Electrical environment can be altered at 1 km distances from high voltage power lines.

High voltage powerlines emit electrical charges into the atmosphere which can then attach to aerosols. This space charge above ground can be measured directly using ion spectrometers or indirectly through perturbations of the Earth’s potential gradient using field mills. Several publications are reviewed to find evidence of aerosol charging at a distance from power lines. Field measurements of charge state near to high voltage power lines selected due to their high emissions of ions measured a small positive enhancement of electrical charge on aerosols at distances greater than 300 m, corresponding to a transit time of up to 400 s A quasi one-dimensional model of ion-aerosol interactions from a high voltage powerlines found that the addition of new ions to an aerosol population will result in those ions transferring charge to the aerosol which would then remain the dominant carrier of charge several hundred meters downwind. 10-min PG measurements from a fixed site measuring in 2008 compared measurements when the site was downwind of a 275 kV powerline to times with no wind and found evidence of space charge overhead through greater fields and variability at distances over 800 m These studies combined show evidence that the electrical environment near to power lines can be altered beyond 1 km from AC high voltage power lines, with excess charges likely to be concentrated on aerosol.

The electrical environment downwind of high voltage power lines (HVPL) has been the focus of numerous epidemiological and mechanistic studies on the effect to human health and the environment [1], often there is an assumption that populations are no longer exposed after a distance between 50 m and 300 m where magnetic fields are below background [2], however, HVPL can produce ions through corona which can be transported far beyond these distances and detected directly through ion counters [3] or indirectly by changes in the Earth's potential gradient (PG) [4].Ions progressively attach to aerosol as they travel, but still contribute to space charge.The Earth's natural potential gradient exists due to electrical activity transferring charge to the ionosphere, but space charge above ground, such as ions and charged aerosol near power lines, can perturb this field and are detectable by field mills [4].
1. Measurements of PG, ions and aerosols near HVPL 1.1.Corona ions production and detection Electricity is transmitted large distances at high power through high voltage transmission networks, which can be either alternating or direct current, alternating current being predominant over shorter distances.When the electric field surrounding the conductors is high enough, electrons accelerated by the line will collide with and ionise particles in the air, causing an ion-electron pair which will continue to ionise other particles causing the process known as corona ionisation.
A newly produced ion will very quickly form a cluster, with the central ion surrounded by polar water molecules, reducing the electrical mobility.Once the ion has travelled far enough from the electrical conductors that it is no longer moved by the electrical field it will be propagated by the wind until it either combines with an ion of the opposite polarity (ion-ion recombination) or transfers its charge to an aerosol (ion-aerosol attachment) [5].Ions can be measured near power lines through the use of ion counters [6] or ion mobility spectrometers [3,7].
Ion mobility spectra were measured using an aspirated condenser ion mobility spectrometer (ACIMS) [7] at 11 sites and found a significant increase in ion concentrations downwind of the lie in 9 of those sites (significantly increased ion concentrations had measurements > 2 standard deviations of the response to positive and negative mobilities, 14.8% and 15.7% respectively).Of the individual measurements, 33 out of 46 had increased ion concentrations, 17 positive only and 15 having both elevated positive and negative ions, negative only occurred once.all measured at a distance between 40 and 640 m.The spectra showed a consistent, but non-statistically significant, increase of mobility at sites DW of HVPL, for both positive and negative ions [3].
Ion concentrations could be measured by leaving the ACIMS at a single voltage which can provide the ion concentration over a large length of time [8].The ACIMS measured both positive and negative ion concentration at the same time at 6 sites shown in figure 1. Ion concentrations are greater beyond 100 m but there are limited measurements beyond 300 m, where the only measurement shows a return to background levels.

