms in recent field experiments.  This would suggest the clouds might have had an inverted charge distribution like that shown below at right.  And one might wonder whether a different cloud electrification mechanism is at work (different from the Reynolds, Brook, Gourley process discussed last Thursday).


Other than the high incidence of positive CG discharge, E field soundings in thunderstorms have provided evidence of the inverted charge structure.  This is discussed in a fairly recent paper that you can find here.  The two figures below illustrate E field soundings in "normal" clouds and in clouds with inverted charge distributions.


As you move upward from the bottom toward the top of the cloud you encounter a positively charged screening layer, the main negative and positive charge centers, and a positive screening layer at the top of the cloud.


Just the opposite situation was observed in the inverted polarity clouds.  At first glance these clouds would seem to invalidate the Reynolds, Brook, Gourley process we have been studying or at least suggest there might be another process that sometimes is at work. 

Follow up studies, however, seem to indicate that the non-inductive electrication mechanism  is capable of explaining clouds with inverted charge distributions.  This is shown on the figure below


In the center of the figure is the graph that we examined in some detail earlier in this lecture.  It shows the polarity of charge acquired by a graupel particle colliding with ice crystals in a laboratory simulation of a cloud environment containing supercooled water droplets.  The liquid water content on the vertical axis is a measure of the supercooled droplet concentration.

Liquid water content in a typical cloud would be found near the level of the lower dotted line on the graph.  The tripolar cloud charge distribution in a typical cloud is shown at left.  High in the cloud where the temperature is cold, the graupel particle acquires negative charge and the ice crystals positive charge (region B on the graph).  The ice crystals are carried upward and form the main positive charge center.  The heavier graupel particle descends to form the main negative charge center.  Somewhat lower in the cloud where temperatures are warmer than TR (the reversal temperature) but still below freezing the polarity of the charging changes (region C on the graph).  The graupel ends up with positive charge and the ice crystals with negative charge.  The positively charged graupel form the lower positive charge centers.

It seems that the unusual Central Plains storms have very high liquid water contents.  If we look at the level of the upper dotted line on the charging graph in the center of the figure we see that the graupel particle always ends up with positive charge in this high LWC environment.  There is no charge reversal temperature.  The Reynolds, Brook, Gourley mechanism can account for clouds with inverted charge distributions.