We must be conscious of the fact that there are no safe places outdoors with nearby thunderstorms and probably the most dangerous lightning strikes are the first ones hitting ground. They catch people exposed and unwary, even though in many cases the first lightning strikes are intra-cloud (IC) flashes. Most lightning victims are struck while going to a safer place, standing under a tree or while doing outdoor activities like golfing, fishing, mountain climbing and even riding on a lawnmower.
For improving safety and reducing fatalities we need to stay ahead of the storm. Weather forecasting based on physical models fed with real time and historical data coming from atmospheric sensors (including balloons) and RADAR have attained a very high degree of accuracy . But locally severe thunderstorm prediction is still difficult for many areas.
We can reduce risk by taking advantage of the existence of several necessary preconditions for lightning to occur. The most noticeable of them is the presence of cumulonimbus clouds in the mature stage with well-defined centres of electrostatic charge. This gives us a clue about how we can anticipate the occurrence of lightning:
- We can detect the existence of all necessary preconditions by means of instruments like RADAR, LIDAR, electrostatic field meters and others or,
- We can keep track of current lightning activity for surrounding areas.
The first one allows for anticipation of first strikes and is the most complex to achieve. The latter is much simpler, to some extent, by means of lightning location networks (LLN). I will cover LLNs in next post.
Measurement of the atmospheric electrical field strength
Forthcoming electrified storm clouds that could eventually produce lightning produce detectable variations in the atmospheric electrical field strength as they pass by. Electrostatic field meters can take advantage of this as charged particles are subject to a force when exposed to an electric field.
The force experienced by a charged particle of net charge q , when exposed to an electric field of magnitude E , can be expressed in the form: F = q x E . Both F and E are vectorial magnitudes and E is usually expressed in V x m-1 . (Note the above expression intentionally neglects the effects of the particle’s speed and the magnetic field- see . It is given here for the sake of simplifying the operating principle description coming next. )
In the case of an isolated conductive plate exposed to an electric field, a charge redistribution occurs (see Figure 1a). Charges are allowed to move under the force exerted by the electric field (Figure 1b) by providing one conduction path between the sensing plate and the reference potential (ground is depicted positive here). After equilibrium is reached (Figure 1c), the sensing plate will be at the same reference potential. From this point on, the magnitude of the flowing current will be a function of the variation in time of the electric field and several sensor construction parameters.
Conductive plate in the presence of an electric field (the arrow shows electrons leaving the plate).
One method for obtaining the value of the electric field strength is the integration in time of the current flowing from/to the sensing plate (Figure 2).
Electrostatic field strength measurement by integration.
This method has several aspects to consider: the current-to-voltage convertor must use a high precision electrometer grade amplifier, the integrator must be almost “ideal”, and special reset, calibration and biasing mechanisms must be implemented. Instruments based on this method still show one major advantage, no moving parts are required.
The field mill
Another electrostatic field meter implementation is what is commonly known as a Field Mill .
I spent a couple of days at last year’s Meteorological Technology World Expo 2015 with one of our distributors for North Europe and one colleague from our Export Department. We received many expo visitors and spend a lot of time talking with them about storm tracking and lightning prevention and protection. One visitor who caught our curiosity was an amateur weather enthusiast and hobbyist. His introductory phrase was “This is the field mill reborn.” He was referring to our electrostatic field meter, a product that has been in our products portfolio for many years now and, of course, something absolutely not new.
The field mill uses a moveable plate, a rotating vane in most cases, which is connected to the reference potential. The movable plate acts as a mechanical shutter that controls the sensor plate exposure to the electric field. By controlling the sensor area which is exposed to the electric field, the current is forced to move in one direction or the other. As an example consider the arrangement depicted in its steady state in Figure 3a. In this state there is no electric field between the shutter and the reference potential, or at least it can be neglected thanks to careful mechanical design.
The shutter, the sensing plate and the reference (ground) are at the same potential. When the shutter opens (Figure 3b) the sensing plate is exposed to the electric field and acquires a net charge due to its effect on available free charge carriers. After the shutter is closed (Figure 3c) the process reverses and current flows in the opposite direction. The polarity of the electric field can be determined by sensing the shutter position.
Field mill operation.
For improving signal to noise ratio and sensitivity, many modern commercial field mills use a differential sensor plate configuration. The differential sensor configuration utilizes two sets of sensor plates connected in a configuration similar to the one shown in Wikipedia .
Figure 4 shows an atmospheric electrostatic field strength graph plot obtained with one field mill at a test site. There are three relevant points highlighted in the graph: 1) the fair weather field (about 195V/m-1 ), 2) the passing of one charged cloud and 3) the passing of one well developed and electrically active (presumably of cumulonimbus type) cloud. Note the field inversion occurred as the second cloud passed by. It is a very interesting effect. Why?
Electrostatic field as measured by the field mill.
Figure 5 shows the electrical field strength graph corresponding to the passing of one highly active thunderstorm. Over-imposed are red marks corresponding to lightning impacts detected by the lightning location network at distances up to 25km from the sensor. The same information is shown in Figure 6, but only impacts occurring at distances of at most 5km are shown.
Electrostatic field as measured by the field mill with lightning impact marks.
Electrostatic field as measured by the field mill with closest lightning impact marks.
During this storm no lightning impact was registered at the test site. The distances of the two closest impacts are shown in the graph. The first one was an intra-cloud flash occurred at an altitude of 15km and the second one was a cloud-to-ground flash.
Usability of electrostatic field strength signals for forecasting
There has been a lot of discussion about whether the electrostatic field strength information is of some value or not. After many campaigns and studies there seems to be a consensus in at least a few things.
The electrostatic field strength measurements localized character poses accuracy limits to forecasting, especially when using the information provided by a limited number of sensors. For best results the number of deployed sensors and their relative positions must be carefully planned. Taking into account the typical base area and column height of storm clouds as well as their most probable approaching trajectory also helps improving detection efficiency and forecasting accuracy.
Field mills have been extensively used by NASA for improving rocket launch safety in both flavours: as fixed ground installations as well as airborne. Kennedy Space Center (KSC) field mill data can be viewed and downloaded from the NASA Global Hydrology Resource Center DAAC site at  (registration required). Figure 7 shows a 2D rendering generated from field mill data collected by The Advanced Ground Based Field Mill (AGBFM) Network .
Image source: NASA Global Hydrology Resource Center DAAC.
It should be noted that, even when it is a necessary condition for lightning to occur when an electrostatically charged cloud must have developed over the area, the opposite is not always true. Figure 6 is an example of the latter. It should be said that this test site has several peculiarities. It is at over 1500m above sea level and a considerable number of lightning strikes occurring in the area are triggered by existing tall structures.
One last thing to say is that field mills in general produce a relatively high number of false-positive alarm triggers. But there are many situations where having some false-positives is far less problematic than failing to warn.
The construction of one field mill requires exercising various skills and poses some challenges. It involves the integration of several distinct modules like an analogue front-end with variable gain, a digital processor (CPU), at least one communications interface, motor control hardware and algorithms and a well-designed housing for containing the electronics and moving parts. As a DIY enthusiast, I highly recommend hobbyists to engage in building projects of this type. It is a great exercise in many senses and it is possible to build one with readily available low cost development kits and a few extra parts.
The next part of this series will focus on lightning location networks (LLN), how they work and sample data as provided by state-of-the-art commercial grade LLN.
Do you find useful the information provided by field mills for thunderstorm forecasting? What other sources of information do you find useful for improving lightning safety?
 – NOAA, Graphical Hazardous Weather Outlook.
 – Wikipedia article, Lorentz force.
 – Wikipedia article, Field Mill.