Aerosol charging by power lines.
Once produced, ions will progressively become attached to aerosol in the atmosphere, altering the charge distribution of the aerosol present until it reaches a steady state.Hendrickson [5] measured aerosol and small ion concentrations downwind of a DC transmission line, a negative correlation between small ion concentration and aerosol concentration confirmed that the aerosol were a sink of ions, and inferred that charge is transferred to those aerosol.Bailey et al [9] measured the number of particles between 0.065 to and 1 µm containing charge out to 500 m from a 500 kV DC power lines and away from lines by comparing aerosols measured by aerosol laser spectrometer and the number measured by a differential mobility analyser column.They showed a change in distribution downwind with more negative charge bias overall, but concluded that the proportion of charged particles compared to neutral particles was not enhanced to a large degree by the DC lines.Charged aerosol have been measured up to 160 m downwind of three AC HVPL in Australia, where approximately seven per cent of particles were charged [6].The aerosol charge distribution is the probability of an aerosol of a given size holding a number of charges as described by the distribution described in equation ( 1) from Clement and Harrison [10] where j is the number of electrical charges expected to be found on a particle of diameter d where, n is the concentration of small ions and µ the average mobility.T is the temperature, k is Boltzmann's constant, ε0 the permittivity of free space and e the charge of one electron.A useful single parameter that has been used to describe an aerosol charge distribution in steady state is the charge asymmetry ratio, x which is equivalent to the ratio of the concentration c multiplied by the mobility µ of positive to negative ions as given by equation ( 2).

𝑥 =
The charge state of aerosols in the atmosphere can be estimated using the techniques described in Buckley et al [11].If an aerosol size distribution is known through measurement and an ion mobility spectrum is also measured, then the charge distribution according to equation ( 1) can be found by progressively changing the charge asymmetry ratio in equation ( 1) to create an estimated charge distribution, then fitting that theoretically to the measured size distribution until the mobility distribution measured is found.
To better understand the effects to the environment, Wright et al [12] ran a field campaign of charge state measurements near to HVPL between October 2007 and September 2008 in the south western part of England at seven different powerline sites.Using the charge state estimation technique, aerosol charge state was measured up to 640 m downwind of power lines, including some sites known to produce high concentrations of ions from previous measurements [3,8,11].
Measurements were made with two Grimm Scanning Mobility Particle Sizer (SMPC+C) systems, one with an Americium 241 source to bring the sample to a charge neutral position, the other with a dummy source with no neutraliser within it.Particles size 10 to 1000 nm measured, and a charge state of +/-20e estimated within the fit.Meteorology was measured at 1-minute samples with a Davies Instruments weather station at 2 m height.Results from all sites are shown in figure 2, most measurements show a small deviation from background concentrations such as those measured upwind of the power lines in the study.At the sites with large emissions of ions, a larger enhancement of charge was shown.Background aerosol show a slight negative bias in charging, so positive corona first neutralises aerosols before increasing charge.Negative emissions (such as during rain) do show an increase in charge.At one site with high corona emissions, enhanced charging was not shown at a site 24 s transit time away but was shown at a further 84 s away.In their results there is a small positive enhancement of electrical charge on aerosols at distances greater than 300 m, corresponding to a transit time of up to 400 s [12].
The creation, attachment and recombination of ions occurs constantly in the atmosphere in a nonequilibrium steady state system.This can be described by the ion balance equation (3), showing the rate of change of positive or negative ions (dn±/dt), which is determined by the creation rate of ions, the ion recombination coefficient α, and the ion-aerosol attachment coefficient β, which is a function of size but simplified here to a single value [13].
A simple approximation of the lifetime of the lesser ion once a steady state has been reached was given by Jones and Jennings [14] in the case of no aerosols present, equation ( 4), and similarly the lifetime of ions in a unipolar cloud in the presence of aerosol can be described by equation (5).

𝑡 =
(4) Wright et al [15] used existing aerosol theory and some knowledge of local plume dispersion characteristics to create a quasi 1-dimensional model of corona ion charging downwind of HVPL.The model used an elevated number of ions to represent the influx at the point of the power line.Background ions and background were considered constant and replenishing as the plume expanded, while the corona ions in the plume were diluted, and new aerosol introduced, based on the corona ions expanding within a Gaussian plume dilution model [16].Electrostatic effects on the plume dispersion were not included in this model.Similarly, effects based on coagulation, particle deposition and ion induced aerosol formation were not included in the model.The model used input parameters of between 0 and 10 cm 3 s -1 ionisation rate, between 0 and 100,000 ions cm -3 and windspeeds between 1 and 10 ms -1 , the aerosol size distribution was taken from values measured in the field and multiplied or divided by a factor of three.Results were validated by comparison with field studies in the UK [12] and in Australia [17].
The initial stages show a loss of the non-dominant polarity of ions and dispersion is the primary factor affecting ion concentrations.At longer times and distances, ion-aerosol attachment became the most important loss mechanism.A quasi-steady state is reached, and greater background aerosol concentrations approach that point earlier.
The net charge on aerosol at long distance, once a steady state has been reached, remains high in the model even up to a distance of 6 km, with greater charge states found at higher wind speeds (this does not account for differences in ion production at higher wind speeds.However, higher wind speeds take longer for the plume to reach ground level.As deposition of particles may be enhanced by charge, and were not included in the study, this may be an important loss process.

Potential Gradient perturbations
The passage of space charge over ground can cause perturbations in PG measured underneath, which can then be measured at ground level using instrumentation such as electrical field mills [18].Early measurement of potential gradient in mist and fog by Chalmers using a parallel plate system on top of a vehicle identified high and variable fields which were ascribed to the presence of HV power lines downwind of the measurement position [19].It was remarked as surprising at the time that fields remained elevated up to several km.Similar experiments by Mühleisen [20] and Bent and Hutchinson [21] also showed changes of field in mist and fog at long distance from the powerline.
Fews et al [22] used electric field mills near to HVPL to make distance profile measurements near to several HV power lines.Of the 12 separate power line sites, 8 showed modifications of the Earth's atmospheric PG defined here as changes > 100 V m -1 from the natural PG.In a subsequent publication, Fews et al. [23] made contemporaneous time series measurements of PG either side of HV power lines in the Bristol area. of 34 measurements almost all sites examined exhibited more variability and either elevated or reduced.In one instance, an elevated field was found at 7 km from the power line.
Both ion concentration and PG were measured simultaneously at 5 HVPL sites near to Bristol, UK, using the ACIMS in scan mode and JCI 140 Electric Field mills [8].In three out of eight measurements downwind of the HVPL, ion polarity measured at ground level were one polarity while the PG measured overhead implied that space charge was the opposite polarity.
In 2007, a fixed site monitoring station was constructed in a field in South Gloucestershire near to two HVPL [4,24].To the south west was a 400 kV power line which was 200 m at its closest point, to the south east was a 275 kV power line further away.PG was measured by a JCI 131F field mill at 1 s sampling interval, which local weather was recorded using a Davis Instruments weather station every 10 minutes, PG was averaged to 10 minutes and summary data recorded.To separate the effects of the two powerlines and control measurements, data was binned into three zones, Z1 downwind of the 400 kv Line, Z2 downwind of the 275 kV line and Z3 upwind of both powerlines.A change in both central tendency and variability of PG was evident in Z2 and Z3, downwind of the HVPL.A long analysis of the PG measurements at the FSMS showed that PG measured downwind of HVPL was more variable and more negative downwind overnight and around sunrise and sunset when downwind of the HVPL [25], and that increasing relative humidity and wind speed resulted in a decrease in measured values of PG, implying negative corona ion production [26].As the model of Wright 2023 has shown that aerosol could retain charge for a long distance, the data from the FSMS was reanalysed to see whether the effects of the HVPL were still detectable in wind directions from Z2, with distances of 1118 m, 798 m, 736 m and 705 m respectively from WSW to S clockwise.
The 10-minute mean and standard deviation were sorted by transit time from the 275 kV line and are shown in figure 4(a) and 4(b).The standard deviation or interquartile range can be used as a proxy for charge carriers, indicating both negative and positive charge present, for the control wind directions the interquartile range of all PG means was 80 Vm -1 , and showed a decreasing trend from 0-10 min (200 Vm -1 ) to 30-40 min (132 Vm -1 ) showing a change to the electrical environment at distances between 700 m and 1120 m from the HVPL.The 10-minute standard deviations are highest between 0 and 20 minutes from the line, but are reduced after that, with the central tendency closer to that of the low wind speed and control wind directions.
It is possible that the change in PG seen at the FSMS is carried from the closer line with space charge carried by turbulence.Figure 4(c) and 4(d) show the 10 minute mean and 10 minute standard deviation for each wind direction.The change in variability is clear between the upwind and downwind directions, but there is a degree of similarity at the wind directions in the border between Z2 and Z3.In the data collected at the FSMS, only 10-minute averages of wind speed and direction were measured.Any future measurements should include higher resolution wind data to understand the possible effects of turbulent dispersion.

Epidemiological considerations
The electrical environment downwind of high voltage power lines (HVPL) has been the focus of numerous epidemiological and mechanistic studies on the effect to human health and the environment, often there is an assumption that populations are no longer exposed after a distance where magnetic fields are below background.Some epidemiological studies assume an unexposed distance of 50 m, with cases selected from within 50 m and controls outside that distance.Some studies where exposed cases are withing 100 m of the HVPL and unexposed > 100 m individually showed a non-significant increase in risk for leukaemias [e.g.[27][28][29][30][31][32][33] or a non-significant decrease [e.g.[34][35].
Some studies have looked at effects beyond 50 m and still seen a small but significant increase in leukaemia, childhood leukaemia.Draper et al [32] found a significant increase of risk of children who lived within 200m of HVPL, and also for those living between 200 m and 600 m, though this distance did not show an increase in risk in a later study in by Sermage-Faure et al [36].Lowenthal et al found an increase in risk of lymphoproliferative disorders for those living up to 300 m away.Other studies have found significant risks for living within 200 m [37], 500 m [38] or 600 m [39].These studies combined show evidence that the electrical environment near to HVPL can be altered well beyond 1 km from HVPL, with excess charges likely to be concentrated on aerosol.Any investigation into health or environmental impact of HVPL should consider these distances.A full systematic review and meta-analysis of all studies to date looking wider exposed distance from HVPL should be considered if any mechanism related to the charge on aerosol is to be under consideration.One such mechanism was considered by Fews et al [22].While deposition in childhood lungs was not studied, a study in adult volunteers with particles charged to a charge state greater than that measured in the studies above [12] did not show a significant increase in deposition [40].Fluctuating electric fields have themselves been suggested as a possible mechanism [41] but those mechanisms require further investigation.

Implications for other measurements
Chalmers [19] suggested that the charge seemed to be carried within a valley and there is some evidence of one polarity being measured at ground level, while PG shows the opposite polarity aloft [8].These implicate that charge might affect dispersion patterns, and might therefore be required as a consideration in aerosol dispersion considerations.Some research [14,42] have shown that electrostatic effects can affect the plume and modelled three distinct effects: the outward radial force of electrostatic repulsion, longitudinal forces of repulsion and downward electrical image forces.The increase of all three is dependent of the electrical mobility of the plume, and therefore if the majority of space charge is attached to larger aerosols, the plume of charged aerosol might be less affected than any unattached ions in the area.
Other sources of space charge include steam locomotives [21] and traffic [43,17] where aerosol is also elevated and may be produced charged.It may be necessary to consider elevated charge within particles in cities, and measurements to test whether elevated charge is retained within cities or near to busy roads is recommended.A city is a complex electrostatic environment where measured electric fields can vary within a few km due to the affect of increased aerosol adding resistivity to the air [44].There may also be an enhancement or reduction of electric fields due to the complex geometry of tall buildings [45].The transport of a charged plume will therefore be affected to different degrees in different parts of a city.

Figure 1 .
Figure 1.Ion concentrations measured by the ACIMS in 'single voltage' mode.At 6 different sites in SW England at distances up to 500 downwind of an HVPL.Noncontemporaneous upwind measurements are represented at -100 m [7, 8].

Figure 2 .
Figure 2. Examples of mean average charge asymmetry ratio (and standard deviation) measured upwind and downwind of an HVPL at position 0, each colour represents a different power line, though measurements may be separate days.Data from [12].

Figure 3 .
Figure 3. Map of the FSMS shown in context of distance to HVPL, to the South, distances of 1118 m, 798 m, 736 m and 705 m respectively from WSW to S clockwise from the 275 kV are indicated by purple lines.

Figure 4 .
Figure 4. Box plots of the 10-minute means (a) and the 10-minute standard deviations (b) of PG measured downwind of the 275 kV powerline, by transit time from the line.Box plots of (c) 10minute mean and (d) 10-minute standard deviation of PG measured near the HVPL from each wind direction.